Series comparison

-[PULL 00/10] riscv-to-apply queue
+[PULL v2 00/45] riscv-to-apply queue
-From: Alistair Francis <alistair.francis@wdc.com>
+The following changes since commit c5ea91da443b458352c1b629b490ee6631775cb4:
-The following changes since commit 2ba341b3694cf3cff7b8a1df4cc765900d5c4f60:
+  Merge tag 'pull-trivial-patches' of https://gitlab.com/mjt0k/qemu into staging (2023-09-08 10:06:25 -0400)
   Merge tag 'kraxel-20221013-pull-request' of https://gitlab.com/kraxel/qemu into staging (2022-10-13 13:55:53 -0400)
 are available in the Git repository at:
-  https://github.com/alistair23/qemu.git tags/pull-riscv-to-apply-20221014
+  https://github.com/alistair23/qemu.git tags/pull-riscv-to-apply-20230911
-for you to fetch changes up to 47566421f029b0a489b63f8195b3ff944e017056:
+for you to fetch changes up to e7a03409f29e2da59297d55afbaec98c96e43e3a:
-  target/riscv: pmp: Fixup TLB size calculation (2022-10-14 14:36:19 +1000)
+  target/riscv: don't read CSR in riscv_csrrw_do64 (2023-09-11 11:45:55 +1000)
 ----------------------------------------------------------------
-Third RISC-V PR for QEMU 7.2
+First RISC-V PR for 8.2
-* Update qtest comment
+ * Remove 'host' CPU from TCG
-* Fix coverity issue with Ibex SPI
+ * riscv_htif Fixup printing on big endian hosts
-* Move load_image_to_fw_cfg() to common location
+ * Add zmmul isa string
-* Enable booting S-mode firmware from pflash on virt machine
+ * Add smepmp isa string
-* Add disas support for vector instructions
+ * Fix page_check_range use in fault-only-first
-* Priority level fixes for PLIC
+ * Use existing lookup tables for MixColumns
-* Fixup TLB size calculation when using PMP
+ * Add RISC-V vector cryptographic instruction set support
  * Implement WARL behaviour for mcountinhibit/mcounteren
  * Add Zihintntl extension ISA string to DTS
  * Fix zfa fleq.d and fltq.d
  * Fix upper/lower mtime write calculation
  * Make rtc variable names consistent
  * Use abi type for linux-user target_ucontext
  * Add RISC-V KVM AIA Support
  * Fix riscv,pmu DT node path in the virt machine
  * Update CSR bits name for svadu extension
  * Mark zicond non-experimental
  * Fix satp_mode_finalize() when satp_mode.supported = 0
  * Fix non-KVM --enable-debug build
  * Add new extensions to hwprobe
  * Use accelerated helper for AES64KS1I
  * Allocate itrigger timers only once
  * Respect mseccfg.RLB for pmpaddrX changes
  * Align the AIA model to v1.0 ratified spec
  * Don't read the CSR in riscv_csrrw_do64
 ----------------------------------------------------------------
-Alistair Francis (1):
+Akihiko Odaki (1):
-      target/riscv: pmp: Fixup TLB size calculation
+      target/riscv: Allocate itrigger timers only once
-Bin Meng (1):
+Ard Biesheuvel (2):
-      hw/riscv: Update comment for qtest check in riscv_find_firmware()
+      target/riscv: Use existing lookup tables for MixColumns
       target/riscv: Use accelerated helper for AES64KS1I
-Jim Shu (2):
+Conor Dooley (1):
-      hw/intc: sifive_plic: fix hard-coded max priority level
+      hw/riscv: virt: Fix riscv,pmu DT node path
       hw/intc: sifive_plic: change interrupt priority register to WARL field
-Sunil V L (3):
+Daniel Henrique Barboza (6):
-      hw/arm, loongarch: Move load_image_to_fw_cfg() to common location
+      target/riscv/cpu.c: do not run 'host' CPU with TCG
-      hw/riscv: virt: Move create_fw_cfg() prior to loading kernel
+      target/riscv/cpu.c: add zmmul isa string
-      hw/riscv: virt: Enable booting S-mode firmware from pflash
+      target/riscv/cpu.c: add smepmp isa string
       target/riscv: fix satp_mode_finalize() when satp_mode.supported = 0
       hw/riscv/virt.c: fix non-KVM --enable-debug build
       hw/intc/riscv_aplic.c fix non-KVM --enable-debug build
-Wilfred Mallawa (2):
+Dickon Hood (2):
-      hw/ssi: ibex_spi: fixup coverity issue
+      target/riscv: Refactor translation of vector-widening instruction
-      hw/ssi: ibex_spi: fixup/add rw1c functionality
+      target/riscv: Add Zvbb ISA extension support
-Yang Liu (1):
+Jason Chien (3):
-      disas/riscv.c: rvv: Add disas support for vector instructions
+      target/riscv: Add Zihintntl extension ISA string to DTS
       hw/intc: Fix upper/lower mtime write calculation
       hw/intc: Make rtc variable names consistent
- include/hw/nvram/fw_cfg.h      |   21 +
+Kiran Ostrolenk (4):
- include/hw/riscv/boot.h        |    1 +
+      target/riscv: Refactor some of the generic vector functionality
- include/hw/ssi/ibex_spi_host.h |    4 +-
+      target/riscv: Refactor vector-vector translation macro
- disas/riscv.c                  | 1432 +++++++++++++++++++++++++++++++++++++++-
+      target/riscv: Refactor some of the generic vector functionality
- hw/arm/boot.c                  |   49 --
+      target/riscv: Add Zvknh ISA extension support
- hw/intc/sifive_plic.c          |   25 +-
- hw/loongarch/virt.c            |   33 -
+LIU Zhiwei (3):
- hw/nvram/fw_cfg.c              |   32 +
+      target/riscv: Fix page_check_range use in fault-only-first
- hw/riscv/boot.c                |   33 +-
+      target/riscv: Fix zfa fleq.d and fltq.d
- hw/riscv/virt.c                |   32 +-
+      linux-user/riscv: Use abi type for target_ucontext
- hw/ssi/ibex_spi_host.c         |  166 +++--
- target/riscv/pmp.c             |   12 +
+Lawrence Hunter (2):
-files changed, 1675 insertions(+), 165 deletions(-)
+      target/riscv: Add Zvbc ISA extension support
       target/riscv: Add Zvksh ISA extension support
 Leon Schuermann (1):
       target/riscv/pmp.c: respect mseccfg.RLB for pmpaddrX changes
 Max Chou (3):
       crypto: Create sm4_subword
       crypto: Add SM4 constant parameter CK
       target/riscv: Add Zvksed ISA extension support
 Nazar Kazakov (4):
       target/riscv: Remove redundant "cpu_vl == 0" checks
       target/riscv: Move vector translation checks
       target/riscv: Add Zvkned ISA extension support
       target/riscv: Add Zvkg ISA extension support
 Nikita Shubin (1):
       target/riscv: don't read CSR in riscv_csrrw_do64
 Rob Bradford (1):
       target/riscv: Implement WARL behaviour for mcountinhibit/mcounteren
 Robbin Ehn (1):
       linux-user/riscv: Add new extensions to hwprobe
 Thomas Huth (2):
       hw/char/riscv_htif: Fix printing of console characters on big endian hosts
       hw/char/riscv_htif: Fix the console syscall on big endian hosts
 Tommy Wu (1):
       target/riscv: Align the AIA model to v1.0 ratified spec
 Vineet Gupta (1):
       riscv: zicond: make non-experimental
 Weiwei Li (1):
       target/riscv: Update CSR bits name for svadu extension
 Yong-Xuan Wang (5):
       target/riscv: support the AIA device emulation with KVM enabled
       target/riscv: check the in-kernel irqchip support
       target/riscv: Create an KVM AIA irqchip
       target/riscv: update APLIC and IMSIC to support KVM AIA
       target/riscv: select KVM AIA in riscv virt machine
  include/crypto/aes.h                      |   7 +
  include/crypto/sm4.h                      |   9 +
  target/riscv/cpu_bits.h                   |   8 +-
  target/riscv/cpu_cfg.h                    |   9 +
  target/riscv/debug.h                      |   3 +-
  target/riscv/helper.h                     |  98 +++
  target/riscv/kvm_riscv.h                  |   5 +
  target/riscv/vector_internals.h           | 228 +++++++
  target/riscv/insn32.decode                |  58 ++
  crypto/aes.c                              |   4 +-
  crypto/sm4.c                              |  10 +
  hw/char/riscv_htif.c                      |  12 +-
  hw/intc/riscv_aclint.c                    |  11 +-
  hw/intc/riscv_aplic.c                     |  52 +-
  hw/intc/riscv_imsic.c                     |  25 +-
  hw/riscv/virt.c                           | 374 ++++++------
  linux-user/riscv/signal.c                 |   4 +-
  linux-user/syscall.c                      |  14 +-
  target/arm/tcg/crypto_helper.c            |  10 +-
  target/riscv/cpu.c                        |  83 ++-
  target/riscv/cpu_helper.c                 |   6 +-
  target/riscv/crypto_helper.c              |  51 +-
  target/riscv/csr.c                        |  54 +-
  target/riscv/debug.c                      |  15 +-
  target/riscv/kvm.c                        | 201 ++++++-
  target/riscv/pmp.c                        |   4 +
  target/riscv/translate.c                  |   1 +
  target/riscv/vcrypto_helper.c             | 970 ++++++++++++++++++++++++++++++
  target/riscv/vector_helper.c              | 245 +-------
  target/riscv/vector_internals.c           |  81 +++
  target/riscv/insn_trans/trans_rvv.c.inc   | 171 +++---
  target/riscv/insn_trans/trans_rvvk.c.inc  | 606 +++++++++++++++++++
  target/riscv/insn_trans/trans_rvzfa.c.inc |   4 +-
  target/riscv/meson.build                  |   4 +-
 files changed, 2785 insertions(+), 652 deletions(-)
  create mode 100644 target/riscv/vector_internals.h
  create mode 100644 target/riscv/vcrypto_helper.c
  create mode 100644 target/riscv/vector_internals.c
  create mode 100644 target/riscv/insn_trans/trans_rvvk.c.inc

-New patch
+[PULL v2 01/45] target/riscv/cpu.c: do not run 'host' CPU with TCG
+From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+The 'host' CPU is available in a CONFIG_KVM build and it's currently
+available for all accels, but is a KVM only CPU. This means that in a
+RISC-V KVM capable host we can do things like this:
+$ ./build/qemu-system-riscv64 -M virt,accel=tcg -cpu host --nographic
+qemu-system-riscv64: H extension requires priv spec 1.12.0
+This CPU does not have a priv spec because we don't filter its extensions
+via priv spec. We shouldn't be reaching riscv_cpu_realize_tcg() at all
+with the 'host' CPU.
+We don't have a way to filter the 'host' CPU out of the available CPU
+options (-cpu help) if the build includes both KVM and TCG. What we can
+do is to error out during riscv_cpu_realize_tcg() if the user chooses
+the 'host' CPU with accel=tcg:
+$ ./build/qemu-system-riscv64 -M virt,accel=tcg -cpu host --nographic
+qemu-system-riscv64: 'host' CPU is not compatible with TCG acceleration
+Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
+Message-Id: <20230721133411.474105-1-dbarboza@ventanamicro.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/cpu.c | 5 +++++
+file changed, 5 insertions(+)
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu.c
++++ b/target/riscv/cpu.c
+@@ -XXX,XX +XXX,XX @@ static void riscv_cpu_realize_tcg(DeviceState *dev, Error **errp)
+     CPURISCVState *env = &cpu->env;
+     Error *local_err = NULL;
++    if (object_dynamic_cast(OBJECT(dev), TYPE_RISCV_CPU_HOST)) {
++        error_setg(errp, "'host' CPU is not compatible with TCG acceleration");
++        return;
++    }
++
+     riscv_cpu_validate_misa_mxl(cpu, &local_err);
+     if (local_err != NULL) {
+         error_propagate(errp, local_err);
+--
+.41.0

-New patch
+[PULL v2 02/45] hw/char/riscv_htif: Fix printing of console characters on big endian hosts
+From: Thomas Huth <thuth@redhat.com>
+The character that should be printed is stored in the 64 bit "payload"
+variable. The code currently tries to print it by taking the address
+of the variable and passing this pointer to qemu_chr_fe_write(). However,
+this only works on little endian hosts where the least significant bits
+are stored on the lowest address. To do this in a portable way, we have
+to store the value in an uint8_t variable instead.
+Fixes: 5033606780 ("RISC-V HTIF Console")
+Signed-off-by: Thomas Huth <thuth@redhat.com>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Bin Meng <bmeng@tinylab.org>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
+Message-Id: <20230721094720.902454-2-thuth@redhat.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ hw/char/riscv_htif.c | 3 ++-
+file changed, 2 insertions(+), 1 deletion(-)
+diff --git a/hw/char/riscv_htif.c b/hw/char/riscv_htif.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/char/riscv_htif.c
++++ b/hw/char/riscv_htif.c
+@@ -XXX,XX +XXX,XX @@ static void htif_handle_tohost_write(HTIFState *s, uint64_t val_written)
+             s->tohost = 0; /* clear to indicate we read */
+             return;
+         } else if (cmd == HTIF_CONSOLE_CMD_PUTC) {
+-            qemu_chr_fe_write(&s->chr, (uint8_t *)&payload, 1);
++            uint8_t ch = (uint8_t)payload;
++            qemu_chr_fe_write(&s->chr, &ch, 1);
+             resp = 0x100 | (uint8_t)payload;
+         } else {
+             qemu_log("HTIF device %d: unknown command\n", device);
+--
+.41.0

-New patch
+[PULL v2 03/45] hw/char/riscv_htif: Fix the console syscall on big endian hosts
+From: Thomas Huth <thuth@redhat.com>
+Values that have been read via cpu_physical_memory_read() from the
+guest's memory have to be swapped in case the host endianess differs
+from the guest.
+Fixes: a6e13e31d5 ("riscv_htif: Support console output via proxy syscall")
+Signed-off-by: Thomas Huth <thuth@redhat.com>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Bin Meng <bmeng@tinylab.org>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Message-Id: <20230721094720.902454-3-thuth@redhat.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ hw/char/riscv_htif.c | 9 +++++----
+file changed, 5 insertions(+), 4 deletions(-)
+diff --git a/hw/char/riscv_htif.c b/hw/char/riscv_htif.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/char/riscv_htif.c
++++ b/hw/char/riscv_htif.c
+@@ -XXX,XX +XXX,XX @@
+ #include "qemu/timer.h"
+ #include "qemu/error-report.h"
+ #include "exec/address-spaces.h"
++#include "exec/tswap.h"
+ #include "sysemu/dma.h"
+ #define RISCV_DEBUG_HTIF 0
+@@ -XXX,XX +XXX,XX @@ static void htif_handle_tohost_write(HTIFState *s, uint64_t val_written)
+             } else {
+                 uint64_t syscall[8];
+                 cpu_physical_memory_read(payload, syscall, sizeof(syscall));
+-                if (syscall[0] == PK_SYS_WRITE &&
+-                    syscall[1] == HTIF_DEV_CONSOLE &&
+-                    syscall[3] == HTIF_CONSOLE_CMD_PUTC) {
++                if (tswap64(syscall[0]) == PK_SYS_WRITE &&
++                    tswap64(syscall[1]) == HTIF_DEV_CONSOLE &&
++                    tswap64(syscall[3]) == HTIF_CONSOLE_CMD_PUTC) {
+                     uint8_t ch;
+-                    cpu_physical_memory_read(syscall[2], &ch, 1);
++                    cpu_physical_memory_read(tswap64(syscall[2]), &ch, 1);
+                     qemu_chr_fe_write(&s->chr, &ch, 1);
+                     resp = 0x100 | (uint8_t)payload;
+                 } else {
+--
+.41.0

-New patch
+[PULL v2 04/45] target/riscv/cpu.c: add zmmul isa string
+From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+zmmul was promoted from experimental to ratified in commit 6d00ffad4e95.
+Add a riscv,isa string for it.
+Fixes: 6d00ffad4e95 ("target/riscv: move zmmul out of the experimental properties")
+Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Message-Id: <20230720132424.371132-2-dbarboza@ventanamicro.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/cpu.c | 1 +
+file changed, 1 insertion(+)
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu.c
++++ b/target/riscv/cpu.c
+@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
+     ISA_EXT_DATA_ENTRY(zicsr, PRIV_VERSION_1_10_0, ext_icsr),
+     ISA_EXT_DATA_ENTRY(zifencei, PRIV_VERSION_1_10_0, ext_ifencei),
+     ISA_EXT_DATA_ENTRY(zihintpause, PRIV_VERSION_1_10_0, ext_zihintpause),
++    ISA_EXT_DATA_ENTRY(zmmul, PRIV_VERSION_1_12_0, ext_zmmul),
+     ISA_EXT_DATA_ENTRY(zawrs, PRIV_VERSION_1_12_0, ext_zawrs),
+     ISA_EXT_DATA_ENTRY(zfa, PRIV_VERSION_1_12_0, ext_zfa),
+     ISA_EXT_DATA_ENTRY(zfbfmin, PRIV_VERSION_1_12_0, ext_zfbfmin),
+--
+.41.0

-New patch
+[PULL v2 05/45] target/riscv/cpu.c: add smepmp isa string
+From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+The cpu->cfg.epmp extension is still experimental, but it already has a
+'smepmp' riscv,isa string. Add it.
+Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Message-Id: <20230720132424.371132-3-dbarboza@ventanamicro.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/cpu.c | 1 +
+file changed, 1 insertion(+)
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu.c
++++ b/target/riscv/cpu.c
+@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
+     ISA_EXT_DATA_ENTRY(zhinx, PRIV_VERSION_1_12_0, ext_zhinx),
+     ISA_EXT_DATA_ENTRY(zhinxmin, PRIV_VERSION_1_12_0, ext_zhinxmin),
+     ISA_EXT_DATA_ENTRY(smaia, PRIV_VERSION_1_12_0, ext_smaia),
++    ISA_EXT_DATA_ENTRY(smepmp, PRIV_VERSION_1_12_0, epmp),
+     ISA_EXT_DATA_ENTRY(smstateen, PRIV_VERSION_1_12_0, ext_smstateen),
+     ISA_EXT_DATA_ENTRY(ssaia, PRIV_VERSION_1_12_0, ext_ssaia),
+     ISA_EXT_DATA_ENTRY(sscofpmf, PRIV_VERSION_1_12_0, ext_sscofpmf),
+--
+.41.0

-New patch
+[PULL v2 06/45] target/riscv: Fix page_check_range use in fault-only-first
+From: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
+Commit bef6f008b98(accel/tcg: Return bool from page_check_range) converts
+integer return value to bool type. However, it wrongly converted the use
+of the API in riscv fault-only-first, where page_check_range < = 0, should
+be converted to !page_check_range.
+Signed-off-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Message-ID: <20230729031618.821-1-zhiwei_liu@linux.alibaba.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/vector_helper.c | 2 +-
+file changed, 1 insertion(+), 1 deletion(-)
+diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/vector_helper.c
++++ b/target/riscv/vector_helper.c
+@@ -XXX,XX +XXX,XX @@ vext_ldff(void *vd, void *v0, target_ulong base,
+                                          cpu_mmu_index(env, false));
+                 if (host) {
+ #ifdef CONFIG_USER_ONLY
+-                    if (page_check_range(addr, offset, PAGE_READ)) {
++                    if (!page_check_range(addr, offset, PAGE_READ)) {
+                         vl = i;
+                         goto ProbeSuccess;
+                     }
+--
+.41.0

-New patch
+[PULL v2 07/45] target/riscv: Use existing lookup tables for MixColumns
+From: Ard Biesheuvel <ardb@kernel.org>
+The AES MixColumns and InvMixColumns operations are relatively
+expensive 4x4 matrix multiplications in GF(2^8), which is why C
+implementations usually rely on precomputed lookup tables rather than
+performing the calculations on demand.
+Given that we already carry those tables in QEMU, we can just grab the
+right value in the implementation of the RISC-V AES32 instructions. Note
+that the tables in question are permuted according to the respective
+Sbox, so we can omit the Sbox lookup as well in this case.
+Cc: Richard Henderson <richard.henderson@linaro.org>
+Cc: Philippe Mathieu-Daudé <philmd@linaro.org>
+Cc: Zewen Ye <lustrew@foxmail.com>
+Cc: Weiwei Li <liweiwei@iscas.ac.cn>
+Cc: Junqiang Wang <wangjunqiang@iscas.ac.cn>
+Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Message-ID: <20230731084043.1791984-1-ardb@kernel.org>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ include/crypto/aes.h         |  7 +++++++
+ crypto/aes.c                 |  4 ++--
+ target/riscv/crypto_helper.c | 34 ++++------------------------------
+files changed, 13 insertions(+), 32 deletions(-)
+diff --git a/include/crypto/aes.h b/include/crypto/aes.h
+index XXXXXXX..XXXXXXX 100644
+--- a/include/crypto/aes.h
++++ b/include/crypto/aes.h
+@@ -XXX,XX +XXX,XX @@ void AES_decrypt(const unsigned char *in, unsigned char *out,
+ extern const uint8_t AES_sbox[256];
+ extern const uint8_t AES_isbox[256];
++/*
++AES_Te0[x] = S [x].[02, 01, 01, 03];
++AES_Td0[x] = Si[x].[0e, 09, 0d, 0b];
++*/
++
++extern const uint32_t AES_Te0[256], AES_Td0[256];
++
+ #endif
+diff --git a/crypto/aes.c b/crypto/aes.c
+index XXXXXXX..XXXXXXX 100644
+--- a/crypto/aes.c
++++ b/crypto/aes.c
+@@ -XXX,XX +XXX,XX @@ AES_Td3[x] = Si[x].[09, 0d, 0b, 0e];
+ AES_Td4[x] = Si[x].[01, 01, 01, 01];
+ */
+-static const uint32_t AES_Te0[256] = {
++const uint32_t AES_Te0[256] = {
+xc66363a5U, 0xf87c7c84U, 0xee777799U, 0xf67b7b8dU,
+xfff2f20dU, 0xd66b6bbdU, 0xde6f6fb1U, 0x91c5c554U,
+x60303050U, 0x02010103U, 0xce6767a9U, 0x562b2b7dU,
+@@ -XXX,XX +XXX,XX @@ static const uint32_t AES_Te4[256] = {
+xb0b0b0b0U, 0x54545454U, 0xbbbbbbbbU, 0x16161616U,
+ };
+-static const uint32_t AES_Td0[256] = {
++const uint32_t AES_Td0[256] = {
+x51f4a750U, 0x7e416553U, 0x1a17a4c3U, 0x3a275e96U,
+x3bab6bcbU, 0x1f9d45f1U, 0xacfa58abU, 0x4be30393U,
+x2030fa55U, 0xad766df6U, 0x88cc7691U, 0xf5024c25U,
+diff --git a/target/riscv/crypto_helper.c b/target/riscv/crypto_helper.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/crypto_helper.c
++++ b/target/riscv/crypto_helper.c
+@@ -XXX,XX +XXX,XX @@
+ #include "crypto/aes-round.h"
+ #include "crypto/sm4.h"
+-#define AES_XTIME(a) \
+-    ((a << 1) ^ ((a & 0x80) ? 0x1b : 0))
+-
+-#define AES_GFMUL(a, b) (( \
+-    (((b) & 0x1) ? (a) : 0) ^ \
+-    (((b) & 0x2) ? AES_XTIME(a) : 0) ^ \
+-    (((b) & 0x4) ? AES_XTIME(AES_XTIME(a)) : 0) ^ \
+-    (((b) & 0x8) ? AES_XTIME(AES_XTIME(AES_XTIME(a))) : 0)) & 0xFF)
+-
+-static inline uint32_t aes_mixcolumn_byte(uint8_t x, bool fwd)
+-{
+-    uint32_t u;
+-
+-    if (fwd) {
+-        u = (AES_GFMUL(x, 3) << 24) | (x << 16) | (x << 8) |
+-            (AES_GFMUL(x, 2) << 0);
+-    } else {
+-        u = (AES_GFMUL(x, 0xb) << 24) | (AES_GFMUL(x, 0xd) << 16) |
+-            (AES_GFMUL(x, 0x9) << 8) | (AES_GFMUL(x, 0xe) << 0);
+-    }
+-    return u;
+-}
+-
+ #define sext32_xlen(x) (target_ulong)(int32_t)(x)
+ static inline target_ulong aes32_operation(target_ulong shamt,
+@@ -XXX,XX +XXX,XX @@ static inline target_ulong aes32_operation(target_ulong shamt,
+                                            bool enc, bool mix)
+ {
+     uint8_t si = rs2 >> shamt;
+-    uint8_t so;
+     uint32_t mixed;
+     target_ulong res;
+     if (enc) {
+-        so = AES_sbox[si];
+         if (mix) {
+-            mixed = aes_mixcolumn_byte(so, true);
++            mixed = be32_to_cpu(AES_Te0[si]);
+         } else {
+-            mixed = so;
++            mixed = AES_sbox[si];
+         }
+     } else {
+-        so = AES_isbox[si];
+         if (mix) {
+-            mixed = aes_mixcolumn_byte(so, false);
++            mixed = be32_to_cpu(AES_Td0[si]);
+         } else {
+-            mixed = so;
++            mixed = AES_isbox[si];
+         }
+     }
+     mixed = rol32(mixed, shamt);
+--
+.41.0

-New patch
+[PULL v2 08/45] target/riscv: Refactor some of the generic vector functionality
+From: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
+Take some functions/macros out of `vector_helper` and put them in a new
+module called `vector_internals`. This ensures they can be used by both
+vector and vector-crypto helpers (latter implemented in proceeding
+commits).
+Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
+Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Acked-by: Alistair Francis <alistair.francis@wdc.com>
+Message-ID: <20230711165917.2629866-2-max.chou@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/vector_internals.h | 182 +++++++++++++++++++++++++++++
+ target/riscv/vector_helper.c    | 201 +-------------------------------
+ target/riscv/vector_internals.c |  81 +++++++++++++
+ target/riscv/meson.build        |   1 +
+files changed, 265 insertions(+), 200 deletions(-)
+ create mode 100644 target/riscv/vector_internals.h
+ create mode 100644 target/riscv/vector_internals.c
+diff --git a/target/riscv/vector_internals.h b/target/riscv/vector_internals.h
+new file mode 100644
+index XXXXXXX..XXXXXXX
+--- /dev/null
++++ b/target/riscv/vector_internals.h
+@@ -XXX,XX +XXX,XX @@
++/*
++ * RISC-V Vector Extension Internals
++ *
++ * Copyright (c) 2020 T-Head Semiconductor Co., Ltd. All rights reserved.
++ *
++ * This program is free software; you can redistribute it and/or modify it
++ * under the terms and conditions of the GNU General Public License,
++ * version 2 or later, as published by the Free Software Foundation.
++ *
++ * This program is distributed in the hope it will be useful, but WITHOUT
++ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
++ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
++ * more details.
++ *
++ * You should have received a copy of the GNU General Public License along with
++ * this program.  If not, see <http://www.gnu.org/licenses/>.
++ */
++
++#ifndef TARGET_RISCV_VECTOR_INTERNALS_H
++#define TARGET_RISCV_VECTOR_INTERNALS_H
++
++#include "qemu/osdep.h"
++#include "qemu/bitops.h"
++#include "cpu.h"
++#include "tcg/tcg-gvec-desc.h"
++#include "internals.h"
++
++static inline uint32_t vext_nf(uint32_t desc)
++{
++    return FIELD_EX32(simd_data(desc), VDATA, NF);
++}
++
++/*
++ * Note that vector data is stored in host-endian 64-bit chunks,
++ * so addressing units smaller than that needs a host-endian fixup.
++ */
++#if HOST_BIG_ENDIAN
++#define H1(x)   ((x) ^ 7)
++#define H1_2(x) ((x) ^ 6)
++#define H1_4(x) ((x) ^ 4)
++#define H2(x)   ((x) ^ 3)
++#define H4(x)   ((x) ^ 1)
++#define H8(x)   ((x))
++#else
++#define H1(x)   (x)
++#define H1_2(x) (x)
++#define H1_4(x) (x)
++#define H2(x)   (x)
++#define H4(x)   (x)
++#define H8(x)   (x)
++#endif
++
++/*
++ * Encode LMUL to lmul as following:
++ *     LMUL    vlmul    lmul
++ *      1       000       0
++ *      2       001       1
++ *      4       010       2
++ *      8       011       3
++ *      -       100       -
++ *     1/8      101      -3
++ *     1/4      110      -2
++ *     1/2      111      -1
++ */
++static inline int32_t vext_lmul(uint32_t desc)
++{
++    return sextract32(FIELD_EX32(simd_data(desc), VDATA, LMUL), 0, 3);
++}
++
++static inline uint32_t vext_vm(uint32_t desc)
++{
++    return FIELD_EX32(simd_data(desc), VDATA, VM);
++}
++
++static inline uint32_t vext_vma(uint32_t desc)
++{
++    return FIELD_EX32(simd_data(desc), VDATA, VMA);
++}
++
++static inline uint32_t vext_vta(uint32_t desc)
++{
++    return FIELD_EX32(simd_data(desc), VDATA, VTA);
++}
++
++static inline uint32_t vext_vta_all_1s(uint32_t desc)
++{
++    return FIELD_EX32(simd_data(desc), VDATA, VTA_ALL_1S);
++}
++
++/*
++ * Earlier designs (pre-0.9) had a varying number of bits
++ * per mask value (MLEN). In the 0.9 design, MLEN=1.
++ * (Section 4.5)
++ */
++static inline int vext_elem_mask(void *v0, int index)
++{
++    int idx = index / 64;
++    int pos = index  % 64;
++    return (((uint64_t *)v0)[idx] >> pos) & 1;
++}
++
++/*
++ * Get number of total elements, including prestart, body and tail elements.
++ * Note that when LMUL < 1, the tail includes the elements past VLMAX that
++ * are held in the same vector register.
++ */
++static inline uint32_t vext_get_total_elems(CPURISCVState *env, uint32_t desc,
++                                            uint32_t esz)
++{
++    uint32_t vlenb = simd_maxsz(desc);
++    uint32_t sew = 1 << FIELD_EX64(env->vtype, VTYPE, VSEW);
++    int8_t emul = ctzl(esz) - ctzl(sew) + vext_lmul(desc) < 0 ? 0 :
++                  ctzl(esz) - ctzl(sew) + vext_lmul(desc);
++    return (vlenb << emul) / esz;
++}
++
++/* set agnostic elements to 1s */
++void vext_set_elems_1s(void *base, uint32_t is_agnostic, uint32_t cnt,
++                       uint32_t tot);
++
++/* expand macro args before macro */
++#define RVVCALL(macro, ...)  macro(__VA_ARGS__)
++
++/* (TD, T1, T2, TX1, TX2) */
++#define OP_UUU_B uint8_t, uint8_t, uint8_t, uint8_t, uint8_t
++#define OP_UUU_H uint16_t, uint16_t, uint16_t, uint16_t, uint16_t
++#define OP_UUU_W uint32_t, uint32_t, uint32_t, uint32_t, uint32_t
++#define OP_UUU_D uint64_t, uint64_t, uint64_t, uint64_t, uint64_t
++
++/* operation of two vector elements */
++typedef void opivv2_fn(void *vd, void *vs1, void *vs2, int i);
++
++#define OPIVV2(NAME, TD, T1, T2, TX1, TX2, HD, HS1, HS2, OP)    \
++static void do_##NAME(void *vd, void *vs1, void *vs2, int i)    \
++{                                                               \
++    TX1 s1 = *((T1 *)vs1 + HS1(i));                             \
++    TX2 s2 = *((T2 *)vs2 + HS2(i));                             \
++    *((TD *)vd + HD(i)) = OP(s2, s1);                           \
++}
++
++void do_vext_vv(void *vd, void *v0, void *vs1, void *vs2,
++                CPURISCVState *env, uint32_t desc,
++                opivv2_fn *fn, uint32_t esz);
++
++/* generate the helpers for OPIVV */
++#define GEN_VEXT_VV(NAME, ESZ)                            \
++void HELPER(NAME)(void *vd, void *v0, void *vs1,          \
++                  void *vs2, CPURISCVState *env,          \
++                  uint32_t desc)                          \
++{                                                         \
++    do_vext_vv(vd, v0, vs1, vs2, env, desc,               \
++               do_##NAME, ESZ);                           \
++}
++
++typedef void opivx2_fn(void *vd, target_long s1, void *vs2, int i);
++
++/*
++ * (T1)s1 gives the real operator type.
++ * (TX1)(T1)s1 expands the operator type of widen or narrow operations.
++ */
++#define OPIVX2(NAME, TD, T1, T2, TX1, TX2, HD, HS2, OP)             \
++static void do_##NAME(void *vd, target_long s1, void *vs2, int i)   \
++{                                                                   \
++    TX2 s2 = *((T2 *)vs2 + HS2(i));                                 \
++    *((TD *)vd + HD(i)) = OP(s2, (TX1)(T1)s1);                      \
++}
++
++void do_vext_vx(void *vd, void *v0, target_long s1, void *vs2,
++                CPURISCVState *env, uint32_t desc,
++                opivx2_fn fn, uint32_t esz);
++
++/* generate the helpers for OPIVX */
++#define GEN_VEXT_VX(NAME, ESZ)                            \
++void HELPER(NAME)(void *vd, void *v0, target_ulong s1,    \
++                  void *vs2, CPURISCVState *env,          \
++                  uint32_t desc)                          \
++{                                                         \
++    do_vext_vx(vd, v0, s1, vs2, env, desc,                \
++               do_##NAME, ESZ);                           \
++}
++
++#endif /* TARGET_RISCV_VECTOR_INTERNALS_H */
+diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/vector_helper.c
++++ b/target/riscv/vector_helper.c
+@@ -XXX,XX +XXX,XX @@
+ #include "fpu/softfloat.h"
+ #include "tcg/tcg-gvec-desc.h"
+ #include "internals.h"
++#include "vector_internals.h"
+ #include <math.h>
+ target_ulong HELPER(vsetvl)(CPURISCVState *env, target_ulong s1,
+@@ -XXX,XX +XXX,XX @@ target_ulong HELPER(vsetvl)(CPURISCVState *env, target_ulong s1,
+     return vl;
+ }
+-/*
+- * Note that vector data is stored in host-endian 64-bit chunks,
+- * so addressing units smaller than that needs a host-endian fixup.
+- */
+-#if HOST_BIG_ENDIAN
+-#define H1(x)   ((x) ^ 7)
+-#define H1_2(x) ((x) ^ 6)
+-#define H1_4(x) ((x) ^ 4)
+-#define H2(x)   ((x) ^ 3)
+-#define H4(x)   ((x) ^ 1)
+-#define H8(x)   ((x))
+-#else
+-#define H1(x)   (x)
+-#define H1_2(x) (x)
+-#define H1_4(x) (x)
+-#define H2(x)   (x)
+-#define H4(x)   (x)
+-#define H8(x)   (x)
+-#endif
+-
+-static inline uint32_t vext_nf(uint32_t desc)
+-{
+-    return FIELD_EX32(simd_data(desc), VDATA, NF);
+-}
+-
+-static inline uint32_t vext_vm(uint32_t desc)
+-{
+-    return FIELD_EX32(simd_data(desc), VDATA, VM);
+-}
+-
+-/*
+- * Encode LMUL to lmul as following:
+- *     LMUL    vlmul    lmul
+- *      1       000       0
+- *      2       001       1
+- *      4       010       2
+- *      8       011       3
+- *      -       100       -
+- *     1/8      101      -3
+- *     1/4      110      -2
+- *     1/2      111      -1
+- */
+-static inline int32_t vext_lmul(uint32_t desc)
+-{
+-    return sextract32(FIELD_EX32(simd_data(desc), VDATA, LMUL), 0, 3);
+-}
+-
+-static inline uint32_t vext_vta(uint32_t desc)
+-{
+-    return FIELD_EX32(simd_data(desc), VDATA, VTA);
+-}
+-
+-static inline uint32_t vext_vma(uint32_t desc)
+-{
+-    return FIELD_EX32(simd_data(desc), VDATA, VMA);
+-}
+-
+-static inline uint32_t vext_vta_all_1s(uint32_t desc)
+-{
+-    return FIELD_EX32(simd_data(desc), VDATA, VTA_ALL_1S);
+-}
+-
+ /*
+  * Get the maximum number of elements can be operated.
+  *
+@@ -XXX,XX +XXX,XX @@ static inline uint32_t vext_max_elems(uint32_t desc, uint32_t log2_esz)
+     return scale < 0 ? vlenb >> -scale : vlenb << scale;
+ }
+-/*
+- * Get number of total elements, including prestart, body and tail elements.
+- * Note that when LMUL < 1, the tail includes the elements past VLMAX that
+- * are held in the same vector register.
+- */
+-static inline uint32_t vext_get_total_elems(CPURISCVState *env, uint32_t desc,
+-                                            uint32_t esz)
+-{
+-    uint32_t vlenb = simd_maxsz(desc);
+-    uint32_t sew = 1 << FIELD_EX64(env->vtype, VTYPE, VSEW);
+-    int8_t emul = ctzl(esz) - ctzl(sew) + vext_lmul(desc) < 0 ? 0 :
+-                  ctzl(esz) - ctzl(sew) + vext_lmul(desc);
+-    return (vlenb << emul) / esz;
+-}
+-
+ static inline target_ulong adjust_addr(CPURISCVState *env, target_ulong addr)
+ {
+     return (addr & ~env->cur_pmmask) | env->cur_pmbase;
+@@ -XXX,XX +XXX,XX @@ static void probe_pages(CPURISCVState *env, target_ulong addr,
+     }
+ }
+-/* set agnostic elements to 1s */
+-static void vext_set_elems_1s(void *base, uint32_t is_agnostic, uint32_t cnt,
+-                              uint32_t tot)
+-{
+-    if (is_agnostic == 0) {
+-        /* policy undisturbed */
+-        return;
+-    }
+-    if (tot - cnt == 0) {
+-        return;
+-    }
+-    memset(base + cnt, -1, tot - cnt);
+-}
+-
+ static inline void vext_set_elem_mask(void *v0, int index,
+                                       uint8_t value)
+ {
+@@ -XXX,XX +XXX,XX @@ static inline void vext_set_elem_mask(void *v0, int index,
+     ((uint64_t *)v0)[idx] = deposit64(old, pos, 1, value);
+ }
+-/*
+- * Earlier designs (pre-0.9) had a varying number of bits
+- * per mask value (MLEN). In the 0.9 design, MLEN=1.
+- * (Section 4.5)
+- */
+-static inline int vext_elem_mask(void *v0, int index)
+-{
+-    int idx = index / 64;
+-    int pos = index  % 64;
+-    return (((uint64_t *)v0)[idx] >> pos) & 1;
+-}
+-
+ /* elements operations for load and store */
+ typedef void vext_ldst_elem_fn(CPURISCVState *env, abi_ptr addr,
+                                uint32_t idx, void *vd, uintptr_t retaddr);
+@@ -XXX,XX +XXX,XX @@ GEN_VEXT_ST_WHOLE(vs8r_v, int8_t, ste_b)
+  * Vector Integer Arithmetic Instructions
+  */
+-/* expand macro args before macro */
+-#define RVVCALL(macro, ...)  macro(__VA_ARGS__)
+-
+ /* (TD, T1, T2, TX1, TX2) */
+ #define OP_SSS_B int8_t, int8_t, int8_t, int8_t, int8_t
+ #define OP_SSS_H int16_t, int16_t, int16_t, int16_t, int16_t
+ #define OP_SSS_W int32_t, int32_t, int32_t, int32_t, int32_t
+ #define OP_SSS_D int64_t, int64_t, int64_t, int64_t, int64_t
+-#define OP_UUU_B uint8_t, uint8_t, uint8_t, uint8_t, uint8_t
+-#define OP_UUU_H uint16_t, uint16_t, uint16_t, uint16_t, uint16_t
+-#define OP_UUU_W uint32_t, uint32_t, uint32_t, uint32_t, uint32_t
+-#define OP_UUU_D uint64_t, uint64_t, uint64_t, uint64_t, uint64_t
+ #define OP_SUS_B int8_t, uint8_t, int8_t, uint8_t, int8_t
+ #define OP_SUS_H int16_t, uint16_t, int16_t, uint16_t, int16_t
+ #define OP_SUS_W int32_t, uint32_t, int32_t, uint32_t, int32_t
+@@ -XXX,XX +XXX,XX @@ GEN_VEXT_ST_WHOLE(vs8r_v, int8_t, ste_b)
+ #define NOP_UUU_H uint16_t, uint16_t, uint32_t, uint16_t, uint32_t
+ #define NOP_UUU_W uint32_t, uint32_t, uint64_t, uint32_t, uint64_t
+-/* operation of two vector elements */
+-typedef void opivv2_fn(void *vd, void *vs1, void *vs2, int i);
+-
+-#define OPIVV2(NAME, TD, T1, T2, TX1, TX2, HD, HS1, HS2, OP)    \
+-static void do_##NAME(void *vd, void *vs1, void *vs2, int i)    \
+-{                                                               \
+-    TX1 s1 = *((T1 *)vs1 + HS1(i));                             \
+-    TX2 s2 = *((T2 *)vs2 + HS2(i));                             \
+-    *((TD *)vd + HD(i)) = OP(s2, s1);                           \
+-}
+ #define DO_SUB(N, M) (N - M)
+ #define DO_RSUB(N, M) (M - N)
+@@ -XXX,XX +XXX,XX @@ RVVCALL(OPIVV2, vsub_vv_h, OP_SSS_H, H2, H2, H2, DO_SUB)
+ RVVCALL(OPIVV2, vsub_vv_w, OP_SSS_W, H4, H4, H4, DO_SUB)
+ RVVCALL(OPIVV2, vsub_vv_d, OP_SSS_D, H8, H8, H8, DO_SUB)
+-static void do_vext_vv(void *vd, void *v0, void *vs1, void *vs2,
+-                       CPURISCVState *env, uint32_t desc,
+-                       opivv2_fn *fn, uint32_t esz)
+-{
+-    uint32_t vm = vext_vm(desc);
+-    uint32_t vl = env->vl;
+-    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
+-    uint32_t vta = vext_vta(desc);
+-    uint32_t vma = vext_vma(desc);
+-    uint32_t i;
+-
+-    for (i = env->vstart; i < vl; i++) {
+-        if (!vm && !vext_elem_mask(v0, i)) {
+-            /* set masked-off elements to 1s */
+-            vext_set_elems_1s(vd, vma, i * esz, (i + 1) * esz);
+-            continue;
+-        }
+-        fn(vd, vs1, vs2, i);
+-    }
+-    env->vstart = 0;
+-    /* set tail elements to 1s */
+-    vext_set_elems_1s(vd, vta, vl * esz, total_elems * esz);
+-}
+-
+-/* generate the helpers for OPIVV */
+-#define GEN_VEXT_VV(NAME, ESZ)                            \
+-void HELPER(NAME)(void *vd, void *v0, void *vs1,          \
+-                  void *vs2, CPURISCVState *env,          \
+-                  uint32_t desc)                          \
+-{                                                         \
+-    do_vext_vv(vd, v0, vs1, vs2, env, desc,               \
+-               do_##NAME, ESZ);                           \
+-}
+-
+ GEN_VEXT_VV(vadd_vv_b, 1)
+ GEN_VEXT_VV(vadd_vv_h, 2)
+ GEN_VEXT_VV(vadd_vv_w, 4)
+@@ -XXX,XX +XXX,XX @@ GEN_VEXT_VV(vsub_vv_h, 2)
+ GEN_VEXT_VV(vsub_vv_w, 4)
+ GEN_VEXT_VV(vsub_vv_d, 8)
+-typedef void opivx2_fn(void *vd, target_long s1, void *vs2, int i);
+-
+-/*
+- * (T1)s1 gives the real operator type.
+- * (TX1)(T1)s1 expands the operator type of widen or narrow operations.
+- */
+-#define OPIVX2(NAME, TD, T1, T2, TX1, TX2, HD, HS2, OP)             \
+-static void do_##NAME(void *vd, target_long s1, void *vs2, int i)   \
+-{                                                                   \
+-    TX2 s2 = *((T2 *)vs2 + HS2(i));                                 \
+-    *((TD *)vd + HD(i)) = OP(s2, (TX1)(T1)s1);                      \
+-}
+ RVVCALL(OPIVX2, vadd_vx_b, OP_SSS_B, H1, H1, DO_ADD)
+ RVVCALL(OPIVX2, vadd_vx_h, OP_SSS_H, H2, H2, DO_ADD)
+@@ -XXX,XX +XXX,XX @@ RVVCALL(OPIVX2, vrsub_vx_h, OP_SSS_H, H2, H2, DO_RSUB)
+ RVVCALL(OPIVX2, vrsub_vx_w, OP_SSS_W, H4, H4, DO_RSUB)
+ RVVCALL(OPIVX2, vrsub_vx_d, OP_SSS_D, H8, H8, DO_RSUB)
+-static void do_vext_vx(void *vd, void *v0, target_long s1, void *vs2,
+-                       CPURISCVState *env, uint32_t desc,
+-                       opivx2_fn fn, uint32_t esz)
+-{
+-    uint32_t vm = vext_vm(desc);
+-    uint32_t vl = env->vl;
+-    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
+-    uint32_t vta = vext_vta(desc);
+-    uint32_t vma = vext_vma(desc);
+-    uint32_t i;
+-
+-    for (i = env->vstart; i < vl; i++) {
+-        if (!vm && !vext_elem_mask(v0, i)) {
+-            /* set masked-off elements to 1s */
+-            vext_set_elems_1s(vd, vma, i * esz, (i + 1) * esz);
+-            continue;
+-        }
+-        fn(vd, s1, vs2, i);
+-    }
+-    env->vstart = 0;
+-    /* set tail elements to 1s */
+-    vext_set_elems_1s(vd, vta, vl * esz, total_elems * esz);
+-}
+-
+-/* generate the helpers for OPIVX */
+-#define GEN_VEXT_VX(NAME, ESZ)                            \
+-void HELPER(NAME)(void *vd, void *v0, target_ulong s1,    \
+-                  void *vs2, CPURISCVState *env,          \
+-                  uint32_t desc)                          \
+-{                                                         \
+-    do_vext_vx(vd, v0, s1, vs2, env, desc,                \
+-               do_##NAME, ESZ);                           \
+-}
+-
+ GEN_VEXT_VX(vadd_vx_b, 1)
+ GEN_VEXT_VX(vadd_vx_h, 2)
+ GEN_VEXT_VX(vadd_vx_w, 4)
+diff --git a/target/riscv/vector_internals.c b/target/riscv/vector_internals.c
+new file mode 100644
+index XXXXXXX..XXXXXXX
+--- /dev/null
++++ b/target/riscv/vector_internals.c
+@@ -XXX,XX +XXX,XX @@
++/*
++ * RISC-V Vector Extension Internals
++ *
++ * Copyright (c) 2020 T-Head Semiconductor Co., Ltd. All rights reserved.
++ *
++ * This program is free software; you can redistribute it and/or modify it
++ * under the terms and conditions of the GNU General Public License,
++ * version 2 or later, as published by the Free Software Foundation.
++ *
++ * This program is distributed in the hope it will be useful, but WITHOUT
++ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
++ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
++ * more details.
++ *
++ * You should have received a copy of the GNU General Public License along with
++ * this program.  If not, see <http://www.gnu.org/licenses/>.
++ */
++
++#include "vector_internals.h"
++
++/* set agnostic elements to 1s */
++void vext_set_elems_1s(void *base, uint32_t is_agnostic, uint32_t cnt,
++                       uint32_t tot)
++{
++    if (is_agnostic == 0) {
++        /* policy undisturbed */
++        return;
++    }
++    if (tot - cnt == 0) {
++        return ;
++    }
++    memset(base + cnt, -1, tot - cnt);
++}
++
++void do_vext_vv(void *vd, void *v0, void *vs1, void *vs2,
++                CPURISCVState *env, uint32_t desc,
++                opivv2_fn *fn, uint32_t esz)
++{
++    uint32_t vm = vext_vm(desc);
++    uint32_t vl = env->vl;
++    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
++    uint32_t vta = vext_vta(desc);
++    uint32_t vma = vext_vma(desc);
++    uint32_t i;
++
++    for (i = env->vstart; i < vl; i++) {
++        if (!vm && !vext_elem_mask(v0, i)) {
++            /* set masked-off elements to 1s */
++            vext_set_elems_1s(vd, vma, i * esz, (i + 1) * esz);
++            continue;
++        }
++        fn(vd, vs1, vs2, i);
++    }
++    env->vstart = 0;
++    /* set tail elements to 1s */
++    vext_set_elems_1s(vd, vta, vl * esz, total_elems * esz);
++}
++
++void do_vext_vx(void *vd, void *v0, target_long s1, void *vs2,
++                CPURISCVState *env, uint32_t desc,
++                opivx2_fn fn, uint32_t esz)
++{
++    uint32_t vm = vext_vm(desc);
++    uint32_t vl = env->vl;
++    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
++    uint32_t vta = vext_vta(desc);
++    uint32_t vma = vext_vma(desc);
++    uint32_t i;
++
++    for (i = env->vstart; i < vl; i++) {
++        if (!vm && !vext_elem_mask(v0, i)) {
++            /* set masked-off elements to 1s */
++            vext_set_elems_1s(vd, vma, i * esz, (i + 1) * esz);
++            continue;
++        }
++        fn(vd, s1, vs2, i);
++    }
++    env->vstart = 0;
++    /* set tail elements to 1s */
++    vext_set_elems_1s(vd, vta, vl * esz, total_elems * esz);
++}
+diff --git a/target/riscv/meson.build b/target/riscv/meson.build
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/meson.build
++++ b/target/riscv/meson.build
+@@ -XXX,XX +XXX,XX @@ riscv_ss.add(files(
+   'gdbstub.c',
+   'op_helper.c',
+   'vector_helper.c',
++  'vector_internals.c',
+   'bitmanip_helper.c',
+   'translate.c',
+   'm128_helper.c',
+--
+.41.0

-New patch
+[PULL v2 09/45] target/riscv: Refactor vector-vector translation macro
+From: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
+Refactor the non SEW-specific stuff out of `GEN_OPIVV_TRANS` into
+function `opivv_trans` (similar to `opivi_trans`). `opivv_trans` will be
+used in proceeding vector-crypto commits.
+Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Message-ID: <20230711165917.2629866-3-max.chou@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/insn_trans/trans_rvv.c.inc | 62 +++++++++++++------------
+file changed, 32 insertions(+), 30 deletions(-)
+diff --git a/target/riscv/insn_trans/trans_rvv.c.inc b/target/riscv/insn_trans/trans_rvv.c.inc
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/insn_trans/trans_rvv.c.inc
++++ b/target/riscv/insn_trans/trans_rvv.c.inc
+@@ -XXX,XX +XXX,XX @@ GEN_OPIWX_WIDEN_TRANS(vwadd_wx)
+ GEN_OPIWX_WIDEN_TRANS(vwsubu_wx)
+ GEN_OPIWX_WIDEN_TRANS(vwsub_wx)
++static bool opivv_trans(uint32_t vd, uint32_t vs1, uint32_t vs2, uint32_t vm,
++                        gen_helper_gvec_4_ptr *fn, DisasContext *s)
++{
++    uint32_t data = 0;
++    TCGLabel *over = gen_new_label();
++    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
++    tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
++
++    data = FIELD_DP32(data, VDATA, VM, vm);
++    data = FIELD_DP32(data, VDATA, LMUL, s->lmul);
++    data = FIELD_DP32(data, VDATA, VTA, s->vta);
++    data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);
++    data = FIELD_DP32(data, VDATA, VMA, s->vma);
++    tcg_gen_gvec_4_ptr(vreg_ofs(s, vd), vreg_ofs(s, 0), vreg_ofs(s, vs1),
++                       vreg_ofs(s, vs2), cpu_env, s->cfg_ptr->vlen / 8,
++                       s->cfg_ptr->vlen / 8, data, fn);
++    mark_vs_dirty(s);
++    gen_set_label(over);
++    return true;
++}
++
+ /* Vector Integer Add-with-Carry / Subtract-with-Borrow Instructions */
+ /* OPIVV without GVEC IR */
+-#define GEN_OPIVV_TRANS(NAME, CHECK)                               \
+-static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
+-{                                                                  \
+-    if (CHECK(s, a)) {                                             \
+-        uint32_t data = 0;                                         \
+-        static gen_helper_gvec_4_ptr * const fns[4] = {            \
+-            gen_helper_##NAME##_b, gen_helper_##NAME##_h,          \
+-            gen_helper_##NAME##_w, gen_helper_##NAME##_d,          \
+-        };                                                         \
+-        TCGLabel *over = gen_new_label();                          \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
+-        tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
+-                                                                   \
+-        data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
+-        data = FIELD_DP32(data, VDATA, LMUL, s->lmul);             \
+-        data = FIELD_DP32(data, VDATA, VTA, s->vta);               \
+-        data =                                                     \
+-            FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);\
+-        data = FIELD_DP32(data, VDATA, VMA, s->vma);               \
+-        tcg_gen_gvec_4_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, 0),     \
+-                           vreg_ofs(s, a->rs1),                    \
+-                           vreg_ofs(s, a->rs2), cpu_env,           \
+-                           s->cfg_ptr->vlen / 8,                   \
+-                           s->cfg_ptr->vlen / 8, data,             \
+-                           fns[s->sew]);                           \
+-        mark_vs_dirty(s);                                          \
+-        gen_set_label(over);                                       \
+-        return true;                                               \
+-    }                                                              \
+-    return false;                                                  \
++#define GEN_OPIVV_TRANS(NAME, CHECK)                                     \
++static bool trans_##NAME(DisasContext *s, arg_rmrr *a)                   \
++{                                                                        \
++    if (CHECK(s, a)) {                                                   \
++        static gen_helper_gvec_4_ptr * const fns[4] = {                  \
++            gen_helper_##NAME##_b, gen_helper_##NAME##_h,                \
++            gen_helper_##NAME##_w, gen_helper_##NAME##_d,                \
++        };                                                               \
++        return opivv_trans(a->rd, a->rs1, a->rs2, a->vm, fns[s->sew], s);\
++    }                                                                    \
++    return false;                                                        \
+ }
+ /*
+--
+.41.0

-New patch
+[PULL v2 10/45] target/riscv: Remove redundant "cpu_vl == 0" checks
+From: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
+Remove the redundant "vl == 0" check which is already included within the  vstart >= vl check, when vl == 0.
+Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
+Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Acked-by: Alistair Francis <alistair.francis@wdc.com>
+Message-ID: <20230711165917.2629866-4-max.chou@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/insn_trans/trans_rvv.c.inc | 31 +------------------------
+file changed, 1 insertion(+), 30 deletions(-)
+diff --git a/target/riscv/insn_trans/trans_rvv.c.inc b/target/riscv/insn_trans/trans_rvv.c.inc
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/insn_trans/trans_rvv.c.inc
++++ b/target/riscv/insn_trans/trans_rvv.c.inc
+@@ -XXX,XX +XXX,XX @@ static bool ldst_us_trans(uint32_t vd, uint32_t rs1, uint32_t data,
+     TCGv_i32 desc;
+     TCGLabel *over = gen_new_label();
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     dest = tcg_temp_new_ptr();
+@@ -XXX,XX +XXX,XX @@ static bool ldst_stride_trans(uint32_t vd, uint32_t rs1, uint32_t rs2,
+     TCGv_i32 desc;
+     TCGLabel *over = gen_new_label();
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     dest = tcg_temp_new_ptr();
+@@ -XXX,XX +XXX,XX @@ static bool ldst_index_trans(uint32_t vd, uint32_t rs1, uint32_t vs2,
+     TCGv_i32 desc;
+     TCGLabel *over = gen_new_label();
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     dest = tcg_temp_new_ptr();
+@@ -XXX,XX +XXX,XX @@ static bool ldff_trans(uint32_t vd, uint32_t rs1, uint32_t data,
+     TCGv_i32 desc;
+     TCGLabel *over = gen_new_label();
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     dest = tcg_temp_new_ptr();
+@@ -XXX,XX +XXX,XX @@ do_opivv_gvec(DisasContext *s, arg_rmrr *a, GVecGen3Fn *gvec_fn,
+         return false;
+     }
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     if (a->vm && s->vl_eq_vlmax && !(s->vta && s->lmul < 0)) {
+@@ -XXX,XX +XXX,XX @@ static bool opivx_trans(uint32_t vd, uint32_t rs1, uint32_t vs2, uint32_t vm,
+     uint32_t data = 0;
+     TCGLabel *over = gen_new_label();
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     dest = tcg_temp_new_ptr();
+@@ -XXX,XX +XXX,XX @@ static bool opivi_trans(uint32_t vd, uint32_t imm, uint32_t vs2, uint32_t vm,
+     uint32_t data = 0;
+     TCGLabel *over = gen_new_label();
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     dest = tcg_temp_new_ptr();
+@@ -XXX,XX +XXX,XX @@ static bool do_opivv_widen(DisasContext *s, arg_rmrr *a,
+     if (checkfn(s, a)) {
+         uint32_t data = 0;
+         TCGLabel *over = gen_new_label();
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+         data = FIELD_DP32(data, VDATA, VM, a->vm);
+@@ -XXX,XX +XXX,XX @@ static bool do_opiwv_widen(DisasContext *s, arg_rmrr *a,
+     if (opiwv_widen_check(s, a)) {
+         uint32_t data = 0;
+         TCGLabel *over = gen_new_label();
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+         data = FIELD_DP32(data, VDATA, VM, a->vm);
+@@ -XXX,XX +XXX,XX @@ static bool opivv_trans(uint32_t vd, uint32_t vs1, uint32_t vs2, uint32_t vm,
+ {
+     uint32_t data = 0;
+     TCGLabel *over = gen_new_label();
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     data = FIELD_DP32(data, VDATA, VM, vm);
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
+             gen_helper_##NAME##_w,                                 \
+         };                                                         \
+         TCGLabel *over = gen_new_label();                          \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
+                                                                    \
+         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
+@@ -XXX,XX +XXX,XX @@ static bool trans_vmv_v_v(DisasContext *s, arg_vmv_v_v *a)
+                 gen_helper_vmv_v_v_w, gen_helper_vmv_v_v_d,
+             };
+             TCGLabel *over = gen_new_label();
+-            tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+             tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+             tcg_gen_gvec_2_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, a->rs1),
+@@ -XXX,XX +XXX,XX @@ static bool trans_vmv_v_x(DisasContext *s, arg_vmv_v_x *a)
+         vext_check_ss(s, a->rd, 0, 1)) {
+         TCGv s1;
+         TCGLabel *over = gen_new_label();
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+         s1 = get_gpr(s, a->rs1, EXT_SIGN);
+@@ -XXX,XX +XXX,XX @@ static bool trans_vmv_v_i(DisasContext *s, arg_vmv_v_i *a)
+                 gen_helper_vmv_v_x_w, gen_helper_vmv_v_x_d,
+             };
+             TCGLabel *over = gen_new_label();
+-            tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+             tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+             s1 = tcg_constant_i64(simm);
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
+         };                                                         \
+         TCGLabel *over = gen_new_label();                          \
+         gen_set_rm(s, RISCV_FRM_DYN);                              \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
+                                                                    \
+         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
+@@ -XXX,XX +XXX,XX @@ static bool opfvf_trans(uint32_t vd, uint32_t rs1, uint32_t vs2,
+     TCGv_i64 t1;
+     TCGLabel *over = gen_new_label();
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     dest = tcg_temp_new_ptr();
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)           \
+         };                                                       \
+         TCGLabel *over = gen_new_label();                        \
+         gen_set_rm(s, RISCV_FRM_DYN);                            \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);        \
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);\
+                                                                  \
+         data = FIELD_DP32(data, VDATA, VM, a->vm);               \
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
+         };                                                         \
+         TCGLabel *over = gen_new_label();                          \
+         gen_set_rm(s, RISCV_FRM_DYN);                              \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
+                                                                    \
+         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
+@@ -XXX,XX +XXX,XX @@ static bool do_opfv(DisasContext *s, arg_rmr *a,
+         uint32_t data = 0;
+         TCGLabel *over = gen_new_label();
+         gen_set_rm_chkfrm(s, rm);
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+         data = FIELD_DP32(data, VDATA, VM, a->vm);
+@@ -XXX,XX +XXX,XX @@ static bool trans_vfmv_v_f(DisasContext *s, arg_vfmv_v_f *a)
+                 gen_helper_vmv_v_x_d,
+             };
+             TCGLabel *over = gen_new_label();
+-            tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+             tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+             t1 = tcg_temp_new_i64();
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a)              \
+         };                                                         \
+         TCGLabel *over = gen_new_label();                          \
+         gen_set_rm_chkfrm(s, FRM);                                 \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
+                                                                    \
+         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a)              \
+         };                                                         \
+         TCGLabel *over = gen_new_label();                          \
+         gen_set_rm(s, RISCV_FRM_DYN);                              \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
+                                                                    \
+         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a)              \
+         };                                                         \
+         TCGLabel *over = gen_new_label();                          \
+         gen_set_rm_chkfrm(s, FRM);                                 \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
+                                                                    \
+         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a)              \
+         };                                                         \
+         TCGLabel *over = gen_new_label();                          \
+         gen_set_rm_chkfrm(s, FRM);                                 \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
+                                                                    \
+         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_r *a)                \
+         uint32_t data = 0;                                         \
+         gen_helper_gvec_4_ptr *fn = gen_helper_##NAME;             \
+         TCGLabel *over = gen_new_label();                          \
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
+                                                                    \
+         data = FIELD_DP32(data, VDATA, LMUL, s->lmul);             \
+@@ -XXX,XX +XXX,XX @@ static bool trans_vid_v(DisasContext *s, arg_vid_v *a)
+         require_vm(a->vm, a->rd)) {
+         uint32_t data = 0;
+         TCGLabel *over = gen_new_label();
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+         data = FIELD_DP32(data, VDATA, VM, a->vm);
+@@ -XXX,XX +XXX,XX @@ static bool trans_vmv_s_x(DisasContext *s, arg_vmv_s_x *a)
+         TCGv s1;
+         TCGLabel *over = gen_new_label();
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+         t1 = tcg_temp_new_i64();
+@@ -XXX,XX +XXX,XX @@ static bool trans_vfmv_s_f(DisasContext *s, arg_vfmv_s_f *a)
+         TCGv_i64 t1;
+         TCGLabel *over = gen_new_label();
+-        /* if vl == 0 or vstart >= vl, skip vector register write back */
+-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
++        /* if vstart >= vl, skip vector register write back */
+         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+         /* NaN-box f[rs1] */
+@@ -XXX,XX +XXX,XX @@ static bool int_ext_op(DisasContext *s, arg_rmr *a, uint8_t seq)
+     uint32_t data = 0;
+     gen_helper_gvec_3_ptr *fn;
+     TCGLabel *over = gen_new_label();
+-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+     static gen_helper_gvec_3_ptr * const fns[6][4] = {
+--
+.41.0

-New patch
+[PULL v2 11/45] target/riscv: Add Zvbc ISA extension support
+From: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
 This commit adds support for the Zvbc vector-crypto extension, which
 consists of the following instructions:
 * vclmulh.[vx,vv]
 * vclmul.[vx,vv]
 Translation functions are defined in
 `target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
 `target/riscv/vcrypto_helper.c`.
 Co-authored-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
 Co-authored-by: Max Chou <max.chou@sifive.com>
 Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
 Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
 Signed-off-by: Max Chou <max.chou@sifive.com>
 [max.chou@sifive.com: Exposed x-zvbc property]
 Message-ID: <20230711165917.2629866-5-max.chou@sifive.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
  target/riscv/cpu_cfg.h                   |  1 +
  target/riscv/helper.h                    |  6 +++
  target/riscv/insn32.decode               |  6 +++
  target/riscv/cpu.c                       |  9 ++++
  target/riscv/translate.c                 |  1 +
  target/riscv/vcrypto_helper.c            | 59 ++++++++++++++++++++++
  target/riscv/insn_trans/trans_rvvk.c.inc | 62 ++++++++++++++++++++++++
  target/riscv/meson.build                 |  3 +-
 files changed, 146 insertions(+), 1 deletion(-)
  create mode 100644 target/riscv/vcrypto_helper.c
  create mode 100644 target/riscv/insn_trans/trans_rvvk.c.inc
 diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/cpu_cfg.h
 +++ b/target/riscv/cpu_cfg.h
@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
      bool ext_zve32f;
      bool ext_zve64f;
      bool ext_zve64d;
 +    bool ext_zvbc;
      bool ext_zmmul;
      bool ext_zvfbfmin;
      bool ext_zvfbfwma;
 diff --git a/target/riscv/helper.h b/target/riscv/helper.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/helper.h
 +++ b/target/riscv/helper.h
@@ -XXX,XX +XXX,XX @@ DEF_HELPER_5(vfwcvtbf16_f_f_v, void, ptr, ptr, ptr, env, i32)
  DEF_HELPER_6(vfwmaccbf16_vv, void, ptr, ptr, ptr, ptr, env, i32)
  DEF_HELPER_6(vfwmaccbf16_vf, void, ptr, ptr, i64, ptr, env, i32)
 +
 +/* Vector crypto functions */
 +DEF_HELPER_6(vclmul_vv, void, ptr, ptr, ptr, ptr, env, i32)
 +DEF_HELPER_6(vclmul_vx, void, ptr, ptr, tl, ptr, env, i32)
 +DEF_HELPER_6(vclmulh_vv, void, ptr, ptr, ptr, ptr, env, i32)
 +DEF_HELPER_6(vclmulh_vx, void, ptr, ptr, tl, ptr, env, i32)
 diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/insn32.decode
 +++ b/target/riscv/insn32.decode
@@ -XXX,XX +XXX,XX @@ vfwcvtbf16_f_f_v  010010 . ..... 01101 001 ..... 1010111 @r2_vm
  # *** Zvfbfwma Standard Extension ***
  vfwmaccbf16_vv    111011 . ..... ..... 001 ..... 1010111 @r_vm
  vfwmaccbf16_vf    111011 . ..... ..... 101 ..... 1010111 @r_vm
 +
 +# *** Zvbc vector crypto extension ***
 +vclmul_vv   001100 . ..... ..... 010 ..... 1010111 @r_vm
 +vclmul_vx   001100 . ..... ..... 110 ..... 1010111 @r_vm
 +vclmulh_vv  001101 . ..... ..... 010 ..... 1010111 @r_vm
 +vclmulh_vx  001101 . ..... ..... 110 ..... 1010111 @r_vm
 diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/cpu.c
 +++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
      ISA_EXT_DATA_ENTRY(zksed, PRIV_VERSION_1_12_0, ext_zksed),
      ISA_EXT_DATA_ENTRY(zksh, PRIV_VERSION_1_12_0, ext_zksh),
      ISA_EXT_DATA_ENTRY(zkt, PRIV_VERSION_1_12_0, ext_zkt),
 +    ISA_EXT_DATA_ENTRY(zvbc, PRIV_VERSION_1_12_0, ext_zvbc),
      ISA_EXT_DATA_ENTRY(zve32f, PRIV_VERSION_1_10_0, ext_zve32f),
      ISA_EXT_DATA_ENTRY(zve64f, PRIV_VERSION_1_10_0, ext_zve64f),
      ISA_EXT_DATA_ENTRY(zve64d, PRIV_VERSION_1_10_0, ext_zve64d),
@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
          return;
      }
 +    if (cpu->cfg.ext_zvbc && !cpu->cfg.ext_zve64f) {
 +        error_setg(errp, "Zvbc extension requires V or Zve64{f,d} extensions");
 +        return;
 +    }
 +
      if (cpu->cfg.ext_zk) {
          cpu->cfg.ext_zkn = true;
          cpu->cfg.ext_zkr = true;
@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
      DEFINE_PROP_BOOL("x-zvfbfmin", RISCVCPU, cfg.ext_zvfbfmin, false),
      DEFINE_PROP_BOOL("x-zvfbfwma", RISCVCPU, cfg.ext_zvfbfwma, false),
 +    /* Vector cryptography extensions */
 +    DEFINE_PROP_BOOL("x-zvbc", RISCVCPU, cfg.ext_zvbc, false),
 +
      DEFINE_PROP_END_OF_LIST(),
  };
 diff --git a/target/riscv/translate.c b/target/riscv/translate.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/translate.c
 +++ b/target/riscv/translate.c
@@ -XXX,XX +XXX,XX @@ static uint32_t opcode_at(DisasContextBase *dcbase, target_ulong pc)
  #include "insn_trans/trans_rvzfa.c.inc"
  #include "insn_trans/trans_rvzfh.c.inc"
  #include "insn_trans/trans_rvk.c.inc"
 +#include "insn_trans/trans_rvvk.c.inc"
  #include "insn_trans/trans_privileged.c.inc"
  #include "insn_trans/trans_svinval.c.inc"
  #include "insn_trans/trans_rvbf16.c.inc"
 diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/target/riscv/vcrypto_helper.c
@@ -XXX,XX +XXX,XX @@
 +/*
 + * RISC-V Vector Crypto Extension Helpers for QEMU.
 + *
 + * Copyright (C) 2023 SiFive, Inc.
 + * Written by Codethink Ltd and SiFive.
 + *
 + * This program is free software; you can redistribute it and/or modify it
 + * under the terms and conditions of the GNU General Public License,
 + * version 2 or later, as published by the Free Software Foundation.
 + *
 + * This program is distributed in the hope it will be useful, but WITHOUT
 + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 + * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
 + * more details.
 + *
 + * You should have received a copy of the GNU General Public License along with
 + * this program.  If not, see <http://www.gnu.org/licenses/>.
 + */
 +
 +#include "qemu/osdep.h"
 +#include "qemu/host-utils.h"
 +#include "qemu/bitops.h"
 +#include "cpu.h"
 +#include "exec/memop.h"
 +#include "exec/exec-all.h"
 +#include "exec/helper-proto.h"
 +#include "internals.h"
 +#include "vector_internals.h"
 +
 +static uint64_t clmul64(uint64_t y, uint64_t x)
 +{
 +    uint64_t result = 0;
 +    for (int j = 63; j >= 0; j--) {
 +        if ((y >> j) & 1) {
 +            result ^= (x << j);
 +        }
 +    }
 +    return result;
 +}
 +
 +static uint64_t clmulh64(uint64_t y, uint64_t x)
 +{
 +    uint64_t result = 0;
 +    for (int j = 63; j >= 1; j--) {
 +        if ((y >> j) & 1) {
 +            result ^= (x >> (64 - j));
 +        }
 +    }
 +    return result;
 +}
 +
 +RVVCALL(OPIVV2, vclmul_vv, OP_UUU_D, H8, H8, H8, clmul64)
 +GEN_VEXT_VV(vclmul_vv, 8)
 +RVVCALL(OPIVX2, vclmul_vx, OP_UUU_D, H8, H8, clmul64)
 +GEN_VEXT_VX(vclmul_vx, 8)
 +RVVCALL(OPIVV2, vclmulh_vv, OP_UUU_D, H8, H8, H8, clmulh64)
 +GEN_VEXT_VV(vclmulh_vv, 8)
 +RVVCALL(OPIVX2, vclmulh_vx, OP_UUU_D, H8, H8, clmulh64)
 +GEN_VEXT_VX(vclmulh_vx, 8)
 diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/target/riscv/insn_trans/trans_rvvk.c.inc
@@ -XXX,XX +XXX,XX @@
 +/*
 + * RISC-V translation routines for the vector crypto extension.
 + *
 + * Copyright (C) 2023 SiFive, Inc.
 + * Written by Codethink Ltd and SiFive.
 + *
 + * This program is free software; you can redistribute it and/or modify it
 + * under the terms and conditions of the GNU General Public License,
 + * version 2 or later, as published by the Free Software Foundation.
 + *
 + * This program is distributed in the hope it will be useful, but WITHOUT
 + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 + * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
 + * more details.
 + *
 + * You should have received a copy of the GNU General Public License along with
 + * this program.  If not, see <http://www.gnu.org/licenses/>.
 + */
 +
 +/*
 + * Zvbc
 + */
 +
 +#define GEN_VV_MASKED_TRANS(NAME, CHECK)                     \
 +    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)   \
 +    {                                                        \
 +        if (CHECK(s, a)) {                                   \
 +            return opivv_trans(a->rd, a->rs1, a->rs2, a->vm, \
 +                               gen_helper_##NAME, s);        \
 +        }                                                    \
 +        return false;                                        \
 +    }
 +
 +static bool vclmul_vv_check(DisasContext *s, arg_rmrr *a)
 +{
 +    return opivv_check(s, a) &&
 +           s->cfg_ptr->ext_zvbc == true &&
 +           s->sew == MO_64;
 +}
 +
 +GEN_VV_MASKED_TRANS(vclmul_vv, vclmul_vv_check)
 +GEN_VV_MASKED_TRANS(vclmulh_vv, vclmul_vv_check)
 +
 +#define GEN_VX_MASKED_TRANS(NAME, CHECK)                     \
 +    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)   \
 +    {                                                        \
 +        if (CHECK(s, a)) {                                   \
 +            return opivx_trans(a->rd, a->rs1, a->rs2, a->vm, \
 +                               gen_helper_##NAME, s);        \
 +        }                                                    \
 +        return false;                                        \
 +    }
 +
 +static bool vclmul_vx_check(DisasContext *s, arg_rmrr *a)
 +{
 +    return opivx_check(s, a) &&
 +           s->cfg_ptr->ext_zvbc == true &&
 +           s->sew == MO_64;
 +}
 +
 +GEN_VX_MASKED_TRANS(vclmul_vx, vclmul_vx_check)
 +GEN_VX_MASKED_TRANS(vclmulh_vx, vclmul_vx_check)
 diff --git a/target/riscv/meson.build b/target/riscv/meson.build
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/meson.build
 +++ b/target/riscv/meson.build
@@ -XXX,XX +XXX,XX @@ riscv_ss.add(files(
    'translate.c',
    'm128_helper.c',
    'crypto_helper.c',
 -  'zce_helper.c'
 +  'zce_helper.c',
 +  'vcrypto_helper.c'
  ))
  riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'), if_false: files('kvm-stub.c'))
 --
 .41.0

-[PULL 04/10] hw/arm, loongarch: Move load_image_to_fw_cfg() to common location
+[PULL v2 12/45] target/riscv: Move vector translation checks
-From: Sunil V L <sunilvl@ventanamicro.com>
+From: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
-load_image_to_fw_cfg() is duplicated by both arm and loongarch. The same
+Move the checks out of `do_opiv{v,x,i}_gvec{,_shift}` functions
-function will be required by riscv too. So, it's time to refactor and
+and into the corresponding macros. This enables the functions to be
-move this function to a common path.
+reused in proceeding commits without check duplication.
-Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
+Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
-Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
-Reviewed-by: Song Gao <gaosong@loongson.cn>
+Signed-off-by: Max Chou <max.chou@sifive.com>
-Message-Id: <20221004092351.18209-2-sunilvl@ventanamicro.com>
+Message-ID: <20230711165917.2629866-6-max.chou@sifive.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
- include/hw/nvram/fw_cfg.h | 21 +++++++++++++++++
+ target/riscv/insn_trans/trans_rvv.c.inc | 28 +++++++++++--------------
- hw/arm/boot.c             | 49 ---------------------------------------
+file changed, 12 insertions(+), 16 deletions(-)
  hw/loongarch/virt.c       | 33 --------------------------
  hw/nvram/fw_cfg.c         | 32 +++++++++++++++++++++++++
 files changed, 53 insertions(+), 82 deletions(-)
-diff --git a/include/hw/nvram/fw_cfg.h b/include/hw/nvram/fw_cfg.h
+diff --git a/target/riscv/insn_trans/trans_rvv.c.inc b/target/riscv/insn_trans/trans_rvv.c.inc
 index XXXXXXX..XXXXXXX 100644
---- a/include/hw/nvram/fw_cfg.h
+--- a/target/riscv/insn_trans/trans_rvv.c.inc
-+++ b/include/hw/nvram/fw_cfg.h
++++ b/target/riscv/insn_trans/trans_rvv.c.inc
-@@ -XXX,XX +XXX,XX @@ bool fw_cfg_dma_enabled(void *opaque);
+@@ -XXX,XX +XXX,XX @@ do_opivv_gvec(DisasContext *s, arg_rmrr *a, GVecGen3Fn *gvec_fn,
-  */
+               gen_helper_gvec_4_ptr *fn)
- const char *fw_cfg_arch_key_name(uint16_t key);
+ {
+     TCGLabel *over = gen_new_label();
-+/**
+-    if (!opivv_check(s, a)) {
-+ * load_image_to_fw_cfg() - Load an image file into an fw_cfg entry identified
+-        return false;
-+ *                          by key.
+-    }
-+ * @fw_cfg:         The firmware config instance to store the data in.
-+ * @size_key:       The firmware config key to store the size of the loaded
+     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
-+ *                  data under, with fw_cfg_add_i32().
-+ * @data_key:       The firmware config key to store the loaded data under,
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
-+ *                  with fw_cfg_add_bytes().
+         gen_helper_##NAME##_b, gen_helper_##NAME##_h,              \
-+ * @image_name:     The name of the image file to load. If it is NULL, the
+         gen_helper_##NAME##_w, gen_helper_##NAME##_d,              \
-+ *                  function returns without doing anything.
+     };                                                             \
-+ * @try_decompress: Whether the image should be decompressed (gunzipped) before
++    if (!opivv_check(s, a)) {                                      \
-+ *                  adding it to fw_cfg. If decompression fails, the image is
++        return false;                                              \
-+ *                  loaded as-is.
++    }                                                              \
-+ *
+     return do_opivv_gvec(s, a, tcg_gen_gvec_##SUF, fns[s->sew]);   \
 + * In case of failure, the function prints an error message to stderr and the
 + * process exits with status 1.
 + */
 +void load_image_to_fw_cfg(FWCfgState *fw_cfg, uint16_t size_key,
 +                          uint16_t data_key, const char *image_name,
 +                          bool try_decompress);
 +
  #endif
 diff --git a/hw/arm/boot.c b/hw/arm/boot.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/arm/boot.c
 +++ b/hw/arm/boot.c
@@ -XXX,XX +XXX,XX @@ static void do_cpu_reset(void *opaque)
      }
  }
--/**
+@@ -XXX,XX +XXX,XX @@ static inline bool
-- * load_image_to_fw_cfg() - Load an image file into an fw_cfg entry identified
+ do_opivx_gvec(DisasContext *s, arg_rmrr *a, GVecGen2sFn *gvec_fn,
-- *                          by key.
+               gen_helper_opivx *fn)
-- * @fw_cfg:         The firmware config instance to store the data in.
+ {
-- * @size_key:       The firmware config key to store the size of the loaded
+-    if (!opivx_check(s, a)) {
-- *                  data under, with fw_cfg_add_i32().
+-        return false;
 - * @data_key:       The firmware config key to store the loaded data under,
 - *                  with fw_cfg_add_bytes().
 - * @image_name:     The name of the image file to load. If it is NULL, the
 - *                  function returns without doing anything.
 - * @try_decompress: Whether the image should be decompressed (gunzipped) before
 - *                  adding it to fw_cfg. If decompression fails, the image is
 - *                  loaded as-is.
 - *
 - * In case of failure, the function prints an error message to stderr and the
 - * process exits with status 1.
 - */
 -static void load_image_to_fw_cfg(FWCfgState *fw_cfg, uint16_t size_key,
 -                                 uint16_t data_key, const char *image_name,
 -                                 bool try_decompress)
 -{
 -    size_t size = -1;
 -    uint8_t *data;
 -
 -    if (image_name == NULL) {
 -        return;
 -    }
 -
--    if (try_decompress) {
+     if (a->vm && s->vl_eq_vlmax && !(s->vta && s->lmul < 0)) {
--        size = load_image_gzipped_buffer(image_name,
+         TCGv_i64 src1 = tcg_temp_new_i64();
--                                         LOAD_IMAGE_MAX_GUNZIP_BYTES, &data);
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
          gen_helper_##NAME##_b, gen_helper_##NAME##_h,              \
          gen_helper_##NAME##_w, gen_helper_##NAME##_d,              \
      };                                                             \
 +    if (!opivx_check(s, a)) {                                      \
 +        return false;                                              \
 +    }                                                              \
      return do_opivx_gvec(s, a, tcg_gen_gvec_##SUF, fns[s->sew]);   \
  }
@@ -XXX,XX +XXX,XX @@ static inline bool
  do_opivi_gvec(DisasContext *s, arg_rmrr *a, GVecGen2iFn *gvec_fn,
                gen_helper_opivx *fn, imm_mode_t imm_mode)
  {
 -    if (!opivx_check(s, a)) {
 -        return false;
 -    }
 -
--    if (size == (size_t)-1) {
+     if (a->vm && s->vl_eq_vlmax && !(s->vta && s->lmul < 0)) {
--        gchar *contents;
+         gvec_fn(s->sew, vreg_ofs(s, a->rd), vreg_ofs(s, a->rs2),
--        gsize length;
+                 extract_imm(s, a->rs1, imm_mode), MAXSZ(s), MAXSZ(s));
--
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
--        if (!g_file_get_contents(image_name, &contents, &length, NULL)) {
+         gen_helper_##OPIVX##_b, gen_helper_##OPIVX##_h,            \
--            error_report("failed to load \"%s\"", image_name);
+         gen_helper_##OPIVX##_w, gen_helper_##OPIVX##_d,            \
--            exit(1);
+     };                                                             \
--        }
++    if (!opivx_check(s, a)) {                                      \
--        size = length;
++        return false;                                              \
--        data = (uint8_t *)contents;
++    }                                                              \
      return do_opivi_gvec(s, a, tcg_gen_gvec_##SUF,                 \
                           fns[s->sew], IMM_MODE);                   \
  }
@@ -XXX,XX +XXX,XX @@ static inline bool
  do_opivx_gvec_shift(DisasContext *s, arg_rmrr *a, GVecGen2sFn32 *gvec_fn,
                      gen_helper_opivx *fn)
  {
 -    if (!opivx_check(s, a)) {
 -        return false;
 -    }
 -
--    fw_cfg_add_i32(fw_cfg, size_key, size);
+     if (a->vm && s->vl_eq_vlmax && !(s->vta && s->lmul < 0)) {
--    fw_cfg_add_bytes(fw_cfg, data_key, data, size);
+         TCGv_i32 src1 = tcg_temp_new_i32();
--}
--
+@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)                    \
- static int do_arm_linux_init(Object *obj, void *opaque)
+         gen_helper_##NAME##_b, gen_helper_##NAME##_h,                     \
- {
+         gen_helper_##NAME##_w, gen_helper_##NAME##_d,                     \
-     if (object_dynamic_cast(obj, TYPE_ARM_LINUX_BOOT_IF)) {
+     };                                                                    \
-diff --git a/hw/loongarch/virt.c b/hw/loongarch/virt.c
+-                                                                          \
-index XXXXXXX..XXXXXXX 100644
++    if (!opivx_check(s, a)) {                                             \
---- a/hw/loongarch/virt.c
++        return false;                                                     \
-+++ b/hw/loongarch/virt.c
++    }                                                                     \
-@@ -XXX,XX +XXX,XX @@ static void reset_load_elf(void *opaque)
+     return do_opivx_gvec_shift(s, a, tcg_gen_gvec_##SUF, fns[s->sew]);    \
      }
  }
--/* Load an image file into an fw_cfg entry identified by key. */
--static void load_image_to_fw_cfg(FWCfgState *fw_cfg, uint16_t size_key,
--                                 uint16_t data_key, const char *image_name,
--                                 bool try_decompress)
--{
--    size_t size = -1;
--    uint8_t *data;
--
--    if (image_name == NULL) {
--        return;
--    }
--
--    if (try_decompress) {
--        size = load_image_gzipped_buffer(image_name,
--                                         LOAD_IMAGE_MAX_GUNZIP_BYTES, &data);
--    }
--
--    if (size == (size_t)-1) {
--        gchar *contents;
--        gsize length;
--
--        if (!g_file_get_contents(image_name, &contents, &length, NULL)) {
--            error_report("failed to load \"%s\"", image_name);
--            exit(1);
--        }
--        size = length;
--        data = (uint8_t *)contents;
--    }
--
--    fw_cfg_add_i32(fw_cfg, size_key, size);
--    fw_cfg_add_bytes(fw_cfg, data_key, data, size);
--}
--
- static void fw_cfg_add_kernel_info(FWCfgState *fw_cfg)
- {
-     /*
-diff --git a/hw/nvram/fw_cfg.c b/hw/nvram/fw_cfg.c
-index XXXXXXX..XXXXXXX 100644
---- a/hw/nvram/fw_cfg.c
-+++ b/hw/nvram/fw_cfg.c
-@@ -XXX,XX +XXX,XX @@
- #include "qapi/error.h"
- #include "hw/acpi/aml-build.h"
- #include "hw/pci/pci_bus.h"
-+#include "hw/loader.h"
- #define FW_CFG_FILE_SLOTS_DFLT 0x20
-@@ -XXX,XX +XXX,XX @@ FWCfgState *fw_cfg_find(void)
-     return FW_CFG(object_resolve_path_type("", TYPE_FW_CFG, NULL));
- }
-+void load_image_to_fw_cfg(FWCfgState *fw_cfg, uint16_t size_key,
-+                          uint16_t data_key, const char *image_name,
-+                          bool try_decompress)
-+{
-+    size_t size = -1;
-+    uint8_t *data;
-+
-+    if (image_name == NULL) {
-+        return;
-+    }
-+
-+    if (try_decompress) {
-+        size = load_image_gzipped_buffer(image_name,
-+                                         LOAD_IMAGE_MAX_GUNZIP_BYTES, &data);
-+    }
-+
-+    if (size == (size_t)-1) {
-+        gchar *contents;
-+        gsize length;
-+
-+        if (!g_file_get_contents(image_name, &contents, &length, NULL)) {
-+            error_report("failed to load \"%s\"", image_name);
-+            exit(1);
-+        }
-+        size = length;
-+        data = (uint8_t *)contents;
-+    }
-+
-+    fw_cfg_add_i32(fw_cfg, size_key, size);
-+    fw_cfg_add_bytes(fw_cfg, data_key, data, size);
-+}
- static void fw_cfg_class_init(ObjectClass *klass, void *data)
- {
 --
-.37.3
+.41.0

-New patch
+[PULL v2 13/45] target/riscv: Refactor translation of vector-widening instruction
+From: Dickon Hood <dickon.hood@codethink.co.uk>
+Zvbb (implemented in later commit) has a widening instruction, which
+requires an extra check on the enabled extensions.  Refactor
+GEN_OPIVX_WIDEN_TRANS() to take a check function to avoid reimplementing
+it.
+Signed-off-by: Dickon Hood <dickon.hood@codethink.co.uk>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Message-ID: <20230711165917.2629866-7-max.chou@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/insn_trans/trans_rvv.c.inc | 52 +++++++++++--------------
+file changed, 23 insertions(+), 29 deletions(-)
+diff --git a/target/riscv/insn_trans/trans_rvv.c.inc b/target/riscv/insn_trans/trans_rvv.c.inc
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/insn_trans/trans_rvv.c.inc
++++ b/target/riscv/insn_trans/trans_rvv.c.inc
+@@ -XXX,XX +XXX,XX @@ static bool opivx_widen_check(DisasContext *s, arg_rmrr *a)
+            vext_check_ds(s, a->rd, a->rs2, a->vm);
+ }
+-static bool do_opivx_widen(DisasContext *s, arg_rmrr *a,
+-                           gen_helper_opivx *fn)
+-{
+-    if (opivx_widen_check(s, a)) {
+-        return opivx_trans(a->rd, a->rs1, a->rs2, a->vm, fn, s);
+-    }
+-    return false;
+-}
+-
+-#define GEN_OPIVX_WIDEN_TRANS(NAME) \
+-static bool trans_##NAME(DisasContext *s, arg_rmrr *a)       \
+-{                                                            \
+-    static gen_helper_opivx * const fns[3] = {               \
+-        gen_helper_##NAME##_b,                               \
+-        gen_helper_##NAME##_h,                               \
+-        gen_helper_##NAME##_w                                \
+-    };                                                       \
+-    return do_opivx_widen(s, a, fns[s->sew]);                \
++#define GEN_OPIVX_WIDEN_TRANS(NAME, CHECK) \
++static bool trans_##NAME(DisasContext *s, arg_rmrr *a)                    \
++{                                                                         \
++    if (CHECK(s, a)) {                                                    \
++        static gen_helper_opivx * const fns[3] = {                        \
++            gen_helper_##NAME##_b,                                        \
++            gen_helper_##NAME##_h,                                        \
++            gen_helper_##NAME##_w                                         \
++        };                                                                \
++        return opivx_trans(a->rd, a->rs1, a->rs2, a->vm, fns[s->sew], s); \
++    }                                                                     \
++    return false;                                                         \
+ }
+-GEN_OPIVX_WIDEN_TRANS(vwaddu_vx)
+-GEN_OPIVX_WIDEN_TRANS(vwadd_vx)
+-GEN_OPIVX_WIDEN_TRANS(vwsubu_vx)
+-GEN_OPIVX_WIDEN_TRANS(vwsub_vx)
++GEN_OPIVX_WIDEN_TRANS(vwaddu_vx, opivx_widen_check)
++GEN_OPIVX_WIDEN_TRANS(vwadd_vx, opivx_widen_check)
++GEN_OPIVX_WIDEN_TRANS(vwsubu_vx, opivx_widen_check)
++GEN_OPIVX_WIDEN_TRANS(vwsub_vx, opivx_widen_check)
+ /* WIDEN OPIVV with WIDEN */
+ static bool opiwv_widen_check(DisasContext *s, arg_rmrr *a)
+@@ -XXX,XX +XXX,XX @@ GEN_OPIVX_TRANS(vrem_vx, opivx_check)
+ GEN_OPIVV_WIDEN_TRANS(vwmul_vv, opivv_widen_check)
+ GEN_OPIVV_WIDEN_TRANS(vwmulu_vv, opivv_widen_check)
+ GEN_OPIVV_WIDEN_TRANS(vwmulsu_vv, opivv_widen_check)
+-GEN_OPIVX_WIDEN_TRANS(vwmul_vx)
+-GEN_OPIVX_WIDEN_TRANS(vwmulu_vx)
+-GEN_OPIVX_WIDEN_TRANS(vwmulsu_vx)
++GEN_OPIVX_WIDEN_TRANS(vwmul_vx, opivx_widen_check)
++GEN_OPIVX_WIDEN_TRANS(vwmulu_vx, opivx_widen_check)
++GEN_OPIVX_WIDEN_TRANS(vwmulsu_vx, opivx_widen_check)
+ /* Vector Single-Width Integer Multiply-Add Instructions */
+ GEN_OPIVV_TRANS(vmacc_vv, opivv_check)
+@@ -XXX,XX +XXX,XX @@ GEN_OPIVX_TRANS(vnmsub_vx, opivx_check)
+ GEN_OPIVV_WIDEN_TRANS(vwmaccu_vv, opivv_widen_check)
+ GEN_OPIVV_WIDEN_TRANS(vwmacc_vv, opivv_widen_check)
+ GEN_OPIVV_WIDEN_TRANS(vwmaccsu_vv, opivv_widen_check)
+-GEN_OPIVX_WIDEN_TRANS(vwmaccu_vx)
+-GEN_OPIVX_WIDEN_TRANS(vwmacc_vx)
+-GEN_OPIVX_WIDEN_TRANS(vwmaccsu_vx)
+-GEN_OPIVX_WIDEN_TRANS(vwmaccus_vx)
++GEN_OPIVX_WIDEN_TRANS(vwmaccu_vx, opivx_widen_check)
++GEN_OPIVX_WIDEN_TRANS(vwmacc_vx, opivx_widen_check)
++GEN_OPIVX_WIDEN_TRANS(vwmaccsu_vx, opivx_widen_check)
++GEN_OPIVX_WIDEN_TRANS(vwmaccus_vx, opivx_widen_check)
+ /* Vector Integer Merge and Move Instructions */
+ static bool trans_vmv_v_v(DisasContext *s, arg_vmv_v_v *a)
+--
+.41.0

-New patch
+[PULL v2 14/45] target/riscv: Refactor some of the generic vector functionality
+From: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
+Move some macros out of `vector_helper` and into `vector_internals`.
+This ensures they can be used by both vector and vector-crypto helpers
+(latter implemented in proceeding commits).
+Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
+Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Message-ID: <20230711165917.2629866-8-max.chou@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/vector_internals.h | 46 +++++++++++++++++++++++++++++++++
+ target/riscv/vector_helper.c    | 42 ------------------------------
+files changed, 46 insertions(+), 42 deletions(-)
+diff --git a/target/riscv/vector_internals.h b/target/riscv/vector_internals.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/vector_internals.h
++++ b/target/riscv/vector_internals.h
+@@ -XXX,XX +XXX,XX @@ void vext_set_elems_1s(void *base, uint32_t is_agnostic, uint32_t cnt,
+ /* expand macro args before macro */
+ #define RVVCALL(macro, ...)  macro(__VA_ARGS__)
++/* (TD, T2, TX2) */
++#define OP_UU_B uint8_t, uint8_t, uint8_t
++#define OP_UU_H uint16_t, uint16_t, uint16_t
++#define OP_UU_W uint32_t, uint32_t, uint32_t
++#define OP_UU_D uint64_t, uint64_t, uint64_t
++
+ /* (TD, T1, T2, TX1, TX2) */
+ #define OP_UUU_B uint8_t, uint8_t, uint8_t, uint8_t, uint8_t
+ #define OP_UUU_H uint16_t, uint16_t, uint16_t, uint16_t, uint16_t
+ #define OP_UUU_W uint32_t, uint32_t, uint32_t, uint32_t, uint32_t
+ #define OP_UUU_D uint64_t, uint64_t, uint64_t, uint64_t, uint64_t
++#define OPIVV1(NAME, TD, T2, TX2, HD, HS2, OP)         \
++static void do_##NAME(void *vd, void *vs2, int i)      \
++{                                                      \
++    TX2 s2 = *((T2 *)vs2 + HS2(i));                    \
++    *((TD *)vd + HD(i)) = OP(s2);                      \
++}
++
++#define GEN_VEXT_V(NAME, ESZ)                          \
++void HELPER(NAME)(void *vd, void *v0, void *vs2,       \
++                  CPURISCVState *env, uint32_t desc)   \
++{                                                      \
++    uint32_t vm = vext_vm(desc);                       \
++    uint32_t vl = env->vl;                             \
++    uint32_t total_elems =                             \
++        vext_get_total_elems(env, desc, ESZ);          \
++    uint32_t vta = vext_vta(desc);                     \
++    uint32_t vma = vext_vma(desc);                     \
++    uint32_t i;                                        \
++                                                       \
++    for (i = env->vstart; i < vl; i++) {               \
++        if (!vm && !vext_elem_mask(v0, i)) {           \
++            /* set masked-off elements to 1s */        \
++            vext_set_elems_1s(vd, vma, i * ESZ,        \
++                              (i + 1) * ESZ);          \
++            continue;                                  \
++        }                                              \
++        do_##NAME(vd, vs2, i);                         \
++    }                                                  \
++    env->vstart = 0;                                   \
++    /* set tail elements to 1s */                      \
++    vext_set_elems_1s(vd, vta, vl * ESZ,               \
++                      total_elems * ESZ);              \
++}
++
+ /* operation of two vector elements */
+ typedef void opivv2_fn(void *vd, void *vs1, void *vs2, int i);
+@@ -XXX,XX +XXX,XX @@ void HELPER(NAME)(void *vd, void *v0, target_ulong s1,    \
+                do_##NAME, ESZ);                           \
+ }
++/* Three of the widening shortening macros: */
++/* (TD, T1, T2, TX1, TX2) */
++#define WOP_UUU_B uint16_t, uint8_t, uint8_t, uint16_t, uint16_t
++#define WOP_UUU_H uint32_t, uint16_t, uint16_t, uint32_t, uint32_t
++#define WOP_UUU_W uint64_t, uint32_t, uint32_t, uint64_t, uint64_t
++
+ #endif /* TARGET_RISCV_VECTOR_INTERNALS_H */
+diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/vector_helper.c
++++ b/target/riscv/vector_helper.c
+@@ -XXX,XX +XXX,XX @@ GEN_VEXT_ST_WHOLE(vs8r_v, int8_t, ste_b)
+ #define OP_SUS_H int16_t, uint16_t, int16_t, uint16_t, int16_t
+ #define OP_SUS_W int32_t, uint32_t, int32_t, uint32_t, int32_t
+ #define OP_SUS_D int64_t, uint64_t, int64_t, uint64_t, int64_t
+-#define WOP_UUU_B uint16_t, uint8_t, uint8_t, uint16_t, uint16_t
+-#define WOP_UUU_H uint32_t, uint16_t, uint16_t, uint32_t, uint32_t
+-#define WOP_UUU_W uint64_t, uint32_t, uint32_t, uint64_t, uint64_t
+ #define WOP_SSS_B int16_t, int8_t, int8_t, int16_t, int16_t
+ #define WOP_SSS_H int32_t, int16_t, int16_t, int32_t, int32_t
+ #define WOP_SSS_W int64_t, int32_t, int32_t, int64_t, int64_t
+@@ -XXX,XX +XXX,XX @@ GEN_VEXT_VF(vfwnmsac_vf_h, 4)
+ GEN_VEXT_VF(vfwnmsac_vf_w, 8)
+ /* Vector Floating-Point Square-Root Instruction */
+-/* (TD, T2, TX2) */
+-#define OP_UU_H uint16_t, uint16_t, uint16_t
+-#define OP_UU_W uint32_t, uint32_t, uint32_t
+-#define OP_UU_D uint64_t, uint64_t, uint64_t
+-
+ #define OPFVV1(NAME, TD, T2, TX2, HD, HS2, OP)         \
+ static void do_##NAME(void *vd, void *vs2, int i,      \
+                       CPURISCVState *env)              \
+@@ -XXX,XX +XXX,XX @@ GEN_VEXT_CMP_VF(vmfge_vf_w, uint32_t, H4, vmfge32)
+ GEN_VEXT_CMP_VF(vmfge_vf_d, uint64_t, H8, vmfge64)
+ /* Vector Floating-Point Classify Instruction */
+-#define OPIVV1(NAME, TD, T2, TX2, HD, HS2, OP)         \
+-static void do_##NAME(void *vd, void *vs2, int i)      \
+-{                                                      \
+-    TX2 s2 = *((T2 *)vs2 + HS2(i));                    \
+-    *((TD *)vd + HD(i)) = OP(s2);                      \
+-}
+-
+-#define GEN_VEXT_V(NAME, ESZ)                          \
+-void HELPER(NAME)(void *vd, void *v0, void *vs2,       \
+-                  CPURISCVState *env, uint32_t desc)   \
+-{                                                      \
+-    uint32_t vm = vext_vm(desc);                       \
+-    uint32_t vl = env->vl;                             \
+-    uint32_t total_elems =                             \
+-        vext_get_total_elems(env, desc, ESZ);          \
+-    uint32_t vta = vext_vta(desc);                     \
+-    uint32_t vma = vext_vma(desc);                     \
+-    uint32_t i;                                        \
+-                                                       \
+-    for (i = env->vstart; i < vl; i++) {               \
+-        if (!vm && !vext_elem_mask(v0, i)) {           \
+-            /* set masked-off elements to 1s */        \
+-            vext_set_elems_1s(vd, vma, i * ESZ,        \
+-                              (i + 1) * ESZ);          \
+-            continue;                                  \
+-        }                                              \
+-        do_##NAME(vd, vs2, i);                         \
+-    }                                                  \
+-    env->vstart = 0;                                   \
+-    /* set tail elements to 1s */                      \
+-    vext_set_elems_1s(vd, vta, vl * ESZ,               \
+-                      total_elems * ESZ);              \
+-}
+-
+ target_ulong fclass_h(uint64_t frs1)
+ {
+     float16 f = frs1;
+--
+.41.0

-[PULL 07/10] disas/riscv.c: rvv: Add disas support for vector instructions
+[PULL v2 15/45] target/riscv: Add Zvbb ISA extension support
-From: Yang Liu <liuyang22@iscas.ac.cn>
+From: Dickon Hood <dickon.hood@codethink.co.uk>
-Tested with https://github.com/ksco/rvv-decoder-tests
+This commit adds support for the Zvbb vector-crypto extension, which
 consists of the following instructions:
-Expected checkpatch errors for consistency and brevity reasons:
+* vrol.[vv,vx]
 * vror.[vv,vx,vi]
 * vbrev8.v
 * vrev8.v
 * vandn.[vv,vx]
 * vbrev.v
 * vclz.v
 * vctz.v
 * vcpop.v
 * vwsll.[vv,vx,vi]
-ERROR: line over 90 characters
+Translation functions are defined in
-ERROR: trailing statements should be on next line
+`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
-ERROR: braces {} are necessary for all arms of this statement
+`target/riscv/vcrypto_helper.c`.
-Signed-off-by: Yang Liu <liuyang22@iscas.ac.cn>
+Co-authored-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
-Acked-by: Alistair Francis <alistair.francis@wdc.com>
+Co-authored-by: William Salmon <will.salmon@codethink.co.uk>
-Message-Id: <20220928051842.16207-1-liuyang22@iscas.ac.cn>
+Co-authored-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
 [max.chou@sifive.com: Fix imm mode of vror.vi]
 Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
 Signed-off-by: William Salmon <will.salmon@codethink.co.uk>
 Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
 Signed-off-by: Dickon Hood <dickon.hood@codethink.co.uk>
 Signed-off-by: Max Chou <max.chou@sifive.com>
 Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
 [max.chou@sifive.com: Exposed x-zvbb property]
 Message-ID: <20230711165917.2629866-9-max.chou@sifive.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
- disas/riscv.c | 1432 ++++++++++++++++++++++++++++++++++++++++++++++++-
+ target/riscv/cpu_cfg.h                   |   1 +
-file changed, 1430 insertions(+), 2 deletions(-)
+ target/riscv/helper.h                    |  62 +++++++++
  target/riscv/insn32.decode               |  20 +++
  target/riscv/cpu.c                       |  12 ++
  target/riscv/vcrypto_helper.c            | 138 +++++++++++++++++++
  target/riscv/insn_trans/trans_rvvk.c.inc | 164 +++++++++++++++++++++++
 files changed, 397 insertions(+)
-diff --git a/disas/riscv.c b/disas/riscv.c
+diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
 index XXXXXXX..XXXXXXX 100644
---- a/disas/riscv.c
+--- a/target/riscv/cpu_cfg.h
-+++ b/disas/riscv.c
++++ b/target/riscv/cpu_cfg.h
-@@ -XXX,XX +XXX,XX @@ typedef enum {
+@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
-     rv_codec_css_sqsp,
+     bool ext_zve32f;
-     rv_codec_k_bs,
+     bool ext_zve64f;
-     rv_codec_k_rnum,
+     bool ext_zve64d;
-+    rv_codec_v_r,
++    bool ext_zvbb;
-+    rv_codec_v_ldst,
+     bool ext_zvbc;
-+    rv_codec_v_i,
+     bool ext_zmmul;
-+    rv_codec_vsetvli,
+     bool ext_zvfbfmin;
-+    rv_codec_vsetivli,
+diff --git a/target/riscv/helper.h b/target/riscv/helper.h
- } rv_codec;
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/helper.h
- typedef enum {
++++ b/target/riscv/helper.h
-@@ -XXX,XX +XXX,XX @@ typedef enum {
+@@ -XXX,XX +XXX,XX @@ DEF_HELPER_6(vclmul_vv, void, ptr, ptr, ptr, ptr, env, i32)
-     rv_op_zip = 396,
+ DEF_HELPER_6(vclmul_vx, void, ptr, ptr, tl, ptr, env, i32)
-     rv_op_xperm4 = 397,
+ DEF_HELPER_6(vclmulh_vv, void, ptr, ptr, ptr, ptr, env, i32)
-     rv_op_xperm8 = 398,
+ DEF_HELPER_6(vclmulh_vx, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vle8_v = 399,
++
-+    rv_op_vle16_v = 400,
++DEF_HELPER_6(vror_vv_b, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vle32_v = 401,
++DEF_HELPER_6(vror_vv_h, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vle64_v = 402,
++DEF_HELPER_6(vror_vv_w, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vse8_v = 403,
++DEF_HELPER_6(vror_vv_d, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vse16_v = 404,
++
-+    rv_op_vse32_v = 405,
++DEF_HELPER_6(vror_vx_b, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vse64_v = 406,
++DEF_HELPER_6(vror_vx_h, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vlm_v = 407,
++DEF_HELPER_6(vror_vx_w, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vsm_v = 408,
++DEF_HELPER_6(vror_vx_d, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vlse8_v = 409,
++
-+    rv_op_vlse16_v = 410,
++DEF_HELPER_6(vrol_vv_b, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vlse32_v = 411,
++DEF_HELPER_6(vrol_vv_h, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vlse64_v = 412,
++DEF_HELPER_6(vrol_vv_w, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vsse8_v = 413,
++DEF_HELPER_6(vrol_vv_d, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vsse16_v = 414,
++
-+    rv_op_vsse32_v = 415,
++DEF_HELPER_6(vrol_vx_b, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vsse64_v = 416,
++DEF_HELPER_6(vrol_vx_h, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vluxei8_v = 417,
++DEF_HELPER_6(vrol_vx_w, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vluxei16_v = 418,
++DEF_HELPER_6(vrol_vx_d, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vluxei32_v = 419,
++
-+    rv_op_vluxei64_v = 420,
++DEF_HELPER_5(vrev8_v_b, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vloxei8_v = 421,
++DEF_HELPER_5(vrev8_v_h, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vloxei16_v = 422,
++DEF_HELPER_5(vrev8_v_w, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vloxei32_v = 423,
++DEF_HELPER_5(vrev8_v_d, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vloxei64_v = 424,
++DEF_HELPER_5(vbrev8_v_b, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vsuxei8_v = 425,
++DEF_HELPER_5(vbrev8_v_h, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vsuxei16_v = 426,
++DEF_HELPER_5(vbrev8_v_w, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vsuxei32_v = 427,
++DEF_HELPER_5(vbrev8_v_d, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vsuxei64_v = 428,
++DEF_HELPER_5(vbrev_v_b, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vsoxei8_v = 429,
++DEF_HELPER_5(vbrev_v_h, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vsoxei16_v = 430,
++DEF_HELPER_5(vbrev_v_w, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vsoxei32_v = 431,
++DEF_HELPER_5(vbrev_v_d, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vsoxei64_v = 432,
++
-+    rv_op_vle8ff_v = 433,
++DEF_HELPER_5(vclz_v_b, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vle16ff_v = 434,
++DEF_HELPER_5(vclz_v_h, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vle32ff_v = 435,
++DEF_HELPER_5(vclz_v_w, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vle64ff_v = 436,
++DEF_HELPER_5(vclz_v_d, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vl1re8_v = 437,
++DEF_HELPER_5(vctz_v_b, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vl1re16_v = 438,
++DEF_HELPER_5(vctz_v_h, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vl1re32_v = 439,
++DEF_HELPER_5(vctz_v_w, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vl1re64_v = 440,
++DEF_HELPER_5(vctz_v_d, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vl2re8_v = 441,
++DEF_HELPER_5(vcpop_v_b, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vl2re16_v = 442,
++DEF_HELPER_5(vcpop_v_h, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vl2re32_v = 443,
++DEF_HELPER_5(vcpop_v_w, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vl2re64_v = 444,
++DEF_HELPER_5(vcpop_v_d, void, ptr, ptr, ptr, env, i32)
-+    rv_op_vl4re8_v = 445,
++
-+    rv_op_vl4re16_v = 446,
++DEF_HELPER_6(vwsll_vv_b, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vl4re32_v = 447,
++DEF_HELPER_6(vwsll_vv_h, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vl4re64_v = 448,
++DEF_HELPER_6(vwsll_vv_w, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vl8re8_v = 449,
++DEF_HELPER_6(vwsll_vx_b, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vl8re16_v = 450,
++DEF_HELPER_6(vwsll_vx_h, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vl8re32_v = 451,
++DEF_HELPER_6(vwsll_vx_w, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vl8re64_v = 452,
++
-+    rv_op_vs1r_v = 453,
++DEF_HELPER_6(vandn_vv_b, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vs2r_v = 454,
++DEF_HELPER_6(vandn_vv_h, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vs4r_v = 455,
++DEF_HELPER_6(vandn_vv_w, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vs8r_v = 456,
++DEF_HELPER_6(vandn_vv_d, void, ptr, ptr, ptr, ptr, env, i32)
-+    rv_op_vadd_vv = 457,
++DEF_HELPER_6(vandn_vx_b, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vadd_vx = 458,
++DEF_HELPER_6(vandn_vx_h, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vadd_vi = 459,
++DEF_HELPER_6(vandn_vx_w, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vsub_vv = 460,
++DEF_HELPER_6(vandn_vx_d, void, ptr, ptr, tl, ptr, env, i32)
-+    rv_op_vsub_vx = 461,
+diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
-+    rv_op_vrsub_vx = 462,
+index XXXXXXX..XXXXXXX 100644
-+    rv_op_vrsub_vi = 463,
+--- a/target/riscv/insn32.decode
-+    rv_op_vwaddu_vv = 464,
++++ b/target/riscv/insn32.decode
-+    rv_op_vwaddu_vx = 465,
+@@ -XXX,XX +XXX,XX @@
-+    rv_op_vwadd_vv = 466,
+ %imm_u    12:s20                 !function=ex_shift_12
-+    rv_op_vwadd_vx = 467,
+ %imm_bs   30:2                   !function=ex_shift_3
-+    rv_op_vwsubu_vv = 468,
+ %imm_rnum 20:4
-+    rv_op_vwsubu_vx = 469,
++%imm_z6   26:1 15:5
-+    rv_op_vwsub_vv = 470,
-+    rv_op_vwsub_vx = 471,
+ # Argument sets:
-+    rv_op_vwaddu_wv = 472,
+ &empty
-+    rv_op_vwaddu_wx = 473,
+@@ -XXX,XX +XXX,XX @@
-+    rv_op_vwadd_wv = 474,
+ @r_vm    ...... vm:1 ..... ..... ... ..... ....... &rmrr %rs2 %rs1 %rd
-+    rv_op_vwadd_wx = 475,
+ @r_vm_1  ...... . ..... ..... ... ..... .......    &rmrr vm=1 %rs2 %rs1 %rd
-+    rv_op_vwsubu_wv = 476,
+ @r_vm_0  ...... . ..... ..... ... ..... .......    &rmrr vm=0 %rs2 %rs1 %rd
-+    rv_op_vwsubu_wx = 477,
++@r2_zimm6  ..... . vm:1 ..... ..... ... ..... .......  &rmrr %rs2 rs1=%imm_z6 %rd
-+    rv_op_vwsub_wv = 478,
+ @r2_zimm11 . zimm:11  ..... ... ..... ....... %rs1 %rd
-+    rv_op_vwsub_wx = 479,
+ @r2_zimm10 .. zimm:10  ..... ... ..... ....... %rs1 %rd
-+    rv_op_vadc_vvm = 480,
+ @r2_s    .......   ..... ..... ... ..... ....... %rs2 %rs1
-+    rv_op_vadc_vxm = 481,
+@@ -XXX,XX +XXX,XX @@ vclmul_vv   001100 . ..... ..... 010 ..... 1010111 @r_vm
-+    rv_op_vadc_vim = 482,
+ vclmul_vx   001100 . ..... ..... 110 ..... 1010111 @r_vm
-+    rv_op_vmadc_vvm = 483,
+ vclmulh_vv  001101 . ..... ..... 010 ..... 1010111 @r_vm
-+    rv_op_vmadc_vxm = 484,
+ vclmulh_vx  001101 . ..... ..... 110 ..... 1010111 @r_vm
-+    rv_op_vmadc_vim = 485,
++
-+    rv_op_vsbc_vvm = 486,
++# *** Zvbb vector crypto extension ***
-+    rv_op_vsbc_vxm = 487,
++vrol_vv     010101 . ..... ..... 000 ..... 1010111 @r_vm
-+    rv_op_vmsbc_vvm = 488,
++vrol_vx     010101 . ..... ..... 100 ..... 1010111 @r_vm
-+    rv_op_vmsbc_vxm = 489,
++vror_vv     010100 . ..... ..... 000 ..... 1010111 @r_vm
-+    rv_op_vand_vv = 490,
++vror_vx     010100 . ..... ..... 100 ..... 1010111 @r_vm
-+    rv_op_vand_vx = 491,
++vror_vi     01010. . ..... ..... 011 ..... 1010111 @r2_zimm6
-+    rv_op_vand_vi = 492,
++vbrev8_v    010010 . ..... 01000 010 ..... 1010111 @r2_vm
-+    rv_op_vor_vv = 493,
++vrev8_v     010010 . ..... 01001 010 ..... 1010111 @r2_vm
-+    rv_op_vor_vx = 494,
++vandn_vv    000001 . ..... ..... 000 ..... 1010111 @r_vm
-+    rv_op_vor_vi = 495,
++vandn_vx    000001 . ..... ..... 100 ..... 1010111 @r_vm
-+    rv_op_vxor_vv = 496,
++vbrev_v     010010 . ..... 01010 010 ..... 1010111 @r2_vm
-+    rv_op_vxor_vx = 497,
++vclz_v      010010 . ..... 01100 010 ..... 1010111 @r2_vm
-+    rv_op_vxor_vi = 498,
++vctz_v      010010 . ..... 01101 010 ..... 1010111 @r2_vm
-+    rv_op_vsll_vv = 499,
++vcpop_v     010010 . ..... 01110 010 ..... 1010111 @r2_vm
-+    rv_op_vsll_vx = 500,
++vwsll_vv    110101 . ..... ..... 000 ..... 1010111 @r_vm
-+    rv_op_vsll_vi = 501,
++vwsll_vx    110101 . ..... ..... 100 ..... 1010111 @r_vm
-+    rv_op_vsrl_vv = 502,
++vwsll_vi    110101 . ..... ..... 011 ..... 1010111 @r_vm
-+    rv_op_vsrl_vx = 503,
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
-+    rv_op_vsrl_vi = 504,
+index XXXXXXX..XXXXXXX 100644
-+    rv_op_vsra_vv = 505,
+--- a/target/riscv/cpu.c
-+    rv_op_vsra_vx = 506,
++++ b/target/riscv/cpu.c
-+    rv_op_vsra_vi = 507,
+@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
-+    rv_op_vnsrl_wv = 508,
+     ISA_EXT_DATA_ENTRY(zksed, PRIV_VERSION_1_12_0, ext_zksed),
-+    rv_op_vnsrl_wx = 509,
+     ISA_EXT_DATA_ENTRY(zksh, PRIV_VERSION_1_12_0, ext_zksh),
-+    rv_op_vnsrl_wi = 510,
+     ISA_EXT_DATA_ENTRY(zkt, PRIV_VERSION_1_12_0, ext_zkt),
-+    rv_op_vnsra_wv = 511,
++    ISA_EXT_DATA_ENTRY(zvbb, PRIV_VERSION_1_12_0, ext_zvbb),
-+    rv_op_vnsra_wx = 512,
+     ISA_EXT_DATA_ENTRY(zvbc, PRIV_VERSION_1_12_0, ext_zvbc),
-+    rv_op_vnsra_wi = 513,
+     ISA_EXT_DATA_ENTRY(zve32f, PRIV_VERSION_1_10_0, ext_zve32f),
-+    rv_op_vmseq_vv = 514,
+     ISA_EXT_DATA_ENTRY(zve64f, PRIV_VERSION_1_10_0, ext_zve64f),
-+    rv_op_vmseq_vx = 515,
+@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
-+    rv_op_vmseq_vi = 516,
+         return;
-+    rv_op_vmsne_vv = 517,
+     }
-+    rv_op_vmsne_vx = 518,
-+    rv_op_vmsne_vi = 519,
++    /*
-+    rv_op_vmsltu_vv = 520,
++     * In principle Zve*x would also suffice here, were they supported
-+    rv_op_vmsltu_vx = 521,
++     * in qemu
-+    rv_op_vmslt_vv = 522,
++     */
-+    rv_op_vmslt_vx = 523,
++    if (cpu->cfg.ext_zvbb && !cpu->cfg.ext_zve32f) {
-+    rv_op_vmsleu_vv = 524,
++        error_setg(errp,
-+    rv_op_vmsleu_vx = 525,
++                   "Vector crypto extensions require V or Zve* extensions");
-+    rv_op_vmsleu_vi = 526,
++        return;
-+    rv_op_vmsle_vv = 527,
++    }
-+    rv_op_vmsle_vx = 528,
++
-+    rv_op_vmsle_vi = 529,
+     if (cpu->cfg.ext_zvbc && !cpu->cfg.ext_zve64f) {
-+    rv_op_vmsgtu_vx = 530,
+         error_setg(errp, "Zvbc extension requires V or Zve64{f,d} extensions");
-+    rv_op_vmsgtu_vi = 531,
+         return;
-+    rv_op_vmsgt_vx = 532,
+@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
-+    rv_op_vmsgt_vi = 533,
+     DEFINE_PROP_BOOL("x-zvfbfwma", RISCVCPU, cfg.ext_zvfbfwma, false),
-+    rv_op_vminu_vv = 534,
-+    rv_op_vminu_vx = 535,
+     /* Vector cryptography extensions */
-+    rv_op_vmin_vv = 536,
++    DEFINE_PROP_BOOL("x-zvbb", RISCVCPU, cfg.ext_zvbb, false),
-+    rv_op_vmin_vx = 537,
+     DEFINE_PROP_BOOL("x-zvbc", RISCVCPU, cfg.ext_zvbc, false),
-+    rv_op_vmaxu_vv = 538,
-+    rv_op_vmaxu_vx = 539,
+     DEFINE_PROP_END_OF_LIST(),
-+    rv_op_vmax_vv = 540,
+diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
-+    rv_op_vmax_vx = 541,
+index XXXXXXX..XXXXXXX 100644
-+    rv_op_vmul_vv = 542,
+--- a/target/riscv/vcrypto_helper.c
-+    rv_op_vmul_vx = 543,
++++ b/target/riscv/vcrypto_helper.c
-+    rv_op_vmulh_vv = 544,
+@@ -XXX,XX +XXX,XX @@
-+    rv_op_vmulh_vx = 545,
+ #include "qemu/osdep.h"
-+    rv_op_vmulhu_vv = 546,
+ #include "qemu/host-utils.h"
-+    rv_op_vmulhu_vx = 547,
+ #include "qemu/bitops.h"
-+    rv_op_vmulhsu_vv = 548,
++#include "qemu/bswap.h"
-+    rv_op_vmulhsu_vx = 549,
+ #include "cpu.h"
-+    rv_op_vdivu_vv = 550,
+ #include "exec/memop.h"
-+    rv_op_vdivu_vx = 551,
+ #include "exec/exec-all.h"
-+    rv_op_vdiv_vv = 552,
+@@ -XXX,XX +XXX,XX @@ RVVCALL(OPIVV2, vclmulh_vv, OP_UUU_D, H8, H8, H8, clmulh64)
-+    rv_op_vdiv_vx = 553,
+ GEN_VEXT_VV(vclmulh_vv, 8)
-+    rv_op_vremu_vv = 554,
+ RVVCALL(OPIVX2, vclmulh_vx, OP_UUU_D, H8, H8, clmulh64)
-+    rv_op_vremu_vx = 555,
+ GEN_VEXT_VX(vclmulh_vx, 8)
-+    rv_op_vrem_vv = 556,
++
-+    rv_op_vrem_vx = 557,
++RVVCALL(OPIVV2, vror_vv_b, OP_UUU_B, H1, H1, H1, ror8)
-+    rv_op_vwmulu_vv = 558,
++RVVCALL(OPIVV2, vror_vv_h, OP_UUU_H, H2, H2, H2, ror16)
-+    rv_op_vwmulu_vx = 559,
++RVVCALL(OPIVV2, vror_vv_w, OP_UUU_W, H4, H4, H4, ror32)
-+    rv_op_vwmulsu_vv = 560,
++RVVCALL(OPIVV2, vror_vv_d, OP_UUU_D, H8, H8, H8, ror64)
-+    rv_op_vwmulsu_vx = 561,
++GEN_VEXT_VV(vror_vv_b, 1)
-+    rv_op_vwmul_vv = 562,
++GEN_VEXT_VV(vror_vv_h, 2)
-+    rv_op_vwmul_vx = 563,
++GEN_VEXT_VV(vror_vv_w, 4)
-+    rv_op_vmacc_vv = 564,
++GEN_VEXT_VV(vror_vv_d, 8)
-+    rv_op_vmacc_vx = 565,
++
-+    rv_op_vnmsac_vv = 566,
++RVVCALL(OPIVX2, vror_vx_b, OP_UUU_B, H1, H1, ror8)
-+    rv_op_vnmsac_vx = 567,
++RVVCALL(OPIVX2, vror_vx_h, OP_UUU_H, H2, H2, ror16)
-+    rv_op_vmadd_vv = 568,
++RVVCALL(OPIVX2, vror_vx_w, OP_UUU_W, H4, H4, ror32)
-+    rv_op_vmadd_vx = 569,
++RVVCALL(OPIVX2, vror_vx_d, OP_UUU_D, H8, H8, ror64)
-+    rv_op_vnmsub_vv = 570,
++GEN_VEXT_VX(vror_vx_b, 1)
-+    rv_op_vnmsub_vx = 571,
++GEN_VEXT_VX(vror_vx_h, 2)
-+    rv_op_vwmaccu_vv = 572,
++GEN_VEXT_VX(vror_vx_w, 4)
-+    rv_op_vwmaccu_vx = 573,
++GEN_VEXT_VX(vror_vx_d, 8)
-+    rv_op_vwmacc_vv = 574,
++
-+    rv_op_vwmacc_vx = 575,
++RVVCALL(OPIVV2, vrol_vv_b, OP_UUU_B, H1, H1, H1, rol8)
-+    rv_op_vwmaccsu_vv = 576,
++RVVCALL(OPIVV2, vrol_vv_h, OP_UUU_H, H2, H2, H2, rol16)
-+    rv_op_vwmaccsu_vx = 577,
++RVVCALL(OPIVV2, vrol_vv_w, OP_UUU_W, H4, H4, H4, rol32)
-+    rv_op_vwmaccus_vx = 578,
++RVVCALL(OPIVV2, vrol_vv_d, OP_UUU_D, H8, H8, H8, rol64)
-+    rv_op_vmv_v_v = 579,
++GEN_VEXT_VV(vrol_vv_b, 1)
-+    rv_op_vmv_v_x = 580,
++GEN_VEXT_VV(vrol_vv_h, 2)
-+    rv_op_vmv_v_i = 581,
++GEN_VEXT_VV(vrol_vv_w, 4)
-+    rv_op_vmerge_vvm = 582,
++GEN_VEXT_VV(vrol_vv_d, 8)
-+    rv_op_vmerge_vxm = 583,
++
-+    rv_op_vmerge_vim = 584,
++RVVCALL(OPIVX2, vrol_vx_b, OP_UUU_B, H1, H1, rol8)
-+    rv_op_vsaddu_vv = 585,
++RVVCALL(OPIVX2, vrol_vx_h, OP_UUU_H, H2, H2, rol16)
-+    rv_op_vsaddu_vx = 586,
++RVVCALL(OPIVX2, vrol_vx_w, OP_UUU_W, H4, H4, rol32)
-+    rv_op_vsaddu_vi = 587,
++RVVCALL(OPIVX2, vrol_vx_d, OP_UUU_D, H8, H8, rol64)
-+    rv_op_vsadd_vv = 588,
++GEN_VEXT_VX(vrol_vx_b, 1)
-+    rv_op_vsadd_vx = 589,
++GEN_VEXT_VX(vrol_vx_h, 2)
-+    rv_op_vsadd_vi = 590,
++GEN_VEXT_VX(vrol_vx_w, 4)
-+    rv_op_vssubu_vv = 591,
++GEN_VEXT_VX(vrol_vx_d, 8)
-+    rv_op_vssubu_vx = 592,
++
-+    rv_op_vssub_vv = 593,
++static uint64_t brev8(uint64_t val)
 +    rv_op_vssub_vx = 594,
 +    rv_op_vaadd_vv = 595,
 +    rv_op_vaadd_vx = 596,
 +    rv_op_vaaddu_vv = 597,
 +    rv_op_vaaddu_vx = 598,
 +    rv_op_vasub_vv = 599,
 +    rv_op_vasub_vx = 600,
 +    rv_op_vasubu_vv = 601,
 +    rv_op_vasubu_vx = 602,
 +    rv_op_vsmul_vv = 603,
 +    rv_op_vsmul_vx = 604,
 +    rv_op_vssrl_vv = 605,
 +    rv_op_vssrl_vx = 606,
 +    rv_op_vssrl_vi = 607,
 +    rv_op_vssra_vv = 608,
 +    rv_op_vssra_vx = 609,
 +    rv_op_vssra_vi = 610,
 +    rv_op_vnclipu_wv = 611,
 +    rv_op_vnclipu_wx = 612,
 +    rv_op_vnclipu_wi = 613,
 +    rv_op_vnclip_wv = 614,
 +    rv_op_vnclip_wx = 615,
 +    rv_op_vnclip_wi = 616,
 +    rv_op_vfadd_vv = 617,
 +    rv_op_vfadd_vf = 618,
 +    rv_op_vfsub_vv = 619,
 +    rv_op_vfsub_vf = 620,
 +    rv_op_vfrsub_vf = 621,
 +    rv_op_vfwadd_vv = 622,
 +    rv_op_vfwadd_vf = 623,
 +    rv_op_vfwadd_wv = 624,
 +    rv_op_vfwadd_wf = 625,
 +    rv_op_vfwsub_vv = 626,
 +    rv_op_vfwsub_vf = 627,
 +    rv_op_vfwsub_wv = 628,
 +    rv_op_vfwsub_wf = 629,
 +    rv_op_vfmul_vv = 630,
 +    rv_op_vfmul_vf = 631,
 +    rv_op_vfdiv_vv = 632,
 +    rv_op_vfdiv_vf = 633,
 +    rv_op_vfrdiv_vf = 634,
 +    rv_op_vfwmul_vv = 635,
 +    rv_op_vfwmul_vf = 636,
 +    rv_op_vfmacc_vv = 637,
 +    rv_op_vfmacc_vf = 638,
 +    rv_op_vfnmacc_vv = 639,
 +    rv_op_vfnmacc_vf = 640,
 +    rv_op_vfmsac_vv = 641,
 +    rv_op_vfmsac_vf = 642,
 +    rv_op_vfnmsac_vv = 643,
 +    rv_op_vfnmsac_vf = 644,
 +    rv_op_vfmadd_vv = 645,
 +    rv_op_vfmadd_vf = 646,
 +    rv_op_vfnmadd_vv = 647,
 +    rv_op_vfnmadd_vf = 648,
 +    rv_op_vfmsub_vv = 649,
 +    rv_op_vfmsub_vf = 650,
 +    rv_op_vfnmsub_vv = 651,
 +    rv_op_vfnmsub_vf = 652,
 +    rv_op_vfwmacc_vv = 653,
 +    rv_op_vfwmacc_vf = 654,
 +    rv_op_vfwnmacc_vv = 655,
 +    rv_op_vfwnmacc_vf = 656,
 +    rv_op_vfwmsac_vv = 657,
 +    rv_op_vfwmsac_vf = 658,
 +    rv_op_vfwnmsac_vv = 659,
 +    rv_op_vfwnmsac_vf = 660,
 +    rv_op_vfsqrt_v = 661,
 +    rv_op_vfrsqrt7_v = 662,
 +    rv_op_vfrec7_v = 663,
 +    rv_op_vfmin_vv = 664,
 +    rv_op_vfmin_vf = 665,
 +    rv_op_vfmax_vv = 666,
 +    rv_op_vfmax_vf = 667,
 +    rv_op_vfsgnj_vv = 668,
 +    rv_op_vfsgnj_vf = 669,
 +    rv_op_vfsgnjn_vv = 670,
 +    rv_op_vfsgnjn_vf = 671,
 +    rv_op_vfsgnjx_vv = 672,
 +    rv_op_vfsgnjx_vf = 673,
 +    rv_op_vfslide1up_vf = 674,
 +    rv_op_vfslide1down_vf = 675,
 +    rv_op_vmfeq_vv = 676,
 +    rv_op_vmfeq_vf = 677,
 +    rv_op_vmfne_vv = 678,
 +    rv_op_vmfne_vf = 679,
 +    rv_op_vmflt_vv = 680,
 +    rv_op_vmflt_vf = 681,
 +    rv_op_vmfle_vv = 682,
 +    rv_op_vmfle_vf = 683,
 +    rv_op_vmfgt_vf = 684,
 +    rv_op_vmfge_vf = 685,
 +    rv_op_vfclass_v = 686,
 +    rv_op_vfmerge_vfm = 687,
 +    rv_op_vfmv_v_f = 688,
 +    rv_op_vfcvt_xu_f_v = 689,
 +    rv_op_vfcvt_x_f_v = 690,
 +    rv_op_vfcvt_f_xu_v = 691,
 +    rv_op_vfcvt_f_x_v = 692,
 +    rv_op_vfcvt_rtz_xu_f_v = 693,
 +    rv_op_vfcvt_rtz_x_f_v = 694,
 +    rv_op_vfwcvt_xu_f_v = 695,
 +    rv_op_vfwcvt_x_f_v = 696,
 +    rv_op_vfwcvt_f_xu_v = 697,
 +    rv_op_vfwcvt_f_x_v = 698,
 +    rv_op_vfwcvt_f_f_v = 699,
 +    rv_op_vfwcvt_rtz_xu_f_v = 700,
 +    rv_op_vfwcvt_rtz_x_f_v = 701,
 +    rv_op_vfncvt_xu_f_w = 702,
 +    rv_op_vfncvt_x_f_w = 703,
 +    rv_op_vfncvt_f_xu_w = 704,
 +    rv_op_vfncvt_f_x_w = 705,
 +    rv_op_vfncvt_f_f_w = 706,
 +    rv_op_vfncvt_rod_f_f_w = 707,
 +    rv_op_vfncvt_rtz_xu_f_w = 708,
 +    rv_op_vfncvt_rtz_x_f_w = 709,
 +    rv_op_vredsum_vs = 710,
 +    rv_op_vredand_vs = 711,
 +    rv_op_vredor_vs = 712,
 +    rv_op_vredxor_vs = 713,
 +    rv_op_vredminu_vs = 714,
 +    rv_op_vredmin_vs = 715,
 +    rv_op_vredmaxu_vs = 716,
 +    rv_op_vredmax_vs = 717,
 +    rv_op_vwredsumu_vs = 718,
 +    rv_op_vwredsum_vs = 719,
 +    rv_op_vfredusum_vs = 720,
 +    rv_op_vfredosum_vs = 721,
 +    rv_op_vfredmin_vs = 722,
 +    rv_op_vfredmax_vs = 723,
 +    rv_op_vfwredusum_vs = 724,
 +    rv_op_vfwredosum_vs = 725,
 +    rv_op_vmand_mm = 726,
 +    rv_op_vmnand_mm = 727,
 +    rv_op_vmandn_mm = 728,
 +    rv_op_vmxor_mm = 729,
 +    rv_op_vmor_mm = 730,
 +    rv_op_vmnor_mm = 731,
 +    rv_op_vmorn_mm = 732,
 +    rv_op_vmxnor_mm = 733,
 +    rv_op_vcpop_m = 734,
 +    rv_op_vfirst_m = 735,
 +    rv_op_vmsbf_m = 736,
 +    rv_op_vmsif_m = 737,
 +    rv_op_vmsof_m = 738,
 +    rv_op_viota_m = 739,
 +    rv_op_vid_v = 740,
 +    rv_op_vmv_x_s = 741,
 +    rv_op_vmv_s_x = 742,
 +    rv_op_vfmv_f_s = 743,
 +    rv_op_vfmv_s_f = 744,
 +    rv_op_vslideup_vx = 745,
 +    rv_op_vslideup_vi = 746,
 +    rv_op_vslide1up_vx = 747,
 +    rv_op_vslidedown_vx = 748,
 +    rv_op_vslidedown_vi = 749,
 +    rv_op_vslide1down_vx = 750,
 +    rv_op_vrgather_vv = 751,
 +    rv_op_vrgatherei16_vv = 752,
 +    rv_op_vrgather_vx = 753,
 +    rv_op_vrgather_vi = 754,
 +    rv_op_vcompress_vm = 755,
 +    rv_op_vmv1r_v = 756,
 +    rv_op_vmv2r_v = 757,
 +    rv_op_vmv4r_v = 758,
 +    rv_op_vmv8r_v = 759,
 +    rv_op_vzext_vf2 = 760,
 +    rv_op_vzext_vf4 = 761,
 +    rv_op_vzext_vf8 = 762,
 +    rv_op_vsext_vf2 = 763,
 +    rv_op_vsext_vf4 = 764,
 +    rv_op_vsext_vf8 = 765,
 +    rv_op_vsetvli = 766,
 +    rv_op_vsetivli = 767,
 +    rv_op_vsetvl = 768,
  } rv_op;
  /* structures */
@@ -XXX,XX +XXX,XX @@ typedef struct {
      uint8_t   rl;
      uint8_t   bs;
      uint8_t   rnum;
 +    uint8_t   vm;
 +    uint32_t  vzimm;
  } rv_decode;
  typedef struct {
@@ -XXX,XX +XXX,XX @@ static const char rv_freg_name_sym[32][5] = {
      "fs8",  "fs9",  "fs10", "fs11", "ft8",  "ft9",  "ft10", "ft11",
  };
 +static const char rv_vreg_name_sym[32][4] = {
 +    "v0",  "v1",  "v2",  "v3",  "v4",  "v5",  "v6",  "v7",
 +    "v8",  "v9",  "v10", "v11", "v12", "v13", "v14", "v15",
 +    "v16", "v17", "v18", "v19", "v20", "v21", "v22", "v23",
 +    "v24", "v25", "v26", "v27", "v28", "v29", "v30", "v31"
 +};
 +
  /* instruction formats */
  #define rv_fmt_none                   "O\t"
@@ -XXX,XX +XXX,XX @@ static const char rv_freg_name_sym[32][5] = {
  #define rv_fmt_rs2_offset             "O\t2,o"
  #define rv_fmt_rs1_rs2_bs             "O\t1,2,b"
  #define rv_fmt_rd_rs1_rnum            "O\t0,1,n"
 +#define rv_fmt_ldst_vd_rs1_vm         "O\tD,(1)m"
 +#define rv_fmt_ldst_vd_rs1_rs2_vm     "O\tD,(1),2m"
 +#define rv_fmt_ldst_vd_rs1_vs2_vm     "O\tD,(1),Fm"
 +#define rv_fmt_vd_vs2_vs1             "O\tD,F,E"
 +#define rv_fmt_vd_vs2_vs1_vl          "O\tD,F,El"
 +#define rv_fmt_vd_vs2_vs1_vm          "O\tD,F,Em"
 +#define rv_fmt_vd_vs2_rs1_vl          "O\tD,F,1l"
 +#define rv_fmt_vd_vs2_fs1_vl          "O\tD,F,4l"
 +#define rv_fmt_vd_vs2_rs1_vm          "O\tD,F,1m"
 +#define rv_fmt_vd_vs2_fs1_vm          "O\tD,F,4m"
 +#define rv_fmt_vd_vs2_imm_vl          "O\tD,F,il"
 +#define rv_fmt_vd_vs2_imm_vm          "O\tD,F,im"
 +#define rv_fmt_vd_vs2_uimm_vm         "O\tD,F,um"
 +#define rv_fmt_vd_vs1_vs2_vm          "O\tD,E,Fm"
 +#define rv_fmt_vd_rs1_vs2_vm          "O\tD,1,Fm"
 +#define rv_fmt_vd_fs1_vs2_vm          "O\tD,4,Fm"
 +#define rv_fmt_vd_vs1                 "O\tD,E"
 +#define rv_fmt_vd_rs1                 "O\tD,1"
 +#define rv_fmt_vd_fs1                 "O\tD,4"
 +#define rv_fmt_vd_imm                 "O\tD,i"
 +#define rv_fmt_vd_vs2                 "O\tD,F"
 +#define rv_fmt_vd_vs2_vm              "O\tD,Fm"
 +#define rv_fmt_rd_vs2_vm              "O\t0,Fm"
 +#define rv_fmt_rd_vs2                 "O\t0,F"
 +#define rv_fmt_fd_vs2                 "O\t3,F"
 +#define rv_fmt_vd_vm                  "O\tDm"
 +#define rv_fmt_vsetvli                "O\t0,1,v"
 +#define rv_fmt_vsetivli               "O\t0,u,v"
  /* pseudo-instruction constraints */
@@ -XXX,XX +XXX,XX @@ const rv_opcode_data opcode_data[] = {
      { "unzip", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0, 0 },
      { "zip", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0, 0 },
      { "xperm4", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 },
 -    { "xperm8", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0, 0 }
 +    { "xperm8", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0, 0 },
 +    { "vle8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle8_v, rv_op_vle8_v, 0 },
 +    { "vle16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle16_v, rv_op_vle16_v, 0 },
 +    { "vle32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle32_v, rv_op_vle32_v, 0 },
 +    { "vle64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle64_v, rv_op_vle64_v, 0 },
 +    { "vse8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vse8_v, rv_op_vse8_v, 0 },
 +    { "vse16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vse16_v, rv_op_vse16_v, 0 },
 +    { "vse32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vse32_v, rv_op_vse32_v, 0 },
 +    { "vse64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vse64_v, rv_op_vse64_v, 0 },
 +    { "vlm.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vlm_v, rv_op_vlm_v, 0 },
 +    { "vsm.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vsm_v, rv_op_vsm_v, 0 },
 +    { "vlse8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vlse8_v, rv_op_vlse8_v, 0 },
 +    { "vlse16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vlse16_v, rv_op_vlse16_v, 0 },
 +    { "vlse32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vlse32_v, rv_op_vlse32_v, 0 },
 +    { "vlse64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vlse64_v, rv_op_vlse64_v, 0 },
 +    { "vsse8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vsse8_v, rv_op_vsse8_v, 0 },
 +    { "vsse16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vsse16_v, rv_op_vsse16_v, 0 },
 +    { "vsse32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vsse32_v, rv_op_vsse32_v, 0 },
 +    { "vsse64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vsse64_v, rv_op_vsse64_v, 0 },
 +    { "vluxei8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vluxei8_v, rv_op_vluxei8_v, 0 },
 +    { "vluxei16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vluxei16_v, rv_op_vluxei16_v, 0 },
 +    { "vluxei32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vluxei32_v, rv_op_vluxei32_v, 0 },
 +    { "vluxei64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vluxei64_v, rv_op_vluxei64_v, 0 },
 +    { "vloxei8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vloxei8_v, rv_op_vloxei8_v, 0 },
 +    { "vloxei16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vloxei16_v, rv_op_vloxei16_v, 0 },
 +    { "vloxei32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vloxei32_v, rv_op_vloxei32_v, 0 },
 +    { "vloxei64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vloxei64_v, rv_op_vloxei64_v, 0 },
 +    { "vsuxei8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsuxei8_v, rv_op_vsuxei8_v, 0 },
 +    { "vsuxei16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsuxei16_v, rv_op_vsuxei16_v, 0 },
 +    { "vsuxei32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsuxei32_v, rv_op_vsuxei32_v, 0 },
 +    { "vsuxei64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsuxei64_v, rv_op_vsuxei64_v, 0 },
 +    { "vsoxei8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsoxei8_v, rv_op_vsoxei8_v, 0 },
 +    { "vsoxei16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsoxei16_v, rv_op_vsoxei16_v, 0 },
 +    { "vsoxei32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsoxei32_v, rv_op_vsoxei32_v, 0 },
 +    { "vsoxei64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsoxei64_v, rv_op_vsoxei64_v, 0 },
 +    { "vle8ff.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle8ff_v, rv_op_vle8ff_v, 0 },
 +    { "vle16ff.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle16ff_v, rv_op_vle16ff_v, 0 },
 +    { "vle32ff.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle32ff_v, rv_op_vle32ff_v, 0 },
 +    { "vle64ff.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle64ff_v, rv_op_vle64ff_v, 0 },
 +    { "vl1re8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl1re8_v, rv_op_vl1re8_v, 0 },
 +    { "vl1re16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl1re16_v, rv_op_vl1re16_v, 0 },
 +    { "vl1re32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl1re32_v, rv_op_vl1re32_v, 0 },
 +    { "vl1re64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl1re64_v, rv_op_vl1re64_v, 0 },
 +    { "vl2re8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl2re8_v, rv_op_vl2re8_v, 0 },
 +    { "vl2re16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl2re16_v, rv_op_vl2re16_v, 0 },
 +    { "vl2re32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl2re32_v, rv_op_vl2re32_v, 0 },
 +    { "vl2re64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl2re64_v, rv_op_vl2re64_v, 0 },
 +    { "vl4re8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl4re8_v, rv_op_vl4re8_v, 0 },
 +    { "vl4re16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl4re16_v, rv_op_vl4re16_v, 0 },
 +    { "vl4re32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl4re32_v, rv_op_vl4re32_v, 0 },
 +    { "vl4re64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl4re64_v, rv_op_vl4re64_v, 0 },
 +    { "vl8re8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl8re8_v, rv_op_vl8re8_v, 0 },
 +    { "vl8re16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl8re16_v, rv_op_vl8re16_v, 0 },
 +    { "vl8re32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl8re32_v, rv_op_vl8re32_v, 0 },
 +    { "vl8re64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl8re64_v, rv_op_vl8re64_v, 0 },
 +    { "vs1r.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vs1r_v, rv_op_vs1r_v, 0 },
 +    { "vs2r.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vs2r_v, rv_op_vs2r_v, 0 },
 +    { "vs4r.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vs4r_v, rv_op_vs4r_v, 0 },
 +    { "vs8r.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vs8r_v, rv_op_vs8r_v, 0 },
 +    { "vadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vadd_vv, rv_op_vadd_vv, 0 },
 +    { "vadd.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vadd_vx, rv_op_vadd_vx, 0 },
 +    { "vadd.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vadd_vi, rv_op_vadd_vi, 0 },
 +    { "vsub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsub_vv, rv_op_vsub_vv, 0 },
 +    { "vsub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsub_vx, rv_op_vsub_vx, 0 },
 +    { "vrsub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vrsub_vx, rv_op_vrsub_vx, 0 },
 +    { "vrsub.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vrsub_vi, rv_op_vrsub_vi, 0 },
 +    { "vwaddu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwaddu_vv, rv_op_vwaddu_vv, 0 },
 +    { "vwaddu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwaddu_vx, rv_op_vwaddu_vx, 0 },
 +    { "vwadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwadd_vv, rv_op_vwadd_vv, 0 },
 +    { "vwadd.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwadd_vx, rv_op_vwadd_vx, 0 },
 +    { "vwsubu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwsubu_vv, rv_op_vwsubu_vv, 0 },
 +    { "vwsubu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwsubu_vx, rv_op_vwsubu_vx, 0 },
 +    { "vwsub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwsub_vv, rv_op_vwsub_vv, 0 },
 +    { "vwsub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwsub_vx, rv_op_vwsub_vx, 0 },
 +    { "vwaddu.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwaddu_wv, rv_op_vwaddu_wv, 0 },
 +    { "vwaddu.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwaddu_wx, rv_op_vwaddu_wx, 0 },
 +    { "vwadd.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwadd_wv, rv_op_vwadd_wv, 0 },
 +    { "vwadd.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwadd_wx, rv_op_vwadd_wx, 0 },
 +    { "vwsubu.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwsubu_wv, rv_op_vwsubu_wv, 0 },
 +    { "vwsubu.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwsubu_wx, rv_op_vwsubu_wx, 0 },
 +    { "vwsub.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwsub_wv, rv_op_vwsub_wv, 0 },
 +    { "vwsub.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwsub_wx, rv_op_vwsub_wx, 0 },
 +    { "vadc.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vadc_vvm, rv_op_vadc_vvm, 0 },
 +    { "vadc.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vadc_vxm, rv_op_vadc_vxm, 0 },
 +    { "vadc.vim", rv_codec_v_i, rv_fmt_vd_vs2_imm_vl, NULL, rv_op_vadc_vim, rv_op_vadc_vim, 0 },
 +    { "vmadc.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vmadc_vvm, rv_op_vmadc_vvm, 0 },
 +    { "vmadc.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vmadc_vxm, rv_op_vmadc_vxm, 0 },
 +    { "vmadc.vim", rv_codec_v_i, rv_fmt_vd_vs2_imm_vl, NULL, rv_op_vmadc_vim, rv_op_vmadc_vim, 0 },
 +    { "vsbc.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vsbc_vvm, rv_op_vsbc_vvm, 0 },
 +    { "vsbc.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vsbc_vxm, rv_op_vsbc_vxm, 0 },
 +    { "vmsbc.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vmsbc_vvm, rv_op_vmsbc_vvm, 0 },
 +    { "vmsbc.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vmsbc_vxm, rv_op_vmsbc_vxm, 0 },
 +    { "vand.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vand_vv, rv_op_vand_vv, 0 },
 +    { "vand.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vand_vx, rv_op_vand_vx, 0 },
 +    { "vand.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vand_vi, rv_op_vand_vi, 0 },
 +    { "vor.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vor_vv, rv_op_vor_vv, 0 },
 +    { "vor.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vor_vx, rv_op_vor_vx, 0 },
 +    { "vor.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vor_vi, rv_op_vor_vi, 0 },
 +    { "vxor.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vxor_vv, rv_op_vxor_vv, 0 },
 +    { "vxor.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vxor_vx, rv_op_vxor_vx, 0 },
 +    { "vxor.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vxor_vi, rv_op_vxor_vi, 0 },
 +    { "vsll.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsll_vv, rv_op_vsll_vv, 0 },
 +    { "vsll.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsll_vx, rv_op_vsll_vx, 0 },
 +    { "vsll.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vsll_vi, rv_op_vsll_vi, 0 },
 +    { "vsrl.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsrl_vv, rv_op_vsrl_vv, 0 },
 +    { "vsrl.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsrl_vx, rv_op_vsrl_vx, 0 },
 +    { "vsrl.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vsrl_vi, rv_op_vsrl_vi, 0 },
 +    { "vsra.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsra_vv, rv_op_vsra_vv, 0 },
 +    { "vsra.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsra_vx, rv_op_vsra_vx, 0 },
 +    { "vsra.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vsra_vi, rv_op_vsra_vi, 0 },
 +    { "vnsrl.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vnsrl_wv, rv_op_vnsrl_wv, 0 },
 +    { "vnsrl.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vnsrl_wx, rv_op_vnsrl_wx, 0 },
 +    { "vnsrl.wi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vnsrl_wi, rv_op_vnsrl_wi, 0 },
 +    { "vnsra.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vnsra_wv, rv_op_vnsra_wv, 0 },
 +    { "vnsra.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vnsra_wx, rv_op_vnsra_wx, 0 },
 +    { "vnsra.wi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vnsra_wi, rv_op_vnsra_wi, 0 },
 +    { "vmseq.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmseq_vv, rv_op_vmseq_vv, 0 },
 +    { "vmseq.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmseq_vx, rv_op_vmseq_vx, 0 },
 +    { "vmseq.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmseq_vi, rv_op_vmseq_vi, 0 },
 +    { "vmsne.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmsne_vv, rv_op_vmsne_vv, 0 },
 +    { "vmsne.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsne_vx, rv_op_vmsne_vx, 0 },
 +    { "vmsne.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsne_vi, rv_op_vmsne_vi, 0 },
 +    { "vmsltu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmsltu_vv, rv_op_vmsltu_vv, 0 },
 +    { "vmsltu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsltu_vx, rv_op_vmsltu_vx, 0 },
 +    { "vmslt.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmslt_vv, rv_op_vmslt_vv, 0 },
 +    { "vmslt.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmslt_vx, rv_op_vmslt_vx, 0 },
 +    { "vmsleu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmsleu_vv, rv_op_vmsleu_vv, 0 },
 +    { "vmsleu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsleu_vx, rv_op_vmsleu_vx, 0 },
 +    { "vmsleu.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsleu_vi, rv_op_vmsleu_vi, 0 },
 +    { "vmsle.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmsle_vv, rv_op_vmsle_vv, 0 },
 +    { "vmsle.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsle_vx, rv_op_vmsle_vx, 0 },
 +    { "vmsle.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsle_vi, rv_op_vmsle_vi, 0 },
 +    { "vmsgtu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsgtu_vx, rv_op_vmsgtu_vx, 0 },
 +    { "vmsgtu.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsgtu_vi, rv_op_vmsgtu_vi, 0 },
 +    { "vmsgt.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsgt_vx, rv_op_vmsgt_vx, 0 },
 +    { "vmsgt.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsgt_vi, rv_op_vmsgt_vi, 0 },
 +    { "vminu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vminu_vv, rv_op_vminu_vv, 0 },
 +    { "vminu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vminu_vx, rv_op_vminu_vx, 0 },
 +    { "vmin.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmin_vv, rv_op_vmin_vv, 0 },
 +    { "vmin.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmin_vx, rv_op_vmin_vx, 0 },
 +    { "vmaxu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmaxu_vv, rv_op_vmaxu_vv, 0 },
 +    { "vmaxu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmaxu_vx, rv_op_vmaxu_vx, 0 },
 +    { "vmax.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmax_vv, rv_op_vmax_vv, 0 },
 +    { "vmax.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmax_vx, rv_op_vmax_vx, 0 },
 +    { "vmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmul_vv, rv_op_vmul_vv, 0 },
 +    { "vmul.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmul_vx, rv_op_vmul_vx, 0 },
 +    { "vmulh.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmulh_vv, rv_op_vmulh_vv, 0 },
 +    { "vmulh.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmulh_vx, rv_op_vmulh_vx, 0 },
 +    { "vmulhu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmulhu_vv, rv_op_vmulhu_vv, 0 },
 +    { "vmulhu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmulhu_vx, rv_op_vmulhu_vx, 0 },
 +    { "vmulhsu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmulhsu_vv, rv_op_vmulhsu_vv, 0 },
 +    { "vmulhsu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmulhsu_vx, rv_op_vmulhsu_vx, 0 },
 +    { "vdivu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vdivu_vv, rv_op_vdivu_vv, 0 },
 +    { "vdivu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vdivu_vx, rv_op_vdivu_vx, 0 },
 +    { "vdiv.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vdiv_vv, rv_op_vdiv_vv, 0 },
 +    { "vdiv.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vdiv_vx, rv_op_vdiv_vx, 0 },
 +    { "vremu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vremu_vv, rv_op_vremu_vv, 0 },
 +    { "vremu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vremu_vx, rv_op_vremu_vx, 0 },
 +    { "vrem.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vrem_vv, rv_op_vrem_vv, 0 },
 +    { "vrem.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vrem_vx, rv_op_vrem_vx, 0 },
 +    { "vwmulu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwmulu_vv, rv_op_vwmulu_vv, 0 },
 +    { "vwmulu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwmulu_vx, rv_op_vwmulu_vx, 0 },
 +    { "vwmulsu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwmulsu_vv, rv_op_vwmulsu_vv, 0 },
 +    { "vwmulsu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwmulsu_vx, rv_op_vwmulsu_vx, 0 },
 +    { "vwmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwmul_vv, rv_op_vwmul_vv, 0 },
 +    { "vwmul.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwmul_vx, rv_op_vwmul_vx, 0 },
 +    { "vmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vmacc_vv, rv_op_vmacc_vv, 0 },
 +    { "vmacc.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vmacc_vx, rv_op_vmacc_vx, 0 },
 +    { "vnmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vnmsac_vv, rv_op_vnmsac_vv, 0 },
 +    { "vnmsac.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vnmsac_vx, rv_op_vnmsac_vx, 0 },
 +    { "vmadd.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vmadd_vv, rv_op_vmadd_vv, 0 },
 +    { "vmadd.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vmadd_vx, rv_op_vmadd_vx, 0 },
 +    { "vnmsub.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vnmsub_vv, rv_op_vnmsub_vv, 0 },
 +    { "vnmsub.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vnmsub_vx, rv_op_vnmsub_vx, 0 },
 +    { "vwmaccu.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vwmaccu_vv, rv_op_vwmaccu_vv, 0 },
 +    { "vwmaccu.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vwmaccu_vx, rv_op_vwmaccu_vx, 0 },
 +    { "vwmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vwmacc_vv, rv_op_vwmacc_vv, 0 },
 +    { "vwmacc.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vwmacc_vx, rv_op_vwmacc_vx, 0 },
 +    { "vwmaccsu.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vwmaccsu_vv, rv_op_vwmaccsu_vv, 0 },
 +    { "vwmaccsu.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vwmaccsu_vx, rv_op_vwmaccsu_vx, 0 },
 +    { "vwmaccus.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vwmaccus_vx, rv_op_vwmaccus_vx, 0 },
 +    { "vmv.v.v", rv_codec_v_r, rv_fmt_vd_vs1, NULL, rv_op_vmv_v_v, rv_op_vmv_v_v, 0 },
 +    { "vmv.v.x", rv_codec_v_r, rv_fmt_vd_rs1, NULL, rv_op_vmv_v_x, rv_op_vmv_v_x, 0 },
 +    { "vmv.v.i", rv_codec_v_i, rv_fmt_vd_imm, NULL, rv_op_vmv_v_i, rv_op_vmv_v_i, 0 },
 +    { "vmerge.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vmerge_vvm, rv_op_vmerge_vvm, 0 },
 +    { "vmerge.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vmerge_vxm, rv_op_vmerge_vxm, 0 },
 +    { "vmerge.vim", rv_codec_v_i, rv_fmt_vd_vs2_imm_vl, NULL, rv_op_vmerge_vim, rv_op_vmerge_vim, 0 },
 +    { "vsaddu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsaddu_vv, rv_op_vsaddu_vv, 0 },
 +    { "vsaddu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsaddu_vx, rv_op_vsaddu_vx, 0 },
 +    { "vsaddu.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vsaddu_vi, rv_op_vsaddu_vi, 0 },
 +    { "vsadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsadd_vv, rv_op_vsadd_vv, 0 },
 +    { "vsadd.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsadd_vx, rv_op_vsadd_vx, 0 },
 +    { "vsadd.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vsadd_vi, rv_op_vsadd_vi, 0 },
 +    { "vssubu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vssubu_vv, rv_op_vssubu_vv, 0 },
 +    { "vssubu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vssubu_vx, rv_op_vssubu_vx, 0 },
 +    { "vssub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vssub_vv, rv_op_vssub_vv, 0 },
 +    { "vssub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vssub_vx, rv_op_vssub_vx, 0 },
 +    { "vaadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vaadd_vv, rv_op_vaadd_vv, 0 },
 +    { "vaadd.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vaadd_vx, rv_op_vaadd_vx, 0 },
 +    { "vaaddu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vaaddu_vv, rv_op_vaaddu_vv, 0 },
 +    { "vaaddu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vaaddu_vx, rv_op_vaaddu_vx, 0 },
 +    { "vasub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vasub_vv, rv_op_vasub_vv, 0 },
 +    { "vasub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vasub_vx, rv_op_vasub_vx, 0 },
 +    { "vasubu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vasubu_vv, rv_op_vasubu_vv, 0 },
 +    { "vasubu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vasubu_vx, rv_op_vasubu_vx, 0 },
 +    { "vsmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsmul_vv, rv_op_vsmul_vv, 0 },
 +    { "vsmul.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsmul_vx, rv_op_vsmul_vx, 0 },
 +    { "vssrl.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vssrl_vv, rv_op_vssrl_vv, 0 },
 +    { "vssrl.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vssrl_vx, rv_op_vssrl_vx, 0 },
 +    { "vssrl.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vssrl_vi, rv_op_vssrl_vi, 0 },
 +    { "vssra.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vssra_vv, rv_op_vssra_vv, 0 },
 +    { "vssra.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vssra_vx, rv_op_vssra_vx, 0 },
 +    { "vssra.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vssra_vi, rv_op_vssra_vi, 0 },
 +    { "vnclipu.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vnclipu_wv, rv_op_vnclipu_wv, 0 },
 +    { "vnclipu.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vnclipu_wx, rv_op_vnclipu_wx, 0 },
 +    { "vnclipu.wi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vnclipu_wi, rv_op_vnclipu_wi, 0 },
 +    { "vnclip.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vnclip_wv, rv_op_vnclip_wv, 0 },
 +    { "vnclip.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vnclip_wx, rv_op_vnclip_wx, 0 },
 +    { "vnclip.wi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vnclip_wi, rv_op_vnclip_wi, 0 },
 +    { "vfadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfadd_vv, rv_op_vfadd_vv, 0 },
 +    { "vfadd.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfadd_vf, rv_op_vfadd_vf, 0 },
 +    { "vfsub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfsub_vv, rv_op_vfsub_vv, 0 },
 +    { "vfsub.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfsub_vf, rv_op_vfsub_vf, 0 },
 +    { "vfrsub.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfrsub_vf, rv_op_vfrsub_vf, 0 },
 +    { "vfwadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwadd_vv, rv_op_vfwadd_vv, 0 },
 +    { "vfwadd.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwadd_vf, rv_op_vfwadd_vf, 0 },
 +    { "vfwadd.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwadd_wv, rv_op_vfwadd_wv, 0 },
 +    { "vfwadd.wf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwadd_wf, rv_op_vfwadd_wf, 0 },
 +    { "vfwsub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwsub_vv, rv_op_vfwsub_vv, 0 },
 +    { "vfwsub.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwsub_vf, rv_op_vfwsub_vf, 0 },
 +    { "vfwsub.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwsub_wv, rv_op_vfwsub_wv, 0 },
 +    { "vfwsub.wf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwsub_wf, rv_op_vfwsub_wf, 0 },
 +    { "vfmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfmul_vv, rv_op_vfmul_vv, 0 },
 +    { "vfmul.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfmul_vf, rv_op_vfmul_vf, 0 },
 +    { "vfdiv.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfdiv_vv, rv_op_vfdiv_vv, 0 },
 +    { "vfdiv.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfdiv_vf, rv_op_vfdiv_vf, 0 },
 +    { "vfrdiv.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfrdiv_vf, rv_op_vfrdiv_vf, 0 },
 +    { "vfwmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwmul_vv, rv_op_vfwmul_vv, 0 },
 +    { "vfwmul.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwmul_vf, rv_op_vfwmul_vf, 0 },
 +    { "vfmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfmacc_vv, rv_op_vfmacc_vv, 0 },
 +    { "vfmacc.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfmacc_vf, rv_op_vfmacc_vf, 0 },
 +    { "vfnmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfnmacc_vv, rv_op_vfnmacc_vv, 0 },
 +    { "vfnmacc.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfnmacc_vf, rv_op_vfnmacc_vf, 0 },
 +    { "vfmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfmsac_vv, rv_op_vfmsac_vv, 0 },
 +    { "vfmsac.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfmsac_vf, rv_op_vfmsac_vf, 0 },
 +    { "vfnmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfnmsac_vv, rv_op_vfnmsac_vv, 0 },
 +    { "vfnmsac.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfnmsac_vf, rv_op_vfnmsac_vf, 0 },
 +    { "vfmadd.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfmadd_vv, rv_op_vfmadd_vv, 0 },
 +    { "vfmadd.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfmadd_vf, rv_op_vfmadd_vf, 0 },
 +    { "vfnmadd.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfnmadd_vv, rv_op_vfnmadd_vv, 0 },
 +    { "vfnmadd.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfnmadd_vf, rv_op_vfnmadd_vf, 0 },
 +    { "vfmsub.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfmsub_vv, rv_op_vfmsub_vv, 0 },
 +    { "vfmsub.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfmsub_vf, rv_op_vfmsub_vf, 0 },
 +    { "vfnmsub.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfnmsub_vv, rv_op_vfnmsub_vv, 0 },
 +    { "vfnmsub.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfnmsub_vf, rv_op_vfnmsub_vf, 0 },
 +    { "vfwmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfwmacc_vv, rv_op_vfwmacc_vv, 0 },
 +    { "vfwmacc.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfwmacc_vf, rv_op_vfwmacc_vf, 0 },
 +    { "vfwnmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfwnmacc_vv, rv_op_vfwnmacc_vv, 0 },
 +    { "vfwnmacc.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfwnmacc_vf, rv_op_vfwnmacc_vf, 0 },
 +    { "vfwmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfwmsac_vv, rv_op_vfwmsac_vv, 0 },
 +    { "vfwmsac.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfwmsac_vf, rv_op_vfwmsac_vf, 0 },
 +    { "vfwnmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfwnmsac_vv, rv_op_vfwnmsac_vv, 0 },
 +    { "vfwnmsac.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfwnmsac_vf, rv_op_vfwnmsac_vf, 0 },
 +    { "vfsqrt.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vfsqrt_v, rv_op_vfsqrt_v, 0 },
 +    { "vfrsqrt7.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vfrsqrt7_v, rv_op_vfrsqrt7_v, 0 },
 +    { "vfrec7.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vfrec7_v, rv_op_vfrec7_v, 0 },
 +    { "vfmin.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfmin_vv, rv_op_vfmin_vv, 0 },
 +    { "vfmin.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfmin_vf, rv_op_vfmin_vf, 0 },
 +    { "vfmax.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfmax_vv, rv_op_vfmax_vv, 0 },
 +    { "vfmax.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfmax_vf, rv_op_vfmax_vf, 0 },
 +    { "vfsgnj.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfsgnj_vv, rv_op_vfsgnj_vv, 0 },
 +    { "vfsgnj.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfsgnj_vf, rv_op_vfsgnj_vf, 0 },
 +    { "vfsgnjn.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfsgnjn_vv, rv_op_vfsgnjn_vv, 0 },
 +    { "vfsgnjn.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfsgnjn_vf, rv_op_vfsgnjn_vf, 0 },
 +    { "vfsgnjx.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfsgnjx_vv, rv_op_vfsgnjx_vv, 0 },
 +    { "vfsgnjx.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfsgnjx_vf, rv_op_vfsgnjx_vf, 0 },
 +    { "vfslide1up.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfslide1up_vf, rv_op_vfslide1up_vf, 0 },
 +    { "vfslide1down.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfslide1down_vf, rv_op_vfslide1down_vf, 0 },
 +    { "vmfeq.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmfeq_vv, rv_op_vmfeq_vv, 0 },
 +    { "vmfeq.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfeq_vf, rv_op_vmfeq_vf, 0 },
 +    { "vmfne.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmfne_vv, rv_op_vmfne_vv, 0 },
 +    { "vmfne.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfne_vf, rv_op_vmfne_vf, 0 },
 +    { "vmflt.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmflt_vv, rv_op_vmflt_vv, 0 },
 +    { "vmflt.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmflt_vf, rv_op_vmflt_vf, 0 },
 +    { "vmfle.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmfle_vv, rv_op_vmfle_vv, 0 },
 +    { "vmfle.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfle_vf, rv_op_vmfle_vf, 0 },
 +    { "vmfgt.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfgt_vf, rv_op_vmfgt_vf, 0 },
 +    { "vmfge.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfge_vf, rv_op_vmfge_vf, 0 },
 +    { "vfclass.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfclass_v, rv_op_vfclass_v, 0 },
 +    { "vfmerge.vfm", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vl, NULL, rv_op_vfmerge_vfm, rv_op_vfmerge_vfm, 0 },
 +    { "vfmv.v.f", rv_codec_v_r, rv_fmt_vd_fs1, NULL, rv_op_vfmv_v_f, rv_op_vfmv_v_f, 0 },
 +    { "vfcvt.xu.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_xu_f_v, rv_op_vfcvt_xu_f_v, 0 },
 +    { "vfcvt.x.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_x_f_v, rv_op_vfcvt_x_f_v, 0 },
 +    { "vfcvt.f.xu.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_f_xu_v, rv_op_vfcvt_f_xu_v, 0 },
 +    { "vfcvt.f.x.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_f_x_v, rv_op_vfcvt_f_x_v, 0 },
 +    { "vfcvt.rtz.xu.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_rtz_xu_f_v, rv_op_vfcvt_rtz_xu_f_v, 0 },
 +    { "vfcvt.rtz.x.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_rtz_x_f_v, rv_op_vfcvt_rtz_x_f_v, 0 },
 +    { "vfwcvt.xu.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_xu_f_v, rv_op_vfwcvt_xu_f_v, 0 },
 +    { "vfwcvt.x.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_x_f_v, rv_op_vfwcvt_x_f_v, 0 },
 +    { "vfwcvt.f.xu.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_f_xu_v, rv_op_vfwcvt_f_xu_v, 0 },
 +    { "vfwcvt.f.x.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_f_x_v, rv_op_vfwcvt_f_x_v, 0 },
 +    { "vfwcvt.f.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_f_f_v, rv_op_vfwcvt_f_f_v, 0 },
 +    { "vfwcvt.rtz.xu.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_rtz_xu_f_v, rv_op_vfwcvt_rtz_xu_f_v, 0 },
 +    { "vfwcvt.rtz.x.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_rtz_x_f_v, rv_op_vfwcvt_rtz_x_f_v, 0 },
 +    { "vfncvt.xu.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_xu_f_w, rv_op_vfncvt_xu_f_w, 0 },
 +    { "vfncvt.x.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_x_f_w, rv_op_vfncvt_x_f_w, 0 },
 +    { "vfncvt.f.xu.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_f_xu_w, rv_op_vfncvt_f_xu_w, 0 },
 +    { "vfncvt.f.x.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_f_x_w, rv_op_vfncvt_f_x_w, 0 },
 +    { "vfncvt.f.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_f_f_w, rv_op_vfncvt_f_f_w, 0 },
 +    { "vfncvt.rod.f.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_rod_f_f_w, rv_op_vfncvt_rod_f_f_w, 0 },
 +    { "vfncvt.rtz.xu.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_rtz_xu_f_w, rv_op_vfncvt_rtz_xu_f_w, 0 },
 +    { "vfncvt.rtz.x.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_rtz_x_f_w, rv_op_vfncvt_rtz_x_f_w, 0 },
 +    { "vredsum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredsum_vs, rv_op_vredsum_vs, 0 },
 +    { "vredand.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredand_vs, rv_op_vredand_vs, 0 },
 +    { "vredor.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredor_vs, rv_op_vredor_vs, 0 },
 +    { "vredxor.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredxor_vs, rv_op_vredxor_vs, 0 },
 +    { "vredminu.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredminu_vs, rv_op_vredminu_vs, 0 },
 +    { "vredmin.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredmin_vs, rv_op_vredmin_vs, 0 },
 +    { "vredmaxu.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredmaxu_vs, rv_op_vredmaxu_vs, 0 },
 +    { "vredmax.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredmax_vs, rv_op_vredmax_vs, 0 },
 +    { "vwredsumu.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwredsumu_vs, rv_op_vwredsumu_vs, 0 },
 +    { "vwredsum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwredsum_vs, rv_op_vwredsum_vs, 0 },
 +    { "vfredusum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfredusum_vs, rv_op_vfredusum_vs, 0 },
 +    { "vfredosum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfredosum_vs, rv_op_vfredosum_vs, 0 },
 +    { "vfredmin.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfredmin_vs, rv_op_vfredmin_vs, 0 },
 +    { "vfredmax.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfredmax_vs, rv_op_vfredmax_vs, 0 },
 +    { "vfwredusum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwredusum_vs, rv_op_vfwredusum_vs, 0 },
 +    { "vfwredosum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwredosum_vs, rv_op_vfwredosum_vs, 0 },
 +    { "vmand.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmand_mm, rv_op_vmand_mm, 0 },
 +    { "vmnand.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmnand_mm, rv_op_vmnand_mm, 0 },
 +    { "vmandn.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmandn_mm, rv_op_vmandn_mm, 0 },
 +    { "vmxor.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmxor_mm, rv_op_vmxor_mm, 0 },
 +    { "vmor.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmor_mm, rv_op_vmor_mm, 0 },
 +    { "vmnor.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmnor_mm, rv_op_vmnor_mm, 0 },
 +    { "vmorn.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmorn_mm, rv_op_vmorn_mm, 0 },
 +    { "vmxnor.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmxnor_mm, rv_op_vmxnor_mm, 0 },
 +    { "vcpop.m", rv_codec_v_r, rv_fmt_rd_vs2_vm, NULL, rv_op_vcpop_m, rv_op_vcpop_m, 0 },
 +    { "vfirst.m", rv_codec_v_r, rv_fmt_rd_vs2_vm, NULL, rv_op_vfirst_m, rv_op_vfirst_m, 0 },
 +    { "vmsbf.m", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vmsbf_m, rv_op_vmsbf_m, 0 },
 +    { "vmsif.m", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vmsif_m, rv_op_vmsif_m, 0 },
 +    { "vmsof.m", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vmsof_m, rv_op_vmsof_m, 0 },
 +    { "viota.m", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_viota_m, rv_op_viota_m, 0 },
 +    { "vid.v", rv_codec_v_r, rv_fmt_vd_vm, NULL, rv_op_vid_v, rv_op_vid_v, 0 },
 +    { "vmv.x.s", rv_codec_v_r, rv_fmt_rd_vs2, NULL, rv_op_vmv_x_s, rv_op_vmv_x_s, 0 },
 +    { "vmv.s.x", rv_codec_v_r, rv_fmt_vd_rs1, NULL, rv_op_vmv_s_x, rv_op_vmv_s_x, 0 },
 +    { "vfmv.f.s", rv_codec_v_r, rv_fmt_fd_vs2, NULL, rv_op_vfmv_f_s, rv_op_vfmv_f_s, 0 },
 +    { "vfmv.s.f", rv_codec_v_r, rv_fmt_vd_fs1, NULL, rv_op_vfmv_s_f, rv_op_vfmv_s_f, 0 },
 +    { "vslideup.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vslideup_vx, rv_op_vslideup_vx, 0 },
 +    { "vslideup.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vslideup_vi, rv_op_vslideup_vi, 0 },
 +    { "vslide1up.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vslide1up_vx, rv_op_vslide1up_vx, 0 },
 +    { "vslidedown.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vslidedown_vx, rv_op_vslidedown_vx, 0 },
 +    { "vslidedown.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vslidedown_vi, rv_op_vslidedown_vi, 0 },
 +    { "vslide1down.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vslide1down_vx, rv_op_vslide1down_vx, 0 },
 +    { "vrgather.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vrgather_vv, rv_op_vrgather_vv, 0 },
 +    { "vrgatherei16.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vrgatherei16_vv, rv_op_vrgatherei16_vv, 0 },
 +    { "vrgather.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vrgather_vx, rv_op_vrgather_vx, 0 },
 +    { "vrgather.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vrgather_vi, rv_op_vrgather_vi, 0 },
 +    { "vcompress.vm", rv_codec_v_r, rv_fmt_vd_vs2_vs1, NULL, rv_op_vcompress_vm, rv_op_vcompress_vm, 0 },
 +    { "vmv1r.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vmv1r_v, rv_op_vmv1r_v, 0 },
 +    { "vmv2r.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vmv2r_v, rv_op_vmv2r_v, 0 },
 +    { "vmv4r.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vmv4r_v, rv_op_vmv4r_v, 0 },
 +    { "vmv8r.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vmv8r_v, rv_op_vmv8r_v, 0 },
 +    { "vzext.vf2", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vzext_vf2, rv_op_vzext_vf2, 0 },
 +    { "vzext.vf4", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vzext_vf4, rv_op_vzext_vf4, 0 },
 +    { "vzext.vf8", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vzext_vf8, rv_op_vzext_vf8, 0 },
 +    { "vsext.vf2", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vsext_vf2, rv_op_vsext_vf2, 0 },
 +    { "vsext.vf4", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vsext_vf4, rv_op_vsext_vf4, 0 },
 +    { "vsext.vf8", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vsext_vf8, rv_op_vsext_vf8, 0 },
 +    { "vsetvli", rv_codec_vsetvli, rv_fmt_vsetvli, NULL, rv_op_vsetvli, rv_op_vsetvli, 0 },
 +    { "vsetivli", rv_codec_vsetivli, rv_fmt_vsetivli, NULL, rv_op_vsetivli, rv_op_vsetivli, 0 },
 +    { "vsetvl", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, rv_op_vsetvl, rv_op_vsetvl, 0 }
  };
  /* CSR names */
@@ -XXX,XX +XXX,XX @@ static const char *csr_name(int csrno)
      case 0x0003: return "fcsr";
      case 0x0004: return "uie";
      case 0x0005: return "utvec";
 +    case 0x0008: return "vstart";
 +    case 0x0009: return "vxsat";
 +    case 0x000a: return "vxrm";
 +    case 0x000f: return "vcsr";
      case 0x0015: return "seed";
      case 0x0040: return "uscratch";
      case 0x0041: return "uepc";
@@ -XXX,XX +XXX,XX @@ static const char *csr_name(int csrno)
      case 0x0c00: return "cycle";
      case 0x0c01: return "time";
      case 0x0c02: return "instret";
 +    case 0x0c20: return "vl";
 +    case 0x0c21: return "vtype";
 +    case 0x0c22: return "vlenb";
      case 0x0c80: return "cycleh";
      case 0x0c81: return "timeh";
      case 0x0c82: return "instreth";
@@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa)
              break;
          case 1:
              switch (((inst >> 12) & 0b111)) {
 +            case 0:
 +                switch (((inst >> 20) & 0b111111111111)) {
 +                case 40: op = rv_op_vl1re8_v; break;
 +                case 552: op = rv_op_vl2re8_v; break;
 +                case 1576: op = rv_op_vl4re8_v; break;
 +                case 3624: op = rv_op_vl8re8_v; break;
 +                }
 +                switch (((inst >> 26) & 0b111)) {
 +                case 0:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: op = rv_op_vle8_v; break;
 +                    case 11: op = rv_op_vlm_v; break;
 +                    case 16: op = rv_op_vle8ff_v; break;
 +                    }
 +                    break;
 +                case 1: op = rv_op_vluxei8_v; break;
 +                case 2: op = rv_op_vlse8_v; break;
 +                case 3: op = rv_op_vloxei8_v; break;
 +                }
 +                break;
              case 2: op = rv_op_flw; break;
              case 3: op = rv_op_fld; break;
              case 4: op = rv_op_flq; break;
 +            case 5:
 +                switch (((inst >> 20) & 0b111111111111)) {
 +                case 40: op = rv_op_vl1re16_v; break;
 +                case 552: op = rv_op_vl2re16_v; break;
 +                case 1576: op = rv_op_vl4re16_v; break;
 +                case 3624: op = rv_op_vl8re16_v; break;
 +                }
 +                switch (((inst >> 26) & 0b111)) {
 +                case 0:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: op = rv_op_vle16_v; break;
 +                    case 16: op = rv_op_vle16ff_v; break;
 +                    }
 +                    break;
 +                case 1: op = rv_op_vluxei16_v; break;
 +                case 2: op = rv_op_vlse16_v; break;
 +                case 3: op = rv_op_vloxei16_v; break;
 +                }
 +                break;
 +            case 6:
 +                switch (((inst >> 20) & 0b111111111111)) {
 +                case 40: op = rv_op_vl1re32_v; break;
 +                case 552: op = rv_op_vl2re32_v; break;
 +                case 1576: op = rv_op_vl4re32_v; break;
 +                case 3624: op = rv_op_vl8re32_v; break;
 +                }
 +                switch (((inst >> 26) & 0b111)) {
 +                case 0:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: op = rv_op_vle32_v; break;
 +                    case 16: op = rv_op_vle32ff_v; break;
 +                    }
 +                    break;
 +                case 1: op = rv_op_vluxei32_v; break;
 +                case 2: op = rv_op_vlse32_v; break;
 +                case 3: op = rv_op_vloxei32_v; break;
 +                }
 +                break;
 +            case 7:
 +                switch (((inst >> 20) & 0b111111111111)) {
 +                case 40: op = rv_op_vl1re64_v; break;
 +                case 552: op = rv_op_vl2re64_v; break;
 +                case 1576: op = rv_op_vl4re64_v; break;
 +                case 3624: op = rv_op_vl8re64_v; break;
 +                }
 +                switch (((inst >> 26) & 0b111)) {
 +                case 0:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: op = rv_op_vle64_v; break;
 +                    case 16: op = rv_op_vle64ff_v; break;
 +                    }
 +                    break;
 +                case 1: op = rv_op_vluxei64_v; break;
 +                case 2: op = rv_op_vlse64_v; break;
 +                case 3: op = rv_op_vloxei64_v; break;
 +                }
 +                break;
              }
              break;
          case 3:
@@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa)
              break;
          case 9:
              switch (((inst >> 12) & 0b111)) {
 +            case 0:
 +                switch (((inst >> 20) & 0b111111111111)) {
 +                case 40: op = rv_op_vs1r_v; break;
 +                case 552: op = rv_op_vs2r_v; break;
 +                case 1576: op = rv_op_vs4r_v; break;
 +                case 3624: op = rv_op_vs8r_v; break;
 +                }
 +                switch (((inst >> 26) & 0b111)) {
 +                case 0:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: op = rv_op_vse8_v; break;
 +                    case 11: op = rv_op_vsm_v; break;
 +                    }
 +                    break;
 +                case 1: op = rv_op_vsuxei8_v; break;
 +                case 2: op = rv_op_vsse8_v; break;
 +                case 3: op = rv_op_vsoxei8_v; break;
 +                }
 +                break;
              case 2: op = rv_op_fsw; break;
              case 3: op = rv_op_fsd; break;
              case 4: op = rv_op_fsq; break;
 +            case 5:
 +                switch (((inst >> 26) & 0b111)) {
 +                case 0:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: op = rv_op_vse16_v; break;
 +                    }
 +                    break;
 +                case 1: op = rv_op_vsuxei16_v; break;
 +                case 2: op = rv_op_vsse16_v; break;
 +                case 3: op = rv_op_vsoxei16_v; break;
 +                }
 +                break;
 +            case 6:
 +                switch (((inst >> 26) & 0b111)) {
 +                case 0:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: op = rv_op_vse32_v; break;
 +                    }
 +                    break;
 +                case 1: op = rv_op_vsuxei32_v; break;
 +                case 2: op = rv_op_vsse32_v; break;
 +                case 3: op = rv_op_vsoxei32_v; break;
 +                }
 +                break;
 +            case 7:
 +                switch (((inst >> 26) & 0b111)) {
 +                case 0:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: op = rv_op_vse64_v; break;
 +                    }
 +                    break;
 +                case 1: op = rv_op_vsuxei64_v; break;
 +                case 2: op = rv_op_vsse64_v; break;
 +                case 3: op = rv_op_vsoxei64_v; break;
 +                }
 +                break;
              }
              break;
          case 11:
@@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa)
                  break;
              }
              break;
 +        case 21:
 +            switch (((inst >> 12) & 0b111)) {
 +            case 0:
 +                switch (((inst >> 26) & 0b111111)) {
 +                case 0: op = rv_op_vadd_vv; break;
 +                case 2: op = rv_op_vsub_vv; break;
 +                case 4: op = rv_op_vminu_vv; break;
 +                case 5: op = rv_op_vmin_vv; break;
 +                case 6: op = rv_op_vmaxu_vv; break;
 +                case 7: op = rv_op_vmax_vv; break;
 +                case 9: op = rv_op_vand_vv; break;
 +                case 10: op = rv_op_vor_vv; break;
 +                case 11: op = rv_op_vxor_vv; break;
 +                case 12: op = rv_op_vrgather_vv; break;
 +                case 14: op = rv_op_vrgatherei16_vv; break;
 +                case 16: if (((inst >> 25) & 1) == 0) op = rv_op_vadc_vvm; break;
 +                case 17: op = rv_op_vmadc_vvm; break;
 +                case 18: if (((inst >> 25) & 1) == 0) op = rv_op_vsbc_vvm; break;
 +                case 19: op = rv_op_vmsbc_vvm; break;
 +                case 23:
 +                    if (((inst >> 20) & 0b111111) == 32)
 +                        op = rv_op_vmv_v_v;
 +                    else if (((inst >> 25) & 1) == 0)
 +                        op = rv_op_vmerge_vvm;
 +                    break;
 +                case 24: op = rv_op_vmseq_vv; break;
 +                case 25: op = rv_op_vmsne_vv; break;
 +                case 26: op = rv_op_vmsltu_vv; break;
 +                case 27: op = rv_op_vmslt_vv; break;
 +                case 28: op = rv_op_vmsleu_vv; break;
 +                case 29: op = rv_op_vmsle_vv; break;
 +                case 32: op = rv_op_vsaddu_vv; break;
 +                case 33: op = rv_op_vsadd_vv; break;
 +                case 34: op = rv_op_vssubu_vv; break;
 +                case 35: op = rv_op_vssub_vv; break;
 +                case 37: op = rv_op_vsll_vv; break;
 +                case 39: op = rv_op_vsmul_vv; break;
 +                case 40: op = rv_op_vsrl_vv; break;
 +                case 41: op = rv_op_vsra_vv; break;
 +                case 42: op = rv_op_vssrl_vv; break;
 +                case 43: op = rv_op_vssra_vv; break;
 +                case 44: op = rv_op_vnsrl_wv; break;
 +                case 45: op = rv_op_vnsra_wv; break;
 +                case 46: op = rv_op_vnclipu_wv; break;
 +                case 47: op = rv_op_vnclip_wv; break;
 +                case 48: op = rv_op_vwredsumu_vs; break;
 +                case 49: op = rv_op_vwredsum_vs; break;
 +                }
 +                break;
 +            case 1:
 +                switch (((inst >> 26) & 0b111111)) {
 +                case 0: op = rv_op_vfadd_vv; break;
 +                case 1: op = rv_op_vfredusum_vs; break;
 +                case 2: op = rv_op_vfsub_vv; break;
 +                case 3: op = rv_op_vfredosum_vs; break;
 +                case 4: op = rv_op_vfmin_vv; break;
 +                case 5: op = rv_op_vfredmin_vs; break;
 +                case 6: op = rv_op_vfmax_vv; break;
 +                case 7: op = rv_op_vfredmax_vs; break;
 +                case 8: op = rv_op_vfsgnj_vv; break;
 +                case 9: op = rv_op_vfsgnjn_vv; break;
 +                case 10: op = rv_op_vfsgnjx_vv; break;
 +                case 16:
 +                    switch (((inst >> 15) & 0b11111)) {
 +                    case 0: if ((inst >> 25) & 1) op = rv_op_vfmv_f_s; break;
 +                    }
 +                    break;
 +                case 18:
 +                    switch (((inst >> 15) & 0b11111)) {
 +                    case 0: op = rv_op_vfcvt_xu_f_v; break;
 +                    case 1: op = rv_op_vfcvt_x_f_v; break;
 +                    case 2: op = rv_op_vfcvt_f_xu_v; break;
 +                    case 3: op = rv_op_vfcvt_f_x_v; break;
 +                    case 6: op = rv_op_vfcvt_rtz_xu_f_v; break;
 +                    case 7: op = rv_op_vfcvt_rtz_x_f_v; break;
 +                    case 8: op = rv_op_vfwcvt_xu_f_v; break;
 +                    case 9: op = rv_op_vfwcvt_x_f_v; break;
 +                    case 10: op = rv_op_vfwcvt_f_xu_v; break;
 +                    case 11: op = rv_op_vfwcvt_f_x_v; break;
 +                    case 12: op = rv_op_vfwcvt_f_f_v; break;
 +                    case 14: op = rv_op_vfwcvt_rtz_xu_f_v; break;
 +                    case 15: op = rv_op_vfwcvt_rtz_x_f_v; break;
 +                    case 16: op = rv_op_vfncvt_xu_f_w; break;
 +                    case 17: op = rv_op_vfncvt_x_f_w; break;
 +                    case 18: op = rv_op_vfncvt_f_xu_w; break;
 +                    case 19: op = rv_op_vfncvt_f_x_w; break;
 +                    case 20: op = rv_op_vfncvt_f_f_w; break;
 +                    case 21: op = rv_op_vfncvt_rod_f_f_w; break;
 +                    case 22: op = rv_op_vfncvt_rtz_xu_f_w; break;
 +                    case 23: op = rv_op_vfncvt_rtz_x_f_w; break;
 +                    }
 +                    break;
 +                case 19:
 +                    switch (((inst >> 15) & 0b11111)) {
 +                    case 0: op = rv_op_vfsqrt_v; break;
 +                    case 4: op = rv_op_vfrsqrt7_v; break;
 +                    case 5: op = rv_op_vfrec7_v; break;
 +                    case 16: op = rv_op_vfclass_v; break;
 +                    }
 +                    break;
 +                case 24: op = rv_op_vmfeq_vv; break;
 +                case 25: op = rv_op_vmfle_vv; break;
 +                case 27: op = rv_op_vmflt_vv; break;
 +                case 28: op = rv_op_vmfne_vv; break;
 +                case 32: op = rv_op_vfdiv_vv; break;
 +                case 36: op = rv_op_vfmul_vv; break;
 +                case 40: op = rv_op_vfmadd_vv; break;
 +                case 41: op = rv_op_vfnmadd_vv; break;
 +                case 42: op = rv_op_vfmsub_vv; break;
 +                case 43: op = rv_op_vfnmsub_vv; break;
 +                case 44: op = rv_op_vfmacc_vv; break;
 +                case 45: op = rv_op_vfnmacc_vv; break;
 +                case 46: op = rv_op_vfmsac_vv; break;
 +                case 47: op = rv_op_vfnmsac_vv; break;
 +                case 48: op = rv_op_vfwadd_vv; break;
 +                case 49: op = rv_op_vfwredusum_vs; break;
 +                case 50: op = rv_op_vfwsub_vv; break;
 +                case 51: op = rv_op_vfwredosum_vs; break;
 +                case 52: op = rv_op_vfwadd_wv; break;
 +                case 54: op = rv_op_vfwsub_wv; break;
 +                case 56: op = rv_op_vfwmul_vv; break;
 +                case 60: op = rv_op_vfwmacc_vv; break;
 +                case 61: op = rv_op_vfwnmacc_vv; break;
 +                case 62: op = rv_op_vfwmsac_vv; break;
 +                case 63: op = rv_op_vfwnmsac_vv; break;
 +                }
 +                break;
 +            case 2:
 +                switch (((inst >> 26) & 0b111111)) {
 +                case 0: op = rv_op_vredsum_vs; break;
 +                case 1: op = rv_op_vredand_vs; break;
 +                case 2: op = rv_op_vredor_vs; break;
 +                case 3: op = rv_op_vredxor_vs; break;
 +                case 4: op = rv_op_vredminu_vs; break;
 +                case 5: op = rv_op_vredmin_vs; break;
 +                case 6: op = rv_op_vredmaxu_vs; break;
 +                case 7: op = rv_op_vredmax_vs; break;
 +                case 8: op = rv_op_vaaddu_vv; break;
 +                case 9: op = rv_op_vaadd_vv; break;
 +                case 10: op = rv_op_vasubu_vv; break;
 +                case 11: op = rv_op_vasub_vv; break;
 +                case 16:
 +                    switch (((inst >> 15) & 0b11111)) {
 +                    case 0: if ((inst >> 25) & 1) op = rv_op_vmv_x_s; break;
 +                    case 16: op = rv_op_vcpop_m; break;
 +                    case 17: op = rv_op_vfirst_m; break;
 +                    }
 +                    break;
 +                case 18:
 +                    switch (((inst >> 15) & 0b11111)) {
 +                    case 2: op = rv_op_vzext_vf8; break;
 +                    case 3: op = rv_op_vsext_vf8; break;
 +                    case 4: op = rv_op_vzext_vf4; break;
 +                    case 5: op = rv_op_vsext_vf4; break;
 +                    case 6: op = rv_op_vzext_vf2; break;
 +                    case 7: op = rv_op_vsext_vf2; break;
 +                    }
 +                    break;
 +                case 20:
 +                    switch (((inst >> 15) & 0b11111)) {
 +                    case 1: op = rv_op_vmsbf_m;  break;
 +                    case 2: op = rv_op_vmsof_m; break;
 +                    case 3: op = rv_op_vmsif_m; break;
 +                    case 16: op = rv_op_viota_m; break;
 +                    case 17: if (((inst >> 20) & 0b11111) == 0) op = rv_op_vid_v; break;
 +                    }
 +                    break;
 +                case 23: if ((inst >> 25) & 1) op = rv_op_vcompress_vm; break;
 +                case 24: if ((inst >> 25) & 1) op = rv_op_vmandn_mm; break;
 +                case 25: if ((inst >> 25) & 1) op = rv_op_vmand_mm; break;
 +                case 26: if ((inst >> 25) & 1) op = rv_op_vmor_mm; break;
 +                case 27: if ((inst >> 25) & 1) op = rv_op_vmxor_mm; break;
 +                case 28: if ((inst >> 25) & 1) op = rv_op_vmorn_mm; break;
 +                case 29: if ((inst >> 25) & 1) op = rv_op_vmnand_mm; break;
 +                case 30: if ((inst >> 25) & 1) op = rv_op_vmnor_mm; break;
 +                case 31: if ((inst >> 25) & 1) op = rv_op_vmxnor_mm; break;
 +                case 32: op = rv_op_vdivu_vv; break;
 +                case 33: op = rv_op_vdiv_vv; break;
 +                case 34: op = rv_op_vremu_vv; break;
 +                case 35: op = rv_op_vrem_vv; break;
 +                case 36: op = rv_op_vmulhu_vv; break;
 +                case 37: op = rv_op_vmul_vv; break;
 +                case 38: op = rv_op_vmulhsu_vv; break;
 +                case 39: op = rv_op_vmulh_vv; break;
 +                case 41: op = rv_op_vmadd_vv; break;
 +                case 43: op = rv_op_vnmsub_vv; break;
 +                case 45: op = rv_op_vmacc_vv; break;
 +                case 47: op = rv_op_vnmsac_vv; break;
 +                case 48: op = rv_op_vwaddu_vv; break;
 +                case 49: op = rv_op_vwadd_vv; break;
 +                case 50: op = rv_op_vwsubu_vv; break;
 +                case 51: op = rv_op_vwsub_vv; break;
 +                case 52: op = rv_op_vwaddu_wv; break;
 +                case 53: op = rv_op_vwadd_wv; break;
 +                case 54: op = rv_op_vwsubu_wv; break;
 +                case 55: op = rv_op_vwsub_wv; break;
 +                case 56: op = rv_op_vwmulu_vv; break;
 +                case 58: op = rv_op_vwmulsu_vv; break;
 +                case 59: op = rv_op_vwmul_vv; break;
 +                case 60: op = rv_op_vwmaccu_vv; break;
 +                case 61: op = rv_op_vwmacc_vv; break;
 +                case 63: op = rv_op_vwmaccsu_vv; break;
 +                }
 +                break;
 +            case 3:
 +                switch (((inst >> 26) & 0b111111)) {
 +                case 0: op = rv_op_vadd_vi; break;
 +                case 3: op = rv_op_vrsub_vi; break;
 +                case 9: op = rv_op_vand_vi; break;
 +                case 10: op = rv_op_vor_vi; break;
 +                case 11: op = rv_op_vxor_vi; break;
 +                case 12: op = rv_op_vrgather_vi; break;
 +                case 14: op = rv_op_vslideup_vi; break;
 +                case 15: op = rv_op_vslidedown_vi; break;
 +                case 16: if (((inst >> 25) & 1) == 0) op = rv_op_vadc_vim; break;
 +                case 17: op = rv_op_vmadc_vim; break;
 +                case 23:
 +                    if (((inst >> 20) & 0b111111) == 32)
 +                        op = rv_op_vmv_v_i;
 +                    else if (((inst >> 25) & 1) == 0)
 +                        op = rv_op_vmerge_vim;
 +                    break;
 +                case 24: op = rv_op_vmseq_vi; break;
 +                case 25: op = rv_op_vmsne_vi; break;
 +                case 28: op = rv_op_vmsleu_vi; break;
 +                case 29: op = rv_op_vmsle_vi; break;
 +                case 30: op = rv_op_vmsgtu_vi; break;
 +                case 31: op = rv_op_vmsgt_vi; break;
 +                case 32: op = rv_op_vsaddu_vi; break;
 +                case 33: op = rv_op_vsadd_vi; break;
 +                case 37: op = rv_op_vsll_vi; break;
 +                case 39:
 +                    switch (((inst >> 15) & 0b11111)) {
 +                    case 0: op = rv_op_vmv1r_v; break;
 +                    case 1: op = rv_op_vmv2r_v; break;
 +                    case 3: op = rv_op_vmv4r_v; break;
 +                    case 7: op = rv_op_vmv8r_v; break;
 +                    }
 +                    break;
 +                case 40: op = rv_op_vsrl_vi; break;
 +                case 41: op = rv_op_vsra_vi; break;
 +                case 42: op = rv_op_vssrl_vi; break;
 +                case 43: op = rv_op_vssra_vi; break;
 +                case 44: op = rv_op_vnsrl_wi; break;
 +                case 45: op = rv_op_vnsra_wi; break;
 +                case 46: op = rv_op_vnclipu_wi; break;
 +                case 47: op = rv_op_vnclip_wi; break;
 +                }
 +                break;
 +            case 4:
 +                switch (((inst >> 26) & 0b111111)) {
 +                case 0: op = rv_op_vadd_vx; break;
 +                case 2: op = rv_op_vsub_vx; break;
 +                case 3: op = rv_op_vrsub_vx; break;
 +                case 4: op = rv_op_vminu_vx; break;
 +                case 5: op = rv_op_vmin_vx; break;
 +                case 6: op = rv_op_vmaxu_vx; break;
 +                case 7: op = rv_op_vmax_vx; break;
 +                case 9: op = rv_op_vand_vx; break;
 +                case 10: op = rv_op_vor_vx; break;
 +                case 11: op = rv_op_vxor_vx; break;
 +                case 12: op = rv_op_vrgather_vx; break;
 +                case 14: op = rv_op_vslideup_vx; break;
 +                case 15: op = rv_op_vslidedown_vx; break;
 +                case 16: if (((inst >> 25) & 1) == 0) op = rv_op_vadc_vxm; break;
 +                case 17: op = rv_op_vmadc_vxm; break;
 +                case 18: if (((inst >> 25) & 1) == 0) op = rv_op_vsbc_vxm; break;
 +                case 19: op = rv_op_vmsbc_vxm; break;
 +                case 23:
 +                    if (((inst >> 20) & 0b111111) == 32)
 +                        op = rv_op_vmv_v_x;
 +                    else if (((inst >> 25) & 1) == 0)
 +                        op = rv_op_vmerge_vxm;
 +                    break;
 +                case 24: op = rv_op_vmseq_vx; break;
 +                case 25: op = rv_op_vmsne_vx; break;
 +                case 26: op = rv_op_vmsltu_vx; break;
 +                case 27: op = rv_op_vmslt_vx; break;
 +                case 28: op = rv_op_vmsleu_vx; break;
 +                case 29: op = rv_op_vmsle_vx; break;
 +                case 30: op = rv_op_vmsgtu_vx; break;
 +                case 31: op = rv_op_vmsgt_vx; break;
 +                case 32: op = rv_op_vsaddu_vx; break;
 +                case 33: op = rv_op_vsadd_vx; break;
 +                case 34: op = rv_op_vssubu_vx; break;
 +                case 35: op = rv_op_vssub_vx; break;
 +                case 37: op = rv_op_vsll_vx; break;
 +                case 39: op = rv_op_vsmul_vx; break;
 +                case 40: op = rv_op_vsrl_vx; break;
 +                case 41: op = rv_op_vsra_vx; break;
 +                case 42: op = rv_op_vssrl_vx; break;
 +                case 43: op = rv_op_vssra_vx; break;
 +                case 44: op = rv_op_vnsrl_wx; break;
 +                case 45: op = rv_op_vnsra_wx; break;
 +                case 46: op = rv_op_vnclipu_wx; break;
 +                case 47: op = rv_op_vnclip_wx; break;
 +                }
 +                break;
 +            case 5:
 +                switch (((inst >> 26) & 0b111111)) {
 +                case 0: op = rv_op_vfadd_vf; break;
 +                case 2: op = rv_op_vfsub_vf; break;
 +                case 4: op = rv_op_vfmin_vf; break;
 +                case 6: op = rv_op_vfmax_vf; break;
 +                case 8: op = rv_op_vfsgnj_vf; break;
 +                case 9: op = rv_op_vfsgnjn_vf; break;
 +                case 10: op = rv_op_vfsgnjx_vf; break;
 +                case 14: op = rv_op_vfslide1up_vf; break;
 +                case 15: op = rv_op_vfslide1down_vf; break;
 +                case 16:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: if ((inst >> 25) & 1) op = rv_op_vfmv_s_f; break;
 +                    }
 +                    break;
 +                case 23:
 +                    if (((inst >> 25) & 1) == 0)
 +                        op = rv_op_vfmerge_vfm;
 +                    else if (((inst >> 20) & 0b111111) == 32)
 +                        op = rv_op_vfmv_v_f;
 +                    break;
 +                case 24: op = rv_op_vmfeq_vf; break;
 +                case 25: op = rv_op_vmfle_vf; break;
 +                case 27: op = rv_op_vmflt_vf; break;
 +                case 28: op = rv_op_vmfne_vf; break;
 +                case 29: op = rv_op_vmfgt_vf; break;
 +                case 31: op = rv_op_vmfge_vf; break;
 +                case 32: op = rv_op_vfdiv_vf; break;
 +                case 33: op = rv_op_vfrdiv_vf; break;
 +                case 36: op = rv_op_vfmul_vf; break;
 +                case 39: op = rv_op_vfrsub_vf; break;
 +                case 40: op = rv_op_vfmadd_vf; break;
 +                case 41: op = rv_op_vfnmadd_vf; break;
 +                case 42: op = rv_op_vfmsub_vf; break;
 +                case 43: op = rv_op_vfnmsub_vf; break;
 +                case 44: op = rv_op_vfmacc_vf; break;
 +                case 45: op = rv_op_vfnmacc_vf; break;
 +                case 46: op = rv_op_vfmsac_vf; break;
 +                case 47: op = rv_op_vfnmsac_vf; break;
 +                case 48: op = rv_op_vfwadd_vf; break;
 +                case 50: op = rv_op_vfwsub_vf; break;
 +                case 52: op = rv_op_vfwadd_wf; break;
 +                case 54: op = rv_op_vfwsub_wf; break;
 +                case 56: op = rv_op_vfwmul_vf; break;
 +                case 60: op = rv_op_vfwmacc_vf; break;
 +                case 61: op = rv_op_vfwnmacc_vf; break;
 +                case 62: op = rv_op_vfwmsac_vf; break;
 +                case 63: op = rv_op_vfwnmsac_vf; break;
 +                }
 +                break;
 +            case 6:
 +                switch (((inst >> 26) & 0b111111)) {
 +                case 8: op = rv_op_vaaddu_vx; break;
 +                case 9: op = rv_op_vaadd_vx; break;
 +                case 10: op = rv_op_vasubu_vx; break;
 +                case 11: op = rv_op_vasub_vx; break;
 +                case 14: op = rv_op_vslide1up_vx; break;
 +                case 15: op = rv_op_vslide1down_vx; break;
 +                case 16:
 +                    switch (((inst >> 20) & 0b11111)) {
 +                    case 0: if ((inst >> 25) & 1) op = rv_op_vmv_s_x; break;
 +                    }
 +                    break;
 +                case 32: op = rv_op_vdivu_vx; break;
 +                case 33: op = rv_op_vdiv_vx; break;
 +                case 34: op = rv_op_vremu_vx; break;
 +                case 35: op = rv_op_vrem_vx; break;
 +                case 36: op = rv_op_vmulhu_vx; break;
 +                case 37: op = rv_op_vmul_vx; break;
 +                case 38: op = rv_op_vmulhsu_vx; break;
 +                case 39: op = rv_op_vmulh_vx; break;
 +                case 41: op = rv_op_vmadd_vx; break;
 +                case 43: op = rv_op_vnmsub_vx; break;
 +                case 45: op = rv_op_vmacc_vx; break;
 +                case 47: op = rv_op_vnmsac_vx; break;
 +                case 48: op = rv_op_vwaddu_vx; break;
 +                case 49: op = rv_op_vwadd_vx; break;
 +                case 50: op = rv_op_vwsubu_vx; break;
 +                case 51: op = rv_op_vwsub_vx; break;
 +                case 52: op = rv_op_vwaddu_wx; break;
 +                case 53: op = rv_op_vwadd_wx; break;
 +                case 54: op = rv_op_vwsubu_wx; break;
 +                case 55: op = rv_op_vwsub_wx; break;
 +                case 56: op = rv_op_vwmulu_vx; break;
 +                case 58: op = rv_op_vwmulsu_vx; break;
 +                case 59: op = rv_op_vwmul_vx; break;
 +                case 60: op = rv_op_vwmaccu_vx; break;
 +                case 61: op = rv_op_vwmacc_vx; break;
 +                case 62: op = rv_op_vwmaccus_vx; break;
 +                case 63: op = rv_op_vwmaccsu_vx; break;
 +                }
 +                break;
 +            case 7:
 +                if (((inst >> 31) & 1) == 0) {
 +                    op = rv_op_vsetvli;
 +                } else if ((inst >> 30) & 1) {
 +                    op = rv_op_vsetivli;
 +                } else if (((inst >> 25) & 0b11111) == 0) {
 +                    op = rv_op_vsetvl;
 +                }
 +                break;
 +            }
 +            break;
          case 22:
              switch (((inst >> 12) & 0b111)) {
              case 0: op = rv_op_addid; break;
@@ -XXX,XX +XXX,XX @@ static uint32_t operand_cimmq(rv_inst inst)
          ((inst << 57) >> 62) << 6;
  }
 +static uint32_t operand_vimm(rv_inst inst)
 +{
-+    return (int64_t)(inst << 44) >> 59;
++    val = ((val & 0x5555555555555555ull) << 1) |
 +          ((val & 0xAAAAAAAAAAAAAAAAull) >> 1);
 +    val = ((val & 0x3333333333333333ull) << 2) |
 +          ((val & 0xCCCCCCCCCCCCCCCCull) >> 2);
 +    val = ((val & 0x0F0F0F0F0F0F0F0Full) << 4) |
 +          ((val & 0xF0F0F0F0F0F0F0F0ull) >> 4);
 +
 +    return val;
 +}
 +
-+static uint32_t operand_vzimm11(rv_inst inst)
++RVVCALL(OPIVV1, vbrev8_v_b, OP_UU_B, H1, H1, brev8)
 +RVVCALL(OPIVV1, vbrev8_v_h, OP_UU_H, H2, H2, brev8)
 +RVVCALL(OPIVV1, vbrev8_v_w, OP_UU_W, H4, H4, brev8)
 +RVVCALL(OPIVV1, vbrev8_v_d, OP_UU_D, H8, H8, brev8)
 +GEN_VEXT_V(vbrev8_v_b, 1)
 +GEN_VEXT_V(vbrev8_v_h, 2)
 +GEN_VEXT_V(vbrev8_v_w, 4)
 +GEN_VEXT_V(vbrev8_v_d, 8)
 +
 +#define DO_IDENTITY(a) (a)
 +RVVCALL(OPIVV1, vrev8_v_b, OP_UU_B, H1, H1, DO_IDENTITY)
 +RVVCALL(OPIVV1, vrev8_v_h, OP_UU_H, H2, H2, bswap16)
 +RVVCALL(OPIVV1, vrev8_v_w, OP_UU_W, H4, H4, bswap32)
 +RVVCALL(OPIVV1, vrev8_v_d, OP_UU_D, H8, H8, bswap64)
 +GEN_VEXT_V(vrev8_v_b, 1)
 +GEN_VEXT_V(vrev8_v_h, 2)
 +GEN_VEXT_V(vrev8_v_w, 4)
 +GEN_VEXT_V(vrev8_v_d, 8)
 +
 +#define DO_ANDN(a, b) ((a) & ~(b))
 +RVVCALL(OPIVV2, vandn_vv_b, OP_UUU_B, H1, H1, H1, DO_ANDN)
 +RVVCALL(OPIVV2, vandn_vv_h, OP_UUU_H, H2, H2, H2, DO_ANDN)
 +RVVCALL(OPIVV2, vandn_vv_w, OP_UUU_W, H4, H4, H4, DO_ANDN)
 +RVVCALL(OPIVV2, vandn_vv_d, OP_UUU_D, H8, H8, H8, DO_ANDN)
 +GEN_VEXT_VV(vandn_vv_b, 1)
 +GEN_VEXT_VV(vandn_vv_h, 2)
 +GEN_VEXT_VV(vandn_vv_w, 4)
 +GEN_VEXT_VV(vandn_vv_d, 8)
 +
 +RVVCALL(OPIVX2, vandn_vx_b, OP_UUU_B, H1, H1, DO_ANDN)
 +RVVCALL(OPIVX2, vandn_vx_h, OP_UUU_H, H2, H2, DO_ANDN)
 +RVVCALL(OPIVX2, vandn_vx_w, OP_UUU_W, H4, H4, DO_ANDN)
 +RVVCALL(OPIVX2, vandn_vx_d, OP_UUU_D, H8, H8, DO_ANDN)
 +GEN_VEXT_VX(vandn_vx_b, 1)
 +GEN_VEXT_VX(vandn_vx_h, 2)
 +GEN_VEXT_VX(vandn_vx_w, 4)
 +GEN_VEXT_VX(vandn_vx_d, 8)
 +
 +RVVCALL(OPIVV1, vbrev_v_b, OP_UU_B, H1, H1, revbit8)
 +RVVCALL(OPIVV1, vbrev_v_h, OP_UU_H, H2, H2, revbit16)
 +RVVCALL(OPIVV1, vbrev_v_w, OP_UU_W, H4, H4, revbit32)
 +RVVCALL(OPIVV1, vbrev_v_d, OP_UU_D, H8, H8, revbit64)
 +GEN_VEXT_V(vbrev_v_b, 1)
 +GEN_VEXT_V(vbrev_v_h, 2)
 +GEN_VEXT_V(vbrev_v_w, 4)
 +GEN_VEXT_V(vbrev_v_d, 8)
 +
 +RVVCALL(OPIVV1, vclz_v_b, OP_UU_B, H1, H1, clz8)
 +RVVCALL(OPIVV1, vclz_v_h, OP_UU_H, H2, H2, clz16)
 +RVVCALL(OPIVV1, vclz_v_w, OP_UU_W, H4, H4, clz32)
 +RVVCALL(OPIVV1, vclz_v_d, OP_UU_D, H8, H8, clz64)
 +GEN_VEXT_V(vclz_v_b, 1)
 +GEN_VEXT_V(vclz_v_h, 2)
 +GEN_VEXT_V(vclz_v_w, 4)
 +GEN_VEXT_V(vclz_v_d, 8)
 +
 +RVVCALL(OPIVV1, vctz_v_b, OP_UU_B, H1, H1, ctz8)
 +RVVCALL(OPIVV1, vctz_v_h, OP_UU_H, H2, H2, ctz16)
 +RVVCALL(OPIVV1, vctz_v_w, OP_UU_W, H4, H4, ctz32)
 +RVVCALL(OPIVV1, vctz_v_d, OP_UU_D, H8, H8, ctz64)
 +GEN_VEXT_V(vctz_v_b, 1)
 +GEN_VEXT_V(vctz_v_h, 2)
 +GEN_VEXT_V(vctz_v_w, 4)
 +GEN_VEXT_V(vctz_v_d, 8)
 +
 +RVVCALL(OPIVV1, vcpop_v_b, OP_UU_B, H1, H1, ctpop8)
 +RVVCALL(OPIVV1, vcpop_v_h, OP_UU_H, H2, H2, ctpop16)
 +RVVCALL(OPIVV1, vcpop_v_w, OP_UU_W, H4, H4, ctpop32)
 +RVVCALL(OPIVV1, vcpop_v_d, OP_UU_D, H8, H8, ctpop64)
 +GEN_VEXT_V(vcpop_v_b, 1)
 +GEN_VEXT_V(vcpop_v_h, 2)
 +GEN_VEXT_V(vcpop_v_w, 4)
 +GEN_VEXT_V(vcpop_v_d, 8)
 +
 +#define DO_SLL(N, M) (N << (M & (sizeof(N) * 8 - 1)))
 +RVVCALL(OPIVV2, vwsll_vv_b, WOP_UUU_B, H2, H1, H1, DO_SLL)
 +RVVCALL(OPIVV2, vwsll_vv_h, WOP_UUU_H, H4, H2, H2, DO_SLL)
 +RVVCALL(OPIVV2, vwsll_vv_w, WOP_UUU_W, H8, H4, H4, DO_SLL)
 +GEN_VEXT_VV(vwsll_vv_b, 2)
 +GEN_VEXT_VV(vwsll_vv_h, 4)
 +GEN_VEXT_VV(vwsll_vv_w, 8)
 +
 +RVVCALL(OPIVX2, vwsll_vx_b, WOP_UUU_B, H2, H1, DO_SLL)
 +RVVCALL(OPIVX2, vwsll_vx_h, WOP_UUU_H, H4, H2, DO_SLL)
 +RVVCALL(OPIVX2, vwsll_vx_w, WOP_UUU_W, H8, H4, DO_SLL)
 +GEN_VEXT_VX(vwsll_vx_b, 2)
 +GEN_VEXT_VX(vwsll_vx_h, 4)
 +GEN_VEXT_VX(vwsll_vx_w, 8)
 diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/insn_trans/trans_rvvk.c.inc
 +++ b/target/riscv/insn_trans/trans_rvvk.c.inc
@@ -XXX,XX +XXX,XX @@ static bool vclmul_vx_check(DisasContext *s, arg_rmrr *a)
  GEN_VX_MASKED_TRANS(vclmul_vx, vclmul_vx_check)
  GEN_VX_MASKED_TRANS(vclmulh_vx, vclmul_vx_check)
 +
 +/*
 + * Zvbb
 + */
 +
 +#define GEN_OPIVI_GVEC_TRANS_CHECK(NAME, IMM_MODE, OPIVX, SUF, CHECK)   \
 +    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)              \
 +    {                                                                   \
 +        if (CHECK(s, a)) {                                              \
 +            static gen_helper_opivx *const fns[4] = {                   \
 +                gen_helper_##OPIVX##_b,                                 \
 +                gen_helper_##OPIVX##_h,                                 \
 +                gen_helper_##OPIVX##_w,                                 \
 +                gen_helper_##OPIVX##_d,                                 \
 +            };                                                          \
 +            return do_opivi_gvec(s, a, tcg_gen_gvec_##SUF, fns[s->sew], \
 +                                 IMM_MODE);                             \
 +        }                                                               \
 +        return false;                                                   \
 +    }
 +
 +#define GEN_OPIVV_GVEC_TRANS_CHECK(NAME, SUF, CHECK)                     \
 +    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)               \
 +    {                                                                    \
 +        if (CHECK(s, a)) {                                               \
 +            static gen_helper_gvec_4_ptr *const fns[4] = {               \
 +                gen_helper_##NAME##_b,                                   \
 +                gen_helper_##NAME##_h,                                   \
 +                gen_helper_##NAME##_w,                                   \
 +                gen_helper_##NAME##_d,                                   \
 +            };                                                           \
 +            return do_opivv_gvec(s, a, tcg_gen_gvec_##SUF, fns[s->sew]); \
 +        }                                                                \
 +        return false;                                                    \
 +    }
 +
 +#define GEN_OPIVX_GVEC_SHIFT_TRANS_CHECK(NAME, SUF, CHECK)       \
 +    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)       \
 +    {                                                            \
 +        if (CHECK(s, a)) {                                       \
 +            static gen_helper_opivx *const fns[4] = {            \
 +                gen_helper_##NAME##_b,                           \
 +                gen_helper_##NAME##_h,                           \
 +                gen_helper_##NAME##_w,                           \
 +                gen_helper_##NAME##_d,                           \
 +            };                                                   \
 +            return do_opivx_gvec_shift(s, a, tcg_gen_gvec_##SUF, \
 +                                       fns[s->sew]);             \
 +        }                                                        \
 +        return false;                                            \
 +    }
 +
 +static bool zvbb_vv_check(DisasContext *s, arg_rmrr *a)
 +{
-+    return (inst << 33) >> 53;
++    return opivv_check(s, a) && s->cfg_ptr->ext_zvbb == true;
 +}
 +
-+static uint32_t operand_vzimm10(rv_inst inst)
++static bool zvbb_vx_check(DisasContext *s, arg_rmrr *a)
 +{
-+    return (inst << 34) >> 54;
++    return opivx_check(s, a) && s->cfg_ptr->ext_zvbb == true;
 +}
 +
- static uint32_t operand_bs(rv_inst inst)
++/* vrol.v[vx] */
- {
++GEN_OPIVV_GVEC_TRANS_CHECK(vrol_vv, rotlv, zvbb_vv_check)
-     return (inst << 32) >> 62;
++GEN_OPIVX_GVEC_SHIFT_TRANS_CHECK(vrol_vx, rotls, zvbb_vx_check)
-@@ -XXX,XX +XXX,XX @@ static uint32_t operand_rnum(rv_inst inst)
++
-     return (inst << 40) >> 60;
++/* vror.v[vxi] */
- }
++GEN_OPIVV_GVEC_TRANS_CHECK(vror_vv, rotrv, zvbb_vv_check)
++GEN_OPIVX_GVEC_SHIFT_TRANS_CHECK(vror_vx, rotrs, zvbb_vx_check)
-+static uint32_t operand_vm(rv_inst inst)
++GEN_OPIVI_GVEC_TRANS_CHECK(vror_vi, IMM_TRUNC_SEW, vror_vx, rotri, zvbb_vx_check)
 +
 +#define GEN_OPIVX_GVEC_TRANS_CHECK(NAME, SUF, CHECK)                     \
 +    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)               \
 +    {                                                                    \
 +        if (CHECK(s, a)) {                                               \
 +            static gen_helper_opivx *const fns[4] = {                    \
 +                gen_helper_##NAME##_b,                                   \
 +                gen_helper_##NAME##_h,                                   \
 +                gen_helper_##NAME##_w,                                   \
 +                gen_helper_##NAME##_d,                                   \
 +            };                                                           \
 +            return do_opivx_gvec(s, a, tcg_gen_gvec_##SUF, fns[s->sew]); \
 +        }                                                                \
 +        return false;                                                    \
 +    }
 +
 +/* vandn.v[vx] */
 +GEN_OPIVV_GVEC_TRANS_CHECK(vandn_vv, andc, zvbb_vv_check)
 +GEN_OPIVX_GVEC_TRANS_CHECK(vandn_vx, andcs, zvbb_vx_check)
 +
 +#define GEN_OPIV_TRANS(NAME, CHECK)                                        \
 +    static bool trans_##NAME(DisasContext *s, arg_rmr *a)                  \
 +    {                                                                      \
 +        if (CHECK(s, a)) {                                                 \
 +            uint32_t data = 0;                                             \
 +            static gen_helper_gvec_3_ptr *const fns[4] = {                 \
 +                gen_helper_##NAME##_b,                                     \
 +                gen_helper_##NAME##_h,                                     \
 +                gen_helper_##NAME##_w,                                     \
 +                gen_helper_##NAME##_d,                                     \
 +            };                                                             \
 +            TCGLabel *over = gen_new_label();                              \
 +            tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);     \
 +                                                                           \
 +            data = FIELD_DP32(data, VDATA, VM, a->vm);                     \
 +            data = FIELD_DP32(data, VDATA, LMUL, s->lmul);                 \
 +            data = FIELD_DP32(data, VDATA, VTA, s->vta);                   \
 +            data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s); \
 +            data = FIELD_DP32(data, VDATA, VMA, s->vma);                   \
 +            tcg_gen_gvec_3_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, 0),         \
 +                               vreg_ofs(s, a->rs2), cpu_env,               \
 +                               s->cfg_ptr->vlen / 8, s->cfg_ptr->vlen / 8, \
 +                               data, fns[s->sew]);                         \
 +            mark_vs_dirty(s);                                              \
 +            gen_set_label(over);                                           \
 +            return true;                                                   \
 +        }                                                                  \
 +        return false;                                                      \
 +    }
 +
 +static bool zvbb_opiv_check(DisasContext *s, arg_rmr *a)
 +{
-+    return (inst << 38) >> 63;
++    return s->cfg_ptr->ext_zvbb == true &&
 +           require_rvv(s) &&
 +           vext_check_isa_ill(s) &&
 +           vext_check_ss(s, a->rd, a->rs2, a->vm);
 +}
 +
- /* decode operands */
++GEN_OPIV_TRANS(vbrev8_v, zvbb_opiv_check)
++GEN_OPIV_TRANS(vrev8_v, zvbb_opiv_check)
- static void decode_inst_operands(rv_decode *dec, rv_isa isa)
++GEN_OPIV_TRANS(vbrev_v, zvbb_opiv_check)
-@@ -XXX,XX +XXX,XX @@ static void decode_inst_operands(rv_decode *dec, rv_isa isa)
++GEN_OPIV_TRANS(vclz_v, zvbb_opiv_check)
-         dec->rs1 = operand_rs1(inst);
++GEN_OPIV_TRANS(vctz_v, zvbb_opiv_check)
-         dec->rnum = operand_rnum(inst);
++GEN_OPIV_TRANS(vcpop_v, zvbb_opiv_check)
-         break;
++
-+    case rv_codec_v_r:
++static bool vwsll_vv_check(DisasContext *s, arg_rmrr *a)
-+        dec->rd = operand_rd(inst);
++{
-+        dec->rs1 = operand_rs1(inst);
++    return s->cfg_ptr->ext_zvbb && opivv_widen_check(s, a);
-+        dec->rs2 = operand_rs2(inst);
++}
-+        dec->vm = operand_vm(inst);
++
-+        break;
++static bool vwsll_vx_check(DisasContext *s, arg_rmrr *a)
-+    case rv_codec_v_ldst:
++{
-+        dec->rd = operand_rd(inst);
++    return s->cfg_ptr->ext_zvbb && opivx_widen_check(s, a);
-+        dec->rs1 = operand_rs1(inst);
++}
-+        dec->vm = operand_vm(inst);
++
-+        break;
++/* OPIVI without GVEC IR */
-+    case rv_codec_v_i:
++#define GEN_OPIVI_WIDEN_TRANS(NAME, IMM_MODE, OPIVX, CHECK)                  \
-+        dec->rd = operand_rd(inst);
++    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)                   \
-+        dec->rs2 = operand_rs2(inst);
++    {                                                                        \
-+        dec->imm = operand_vimm(inst);
++        if (CHECK(s, a)) {                                                   \
-+        dec->vm = operand_vm(inst);
++            static gen_helper_opivx *const fns[3] = {                        \
-+        break;
++                gen_helper_##OPIVX##_b,                                      \
-+    case rv_codec_vsetvli:
++                gen_helper_##OPIVX##_h,                                      \
-+        dec->rd = operand_rd(inst);
++                gen_helper_##OPIVX##_w,                                      \
-+        dec->rs1 = operand_rs1(inst);
++            };                                                               \
-+        dec->vzimm = operand_vzimm11(inst);
++            return opivi_trans(a->rd, a->rs1, a->rs2, a->vm, fns[s->sew], s, \
-+        break;
++                               IMM_MODE);                                    \
-+    case rv_codec_vsetivli:
++        }                                                                    \
-+        dec->rd = operand_rd(inst);
++        return false;                                                        \
-+        dec->imm = operand_vimm(inst);
++    }
-+        dec->vzimm = operand_vzimm10(inst);
++
-+        break;
++GEN_OPIVV_WIDEN_TRANS(vwsll_vv, vwsll_vv_check)
-     };
++GEN_OPIVX_WIDEN_TRANS(vwsll_vx, vwsll_vx_check)
- }
++GEN_OPIVI_WIDEN_TRANS(vwsll_vi, IMM_ZX, vwsll_vx, vwsll_vx_check)
@@ -XXX,XX +XXX,XX @@ static void format_inst(char *buf, size_t buflen, size_t tab, rv_decode *dec)
              snprintf(tmp, sizeof(tmp), "%d", dec->imm);
              append(buf, tmp, buflen);
              break;
 +        case 'u':
 +            snprintf(tmp, sizeof(tmp), "%u", ((uint32_t)dec->imm & 0b11111));
 +            append(buf, tmp, buflen);
 +            break;
          case 'o':
              snprintf(tmp, sizeof(tmp), "%d", dec->imm);
              append(buf, tmp, buflen);
@@ -XXX,XX +XXX,XX @@ static void format_inst(char *buf, size_t buflen, size_t tab, rv_decode *dec)
                  append(buf, ".rl", buflen);
              }
              break;
 +        case 'l':
 +            append(buf, ",v0", buflen);
 +            break;
 +        case 'm':
 +            if (dec->vm == 0) {
 +                append(buf, ",v0.t", buflen);
 +            }
 +            break;
 +        case 'D':
 +            append(buf, rv_vreg_name_sym[dec->rd], buflen);
 +            break;
 +        case 'E':
 +            append(buf, rv_vreg_name_sym[dec->rs1], buflen);
 +            break;
 +        case 'F':
 +            append(buf, rv_vreg_name_sym[dec->rs2], buflen);
 +            break;
 +        case 'G':
 +            append(buf, rv_vreg_name_sym[dec->rs3], buflen);
 +            break;
 +        case 'v': {
 +            char nbuf[32] = {0};
 +            const int sew = 1 << (((dec->vzimm >> 3) & 0b111) + 3);
 +            sprintf(nbuf, "%d", sew);
 +            const int lmul = dec->vzimm & 0b11;
 +            const int flmul = (dec->vzimm >> 2) & 1;
 +            const char *vta = (dec->vzimm >> 6) & 1 ? "ta" : "tu";
 +            const char *vma = (dec->vzimm >> 7) & 1 ? "ma" : "mu";
 +            append(buf, "e", buflen);
 +            append(buf, nbuf, buflen);
 +            append(buf, ",m", buflen);
 +            if (flmul) {
 +                switch (lmul) {
 +                case 3:
 +                    sprintf(nbuf, "f2");
 +                    break;
 +                case 2:
 +                    sprintf(nbuf, "f4");
 +                    break;
 +                case 1:
 +                    sprintf(nbuf, "f8");
 +                break;
 +                }
 +                append(buf, nbuf, buflen);
 +            } else {
 +                sprintf(nbuf, "%d", 1 << lmul);
 +                append(buf, nbuf, buflen);
 +            }
 +            append(buf, ",", buflen);
 +            append(buf, vta, buflen);
 +            append(buf, ",", buflen);
 +            append(buf, vma, buflen);
 +            break;
 +        }
          default:
              break;
          }
@@ -XXX,XX +XXX,XX @@ disasm_inst(char *buf, size_t buflen, rv_isa isa, uint64_t pc, rv_inst inst)
      decode_inst_operands(&dec, isa);
      decode_inst_decompress(&dec, isa);
      decode_inst_lift_pseudo(&dec);
 -    format_inst(buf, buflen, 16, &dec);
 +    format_inst(buf, buflen, 24, &dec);
  }
  #define INST_FMT_2 "%04" PRIx64 "              "
 --
-.37.3
+.41.0

-New patch
+[PULL v2 16/45] target/riscv: Add Zvkned ISA extension support
+From: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
+This commit adds support for the Zvkned vector-crypto extension, which
+consists of the following instructions:
+* vaesef.[vv,vs]
+* vaesdf.[vv,vs]
+* vaesdm.[vv,vs]
+* vaesz.vs
+* vaesem.[vv,vs]
+* vaeskf1.vi
+* vaeskf2.vi
+Translation functions are defined in
+`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
+`target/riscv/vcrypto_helper.c`.
+Co-authored-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
+Co-authored-by: William Salmon <will.salmon@codethink.co.uk>
+[max.chou@sifive.com: Replaced vstart checking by TCG op]
+Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
+Signed-off-by: William Salmon <will.salmon@codethink.co.uk>
+Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+[max.chou@sifive.com: Imported aes-round.h and exposed x-zvkned
+property]
+[max.chou@sifive.com: Fixed endian issues and replaced the vstart & vl
+egs checking by helper function]
+[max.chou@sifive.com: Replaced bswap32 calls in aes key expanding]
+Message-ID: <20230711165917.2629866-10-max.chou@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/cpu_cfg.h                   |   1 +
+ target/riscv/helper.h                    |  14 ++
+ target/riscv/insn32.decode               |  14 ++
+ target/riscv/cpu.c                       |   4 +-
+ target/riscv/vcrypto_helper.c            | 202 +++++++++++++++++++++++
+ target/riscv/insn_trans/trans_rvvk.c.inc | 147 +++++++++++++++++
+files changed, 381 insertions(+), 1 deletion(-)
+diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu_cfg.h
++++ b/target/riscv/cpu_cfg.h
+@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
+     bool ext_zve64d;
+     bool ext_zvbb;
+     bool ext_zvbc;
++    bool ext_zvkned;
+     bool ext_zmmul;
+     bool ext_zvfbfmin;
+     bool ext_zvfbfwma;
+diff --git a/target/riscv/helper.h b/target/riscv/helper.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/helper.h
++++ b/target/riscv/helper.h
+@@ -XXX,XX +XXX,XX @@ DEF_HELPER_6(vandn_vx_b, void, ptr, ptr, tl, ptr, env, i32)
+ DEF_HELPER_6(vandn_vx_h, void, ptr, ptr, tl, ptr, env, i32)
+ DEF_HELPER_6(vandn_vx_w, void, ptr, ptr, tl, ptr, env, i32)
+ DEF_HELPER_6(vandn_vx_d, void, ptr, ptr, tl, ptr, env, i32)
++
++DEF_HELPER_2(egs_check, void, i32, env)
++
++DEF_HELPER_4(vaesef_vv, void, ptr, ptr, env, i32)
++DEF_HELPER_4(vaesef_vs, void, ptr, ptr, env, i32)
++DEF_HELPER_4(vaesdf_vv, void, ptr, ptr, env, i32)
++DEF_HELPER_4(vaesdf_vs, void, ptr, ptr, env, i32)
++DEF_HELPER_4(vaesem_vv, void, ptr, ptr, env, i32)
++DEF_HELPER_4(vaesem_vs, void, ptr, ptr, env, i32)
++DEF_HELPER_4(vaesdm_vv, void, ptr, ptr, env, i32)
++DEF_HELPER_4(vaesdm_vs, void, ptr, ptr, env, i32)
++DEF_HELPER_4(vaesz_vs, void, ptr, ptr, env, i32)
++DEF_HELPER_5(vaeskf1_vi, void, ptr, ptr, i32, env, i32)
++DEF_HELPER_5(vaeskf2_vi, void, ptr, ptr, i32, env, i32)
+diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/insn32.decode
++++ b/target/riscv/insn32.decode
+@@ -XXX,XX +XXX,XX @@
+ @r_rm    .......   ..... ..... ... ..... ....... %rs2 %rs1 %rm %rd
+ @r2_rm   .......   ..... ..... ... ..... ....... %rs1 %rm %rd
+ @r2      .......   ..... ..... ... ..... ....... &r2 %rs1 %rd
++@r2_vm_1 ...... . ..... ..... ... ..... ....... &rmr vm=1 %rs2 %rd
+ @r2_nfvm ... ... vm:1 ..... ..... ... ..... ....... &r2nfvm %nf %rs1 %rd
+ @r2_vm   ...... vm:1 ..... ..... ... ..... ....... &rmr %rs2 %rd
+ @r1_vm   ...... vm:1 ..... ..... ... ..... ....... %rd
+@@ -XXX,XX +XXX,XX @@ vcpop_v     010010 . ..... 01110 010 ..... 1010111 @r2_vm
+ vwsll_vv    110101 . ..... ..... 000 ..... 1010111 @r_vm
+ vwsll_vx    110101 . ..... ..... 100 ..... 1010111 @r_vm
+ vwsll_vi    110101 . ..... ..... 011 ..... 1010111 @r_vm
++
++# *** Zvkned vector crypto extension ***
++vaesef_vv   101000 1 ..... 00011 010 ..... 1110111 @r2_vm_1
++vaesef_vs   101001 1 ..... 00011 010 ..... 1110111 @r2_vm_1
++vaesdf_vv   101000 1 ..... 00001 010 ..... 1110111 @r2_vm_1
++vaesdf_vs   101001 1 ..... 00001 010 ..... 1110111 @r2_vm_1
++vaesem_vv   101000 1 ..... 00010 010 ..... 1110111 @r2_vm_1
++vaesem_vs   101001 1 ..... 00010 010 ..... 1110111 @r2_vm_1
++vaesdm_vv   101000 1 ..... 00000 010 ..... 1110111 @r2_vm_1
++vaesdm_vs   101001 1 ..... 00000 010 ..... 1110111 @r2_vm_1
++vaesz_vs    101001 1 ..... 00111 010 ..... 1110111 @r2_vm_1
++vaeskf1_vi  100010 1 ..... ..... 010 ..... 1110111 @r_vm_1
++vaeskf2_vi  101010 1 ..... ..... 010 ..... 1110111 @r_vm_1
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu.c
++++ b/target/riscv/cpu.c
+@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
+     ISA_EXT_DATA_ENTRY(zvfbfwma, PRIV_VERSION_1_12_0, ext_zvfbfwma),
+     ISA_EXT_DATA_ENTRY(zvfh, PRIV_VERSION_1_12_0, ext_zvfh),
+     ISA_EXT_DATA_ENTRY(zvfhmin, PRIV_VERSION_1_12_0, ext_zvfhmin),
++    ISA_EXT_DATA_ENTRY(zvkned, PRIV_VERSION_1_12_0, ext_zvkned),
+     ISA_EXT_DATA_ENTRY(zhinx, PRIV_VERSION_1_12_0, ext_zhinx),
+     ISA_EXT_DATA_ENTRY(zhinxmin, PRIV_VERSION_1_12_0, ext_zhinxmin),
+     ISA_EXT_DATA_ENTRY(smaia, PRIV_VERSION_1_12_0, ext_smaia),
+@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
+      * In principle Zve*x would also suffice here, were they supported
+      * in qemu
+      */
+-    if (cpu->cfg.ext_zvbb && !cpu->cfg.ext_zve32f) {
++    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned) && !cpu->cfg.ext_zve32f) {
+         error_setg(errp,
+                    "Vector crypto extensions require V or Zve* extensions");
+         return;
+@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
+     /* Vector cryptography extensions */
+     DEFINE_PROP_BOOL("x-zvbb", RISCVCPU, cfg.ext_zvbb, false),
+     DEFINE_PROP_BOOL("x-zvbc", RISCVCPU, cfg.ext_zvbc, false),
++    DEFINE_PROP_BOOL("x-zvkned", RISCVCPU, cfg.ext_zvkned, false),
+     DEFINE_PROP_END_OF_LIST(),
+ };
+diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/vcrypto_helper.c
++++ b/target/riscv/vcrypto_helper.c
+@@ -XXX,XX +XXX,XX @@
+ #include "qemu/bitops.h"
+ #include "qemu/bswap.h"
+ #include "cpu.h"
++#include "crypto/aes.h"
++#include "crypto/aes-round.h"
+ #include "exec/memop.h"
+ #include "exec/exec-all.h"
+ #include "exec/helper-proto.h"
+@@ -XXX,XX +XXX,XX @@ RVVCALL(OPIVX2, vwsll_vx_w, WOP_UUU_W, H8, H4, DO_SLL)
+ GEN_VEXT_VX(vwsll_vx_b, 2)
+ GEN_VEXT_VX(vwsll_vx_h, 4)
+ GEN_VEXT_VX(vwsll_vx_w, 8)
++
++void HELPER(egs_check)(uint32_t egs, CPURISCVState *env)
++{
++    uint32_t vl = env->vl;
++    uint32_t vstart = env->vstart;
++
++    if (vl % egs != 0 || vstart % egs != 0) {
++        riscv_raise_exception(env, RISCV_EXCP_ILLEGAL_INST, GETPC());
++    }
++}
++
++static inline void xor_round_key(AESState *round_state, AESState *round_key)
++{
++    round_state->v = round_state->v ^ round_key->v;
++}
++
++#define GEN_ZVKNED_HELPER_VV(NAME, ...)                                   \
++    void HELPER(NAME)(void *vd, void *vs2, CPURISCVState *env,            \
++                      uint32_t desc)                                      \
++    {                                                                     \
++        uint32_t vl = env->vl;                                            \
++        uint32_t total_elems = vext_get_total_elems(env, desc, 4);        \
++        uint32_t vta = vext_vta(desc);                                    \
++                                                                          \
++        for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {        \
++            AESState round_key;                                           \
++            round_key.d[0] = *((uint64_t *)vs2 + H8(i * 2 + 0));          \
++            round_key.d[1] = *((uint64_t *)vs2 + H8(i * 2 + 1));          \
++            AESState round_state;                                         \
++            round_state.d[0] = *((uint64_t *)vd + H8(i * 2 + 0));         \
++            round_state.d[1] = *((uint64_t *)vd + H8(i * 2 + 1));         \
++            __VA_ARGS__;                                                  \
++            *((uint64_t *)vd + H8(i * 2 + 0)) = round_state.d[0];         \
++            *((uint64_t *)vd + H8(i * 2 + 1)) = round_state.d[1];         \
++        }                                                                 \
++        env->vstart = 0;                                                  \
++        /* set tail elements to 1s */                                     \
++        vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);              \
++    }
++
++#define GEN_ZVKNED_HELPER_VS(NAME, ...)                                   \
++    void HELPER(NAME)(void *vd, void *vs2, CPURISCVState *env,            \
++                      uint32_t desc)                                      \
++    {                                                                     \
++        uint32_t vl = env->vl;                                            \
++        uint32_t total_elems = vext_get_total_elems(env, desc, 4);        \
++        uint32_t vta = vext_vta(desc);                                    \
++                                                                          \
++        for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {        \
++            AESState round_key;                                           \
++            round_key.d[0] = *((uint64_t *)vs2 + H8(0));                  \
++            round_key.d[1] = *((uint64_t *)vs2 + H8(1));                  \
++            AESState round_state;                                         \
++            round_state.d[0] = *((uint64_t *)vd + H8(i * 2 + 0));         \
++            round_state.d[1] = *((uint64_t *)vd + H8(i * 2 + 1));         \
++            __VA_ARGS__;                                                  \
++            *((uint64_t *)vd + H8(i * 2 + 0)) = round_state.d[0];         \
++            *((uint64_t *)vd + H8(i * 2 + 1)) = round_state.d[1];         \
++        }                                                                 \
++        env->vstart = 0;                                                  \
++        /* set tail elements to 1s */                                     \
++        vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);              \
++    }
++
++GEN_ZVKNED_HELPER_VV(vaesef_vv, aesenc_SB_SR_AK(&round_state,
++                                                &round_state,
++                                                &round_key,
++                                                false);)
++GEN_ZVKNED_HELPER_VS(vaesef_vs, aesenc_SB_SR_AK(&round_state,
++                                                &round_state,
++                                                &round_key,
++                                                false);)
++GEN_ZVKNED_HELPER_VV(vaesdf_vv, aesdec_ISB_ISR_AK(&round_state,
++                                                  &round_state,
++                                                  &round_key,
++                                                  false);)
++GEN_ZVKNED_HELPER_VS(vaesdf_vs, aesdec_ISB_ISR_AK(&round_state,
++                                                  &round_state,
++                                                  &round_key,
++                                                  false);)
++GEN_ZVKNED_HELPER_VV(vaesem_vv, aesenc_SB_SR_MC_AK(&round_state,
++                                                   &round_state,
++                                                   &round_key,
++                                                   false);)
++GEN_ZVKNED_HELPER_VS(vaesem_vs, aesenc_SB_SR_MC_AK(&round_state,
++                                                   &round_state,
++                                                   &round_key,
++                                                   false);)
++GEN_ZVKNED_HELPER_VV(vaesdm_vv, aesdec_ISB_ISR_AK_IMC(&round_state,
++                                                      &round_state,
++                                                      &round_key,
++                                                      false);)
++GEN_ZVKNED_HELPER_VS(vaesdm_vs, aesdec_ISB_ISR_AK_IMC(&round_state,
++                                                      &round_state,
++                                                      &round_key,
++                                                      false);)
++GEN_ZVKNED_HELPER_VS(vaesz_vs, xor_round_key(&round_state, &round_key);)
++
++void HELPER(vaeskf1_vi)(void *vd_vptr, void *vs2_vptr, uint32_t uimm,
++                        CPURISCVState *env, uint32_t desc)
++{
++    uint32_t *vd = vd_vptr;
++    uint32_t *vs2 = vs2_vptr;
++    uint32_t vl = env->vl;
++    uint32_t total_elems = vext_get_total_elems(env, desc, 4);
++    uint32_t vta = vext_vta(desc);
++
++    uimm &= 0b1111;
++    if (uimm > 10 || uimm == 0) {
++        uimm ^= 0b1000;
++    }
++
++    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
++        uint32_t rk[8], tmp;
++        static const uint32_t rcon[] = {
++            0x00000001, 0x00000002, 0x00000004, 0x00000008, 0x00000010,
++            0x00000020, 0x00000040, 0x00000080, 0x0000001B, 0x00000036,
++        };
++
++        rk[0] = vs2[i * 4 + H4(0)];
++        rk[1] = vs2[i * 4 + H4(1)];
++        rk[2] = vs2[i * 4 + H4(2)];
++        rk[3] = vs2[i * 4 + H4(3)];
++        tmp = ror32(rk[3], 8);
++
++        rk[4] = rk[0] ^ (((uint32_t)AES_sbox[(tmp >> 24) & 0xff] << 24) |
++                         ((uint32_t)AES_sbox[(tmp >> 16) & 0xff] << 16) |
++                         ((uint32_t)AES_sbox[(tmp >> 8) & 0xff] << 8) |
++                         ((uint32_t)AES_sbox[(tmp >> 0) & 0xff] << 0))
++                      ^ rcon[uimm - 1];
++        rk[5] = rk[1] ^ rk[4];
++        rk[6] = rk[2] ^ rk[5];
++        rk[7] = rk[3] ^ rk[6];
++
++        vd[i * 4 + H4(0)] = rk[4];
++        vd[i * 4 + H4(1)] = rk[5];
++        vd[i * 4 + H4(2)] = rk[6];
++        vd[i * 4 + H4(3)] = rk[7];
++    }
++    env->vstart = 0;
++    /* set tail elements to 1s */
++    vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);
++}
++
++void HELPER(vaeskf2_vi)(void *vd_vptr, void *vs2_vptr, uint32_t uimm,
++                        CPURISCVState *env, uint32_t desc)
++{
++    uint32_t *vd = vd_vptr;
++    uint32_t *vs2 = vs2_vptr;
++    uint32_t vl = env->vl;
++    uint32_t total_elems = vext_get_total_elems(env, desc, 4);
++    uint32_t vta = vext_vta(desc);
++
++    uimm &= 0b1111;
++    if (uimm > 14 || uimm < 2) {
++        uimm ^= 0b1000;
++    }
++
++    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
++        uint32_t rk[12], tmp;
++        static const uint32_t rcon[] = {
++            0x00000001, 0x00000002, 0x00000004, 0x00000008, 0x00000010,
++            0x00000020, 0x00000040, 0x00000080, 0x0000001B, 0x00000036,
++        };
++
++        rk[0] = vd[i * 4 + H4(0)];
++        rk[1] = vd[i * 4 + H4(1)];
++        rk[2] = vd[i * 4 + H4(2)];
++        rk[3] = vd[i * 4 + H4(3)];
++        rk[4] = vs2[i * 4 + H4(0)];
++        rk[5] = vs2[i * 4 + H4(1)];
++        rk[6] = vs2[i * 4 + H4(2)];
++        rk[7] = vs2[i * 4 + H4(3)];
++
++        if (uimm % 2 == 0) {
++            tmp = ror32(rk[7], 8);
++            rk[8] = rk[0] ^ (((uint32_t)AES_sbox[(tmp >> 24) & 0xff] << 24) |
++                             ((uint32_t)AES_sbox[(tmp >> 16) & 0xff] << 16) |
++                             ((uint32_t)AES_sbox[(tmp >> 8) & 0xff] << 8) |
++                             ((uint32_t)AES_sbox[(tmp >> 0) & 0xff] << 0))
++                          ^ rcon[(uimm - 1) / 2];
++        } else {
++            rk[8] = rk[0] ^ (((uint32_t)AES_sbox[(rk[7] >> 24) & 0xff] << 24) |
++                             ((uint32_t)AES_sbox[(rk[7] >> 16) & 0xff] << 16) |
++                             ((uint32_t)AES_sbox[(rk[7] >> 8) & 0xff] << 8) |
++                             ((uint32_t)AES_sbox[(rk[7] >> 0) & 0xff] << 0));
++        }
++        rk[9] = rk[1] ^ rk[8];
++        rk[10] = rk[2] ^ rk[9];
++        rk[11] = rk[3] ^ rk[10];
++
++        vd[i * 4 + H4(0)] = rk[8];
++        vd[i * 4 + H4(1)] = rk[9];
++        vd[i * 4 + H4(2)] = rk[10];
++        vd[i * 4 + H4(3)] = rk[11];
++    }
++    env->vstart = 0;
++    /* set tail elements to 1s */
++    vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);
++}
+diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/insn_trans/trans_rvvk.c.inc
++++ b/target/riscv/insn_trans/trans_rvvk.c.inc
+@@ -XXX,XX +XXX,XX @@ static bool vwsll_vx_check(DisasContext *s, arg_rmrr *a)
+ GEN_OPIVV_WIDEN_TRANS(vwsll_vv, vwsll_vv_check)
+ GEN_OPIVX_WIDEN_TRANS(vwsll_vx, vwsll_vx_check)
+ GEN_OPIVI_WIDEN_TRANS(vwsll_vi, IMM_ZX, vwsll_vx, vwsll_vx_check)
++
++/*
++ * Zvkned
++ */
++
++#define ZVKNED_EGS 4
++
++#define GEN_V_UNMASKED_TRANS(NAME, CHECK, EGS)                                \
++    static bool trans_##NAME(DisasContext *s, arg_##NAME *a)                  \
++    {                                                                         \
++        if (CHECK(s, a)) {                                                    \
++            TCGv_ptr rd_v, rs2_v;                                             \
++            TCGv_i32 desc, egs;                                               \
++            uint32_t data = 0;                                                \
++            TCGLabel *over = gen_new_label();                                 \
++                                                                              \
++            if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {                      \
++                /* save opcode for unwinding in case we throw an exception */ \
++                decode_save_opc(s);                                           \
++                egs = tcg_constant_i32(EGS);                                  \
++                gen_helper_egs_check(egs, cpu_env);                           \
++                tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);    \
++            }                                                                 \
++                                                                              \
++            data = FIELD_DP32(data, VDATA, VM, a->vm);                        \
++            data = FIELD_DP32(data, VDATA, LMUL, s->lmul);                    \
++            data = FIELD_DP32(data, VDATA, VTA, s->vta);                      \
++            data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);    \
++            data = FIELD_DP32(data, VDATA, VMA, s->vma);                      \
++            rd_v = tcg_temp_new_ptr();                                        \
++            rs2_v = tcg_temp_new_ptr();                                       \
++            desc = tcg_constant_i32(                                          \
++                simd_desc(s->cfg_ptr->vlen / 8, s->cfg_ptr->vlen / 8, data)); \
++            tcg_gen_addi_ptr(rd_v, cpu_env, vreg_ofs(s, a->rd));              \
++            tcg_gen_addi_ptr(rs2_v, cpu_env, vreg_ofs(s, a->rs2));            \
++            gen_helper_##NAME(rd_v, rs2_v, cpu_env, desc);                    \
++            mark_vs_dirty(s);                                                 \
++            gen_set_label(over);                                              \
++            return true;                                                      \
++        }                                                                     \
++        return false;                                                         \
++    }
++
++static bool vaes_check_vv(DisasContext *s, arg_rmr *a)
++{
++    int egw_bytes = ZVKNED_EGS << s->sew;
++    return s->cfg_ptr->ext_zvkned == true &&
++           require_rvv(s) &&
++           vext_check_isa_ill(s) &&
++           MAXSZ(s) >= egw_bytes &&
++           require_align(a->rd, s->lmul) &&
++           require_align(a->rs2, s->lmul) &&
++           s->sew == MO_32;
++}
++
++static bool vaes_check_overlap(DisasContext *s, int vd, int vs2)
++{
++    int8_t op_size = s->lmul <= 0 ? 1 : 1 << s->lmul;
++    return !is_overlapped(vd, op_size, vs2, 1);
++}
++
++static bool vaes_check_vs(DisasContext *s, arg_rmr *a)
++{
++    int egw_bytes = ZVKNED_EGS << s->sew;
++    return vaes_check_overlap(s, a->rd, a->rs2) &&
++           MAXSZ(s) >= egw_bytes &&
++           s->cfg_ptr->ext_zvkned == true &&
++           require_rvv(s) &&
++           vext_check_isa_ill(s) &&
++           require_align(a->rd, s->lmul) &&
++           s->sew == MO_32;
++}
++
++GEN_V_UNMASKED_TRANS(vaesef_vv, vaes_check_vv, ZVKNED_EGS)
++GEN_V_UNMASKED_TRANS(vaesef_vs, vaes_check_vs, ZVKNED_EGS)
++GEN_V_UNMASKED_TRANS(vaesdf_vv, vaes_check_vv, ZVKNED_EGS)
++GEN_V_UNMASKED_TRANS(vaesdf_vs, vaes_check_vs, ZVKNED_EGS)
++GEN_V_UNMASKED_TRANS(vaesdm_vv, vaes_check_vv, ZVKNED_EGS)
++GEN_V_UNMASKED_TRANS(vaesdm_vs, vaes_check_vs, ZVKNED_EGS)
++GEN_V_UNMASKED_TRANS(vaesz_vs, vaes_check_vs, ZVKNED_EGS)
++GEN_V_UNMASKED_TRANS(vaesem_vv, vaes_check_vv, ZVKNED_EGS)
++GEN_V_UNMASKED_TRANS(vaesem_vs, vaes_check_vs, ZVKNED_EGS)
++
++#define GEN_VI_UNMASKED_TRANS(NAME, CHECK, EGS)                               \
++    static bool trans_##NAME(DisasContext *s, arg_##NAME *a)                  \
++    {                                                                         \
++        if (CHECK(s, a)) {                                                    \
++            TCGv_ptr rd_v, rs2_v;                                             \
++            TCGv_i32 uimm_v, desc, egs;                                       \
++            uint32_t data = 0;                                                \
++            TCGLabel *over = gen_new_label();                                 \
++                                                                              \
++            if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {                      \
++                /* save opcode for unwinding in case we throw an exception */ \
++                decode_save_opc(s);                                           \
++                egs = tcg_constant_i32(EGS);                                  \
++                gen_helper_egs_check(egs, cpu_env);                           \
++                tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);    \
++            }                                                                 \
++                                                                              \
++            data = FIELD_DP32(data, VDATA, VM, a->vm);                        \
++            data = FIELD_DP32(data, VDATA, LMUL, s->lmul);                    \
++            data = FIELD_DP32(data, VDATA, VTA, s->vta);                      \
++            data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);    \
++            data = FIELD_DP32(data, VDATA, VMA, s->vma);                      \
++                                                                              \
++            rd_v = tcg_temp_new_ptr();                                        \
++            rs2_v = tcg_temp_new_ptr();                                       \
++            uimm_v = tcg_constant_i32(a->rs1);                                \
++            desc = tcg_constant_i32(                                          \
++                simd_desc(s->cfg_ptr->vlen / 8, s->cfg_ptr->vlen / 8, data)); \
++            tcg_gen_addi_ptr(rd_v, cpu_env, vreg_ofs(s, a->rd));              \
++            tcg_gen_addi_ptr(rs2_v, cpu_env, vreg_ofs(s, a->rs2));            \
++            gen_helper_##NAME(rd_v, rs2_v, uimm_v, cpu_env, desc);            \
++            mark_vs_dirty(s);                                                 \
++            gen_set_label(over);                                              \
++            return true;                                                      \
++        }                                                                     \
++        return false;                                                         \
++    }
++
++static bool vaeskf1_check(DisasContext *s, arg_vaeskf1_vi *a)
++{
++    int egw_bytes = ZVKNED_EGS << s->sew;
++    return s->cfg_ptr->ext_zvkned == true &&
++           require_rvv(s) &&
++           vext_check_isa_ill(s) &&
++           MAXSZ(s) >= egw_bytes &&
++           s->sew == MO_32 &&
++           require_align(a->rd, s->lmul) &&
++           require_align(a->rs2, s->lmul);
++}
++
++static bool vaeskf2_check(DisasContext *s, arg_vaeskf2_vi *a)
++{
++    int egw_bytes = ZVKNED_EGS << s->sew;
++    return s->cfg_ptr->ext_zvkned == true &&
++           require_rvv(s) &&
++           vext_check_isa_ill(s) &&
++           MAXSZ(s) >= egw_bytes &&
++           s->sew == MO_32 &&
++           require_align(a->rd, s->lmul) &&
++           require_align(a->rs2, s->lmul);
++}
++
++GEN_VI_UNMASKED_TRANS(vaeskf1_vi, vaeskf1_check, ZVKNED_EGS)
++GEN_VI_UNMASKED_TRANS(vaeskf2_vi, vaeskf2_check, ZVKNED_EGS)
+--
+.41.0

-New patch
+[PULL v2 17/45] target/riscv: Add Zvknh ISA extension support
+From: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
+This commit adds support for the Zvknh vector-crypto extension, which
+consists of the following instructions:
+* vsha2ms.vv
+* vsha2c[hl].vv
+Translation functions are defined in
+`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
+`target/riscv/vcrypto_helper.c`.
+Co-authored-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
+Co-authored-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
+[max.chou@sifive.com: Replaced vstart checking by TCG op]
+Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
+Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
+Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+[max.chou@sifive.com: Exposed x-zvknha & x-zvknhb properties]
+[max.chou@sifive.com: Replaced SEW selection to happened during
+translation]
+Message-ID: <20230711165917.2629866-11-max.chou@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/cpu_cfg.h                   |   2 +
+ target/riscv/helper.h                    |   6 +
+ target/riscv/insn32.decode               |   5 +
+ target/riscv/cpu.c                       |  13 +-
+ target/riscv/vcrypto_helper.c            | 238 +++++++++++++++++++++++
+ target/riscv/insn_trans/trans_rvvk.c.inc | 129 ++++++++++++
+files changed, 390 insertions(+), 3 deletions(-)
+diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu_cfg.h
++++ b/target/riscv/cpu_cfg.h
+@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
+     bool ext_zvbb;
+     bool ext_zvbc;
+     bool ext_zvkned;
++    bool ext_zvknha;
++    bool ext_zvknhb;
+     bool ext_zmmul;
+     bool ext_zvfbfmin;
+     bool ext_zvfbfwma;
+diff --git a/target/riscv/helper.h b/target/riscv/helper.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/helper.h
++++ b/target/riscv/helper.h
+@@ -XXX,XX +XXX,XX @@ DEF_HELPER_4(vaesdm_vs, void, ptr, ptr, env, i32)
+ DEF_HELPER_4(vaesz_vs, void, ptr, ptr, env, i32)
+ DEF_HELPER_5(vaeskf1_vi, void, ptr, ptr, i32, env, i32)
+ DEF_HELPER_5(vaeskf2_vi, void, ptr, ptr, i32, env, i32)
++
++DEF_HELPER_5(vsha2ms_vv, void, ptr, ptr, ptr, env, i32)
++DEF_HELPER_5(vsha2ch32_vv, void, ptr, ptr, ptr, env, i32)
++DEF_HELPER_5(vsha2ch64_vv, void, ptr, ptr, ptr, env, i32)
++DEF_HELPER_5(vsha2cl32_vv, void, ptr, ptr, ptr, env, i32)
++DEF_HELPER_5(vsha2cl64_vv, void, ptr, ptr, ptr, env, i32)
+diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/insn32.decode
++++ b/target/riscv/insn32.decode
+@@ -XXX,XX +XXX,XX @@ vaesdm_vs   101001 1 ..... 00000 010 ..... 1110111 @r2_vm_1
+ vaesz_vs    101001 1 ..... 00111 010 ..... 1110111 @r2_vm_1
+ vaeskf1_vi  100010 1 ..... ..... 010 ..... 1110111 @r_vm_1
+ vaeskf2_vi  101010 1 ..... ..... 010 ..... 1110111 @r_vm_1
++
++# *** Zvknh vector crypto extension ***
++vsha2ms_vv  101101 1 ..... ..... 010 ..... 1110111 @r_vm_1
++vsha2ch_vv  101110 1 ..... ..... 010 ..... 1110111 @r_vm_1
++vsha2cl_vv  101111 1 ..... ..... 010 ..... 1110111 @r_vm_1
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu.c
++++ b/target/riscv/cpu.c
+@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
+     ISA_EXT_DATA_ENTRY(zvfh, PRIV_VERSION_1_12_0, ext_zvfh),
+     ISA_EXT_DATA_ENTRY(zvfhmin, PRIV_VERSION_1_12_0, ext_zvfhmin),
+     ISA_EXT_DATA_ENTRY(zvkned, PRIV_VERSION_1_12_0, ext_zvkned),
++    ISA_EXT_DATA_ENTRY(zvknha, PRIV_VERSION_1_12_0, ext_zvknha),
++    ISA_EXT_DATA_ENTRY(zvknhb, PRIV_VERSION_1_12_0, ext_zvknhb),
+     ISA_EXT_DATA_ENTRY(zhinx, PRIV_VERSION_1_12_0, ext_zhinx),
+     ISA_EXT_DATA_ENTRY(zhinxmin, PRIV_VERSION_1_12_0, ext_zhinxmin),
+     ISA_EXT_DATA_ENTRY(smaia, PRIV_VERSION_1_12_0, ext_smaia),
+@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
+      * In principle Zve*x would also suffice here, were they supported
+      * in qemu
+      */
+-    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned) && !cpu->cfg.ext_zve32f) {
++    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned || cpu->cfg.ext_zvknha) &&
++        !cpu->cfg.ext_zve32f) {
+         error_setg(errp,
+                    "Vector crypto extensions require V or Zve* extensions");
+         return;
+     }
+-    if (cpu->cfg.ext_zvbc && !cpu->cfg.ext_zve64f) {
+-        error_setg(errp, "Zvbc extension requires V or Zve64{f,d} extensions");
++    if ((cpu->cfg.ext_zvbc || cpu->cfg.ext_zvknhb) && !cpu->cfg.ext_zve64f) {
++        error_setg(
++            errp,
++            "Zvbc and Zvknhb extensions require V or Zve64{f,d} extensions");
+         return;
+     }
+@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
+     DEFINE_PROP_BOOL("x-zvbb", RISCVCPU, cfg.ext_zvbb, false),
+     DEFINE_PROP_BOOL("x-zvbc", RISCVCPU, cfg.ext_zvbc, false),
+     DEFINE_PROP_BOOL("x-zvkned", RISCVCPU, cfg.ext_zvkned, false),
++    DEFINE_PROP_BOOL("x-zvknha", RISCVCPU, cfg.ext_zvknha, false),
++    DEFINE_PROP_BOOL("x-zvknhb", RISCVCPU, cfg.ext_zvknhb, false),
+     DEFINE_PROP_END_OF_LIST(),
+ };
+diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/vcrypto_helper.c
++++ b/target/riscv/vcrypto_helper.c
+@@ -XXX,XX +XXX,XX @@ void HELPER(vaeskf2_vi)(void *vd_vptr, void *vs2_vptr, uint32_t uimm,
+     /* set tail elements to 1s */
+     vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);
+ }
++
++static inline uint32_t sig0_sha256(uint32_t x)
++{
++    return ror32(x, 7) ^ ror32(x, 18) ^ (x >> 3);
++}
++
++static inline uint32_t sig1_sha256(uint32_t x)
++{
++    return ror32(x, 17) ^ ror32(x, 19) ^ (x >> 10);
++}
++
++static inline uint64_t sig0_sha512(uint64_t x)
++{
++    return ror64(x, 1) ^ ror64(x, 8) ^ (x >> 7);
++}
++
++static inline uint64_t sig1_sha512(uint64_t x)
++{
++    return ror64(x, 19) ^ ror64(x, 61) ^ (x >> 6);
++}
++
++static inline void vsha2ms_e32(uint32_t *vd, uint32_t *vs1, uint32_t *vs2)
++{
++    uint32_t res[4];
++    res[0] = sig1_sha256(vs1[H4(2)]) + vs2[H4(1)] + sig0_sha256(vd[H4(1)]) +
++             vd[H4(0)];
++    res[1] = sig1_sha256(vs1[H4(3)]) + vs2[H4(2)] + sig0_sha256(vd[H4(2)]) +
++             vd[H4(1)];
++    res[2] =
++        sig1_sha256(res[0]) + vs2[H4(3)] + sig0_sha256(vd[H4(3)]) + vd[H4(2)];
++    res[3] =
++        sig1_sha256(res[1]) + vs1[H4(0)] + sig0_sha256(vs2[H4(0)]) + vd[H4(3)];
++    vd[H4(3)] = res[3];
++    vd[H4(2)] = res[2];
++    vd[H4(1)] = res[1];
++    vd[H4(0)] = res[0];
++}
++
++static inline void vsha2ms_e64(uint64_t *vd, uint64_t *vs1, uint64_t *vs2)
++{
++    uint64_t res[4];
++    res[0] = sig1_sha512(vs1[2]) + vs2[1] + sig0_sha512(vd[1]) + vd[0];
++    res[1] = sig1_sha512(vs1[3]) + vs2[2] + sig0_sha512(vd[2]) + vd[1];
++    res[2] = sig1_sha512(res[0]) + vs2[3] + sig0_sha512(vd[3]) + vd[2];
++    res[3] = sig1_sha512(res[1]) + vs1[0] + sig0_sha512(vs2[0]) + vd[3];
++    vd[3] = res[3];
++    vd[2] = res[2];
++    vd[1] = res[1];
++    vd[0] = res[0];
++}
++
++void HELPER(vsha2ms_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
++                        uint32_t desc)
++{
++    uint32_t sew = FIELD_EX64(env->vtype, VTYPE, VSEW);
++    uint32_t esz = sew == MO_32 ? 4 : 8;
++    uint32_t total_elems;
++    uint32_t vta = vext_vta(desc);
++
++    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
++        if (sew == MO_32) {
++            vsha2ms_e32(((uint32_t *)vd) + i * 4, ((uint32_t *)vs1) + i * 4,
++                        ((uint32_t *)vs2) + i * 4);
++        } else {
++            /* If not 32 then SEW should be 64 */
++            vsha2ms_e64(((uint64_t *)vd) + i * 4, ((uint64_t *)vs1) + i * 4,
++                        ((uint64_t *)vs2) + i * 4);
++        }
++    }
++    /* set tail elements to 1s */
++    total_elems = vext_get_total_elems(env, desc, esz);
++    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
++    env->vstart = 0;
++}
++
++static inline uint64_t sum0_64(uint64_t x)
++{
++    return ror64(x, 28) ^ ror64(x, 34) ^ ror64(x, 39);
++}
++
++static inline uint32_t sum0_32(uint32_t x)
++{
++    return ror32(x, 2) ^ ror32(x, 13) ^ ror32(x, 22);
++}
++
++static inline uint64_t sum1_64(uint64_t x)
++{
++    return ror64(x, 14) ^ ror64(x, 18) ^ ror64(x, 41);
++}
++
++static inline uint32_t sum1_32(uint32_t x)
++{
++    return ror32(x, 6) ^ ror32(x, 11) ^ ror32(x, 25);
++}
++
++#define ch(x, y, z) ((x & y) ^ ((~x) & z))
++
++#define maj(x, y, z) ((x & y) ^ (x & z) ^ (y & z))
++
++static void vsha2c_64(uint64_t *vs2, uint64_t *vd, uint64_t *vs1)
++{
++    uint64_t a = vs2[3], b = vs2[2], e = vs2[1], f = vs2[0];
++    uint64_t c = vd[3], d = vd[2], g = vd[1], h = vd[0];
++    uint64_t W0 = vs1[0], W1 = vs1[1];
++    uint64_t T1 = h + sum1_64(e) + ch(e, f, g) + W0;
++    uint64_t T2 = sum0_64(a) + maj(a, b, c);
++
++    h = g;
++    g = f;
++    f = e;
++    e = d + T1;
++    d = c;
++    c = b;
++    b = a;
++    a = T1 + T2;
++
++    T1 = h + sum1_64(e) + ch(e, f, g) + W1;
++    T2 = sum0_64(a) + maj(a, b, c);
++    h = g;
++    g = f;
++    f = e;
++    e = d + T1;
++    d = c;
++    c = b;
++    b = a;
++    a = T1 + T2;
++
++    vd[0] = f;
++    vd[1] = e;
++    vd[2] = b;
++    vd[3] = a;
++}
++
++static void vsha2c_32(uint32_t *vs2, uint32_t *vd, uint32_t *vs1)
++{
++    uint32_t a = vs2[H4(3)], b = vs2[H4(2)], e = vs2[H4(1)], f = vs2[H4(0)];
++    uint32_t c = vd[H4(3)], d = vd[H4(2)], g = vd[H4(1)], h = vd[H4(0)];
++    uint32_t W0 = vs1[H4(0)], W1 = vs1[H4(1)];
++    uint32_t T1 = h + sum1_32(e) + ch(e, f, g) + W0;
++    uint32_t T2 = sum0_32(a) + maj(a, b, c);
++
++    h = g;
++    g = f;
++    f = e;
++    e = d + T1;
++    d = c;
++    c = b;
++    b = a;
++    a = T1 + T2;
++
++    T1 = h + sum1_32(e) + ch(e, f, g) + W1;
++    T2 = sum0_32(a) + maj(a, b, c);
++    h = g;
++    g = f;
++    f = e;
++    e = d + T1;
++    d = c;
++    c = b;
++    b = a;
++    a = T1 + T2;
++
++    vd[H4(0)] = f;
++    vd[H4(1)] = e;
++    vd[H4(2)] = b;
++    vd[H4(3)] = a;
++}
++
++void HELPER(vsha2ch32_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
++                          uint32_t desc)
++{
++    const uint32_t esz = 4;
++    uint32_t total_elems;
++    uint32_t vta = vext_vta(desc);
++
++    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
++        vsha2c_32(((uint32_t *)vs2) + 4 * i, ((uint32_t *)vd) + 4 * i,
++                  ((uint32_t *)vs1) + 4 * i + 2);
++    }
++
++    /* set tail elements to 1s */
++    total_elems = vext_get_total_elems(env, desc, esz);
++    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
++    env->vstart = 0;
++}
++
++void HELPER(vsha2ch64_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
++                          uint32_t desc)
++{
++    const uint32_t esz = 8;
++    uint32_t total_elems;
++    uint32_t vta = vext_vta(desc);
++
++    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
++        vsha2c_64(((uint64_t *)vs2) + 4 * i, ((uint64_t *)vd) + 4 * i,
++                  ((uint64_t *)vs1) + 4 * i + 2);
++    }
++
++    /* set tail elements to 1s */
++    total_elems = vext_get_total_elems(env, desc, esz);
++    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
++    env->vstart = 0;
++}
++
++void HELPER(vsha2cl32_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
++                          uint32_t desc)
++{
++    const uint32_t esz = 4;
++    uint32_t total_elems;
++    uint32_t vta = vext_vta(desc);
++
++    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
++        vsha2c_32(((uint32_t *)vs2) + 4 * i, ((uint32_t *)vd) + 4 * i,
++                  (((uint32_t *)vs1) + 4 * i));
++    }
++
++    /* set tail elements to 1s */
++    total_elems = vext_get_total_elems(env, desc, esz);
++    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
++    env->vstart = 0;
++}
++
++void HELPER(vsha2cl64_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
++                          uint32_t desc)
++{
++    uint32_t esz = 8;
++    uint32_t total_elems;
++    uint32_t vta = vext_vta(desc);
++
++    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
++        vsha2c_64(((uint64_t *)vs2) + 4 * i, ((uint64_t *)vd) + 4 * i,
++                  (((uint64_t *)vs1) + 4 * i));
++    }
++
++    /* set tail elements to 1s */
++    total_elems = vext_get_total_elems(env, desc, esz);
++    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
++    env->vstart = 0;
++}
+diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/insn_trans/trans_rvvk.c.inc
++++ b/target/riscv/insn_trans/trans_rvvk.c.inc
+@@ -XXX,XX +XXX,XX @@ static bool vaeskf2_check(DisasContext *s, arg_vaeskf2_vi *a)
+ GEN_VI_UNMASKED_TRANS(vaeskf1_vi, vaeskf1_check, ZVKNED_EGS)
+ GEN_VI_UNMASKED_TRANS(vaeskf2_vi, vaeskf2_check, ZVKNED_EGS)
++
++/*
++ * Zvknh
++ */
++
++#define ZVKNH_EGS 4
++
++#define GEN_VV_UNMASKED_TRANS(NAME, CHECK, EGS)                               \
++    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)                    \
++    {                                                                         \
++        if (CHECK(s, a)) {                                                    \
++            uint32_t data = 0;                                                \
++            TCGLabel *over = gen_new_label();                                 \
++            TCGv_i32 egs;                                                     \
++                                                                              \
++            if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {                      \
++                /* save opcode for unwinding in case we throw an exception */ \
++                decode_save_opc(s);                                           \
++                egs = tcg_constant_i32(EGS);                                  \
++                gen_helper_egs_check(egs, cpu_env);                           \
++                tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);    \
++            }                                                                 \
++                                                                              \
++            data = FIELD_DP32(data, VDATA, VM, a->vm);                        \
++            data = FIELD_DP32(data, VDATA, LMUL, s->lmul);                    \
++            data = FIELD_DP32(data, VDATA, VTA, s->vta);                      \
++            data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);    \
++            data = FIELD_DP32(data, VDATA, VMA, s->vma);                      \
++                                                                              \
++            tcg_gen_gvec_3_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, a->rs1),       \
++                               vreg_ofs(s, a->rs2), cpu_env,                  \
++                               s->cfg_ptr->vlen / 8, s->cfg_ptr->vlen / 8,    \
++                               data, gen_helper_##NAME);                      \
++                                                                              \
++            mark_vs_dirty(s);                                                 \
++            gen_set_label(over);                                              \
++            return true;                                                      \
++        }                                                                     \
++        return false;                                                         \
++    }
++
++static bool vsha_check_sew(DisasContext *s)
++{
++    return (s->cfg_ptr->ext_zvknha == true && s->sew == MO_32) ||
++           (s->cfg_ptr->ext_zvknhb == true &&
++            (s->sew == MO_32 || s->sew == MO_64));
++}
++
++static bool vsha_check(DisasContext *s, arg_rmrr *a)
++{
++    int egw_bytes = ZVKNH_EGS << s->sew;
++    int mult = 1 << MAX(s->lmul, 0);
++    return opivv_check(s, a) &&
++           vsha_check_sew(s) &&
++           MAXSZ(s) >= egw_bytes &&
++           !is_overlapped(a->rd, mult, a->rs1, mult) &&
++           !is_overlapped(a->rd, mult, a->rs2, mult) &&
++           s->lmul >= 0;
++}
++
++GEN_VV_UNMASKED_TRANS(vsha2ms_vv, vsha_check, ZVKNH_EGS)
++
++static bool trans_vsha2cl_vv(DisasContext *s, arg_rmrr *a)
++{
++    if (vsha_check(s, a)) {
++        uint32_t data = 0;
++        TCGLabel *over = gen_new_label();
++        TCGv_i32 egs;
++
++        if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {
++            /* save opcode for unwinding in case we throw an exception */
++            decode_save_opc(s);
++            egs = tcg_constant_i32(ZVKNH_EGS);
++            gen_helper_egs_check(egs, cpu_env);
++            tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
++        }
++
++        data = FIELD_DP32(data, VDATA, VM, a->vm);
++        data = FIELD_DP32(data, VDATA, LMUL, s->lmul);
++        data = FIELD_DP32(data, VDATA, VTA, s->vta);
++        data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);
++        data = FIELD_DP32(data, VDATA, VMA, s->vma);
++
++        tcg_gen_gvec_3_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, a->rs1),
++            vreg_ofs(s, a->rs2), cpu_env, s->cfg_ptr->vlen / 8,
++            s->cfg_ptr->vlen / 8, data,
++            s->sew == MO_32 ?
++                gen_helper_vsha2cl32_vv : gen_helper_vsha2cl64_vv);
++
++        mark_vs_dirty(s);
++        gen_set_label(over);
++        return true;
++    }
++    return false;
++}
++
++static bool trans_vsha2ch_vv(DisasContext *s, arg_rmrr *a)
++{
++    if (vsha_check(s, a)) {
++        uint32_t data = 0;
++        TCGLabel *over = gen_new_label();
++        TCGv_i32 egs;
++
++        if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {
++            /* save opcode for unwinding in case we throw an exception */
++            decode_save_opc(s);
++            egs = tcg_constant_i32(ZVKNH_EGS);
++            gen_helper_egs_check(egs, cpu_env);
++            tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
++        }
++
++        data = FIELD_DP32(data, VDATA, VM, a->vm);
++        data = FIELD_DP32(data, VDATA, LMUL, s->lmul);
++        data = FIELD_DP32(data, VDATA, VTA, s->vta);
++        data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);
++        data = FIELD_DP32(data, VDATA, VMA, s->vma);
++
++        tcg_gen_gvec_3_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, a->rs1),
++            vreg_ofs(s, a->rs2), cpu_env, s->cfg_ptr->vlen / 8,
++            s->cfg_ptr->vlen / 8, data,
++            s->sew == MO_32 ?
++                gen_helper_vsha2ch32_vv : gen_helper_vsha2ch64_vv);
++
++        mark_vs_dirty(s);
++        gen_set_label(over);
++        return true;
++    }
++    return false;
++}
+--
+.41.0

-New patch
+[PULL v2 18/45] target/riscv: Add Zvksh ISA extension support
+From: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
 This commit adds support for the Zvksh vector-crypto extension, which
 consists of the following instructions:
 * vsm3me.vv
 * vsm3c.vi
 Translation functions are defined in
 `target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
 `target/riscv/vcrypto_helper.c`.
 Co-authored-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
 [max.chou@sifive.com: Replaced vstart checking by TCG op]
 Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
 Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
 Signed-off-by: Max Chou <max.chou@sifive.com>
 Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
 [max.chou@sifive.com: Exposed x-zvksh property]
 Message-ID: <20230711165917.2629866-12-max.chou@sifive.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
  target/riscv/cpu_cfg.h                   |   1 +
  target/riscv/helper.h                    |   3 +
  target/riscv/insn32.decode               |   4 +
  target/riscv/cpu.c                       |   6 +-
  target/riscv/vcrypto_helper.c            | 134 +++++++++++++++++++++++
  target/riscv/insn_trans/trans_rvvk.c.inc |  31 ++++++
 files changed, 177 insertions(+), 2 deletions(-)
 diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/cpu_cfg.h
 +++ b/target/riscv/cpu_cfg.h
@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
      bool ext_zvkned;
      bool ext_zvknha;
      bool ext_zvknhb;
 +    bool ext_zvksh;
      bool ext_zmmul;
      bool ext_zvfbfmin;
      bool ext_zvfbfwma;
 diff --git a/target/riscv/helper.h b/target/riscv/helper.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/helper.h
 +++ b/target/riscv/helper.h
@@ -XXX,XX +XXX,XX @@ DEF_HELPER_5(vsha2ch32_vv, void, ptr, ptr, ptr, env, i32)
  DEF_HELPER_5(vsha2ch64_vv, void, ptr, ptr, ptr, env, i32)
  DEF_HELPER_5(vsha2cl32_vv, void, ptr, ptr, ptr, env, i32)
  DEF_HELPER_5(vsha2cl64_vv, void, ptr, ptr, ptr, env, i32)
 +
 +DEF_HELPER_5(vsm3me_vv, void, ptr, ptr, ptr, env, i32)
 +DEF_HELPER_5(vsm3c_vi, void, ptr, ptr, i32, env, i32)
 diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/insn32.decode
 +++ b/target/riscv/insn32.decode
@@ -XXX,XX +XXX,XX @@ vaeskf2_vi  101010 1 ..... ..... 010 ..... 1110111 @r_vm_1
  vsha2ms_vv  101101 1 ..... ..... 010 ..... 1110111 @r_vm_1
  vsha2ch_vv  101110 1 ..... ..... 010 ..... 1110111 @r_vm_1
  vsha2cl_vv  101111 1 ..... ..... 010 ..... 1110111 @r_vm_1
 +
 +# *** Zvksh vector crypto extension ***
 +vsm3me_vv   100000 1 ..... ..... 010 ..... 1110111 @r_vm_1
 +vsm3c_vi    101011 1 ..... ..... 010 ..... 1110111 @r_vm_1
 diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/cpu.c
 +++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
      ISA_EXT_DATA_ENTRY(zvkned, PRIV_VERSION_1_12_0, ext_zvkned),
      ISA_EXT_DATA_ENTRY(zvknha, PRIV_VERSION_1_12_0, ext_zvknha),
      ISA_EXT_DATA_ENTRY(zvknhb, PRIV_VERSION_1_12_0, ext_zvknhb),
 +    ISA_EXT_DATA_ENTRY(zvksh, PRIV_VERSION_1_12_0, ext_zvksh),
      ISA_EXT_DATA_ENTRY(zhinx, PRIV_VERSION_1_12_0, ext_zhinx),
      ISA_EXT_DATA_ENTRY(zhinxmin, PRIV_VERSION_1_12_0, ext_zhinxmin),
      ISA_EXT_DATA_ENTRY(smaia, PRIV_VERSION_1_12_0, ext_smaia),
@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
       * In principle Zve*x would also suffice here, were they supported
       * in qemu
       */
 -    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned || cpu->cfg.ext_zvknha) &&
 -        !cpu->cfg.ext_zve32f) {
 +    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned || cpu->cfg.ext_zvknha ||
 +         cpu->cfg.ext_zvksh) && !cpu->cfg.ext_zve32f) {
          error_setg(errp,
                     "Vector crypto extensions require V or Zve* extensions");
          return;
@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
      DEFINE_PROP_BOOL("x-zvkned", RISCVCPU, cfg.ext_zvkned, false),
      DEFINE_PROP_BOOL("x-zvknha", RISCVCPU, cfg.ext_zvknha, false),
      DEFINE_PROP_BOOL("x-zvknhb", RISCVCPU, cfg.ext_zvknhb, false),
 +    DEFINE_PROP_BOOL("x-zvksh", RISCVCPU, cfg.ext_zvksh, false),
      DEFINE_PROP_END_OF_LIST(),
  };
 diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/vcrypto_helper.c
 +++ b/target/riscv/vcrypto_helper.c
@@ -XXX,XX +XXX,XX @@ void HELPER(vsha2cl64_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
      vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
      env->vstart = 0;
  }
 +
 +static inline uint32_t p1(uint32_t x)
 +{
 +    return x ^ rol32(x, 15) ^ rol32(x, 23);
 +}
 +
 +static inline uint32_t zvksh_w(uint32_t m16, uint32_t m9, uint32_t m3,
 +                               uint32_t m13, uint32_t m6)
 +{
 +    return p1(m16 ^ m9 ^ rol32(m3, 15)) ^ rol32(m13, 7) ^ m6;
 +}
 +
 +void HELPER(vsm3me_vv)(void *vd_vptr, void *vs1_vptr, void *vs2_vptr,
 +                       CPURISCVState *env, uint32_t desc)
 +{
 +    uint32_t esz = memop_size(FIELD_EX64(env->vtype, VTYPE, VSEW));
 +    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
 +    uint32_t vta = vext_vta(desc);
 +    uint32_t *vd = vd_vptr;
 +    uint32_t *vs1 = vs1_vptr;
 +    uint32_t *vs2 = vs2_vptr;
 +
 +    for (int i = env->vstart / 8; i < env->vl / 8; i++) {
 +        uint32_t w[24];
 +        for (int j = 0; j < 8; j++) {
 +            w[j] = bswap32(vs1[H4((i * 8) + j)]);
 +            w[j + 8] = bswap32(vs2[H4((i * 8) + j)]);
 +        }
 +        for (int j = 0; j < 8; j++) {
 +            w[j + 16] =
 +                zvksh_w(w[j], w[j + 7], w[j + 13], w[j + 3], w[j + 10]);
 +        }
 +        for (int j = 0; j < 8; j++) {
 +            vd[(i * 8) + j] = bswap32(w[H4(j + 16)]);
 +        }
 +    }
 +    vext_set_elems_1s(vd_vptr, vta, env->vl * esz, total_elems * esz);
 +    env->vstart = 0;
 +}
 +
 +static inline uint32_t ff1(uint32_t x, uint32_t y, uint32_t z)
 +{
 +    return x ^ y ^ z;
 +}
 +
 +static inline uint32_t ff2(uint32_t x, uint32_t y, uint32_t z)
 +{
 +    return (x & y) | (x & z) | (y & z);
 +}
 +
 +static inline uint32_t ff_j(uint32_t x, uint32_t y, uint32_t z, uint32_t j)
 +{
 +    return (j <= 15) ? ff1(x, y, z) : ff2(x, y, z);
 +}
 +
 +static inline uint32_t gg1(uint32_t x, uint32_t y, uint32_t z)
 +{
 +    return x ^ y ^ z;
 +}
 +
 +static inline uint32_t gg2(uint32_t x, uint32_t y, uint32_t z)
 +{
 +    return (x & y) | (~x & z);
 +}
 +
 +static inline uint32_t gg_j(uint32_t x, uint32_t y, uint32_t z, uint32_t j)
 +{
 +    return (j <= 15) ? gg1(x, y, z) : gg2(x, y, z);
 +}
 +
 +static inline uint32_t t_j(uint32_t j)
 +{
 +    return (j <= 15) ? 0x79cc4519 : 0x7a879d8a;
 +}
 +
 +static inline uint32_t p_0(uint32_t x)
 +{
 +    return x ^ rol32(x, 9) ^ rol32(x, 17);
 +}
 +
 +static void sm3c(uint32_t *vd, uint32_t *vs1, uint32_t *vs2, uint32_t uimm)
 +{
 +    uint32_t x0, x1;
 +    uint32_t j;
 +    uint32_t ss1, ss2, tt1, tt2;
 +    x0 = vs2[0] ^ vs2[4];
 +    x1 = vs2[1] ^ vs2[5];
 +    j = 2 * uimm;
 +    ss1 = rol32(rol32(vs1[0], 12) + vs1[4] + rol32(t_j(j), j % 32), 7);
 +    ss2 = ss1 ^ rol32(vs1[0], 12);
 +    tt1 = ff_j(vs1[0], vs1[1], vs1[2], j) + vs1[3] + ss2 + x0;
 +    tt2 = gg_j(vs1[4], vs1[5], vs1[6], j) + vs1[7] + ss1 + vs2[0];
 +    vs1[3] = vs1[2];
 +    vd[3] = rol32(vs1[1], 9);
 +    vs1[1] = vs1[0];
 +    vd[1] = tt1;
 +    vs1[7] = vs1[6];
 +    vd[7] = rol32(vs1[5], 19);
 +    vs1[5] = vs1[4];
 +    vd[5] = p_0(tt2);
 +    j = 2 * uimm + 1;
 +    ss1 = rol32(rol32(vd[1], 12) + vd[5] + rol32(t_j(j), j % 32), 7);
 +    ss2 = ss1 ^ rol32(vd[1], 12);
 +    tt1 = ff_j(vd[1], vs1[1], vd[3], j) + vs1[3] + ss2 + x1;
 +    tt2 = gg_j(vd[5], vs1[5], vd[7], j) + vs1[7] + ss1 + vs2[1];
 +    vd[2] = rol32(vs1[1], 9);
 +    vd[0] = tt1;
 +    vd[6] = rol32(vs1[5], 19);
 +    vd[4] = p_0(tt2);
 +}
 +
 +void HELPER(vsm3c_vi)(void *vd_vptr, void *vs2_vptr, uint32_t uimm,
 +                      CPURISCVState *env, uint32_t desc)
 +{
 +    uint32_t esz = memop_size(FIELD_EX64(env->vtype, VTYPE, VSEW));
 +    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
 +    uint32_t vta = vext_vta(desc);
 +    uint32_t *vd = vd_vptr;
 +    uint32_t *vs2 = vs2_vptr;
 +    uint32_t v1[8], v2[8], v3[8];
 +
 +    for (int i = env->vstart / 8; i < env->vl / 8; i++) {
 +        for (int k = 0; k < 8; k++) {
 +            v2[k] = bswap32(vd[H4(i * 8 + k)]);
 +            v3[k] = bswap32(vs2[H4(i * 8 + k)]);
 +        }
 +        sm3c(v1, v2, v3, uimm);
 +        for (int k = 0; k < 8; k++) {
 +            vd[i * 8 + k] = bswap32(v1[H4(k)]);
 +        }
 +    }
 +    vext_set_elems_1s(vd_vptr, vta, env->vl * esz, total_elems * esz);
 +    env->vstart = 0;
 +}
 diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/insn_trans/trans_rvvk.c.inc
 +++ b/target/riscv/insn_trans/trans_rvvk.c.inc
@@ -XXX,XX +XXX,XX @@ static bool trans_vsha2ch_vv(DisasContext *s, arg_rmrr *a)
      }
      return false;
  }
 +
 +/*
 + * Zvksh
 + */
 +
 +#define ZVKSH_EGS 8
 +
 +static inline bool vsm3_check(DisasContext *s, arg_rmrr *a)
 +{
 +    int egw_bytes = ZVKSH_EGS << s->sew;
 +    int mult = 1 << MAX(s->lmul, 0);
 +    return s->cfg_ptr->ext_zvksh == true &&
 +           require_rvv(s) &&
 +           vext_check_isa_ill(s) &&
 +           !is_overlapped(a->rd, mult, a->rs2, mult) &&
 +           MAXSZ(s) >= egw_bytes &&
 +           s->sew == MO_32;
 +}
 +
 +static inline bool vsm3me_check(DisasContext *s, arg_rmrr *a)
 +{
 +    return vsm3_check(s, a) && vext_check_sss(s, a->rd, a->rs1, a->rs2, a->vm);
 +}
 +
 +static inline bool vsm3c_check(DisasContext *s, arg_rmrr *a)
 +{
 +    return vsm3_check(s, a) && vext_check_ss(s, a->rd, a->rs2, a->vm);
 +}
 +
 +GEN_VV_UNMASKED_TRANS(vsm3me_vv, vsm3me_check, ZVKSH_EGS)
 +GEN_VI_UNMASKED_TRANS(vsm3c_vi, vsm3c_check, ZVKSH_EGS)
 --
 .41.0

-New patch
+[PULL v2 19/45] target/riscv: Add Zvkg ISA extension support
+From: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
 This commit adds support for the Zvkg vector-crypto extension, which
 consists of the following instructions:
 * vgmul.vv
 * vghsh.vv
 Translation functions are defined in
 `target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
 `target/riscv/vcrypto_helper.c`.
 Co-authored-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
 [max.chou@sifive.com: Replaced vstart checking by TCG op]
 Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
 Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
 Signed-off-by: Max Chou <max.chou@sifive.com>
 Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
 [max.chou@sifive.com: Exposed x-zvkg property]
 [max.chou@sifive.com: Replaced uint by int for cross win32 build]
 Message-ID: <20230711165917.2629866-13-max.chou@sifive.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
  target/riscv/cpu_cfg.h                   |  1 +
  target/riscv/helper.h                    |  3 +
  target/riscv/insn32.decode               |  4 ++
  target/riscv/cpu.c                       |  6 +-
  target/riscv/vcrypto_helper.c            | 72 ++++++++++++++++++++++++
  target/riscv/insn_trans/trans_rvvk.c.inc | 30 ++++++++++
 files changed, 114 insertions(+), 2 deletions(-)
 diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/cpu_cfg.h
 +++ b/target/riscv/cpu_cfg.h
@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
      bool ext_zve64d;
      bool ext_zvbb;
      bool ext_zvbc;
 +    bool ext_zvkg;
      bool ext_zvkned;
      bool ext_zvknha;
      bool ext_zvknhb;
 diff --git a/target/riscv/helper.h b/target/riscv/helper.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/helper.h
 +++ b/target/riscv/helper.h
@@ -XXX,XX +XXX,XX @@ DEF_HELPER_5(vsha2cl64_vv, void, ptr, ptr, ptr, env, i32)
  DEF_HELPER_5(vsm3me_vv, void, ptr, ptr, ptr, env, i32)
  DEF_HELPER_5(vsm3c_vi, void, ptr, ptr, i32, env, i32)
 +
 +DEF_HELPER_5(vghsh_vv, void, ptr, ptr, ptr, env, i32)
 +DEF_HELPER_4(vgmul_vv, void, ptr, ptr, env, i32)
 diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/insn32.decode
 +++ b/target/riscv/insn32.decode
@@ -XXX,XX +XXX,XX @@ vsha2cl_vv  101111 1 ..... ..... 010 ..... 1110111 @r_vm_1
  # *** Zvksh vector crypto extension ***
  vsm3me_vv   100000 1 ..... ..... 010 ..... 1110111 @r_vm_1
  vsm3c_vi    101011 1 ..... ..... 010 ..... 1110111 @r_vm_1
 +
 +# *** Zvkg vector crypto extension ***
 +vghsh_vv    101100 1 ..... ..... 010 ..... 1110111 @r_vm_1
 +vgmul_vv    101000 1 ..... 10001 010 ..... 1110111 @r2_vm_1
 diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/cpu.c
 +++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
      ISA_EXT_DATA_ENTRY(zvfbfwma, PRIV_VERSION_1_12_0, ext_zvfbfwma),
      ISA_EXT_DATA_ENTRY(zvfh, PRIV_VERSION_1_12_0, ext_zvfh),
      ISA_EXT_DATA_ENTRY(zvfhmin, PRIV_VERSION_1_12_0, ext_zvfhmin),
 +    ISA_EXT_DATA_ENTRY(zvkg, PRIV_VERSION_1_12_0, ext_zvkg),
      ISA_EXT_DATA_ENTRY(zvkned, PRIV_VERSION_1_12_0, ext_zvkned),
      ISA_EXT_DATA_ENTRY(zvknha, PRIV_VERSION_1_12_0, ext_zvknha),
      ISA_EXT_DATA_ENTRY(zvknhb, PRIV_VERSION_1_12_0, ext_zvknhb),
@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
       * In principle Zve*x would also suffice here, were they supported
       * in qemu
       */
 -    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned || cpu->cfg.ext_zvknha ||
 -         cpu->cfg.ext_zvksh) && !cpu->cfg.ext_zve32f) {
 +    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkg || cpu->cfg.ext_zvkned ||
 +         cpu->cfg.ext_zvknha || cpu->cfg.ext_zvksh) && !cpu->cfg.ext_zve32f) {
          error_setg(errp,
                     "Vector crypto extensions require V or Zve* extensions");
          return;
@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
      /* Vector cryptography extensions */
      DEFINE_PROP_BOOL("x-zvbb", RISCVCPU, cfg.ext_zvbb, false),
      DEFINE_PROP_BOOL("x-zvbc", RISCVCPU, cfg.ext_zvbc, false),
 +    DEFINE_PROP_BOOL("x-zvkg", RISCVCPU, cfg.ext_zvkg, false),
      DEFINE_PROP_BOOL("x-zvkned", RISCVCPU, cfg.ext_zvkned, false),
      DEFINE_PROP_BOOL("x-zvknha", RISCVCPU, cfg.ext_zvknha, false),
      DEFINE_PROP_BOOL("x-zvknhb", RISCVCPU, cfg.ext_zvknhb, false),
 diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/vcrypto_helper.c
 +++ b/target/riscv/vcrypto_helper.c
@@ -XXX,XX +XXX,XX @@ void HELPER(vsm3c_vi)(void *vd_vptr, void *vs2_vptr, uint32_t uimm,
      vext_set_elems_1s(vd_vptr, vta, env->vl * esz, total_elems * esz);
      env->vstart = 0;
  }
 +
 +void HELPER(vghsh_vv)(void *vd_vptr, void *vs1_vptr, void *vs2_vptr,
 +                      CPURISCVState *env, uint32_t desc)
 +{
 +    uint64_t *vd = vd_vptr;
 +    uint64_t *vs1 = vs1_vptr;
 +    uint64_t *vs2 = vs2_vptr;
 +    uint32_t vta = vext_vta(desc);
 +    uint32_t total_elems = vext_get_total_elems(env, desc, 4);
 +
 +    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
 +        uint64_t Y[2] = {vd[i * 2 + 0], vd[i * 2 + 1]};
 +        uint64_t H[2] = {brev8(vs2[i * 2 + 0]), brev8(vs2[i * 2 + 1])};
 +        uint64_t X[2] = {vs1[i * 2 + 0], vs1[i * 2 + 1]};
 +        uint64_t Z[2] = {0, 0};
 +
 +        uint64_t S[2] = {brev8(Y[0] ^ X[0]), brev8(Y[1] ^ X[1])};
 +
 +        for (int j = 0; j < 128; j++) {
 +            if ((S[j / 64] >> (j % 64)) & 1) {
 +                Z[0] ^= H[0];
 +                Z[1] ^= H[1];
 +            }
 +            bool reduce = ((H[1] >> 63) & 1);
 +            H[1] = H[1] << 1 | H[0] >> 63;
 +            H[0] = H[0] << 1;
 +            if (reduce) {
 +                H[0] ^= 0x87;
 +            }
 +        }
 +
 +        vd[i * 2 + 0] = brev8(Z[0]);
 +        vd[i * 2 + 1] = brev8(Z[1]);
 +    }
 +    /* set tail elements to 1s */
 +    vext_set_elems_1s(vd, vta, env->vl * 4, total_elems * 4);
 +    env->vstart = 0;
 +}
 +
 +void HELPER(vgmul_vv)(void *vd_vptr, void *vs2_vptr, CPURISCVState *env,
 +                      uint32_t desc)
 +{
 +    uint64_t *vd = vd_vptr;
 +    uint64_t *vs2 = vs2_vptr;
 +    uint32_t vta = vext_vta(desc);
 +    uint32_t total_elems = vext_get_total_elems(env, desc, 4);
 +
 +    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
 +        uint64_t Y[2] = {brev8(vd[i * 2 + 0]), brev8(vd[i * 2 + 1])};
 +        uint64_t H[2] = {brev8(vs2[i * 2 + 0]), brev8(vs2[i * 2 + 1])};
 +        uint64_t Z[2] = {0, 0};
 +
 +        for (int j = 0; j < 128; j++) {
 +            if ((Y[j / 64] >> (j % 64)) & 1) {
 +                Z[0] ^= H[0];
 +                Z[1] ^= H[1];
 +            }
 +            bool reduce = ((H[1] >> 63) & 1);
 +            H[1] = H[1] << 1 | H[0] >> 63;
 +            H[0] = H[0] << 1;
 +            if (reduce) {
 +                H[0] ^= 0x87;
 +            }
 +        }
 +
 +        vd[i * 2 + 0] = brev8(Z[0]);
 +        vd[i * 2 + 1] = brev8(Z[1]);
 +    }
 +    /* set tail elements to 1s */
 +    vext_set_elems_1s(vd, vta, env->vl * 4, total_elems * 4);
 +    env->vstart = 0;
 +}
 diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/insn_trans/trans_rvvk.c.inc
 +++ b/target/riscv/insn_trans/trans_rvvk.c.inc
@@ -XXX,XX +XXX,XX @@ static inline bool vsm3c_check(DisasContext *s, arg_rmrr *a)
  GEN_VV_UNMASKED_TRANS(vsm3me_vv, vsm3me_check, ZVKSH_EGS)
  GEN_VI_UNMASKED_TRANS(vsm3c_vi, vsm3c_check, ZVKSH_EGS)
 +
 +/*
 + * Zvkg
 + */
 +
 +#define ZVKG_EGS 4
 +
 +static bool vgmul_check(DisasContext *s, arg_rmr *a)
 +{
 +    int egw_bytes = ZVKG_EGS << s->sew;
 +    return s->cfg_ptr->ext_zvkg == true &&
 +           vext_check_isa_ill(s) &&
 +           require_rvv(s) &&
 +           MAXSZ(s) >= egw_bytes &&
 +           vext_check_ss(s, a->rd, a->rs2, a->vm) &&
 +           s->sew == MO_32;
 +}
 +
 +GEN_V_UNMASKED_TRANS(vgmul_vv, vgmul_check, ZVKG_EGS)
 +
 +static bool vghsh_check(DisasContext *s, arg_rmrr *a)
 +{
 +    int egw_bytes = ZVKG_EGS << s->sew;
 +    return s->cfg_ptr->ext_zvkg == true &&
 +           opivv_check(s, a) &&
 +           MAXSZ(s) >= egw_bytes &&
 +           s->sew == MO_32;
 +}
 +
 +GEN_VV_UNMASKED_TRANS(vghsh_vv, vghsh_check, ZVKG_EGS)
 --
 .41.0

-New patch
+[PULL v2 20/45] crypto: Create sm4_subword
+From: Max Chou <max.chou@sifive.com>
+Allows sharing of sm4_subword between different targets.
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Reviewed-by: Frank Chang <frank.chang@sifive.com>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Message-ID: <20230711165917.2629866-14-max.chou@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ include/crypto/sm4.h           |  8 ++++++++
+ target/arm/tcg/crypto_helper.c | 10 ++--------
+files changed, 10 insertions(+), 8 deletions(-)
+diff --git a/include/crypto/sm4.h b/include/crypto/sm4.h
+index XXXXXXX..XXXXXXX 100644
+--- a/include/crypto/sm4.h
++++ b/include/crypto/sm4.h
+@@ -XXX,XX +XXX,XX @@
+ extern const uint8_t sm4_sbox[256];
++static inline uint32_t sm4_subword(uint32_t word)
++{
++    return sm4_sbox[word & 0xff] |
++           sm4_sbox[(word >> 8) & 0xff] << 8 |
++           sm4_sbox[(word >> 16) & 0xff] << 16 |
++           sm4_sbox[(word >> 24) & 0xff] << 24;
++}
++
+ #endif
+diff --git a/target/arm/tcg/crypto_helper.c b/target/arm/tcg/crypto_helper.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/arm/tcg/crypto_helper.c
++++ b/target/arm/tcg/crypto_helper.c
+@@ -XXX,XX +XXX,XX @@ static void do_crypto_sm4e(uint64_t *rd, uint64_t *rn, uint64_t *rm)
+             CR_ST_WORD(d, (i + 3) % 4) ^
+             CR_ST_WORD(n, i);
+-        t = sm4_sbox[t & 0xff] |
+-            sm4_sbox[(t >> 8) & 0xff] << 8 |
+-            sm4_sbox[(t >> 16) & 0xff] << 16 |
+-            sm4_sbox[(t >> 24) & 0xff] << 24;
++        t = sm4_subword(t);
+         CR_ST_WORD(d, i) ^= t ^ rol32(t, 2) ^ rol32(t, 10) ^ rol32(t, 18) ^
+                             rol32(t, 24);
+@@ -XXX,XX +XXX,XX @@ static void do_crypto_sm4ekey(uint64_t *rd, uint64_t *rn, uint64_t *rm)
+             CR_ST_WORD(d, (i + 3) % 4) ^
+             CR_ST_WORD(m, i);
+-        t = sm4_sbox[t & 0xff] |
+-            sm4_sbox[(t >> 8) & 0xff] << 8 |
+-            sm4_sbox[(t >> 16) & 0xff] << 16 |
+-            sm4_sbox[(t >> 24) & 0xff] << 24;
++        t = sm4_subword(t);
+         CR_ST_WORD(d, i) ^= t ^ rol32(t, 13) ^ rol32(t, 23);
+     }
+--
+.41.0

-New patch
+[PULL v2 21/45] crypto: Add SM4 constant parameter CK
+From: Max Chou <max.chou@sifive.com>
+Adds sm4_ck constant for use in sm4 cryptography across different targets.
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Reviewed-by: Frank Chang <frank.chang@sifive.com>
+Signed-off-by: Max Chou <max.chou@sifive.com>
+Message-ID: <20230711165917.2629866-15-max.chou@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ include/crypto/sm4.h |  1 +
+ crypto/sm4.c         | 10 ++++++++++
+files changed, 11 insertions(+)
+diff --git a/include/crypto/sm4.h b/include/crypto/sm4.h
+index XXXXXXX..XXXXXXX 100644
+--- a/include/crypto/sm4.h
++++ b/include/crypto/sm4.h
+@@ -XXX,XX +XXX,XX @@
+ #define QEMU_SM4_H
+ extern const uint8_t sm4_sbox[256];
++extern const uint32_t sm4_ck[32];
+ static inline uint32_t sm4_subword(uint32_t word)
+ {
+diff --git a/crypto/sm4.c b/crypto/sm4.c
+index XXXXXXX..XXXXXXX 100644
+--- a/crypto/sm4.c
++++ b/crypto/sm4.c
+@@ -XXX,XX +XXX,XX @@ uint8_t const sm4_sbox[] = {
+x79, 0xee, 0x5f, 0x3e, 0xd7, 0xcb, 0x39, 0x48,
+ };
++uint32_t const sm4_ck[] = {
++    0x00070e15, 0x1c232a31, 0x383f464d, 0x545b6269,
++    0x70777e85, 0x8c939aa1, 0xa8afb6bd, 0xc4cbd2d9,
++    0xe0e7eef5, 0xfc030a11, 0x181f262d, 0x343b4249,
++    0x50575e65, 0x6c737a81, 0x888f969d, 0xa4abb2b9,
++    0xc0c7ced5, 0xdce3eaf1, 0xf8ff060d, 0x141b2229,
++    0x30373e45, 0x4c535a61, 0x686f767d, 0x848b9299,
++    0xa0a7aeb5, 0xbcc3cad1, 0xd8dfe6ed, 0xf4fb0209,
++    0x10171e25, 0x2c333a41, 0x484f565d, 0x646b7279
++};
+--
+.41.0

-New patch
+[PULL v2 22/45] target/riscv: Add Zvksed ISA extension support
+From: Max Chou <max.chou@sifive.com>
 This commit adds support for the Zvksed vector-crypto extension, which
 consists of the following instructions:
 * vsm4k.vi
 * vsm4r.[vv,vs]
 Translation functions are defined in
 `target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
 `target/riscv/vcrypto_helper.c`.
 Signed-off-by: Max Chou <max.chou@sifive.com>
 Reviewed-by: Frank Chang <frank.chang@sifive.com>
 [lawrence.hunter@codethink.co.uk: Moved SM4 functions from
 crypto_helper.c to vcrypto_helper.c]
 [nazar.kazakov@codethink.co.uk: Added alignment checks, refactored code to
 use macros, and minor style changes]
 Signed-off-by: Max Chou <max.chou@sifive.com>
 Message-ID: <20230711165917.2629866-16-max.chou@sifive.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
  target/riscv/cpu_cfg.h                   |   1 +
  target/riscv/helper.h                    |   4 +
  target/riscv/insn32.decode               |   5 +
  target/riscv/cpu.c                       |   5 +-
  target/riscv/vcrypto_helper.c            | 127 +++++++++++++++++++++++
  target/riscv/insn_trans/trans_rvvk.c.inc |  43 ++++++++
 files changed, 184 insertions(+), 1 deletion(-)
 diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/cpu_cfg.h
 +++ b/target/riscv/cpu_cfg.h
@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
      bool ext_zvkned;
      bool ext_zvknha;
      bool ext_zvknhb;
 +    bool ext_zvksed;
      bool ext_zvksh;
      bool ext_zmmul;
      bool ext_zvfbfmin;
 diff --git a/target/riscv/helper.h b/target/riscv/helper.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/helper.h
 +++ b/target/riscv/helper.h
@@ -XXX,XX +XXX,XX @@ DEF_HELPER_5(vsm3c_vi, void, ptr, ptr, i32, env, i32)
  DEF_HELPER_5(vghsh_vv, void, ptr, ptr, ptr, env, i32)
  DEF_HELPER_4(vgmul_vv, void, ptr, ptr, env, i32)
 +
 +DEF_HELPER_5(vsm4k_vi, void, ptr, ptr, i32, env, i32)
 +DEF_HELPER_4(vsm4r_vv, void, ptr, ptr, env, i32)
 +DEF_HELPER_4(vsm4r_vs, void, ptr, ptr, env, i32)
 diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/insn32.decode
 +++ b/target/riscv/insn32.decode
@@ -XXX,XX +XXX,XX @@ vsm3c_vi    101011 1 ..... ..... 010 ..... 1110111 @r_vm_1
  # *** Zvkg vector crypto extension ***
  vghsh_vv    101100 1 ..... ..... 010 ..... 1110111 @r_vm_1
  vgmul_vv    101000 1 ..... 10001 010 ..... 1110111 @r2_vm_1
 +
 +# *** Zvksed vector crypto extension ***
 +vsm4k_vi    100001 1 ..... ..... 010 ..... 1110111 @r_vm_1
 +vsm4r_vv    101000 1 ..... 10000 010 ..... 1110111 @r2_vm_1
 +vsm4r_vs    101001 1 ..... 10000 010 ..... 1110111 @r2_vm_1
 diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/cpu.c
 +++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
      ISA_EXT_DATA_ENTRY(zvkned, PRIV_VERSION_1_12_0, ext_zvkned),
      ISA_EXT_DATA_ENTRY(zvknha, PRIV_VERSION_1_12_0, ext_zvknha),
      ISA_EXT_DATA_ENTRY(zvknhb, PRIV_VERSION_1_12_0, ext_zvknhb),
 +    ISA_EXT_DATA_ENTRY(zvksed, PRIV_VERSION_1_12_0, ext_zvksed),
      ISA_EXT_DATA_ENTRY(zvksh, PRIV_VERSION_1_12_0, ext_zvksh),
      ISA_EXT_DATA_ENTRY(zhinx, PRIV_VERSION_1_12_0, ext_zhinx),
      ISA_EXT_DATA_ENTRY(zhinxmin, PRIV_VERSION_1_12_0, ext_zhinxmin),
@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
       * in qemu
       */
      if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkg || cpu->cfg.ext_zvkned ||
 -         cpu->cfg.ext_zvknha || cpu->cfg.ext_zvksh) && !cpu->cfg.ext_zve32f) {
 +         cpu->cfg.ext_zvknha || cpu->cfg.ext_zvksed || cpu->cfg.ext_zvksh) &&
 +        !cpu->cfg.ext_zve32f) {
          error_setg(errp,
                     "Vector crypto extensions require V or Zve* extensions");
          return;
@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
      DEFINE_PROP_BOOL("x-zvkned", RISCVCPU, cfg.ext_zvkned, false),
      DEFINE_PROP_BOOL("x-zvknha", RISCVCPU, cfg.ext_zvknha, false),
      DEFINE_PROP_BOOL("x-zvknhb", RISCVCPU, cfg.ext_zvknhb, false),
 +    DEFINE_PROP_BOOL("x-zvksed", RISCVCPU, cfg.ext_zvksed, false),
      DEFINE_PROP_BOOL("x-zvksh", RISCVCPU, cfg.ext_zvksh, false),
      DEFINE_PROP_END_OF_LIST(),
 diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/vcrypto_helper.c
 +++ b/target/riscv/vcrypto_helper.c
@@ -XXX,XX +XXX,XX @@
  #include "cpu.h"
  #include "crypto/aes.h"
  #include "crypto/aes-round.h"
 +#include "crypto/sm4.h"
  #include "exec/memop.h"
  #include "exec/exec-all.h"
  #include "exec/helper-proto.h"
@@ -XXX,XX +XXX,XX @@ void HELPER(vgmul_vv)(void *vd_vptr, void *vs2_vptr, CPURISCVState *env,
      vext_set_elems_1s(vd, vta, env->vl * 4, total_elems * 4);
      env->vstart = 0;
  }
 +
 +void HELPER(vsm4k_vi)(void *vd, void *vs2, uint32_t uimm5, CPURISCVState *env,
 +                      uint32_t desc)
 +{
 +    const uint32_t egs = 4;
 +    uint32_t rnd = uimm5 & 0x7;
 +    uint32_t group_start = env->vstart / egs;
 +    uint32_t group_end = env->vl / egs;
 +    uint32_t esz = sizeof(uint32_t);
 +    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
 +
 +    for (uint32_t i = group_start; i < group_end; ++i) {
 +        uint32_t vstart = i * egs;
 +        uint32_t vend = (i + 1) * egs;
 +        uint32_t rk[4] = {0};
 +        uint32_t tmp[8] = {0};
 +
 +        for (uint32_t j = vstart; j < vend; ++j) {
 +            rk[j - vstart] = *((uint32_t *)vs2 + H4(j));
 +        }
 +
 +        for (uint32_t j = 0; j < egs; ++j) {
 +            tmp[j] = rk[j];
 +        }
 +
 +        for (uint32_t j = 0; j < egs; ++j) {
 +            uint32_t b, s;
 +            b = tmp[j + 1] ^ tmp[j + 2] ^ tmp[j + 3] ^ sm4_ck[rnd * 4 + j];
 +
 +            s = sm4_subword(b);
 +
 +            tmp[j + 4] = tmp[j] ^ (s ^ rol32(s, 13) ^ rol32(s, 23));
 +        }
 +
 +        for (uint32_t j = vstart; j < vend; ++j) {
 +            *((uint32_t *)vd + H4(j)) = tmp[egs + (j - vstart)];
 +        }
 +    }
 +
 +    env->vstart = 0;
 +    /* set tail elements to 1s */
 +    vext_set_elems_1s(vd, vext_vta(desc), env->vl * esz, total_elems * esz);
 +}
 +
 +static void do_sm4_round(uint32_t *rk, uint32_t *buf)
 +{
 +    const uint32_t egs = 4;
 +    uint32_t s, b;
 +
 +    for (uint32_t j = egs; j < egs * 2; ++j) {
 +        b = buf[j - 3] ^ buf[j - 2] ^ buf[j - 1] ^ rk[j - 4];
 +
 +        s = sm4_subword(b);
 +
 +        buf[j] = buf[j - 4] ^ (s ^ rol32(s, 2) ^ rol32(s, 10) ^ rol32(s, 18) ^
 +                               rol32(s, 24));
 +    }
 +}
 +
 +void HELPER(vsm4r_vv)(void *vd, void *vs2, CPURISCVState *env, uint32_t desc)
 +{
 +    const uint32_t egs = 4;
 +    uint32_t group_start = env->vstart / egs;
 +    uint32_t group_end = env->vl / egs;
 +    uint32_t esz = sizeof(uint32_t);
 +    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
 +
 +    for (uint32_t i = group_start; i < group_end; ++i) {
 +        uint32_t vstart = i * egs;
 +        uint32_t vend = (i + 1) * egs;
 +        uint32_t rk[4] = {0};
 +        uint32_t tmp[8] = {0};
 +
 +        for (uint32_t j = vstart; j < vend; ++j) {
 +            rk[j - vstart] = *((uint32_t *)vs2 + H4(j));
 +        }
 +
 +        for (uint32_t j = vstart; j < vend; ++j) {
 +            tmp[j - vstart] = *((uint32_t *)vd + H4(j));
 +        }
 +
 +        do_sm4_round(rk, tmp);
 +
 +        for (uint32_t j = vstart; j < vend; ++j) {
 +            *((uint32_t *)vd + H4(j)) = tmp[egs + (j - vstart)];
 +        }
 +    }
 +
 +    env->vstart = 0;
 +    /* set tail elements to 1s */
 +    vext_set_elems_1s(vd, vext_vta(desc), env->vl * esz, total_elems * esz);
 +}
 +
 +void HELPER(vsm4r_vs)(void *vd, void *vs2, CPURISCVState *env, uint32_t desc)
 +{
 +    const uint32_t egs = 4;
 +    uint32_t group_start = env->vstart / egs;
 +    uint32_t group_end = env->vl / egs;
 +    uint32_t esz = sizeof(uint32_t);
 +    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
 +
 +    for (uint32_t i = group_start; i < group_end; ++i) {
 +        uint32_t vstart = i * egs;
 +        uint32_t vend = (i + 1) * egs;
 +        uint32_t rk[4] = {0};
 +        uint32_t tmp[8] = {0};
 +
 +        for (uint32_t j = 0; j < egs; ++j) {
 +            rk[j] = *((uint32_t *)vs2 + H4(j));
 +        }
 +
 +        for (uint32_t j = vstart; j < vend; ++j) {
 +            tmp[j - vstart] = *((uint32_t *)vd + H4(j));
 +        }
 +
 +        do_sm4_round(rk, tmp);
 +
 +        for (uint32_t j = vstart; j < vend; ++j) {
 +            *((uint32_t *)vd + H4(j)) = tmp[egs + (j - vstart)];
 +        }
 +    }
 +
 +    env->vstart = 0;
 +    /* set tail elements to 1s */
 +    vext_set_elems_1s(vd, vext_vta(desc), env->vl * esz, total_elems * esz);
 +}
 diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/insn_trans/trans_rvvk.c.inc
 +++ b/target/riscv/insn_trans/trans_rvvk.c.inc
@@ -XXX,XX +XXX,XX @@ static bool vghsh_check(DisasContext *s, arg_rmrr *a)
  }
  GEN_VV_UNMASKED_TRANS(vghsh_vv, vghsh_check, ZVKG_EGS)
 +
 +/*
 + * Zvksed
 + */
 +
 +#define ZVKSED_EGS 4
 +
 +static bool zvksed_check(DisasContext *s)
 +{
 +    int egw_bytes = ZVKSED_EGS << s->sew;
 +    return s->cfg_ptr->ext_zvksed == true &&
 +           require_rvv(s) &&
 +           vext_check_isa_ill(s) &&
 +           MAXSZ(s) >= egw_bytes &&
 +           s->sew == MO_32;
 +}
 +
 +static bool vsm4k_vi_check(DisasContext *s, arg_rmrr *a)
 +{
 +    return zvksed_check(s) &&
 +           require_align(a->rd, s->lmul) &&
 +           require_align(a->rs2, s->lmul);
 +}
 +
 +GEN_VI_UNMASKED_TRANS(vsm4k_vi, vsm4k_vi_check, ZVKSED_EGS)
 +
 +static bool vsm4r_vv_check(DisasContext *s, arg_rmr *a)
 +{
 +    return zvksed_check(s) &&
 +           require_align(a->rd, s->lmul) &&
 +           require_align(a->rs2, s->lmul);
 +}
 +
 +GEN_V_UNMASKED_TRANS(vsm4r_vv, vsm4r_vv_check, ZVKSED_EGS)
 +
 +static bool vsm4r_vs_check(DisasContext *s, arg_rmr *a)
 +{
 +    return zvksed_check(s) &&
 +           !is_overlapped(a->rd, 1 << MAX(s->lmul, 0), a->rs2, 1) &&
 +           require_align(a->rd, s->lmul);
 +}
 +
 +GEN_V_UNMASKED_TRANS(vsm4r_vs, vsm4r_vs_check, ZVKSED_EGS)
 --
 .41.0

-New patch
+[PULL v2 23/45] target/riscv: Implement WARL behaviour for mcountinhibit/mcounteren
+From: Rob Bradford <rbradford@rivosinc.com>
+These are WARL fields - zero out the bits for unavailable counters and
+special case the TM bit in mcountinhibit which is hardwired to zero.
+This patch achieves this by modifying the value written so that any use
+of the field will see the correctly masked bits.
+Tested by modifying OpenSBI to write max value to these CSRs and upon
+subsequent read the appropriate number of bits for number of PMUs is
+enabled and the TM bit is zero in mcountinhibit.
+Signed-off-by: Rob Bradford <rbradford@rivosinc.com>
+Acked-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Atish Patra <atishp@rivosinc.com>
+Message-ID: <20230802124906.24197-1-rbradford@rivosinc.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/csr.c | 11 +++++++++--
+file changed, 9 insertions(+), 2 deletions(-)
+diff --git a/target/riscv/csr.c b/target/riscv/csr.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/csr.c
++++ b/target/riscv/csr.c
+@@ -XXX,XX +XXX,XX @@ static RISCVException write_mcountinhibit(CPURISCVState *env, int csrno,
+ {
+     int cidx;
+     PMUCTRState *counter;
++    RISCVCPU *cpu = env_archcpu(env);
+-    env->mcountinhibit = val;
++    /* WARL register - disable unavailable counters; TM bit is always 0 */
++    env->mcountinhibit =
++        val & (cpu->pmu_avail_ctrs | COUNTEREN_CY | COUNTEREN_IR);
+     /* Check if any other counter is also monitoring cycles/instructions */
+     for (cidx = 0; cidx < RV_MAX_MHPMCOUNTERS; cidx++) {
+@@ -XXX,XX +XXX,XX @@ static RISCVException read_mcounteren(CPURISCVState *env, int csrno,
+ static RISCVException write_mcounteren(CPURISCVState *env, int csrno,
+                                        target_ulong val)
+ {
+-    env->mcounteren = val;
++    RISCVCPU *cpu = env_archcpu(env);
++
++    /* WARL register - disable unavailable counters */
++    env->mcounteren = val & (cpu->pmu_avail_ctrs | COUNTEREN_CY | COUNTEREN_TM |
++                             COUNTEREN_IR);
+     return RISCV_EXCP_NONE;
+ }
+--
+.41.0

-New patch
+[PULL v2 24/45] target/riscv: Add Zihintntl extension ISA string to DTS
+From: Jason Chien <jason.chien@sifive.com>
+RVA23 Profiles states:
+The RVA23 profiles are intended to be used for 64-bit application
+processors that will run rich OS stacks from standard binary OS
+distributions and with a substantial number of third-party binary user
+applications that will be supported over a considerable length of time
+in the field.
+The chapter 4 of the unprivileged spec introduces the Zihintntl extension
+and Zihintntl is a mandatory extension presented in RVA23 Profiles, whose
+purpose is to enable application and operating system portability across
+different implementations. Thus the DTS should contain the Zihintntl ISA
+string in order to pass to software.
+The unprivileged spec states:
+Like any HINTs, these instructions may be freely ignored. Hence, although
+they are described in terms of cache-based memory hierarchies, they do not
+mandate the provision of caches.
+These instructions are encoded with non-used opcode, e.g. ADD x0, x0, x2,
+which QEMU already supports, and QEMU does not emulate cache. Therefore
+these instructions can be considered as a no-op, and we only need to add
+a new property for the Zihintntl extension.
+Reviewed-by: Frank Chang <frank.chang@sifive.com>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Signed-off-by: Jason Chien <jason.chien@sifive.com>
+Message-ID: <20230726074049.19505-2-jason.chien@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/cpu_cfg.h | 1 +
+ target/riscv/cpu.c     | 2 ++
+files changed, 3 insertions(+)
+diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu_cfg.h
++++ b/target/riscv/cpu_cfg.h
+@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
+     bool ext_icbom;
+     bool ext_icboz;
+     bool ext_zicond;
++    bool ext_zihintntl;
+     bool ext_zihintpause;
+     bool ext_smstateen;
+     bool ext_sstc;
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu.c
++++ b/target/riscv/cpu.c
+@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
+     ISA_EXT_DATA_ENTRY(zicond, PRIV_VERSION_1_12_0, ext_zicond),
+     ISA_EXT_DATA_ENTRY(zicsr, PRIV_VERSION_1_10_0, ext_icsr),
+     ISA_EXT_DATA_ENTRY(zifencei, PRIV_VERSION_1_10_0, ext_ifencei),
++    ISA_EXT_DATA_ENTRY(zihintntl, PRIV_VERSION_1_10_0, ext_zihintntl),
+     ISA_EXT_DATA_ENTRY(zihintpause, PRIV_VERSION_1_10_0, ext_zihintpause),
+     ISA_EXT_DATA_ENTRY(zmmul, PRIV_VERSION_1_12_0, ext_zmmul),
+     ISA_EXT_DATA_ENTRY(zawrs, PRIV_VERSION_1_12_0, ext_zawrs),
+@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
+     DEFINE_PROP_BOOL("sscofpmf", RISCVCPU, cfg.ext_sscofpmf, false),
+     DEFINE_PROP_BOOL("Zifencei", RISCVCPU, cfg.ext_ifencei, true),
+     DEFINE_PROP_BOOL("Zicsr", RISCVCPU, cfg.ext_icsr, true),
++    DEFINE_PROP_BOOL("Zihintntl", RISCVCPU, cfg.ext_zihintntl, true),
+     DEFINE_PROP_BOOL("Zihintpause", RISCVCPU, cfg.ext_zihintpause, true),
+     DEFINE_PROP_BOOL("Zawrs", RISCVCPU, cfg.ext_zawrs, true),
+     DEFINE_PROP_BOOL("Zfa", RISCVCPU, cfg.ext_zfa, true),
+--
+.41.0

-New patch
+[PULL v2 25/45] target/riscv: Fix zfa fleq.d and fltq.d
+From: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
+Commit a47842d ("riscv: Add support for the Zfa extension") implemented the zfa extension.
+However, it has some typos for fleq.d and fltq.d. Both of them misused the fltq.s
+helper function.
+Fixes: a47842d ("riscv: Add support for the Zfa extension")
+Signed-off-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
+Message-ID: <20230728003906.768-1-zhiwei_liu@linux.alibaba.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/insn_trans/trans_rvzfa.c.inc | 4 ++--
+file changed, 2 insertions(+), 2 deletions(-)
+diff --git a/target/riscv/insn_trans/trans_rvzfa.c.inc b/target/riscv/insn_trans/trans_rvzfa.c.inc
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/insn_trans/trans_rvzfa.c.inc
++++ b/target/riscv/insn_trans/trans_rvzfa.c.inc
+@@ -XXX,XX +XXX,XX @@ bool trans_fleq_d(DisasContext *ctx, arg_fleq_d *a)
+     TCGv_i64 src1 = get_fpr_hs(ctx, a->rs1);
+     TCGv_i64 src2 = get_fpr_hs(ctx, a->rs2);
+-    gen_helper_fltq_s(dest, cpu_env, src1, src2);
++    gen_helper_fleq_d(dest, cpu_env, src1, src2);
+     gen_set_gpr(ctx, a->rd, dest);
+     return true;
+ }
+@@ -XXX,XX +XXX,XX @@ bool trans_fltq_d(DisasContext *ctx, arg_fltq_d *a)
+     TCGv_i64 src1 = get_fpr_hs(ctx, a->rs1);
+     TCGv_i64 src2 = get_fpr_hs(ctx, a->rs2);
+-    gen_helper_fltq_s(dest, cpu_env, src1, src2);
++    gen_helper_fltq_d(dest, cpu_env, src1, src2);
+     gen_set_gpr(ctx, a->rd, dest);
+     return true;
+ }
+--
+.41.0

-New patch
+[PULL v2 26/45] hw/intc: Fix upper/lower mtime write calculation
+From: Jason Chien <jason.chien@sifive.com>
+When writing the upper mtime, we should keep the original lower mtime
+whose value is given by cpu_riscv_read_rtc() instead of
+cpu_riscv_read_rtc_raw(). The same logic applies to writes to lower mtime.
+Signed-off-by: Jason Chien <jason.chien@sifive.com>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Message-ID: <20230728082502.26439-1-jason.chien@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ hw/intc/riscv_aclint.c | 5 +++--
+file changed, 3 insertions(+), 2 deletions(-)
+diff --git a/hw/intc/riscv_aclint.c b/hw/intc/riscv_aclint.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/intc/riscv_aclint.c
++++ b/hw/intc/riscv_aclint.c
+@@ -XXX,XX +XXX,XX @@ static void riscv_aclint_mtimer_write(void *opaque, hwaddr addr,
+         return;
+     } else if (addr == mtimer->time_base || addr == mtimer->time_base + 4) {
+         uint64_t rtc_r = cpu_riscv_read_rtc_raw(mtimer->timebase_freq);
++        uint64_t rtc = cpu_riscv_read_rtc(mtimer);
+         if (addr == mtimer->time_base) {
+             if (size == 4) {
+                 /* time_lo for RV32/RV64 */
+-                mtimer->time_delta = ((rtc_r & ~0xFFFFFFFFULL) | value) - rtc_r;
++                mtimer->time_delta = ((rtc & ~0xFFFFFFFFULL) | value) - rtc_r;
+             } else {
+                 /* time for RV64 */
+                 mtimer->time_delta = value - rtc_r;
+@@ -XXX,XX +XXX,XX @@ static void riscv_aclint_mtimer_write(void *opaque, hwaddr addr,
+         } else {
+             if (size == 4) {
+                 /* time_hi for RV32/RV64 */
+-                mtimer->time_delta = (value << 32 | (rtc_r & 0xFFFFFFFF)) - rtc_r;
++                mtimer->time_delta = (value << 32 | (rtc & 0xFFFFFFFF)) - rtc_r;
+             } else {
+                 qemu_log_mask(LOG_GUEST_ERROR,
+                               "aclint-mtimer: invalid time_hi write: %08x",
+--
+.41.0

-New patch
+[PULL v2 27/45] hw/intc: Make rtc variable names consistent
+From: Jason Chien <jason.chien@sifive.com>
+The variables whose values are given by cpu_riscv_read_rtc() should be named
+"rtc". The variables whose value are given by cpu_riscv_read_rtc_raw()
+should be named "rtc_r".
+Signed-off-by: Jason Chien <jason.chien@sifive.com>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Message-ID: <20230728082502.26439-2-jason.chien@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ hw/intc/riscv_aclint.c | 6 +++---
+file changed, 3 insertions(+), 3 deletions(-)
+diff --git a/hw/intc/riscv_aclint.c b/hw/intc/riscv_aclint.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/intc/riscv_aclint.c
++++ b/hw/intc/riscv_aclint.c
+@@ -XXX,XX +XXX,XX @@ static void riscv_aclint_mtimer_write_timecmp(RISCVAclintMTimerState *mtimer,
+     uint64_t next;
+     uint64_t diff;
+-    uint64_t rtc_r = cpu_riscv_read_rtc(mtimer);
++    uint64_t rtc = cpu_riscv_read_rtc(mtimer);
+     /* Compute the relative hartid w.r.t the socket */
+     hartid = hartid - mtimer->hartid_base;
+     mtimer->timecmp[hartid] = value;
+-    if (mtimer->timecmp[hartid] <= rtc_r) {
++    if (mtimer->timecmp[hartid] <= rtc) {
+         /*
+          * If we're setting an MTIMECMP value in the "past",
+          * immediately raise the timer interrupt
+@@ -XXX,XX +XXX,XX @@ static void riscv_aclint_mtimer_write_timecmp(RISCVAclintMTimerState *mtimer,
+     /* otherwise, set up the future timer interrupt */
+     qemu_irq_lower(mtimer->timer_irqs[hartid]);
+-    diff = mtimer->timecmp[hartid] - rtc_r;
++    diff = mtimer->timecmp[hartid] - rtc;
+     /* back to ns (note args switched in muldiv64) */
+     uint64_t ns_diff = muldiv64(diff, NANOSECONDS_PER_SECOND, timebase_freq);
+--
+.41.0

-[PULL 01/10] hw/riscv: Update comment for qtest check in riscv_find_firmware()
+[PULL v2 28/45] linux-user/riscv: Use abi type for target_ucontext
-From: Bin Meng <bmeng.cn@gmail.com>
+From: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
-Since commit 4211fc553234 ("roms/opensbi: Remove ELF images"), the
+We should not use types dependend on host arch for target_ucontext.
-comment for qtest check in riscv_find_firmware() is out of date.
+This bug is found when run rv32 applications.
 Update it to reflect the latest status.
-Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
+Signed-off-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
-Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
-Message-Id: <BN7PR08MB435525C92550BAC5467BE672BF219@BN7PR08MB4355.namprd08.prod.outlook.com>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
 Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
 Message-ID: <20230811055438.1945-1-zhiwei_liu@linux.alibaba.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
- hw/riscv/boot.c | 4 ++--
+ linux-user/riscv/signal.c | 4 ++--
 file changed, 2 insertions(+), 2 deletions(-)
-diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c
+diff --git a/linux-user/riscv/signal.c b/linux-user/riscv/signal.c
 index XXXXXXX..XXXXXXX 100644
---- a/hw/riscv/boot.c
+--- a/linux-user/riscv/signal.c
-+++ b/hw/riscv/boot.c
++++ b/linux-user/riscv/signal.c
-@@ -XXX,XX +XXX,XX @@ char *riscv_find_firmware(const char *firmware_filename)
+@@ -XXX,XX +XXX,XX @@ struct target_sigcontext {
-     if (filename == NULL) {
+ }; /* cf. riscv-linux:arch/riscv/include/uapi/asm/ptrace.h */
-         if (!qtest_enabled()) {
-             /*
+ struct target_ucontext {
--             * We only ship plain binary bios images in the QEMU source.
+-    unsigned long uc_flags;
--             * With Spike machine that uses ELF images as the default bios,
+-    struct target_ucontext *uc_link;
-+             * We only ship OpenSBI binary bios images in the QEMU source.
++    abi_ulong uc_flags;
-+             * For machines that use images other than the default bios,
++    abi_ptr uc_link;
-              * running QEMU test will complain hence let's suppress the error
+     target_stack_t uc_stack;
-              * report for QEMU testing.
+     target_sigset_t uc_sigmask;
-              */
+     uint8_t   __unused[1024 / 8 - sizeof(target_sigset_t)];
 --
-.37.3
+.41.0

-[PULL 02/10] hw/ssi: ibex_spi: fixup coverity issue
+[PULL v2 29/45] target/riscv: support the AIA device emulation with KVM enabled
-From: Wilfred Mallawa <wilfred.mallawa@wdc.com>
+From: Yong-Xuan Wang <yongxuan.wang@sifive.com>
-This patch addresses the coverity issues specified in [1],
+In this patch, we create the APLIC and IMSIC FDT helper functions and
-as suggested, `FIELD_DP32()`/`FIELD_EX32()` macros have been
+remove M mode AIA devices when using KVM acceleration.
 implemented to clean up the code.
-[1] https://www.mail-archive.com/qemu-devel@nongnu.org/msg887713.html
+Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
+Reviewed-by: Jim Shu <jim.shu@sifive.com>
-Fixes: Coverity CID 1488107
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
 Signed-off-by: Wilfred Mallawa <wilfred.mallawa@wdc.com>
 Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
 Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
-Message-Id: <20220930033241.206581-2-wilfred.mallawa@opensource.wdc.com>
+Message-ID: <20230727102439.22554-2-yongxuan.wang@sifive.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
- hw/ssi/ibex_spi_host.c | 132 +++++++++++++++++++++--------------------
+ hw/riscv/virt.c | 290 +++++++++++++++++++++++-------------------------
-file changed, 68 insertions(+), 64 deletions(-)
+file changed, 137 insertions(+), 153 deletions(-)
-diff --git a/hw/ssi/ibex_spi_host.c b/hw/ssi/ibex_spi_host.c
+diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
 index XXXXXXX..XXXXXXX 100644
---- a/hw/ssi/ibex_spi_host.c
+--- a/hw/riscv/virt.c
-+++ b/hw/ssi/ibex_spi_host.c
++++ b/hw/riscv/virt.c
-@@ -XXX,XX +XXX,XX @@ static inline uint8_t div4_round_up(uint8_t dividend)
+@@ -XXX,XX +XXX,XX @@ static uint32_t imsic_num_bits(uint32_t count)
+     return ret;
- static void ibex_spi_rxfifo_reset(IbexSPIHostState *s)
+ }
 -static void create_fdt_imsic(RISCVVirtState *s, const MemMapEntry *memmap,
 -                             uint32_t *phandle, uint32_t *intc_phandles,
 -                             uint32_t *msi_m_phandle, uint32_t *msi_s_phandle)
 +static void create_fdt_one_imsic(RISCVVirtState *s, hwaddr base_addr,
 +                                 uint32_t *intc_phandles, uint32_t msi_phandle,
 +                                 bool m_mode, uint32_t imsic_guest_bits)
  {
-+    uint32_t data = s->regs[IBEX_SPI_HOST_STATUS];
+     int cpu, socket;
-     /* Empty the RX FIFO and assert RXEMPTY */
+     char *imsic_name;
-     fifo8_reset(&s->rx_fifo);
+     MachineState *ms = MACHINE(s);
--    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_RXFULL_MASK;
+     int socket_count = riscv_socket_count(ms);
--    s->regs[IBEX_SPI_HOST_STATUS] |= R_STATUS_RXEMPTY_MASK;
+-    uint32_t imsic_max_hart_per_socket, imsic_guest_bits;
-+    data = FIELD_DP32(data, STATUS, RXFULL, 0);
++    uint32_t imsic_max_hart_per_socket;
-+    data = FIELD_DP32(data, STATUS, RXEMPTY, 1);
+     uint32_t *imsic_cells, *imsic_regs, imsic_addr, imsic_size;
-+    s->regs[IBEX_SPI_HOST_STATUS] = data;
 -    *msi_m_phandle = (*phandle)++;
 -    *msi_s_phandle = (*phandle)++;
      imsic_cells = g_new0(uint32_t, ms->smp.cpus * 2);
      imsic_regs = g_new0(uint32_t, socket_count * 4);
 -    /* M-level IMSIC node */
      for (cpu = 0; cpu < ms->smp.cpus; cpu++) {
          imsic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
 -        imsic_cells[cpu * 2 + 1] = cpu_to_be32(IRQ_M_EXT);
 +        imsic_cells[cpu * 2 + 1] = cpu_to_be32(m_mode ? IRQ_M_EXT : IRQ_S_EXT);
      }
 -    imsic_max_hart_per_socket = 0;
 -    for (socket = 0; socket < socket_count; socket++) {
 -        imsic_addr = memmap[VIRT_IMSIC_M].base +
 -                     socket * VIRT_IMSIC_GROUP_MAX_SIZE;
 -        imsic_size = IMSIC_HART_SIZE(0) * s->soc[socket].num_harts;
 -        imsic_regs[socket * 4 + 0] = 0;
 -        imsic_regs[socket * 4 + 1] = cpu_to_be32(imsic_addr);
 -        imsic_regs[socket * 4 + 2] = 0;
 -        imsic_regs[socket * 4 + 3] = cpu_to_be32(imsic_size);
 -        if (imsic_max_hart_per_socket < s->soc[socket].num_harts) {
 -            imsic_max_hart_per_socket = s->soc[socket].num_harts;
 -        }
 -    }
 -    imsic_name = g_strdup_printf("/soc/imsics@%lx",
 -        (unsigned long)memmap[VIRT_IMSIC_M].base);
 -    qemu_fdt_add_subnode(ms->fdt, imsic_name);
 -    qemu_fdt_setprop_string(ms->fdt, imsic_name, "compatible",
 -        "riscv,imsics");
 -    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "#interrupt-cells",
 -        FDT_IMSIC_INT_CELLS);
 -    qemu_fdt_setprop(ms->fdt, imsic_name, "interrupt-controller",
 -        NULL, 0);
 -    qemu_fdt_setprop(ms->fdt, imsic_name, "msi-controller",
 -        NULL, 0);
 -    qemu_fdt_setprop(ms->fdt, imsic_name, "interrupts-extended",
 -        imsic_cells, ms->smp.cpus * sizeof(uint32_t) * 2);
 -    qemu_fdt_setprop(ms->fdt, imsic_name, "reg", imsic_regs,
 -        socket_count * sizeof(uint32_t) * 4);
 -    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,num-ids",
 -        VIRT_IRQCHIP_NUM_MSIS);
 -    if (socket_count > 1) {
 -        qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,hart-index-bits",
 -            imsic_num_bits(imsic_max_hart_per_socket));
 -        qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,group-index-bits",
 -            imsic_num_bits(socket_count));
 -        qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,group-index-shift",
 -            IMSIC_MMIO_GROUP_MIN_SHIFT);
 -    }
 -    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "phandle", *msi_m_phandle);
 -
 -    g_free(imsic_name);
 -    /* S-level IMSIC node */
 -    for (cpu = 0; cpu < ms->smp.cpus; cpu++) {
 -        imsic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
 -        imsic_cells[cpu * 2 + 1] = cpu_to_be32(IRQ_S_EXT);
 -    }
 -    imsic_guest_bits = imsic_num_bits(s->aia_guests + 1);
      imsic_max_hart_per_socket = 0;
      for (socket = 0; socket < socket_count; socket++) {
 -        imsic_addr = memmap[VIRT_IMSIC_S].base +
 -                     socket * VIRT_IMSIC_GROUP_MAX_SIZE;
 +        imsic_addr = base_addr + socket * VIRT_IMSIC_GROUP_MAX_SIZE;
          imsic_size = IMSIC_HART_SIZE(imsic_guest_bits) *
                       s->soc[socket].num_harts;
          imsic_regs[socket * 4 + 0] = 0;
@@ -XXX,XX +XXX,XX @@ static void create_fdt_imsic(RISCVVirtState *s, const MemMapEntry *memmap,
              imsic_max_hart_per_socket = s->soc[socket].num_harts;
          }
      }
 -    imsic_name = g_strdup_printf("/soc/imsics@%lx",
 -        (unsigned long)memmap[VIRT_IMSIC_S].base);
 +
 +    imsic_name = g_strdup_printf("/soc/imsics@%lx", (unsigned long)base_addr);
      qemu_fdt_add_subnode(ms->fdt, imsic_name);
 -    qemu_fdt_setprop_string(ms->fdt, imsic_name, "compatible",
 -        "riscv,imsics");
 +    qemu_fdt_setprop_string(ms->fdt, imsic_name, "compatible", "riscv,imsics");
      qemu_fdt_setprop_cell(ms->fdt, imsic_name, "#interrupt-cells",
 -        FDT_IMSIC_INT_CELLS);
 -    qemu_fdt_setprop(ms->fdt, imsic_name, "interrupt-controller",
 -        NULL, 0);
 -    qemu_fdt_setprop(ms->fdt, imsic_name, "msi-controller",
 -        NULL, 0);
 +                          FDT_IMSIC_INT_CELLS);
 +    qemu_fdt_setprop(ms->fdt, imsic_name, "interrupt-controller", NULL, 0);
 +    qemu_fdt_setprop(ms->fdt, imsic_name, "msi-controller", NULL, 0);
      qemu_fdt_setprop(ms->fdt, imsic_name, "interrupts-extended",
 -        imsic_cells, ms->smp.cpus * sizeof(uint32_t) * 2);
 +                     imsic_cells, ms->smp.cpus * sizeof(uint32_t) * 2);
      qemu_fdt_setprop(ms->fdt, imsic_name, "reg", imsic_regs,
 -        socket_count * sizeof(uint32_t) * 4);
 +                     socket_count * sizeof(uint32_t) * 4);
      qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,num-ids",
 -        VIRT_IRQCHIP_NUM_MSIS);
 +                     VIRT_IRQCHIP_NUM_MSIS);
 +
      if (imsic_guest_bits) {
          qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,guest-index-bits",
 -            imsic_guest_bits);
 +                              imsic_guest_bits);
      }
 +
      if (socket_count > 1) {
          qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,hart-index-bits",
 -            imsic_num_bits(imsic_max_hart_per_socket));
 +                              imsic_num_bits(imsic_max_hart_per_socket));
          qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,group-index-bits",
 -            imsic_num_bits(socket_count));
 +                              imsic_num_bits(socket_count));
          qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,group-index-shift",
 -            IMSIC_MMIO_GROUP_MIN_SHIFT);
 +                              IMSIC_MMIO_GROUP_MIN_SHIFT);
      }
 -    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "phandle", *msi_s_phandle);
 -    g_free(imsic_name);
 +    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "phandle", msi_phandle);
 +    g_free(imsic_name);
      g_free(imsic_regs);
      g_free(imsic_cells);
  }
- static void ibex_spi_txfifo_reset(IbexSPIHostState *s)
+-static void create_fdt_socket_aplic(RISCVVirtState *s,
 -                                    const MemMapEntry *memmap, int socket,
 -                                    uint32_t msi_m_phandle,
 -                                    uint32_t msi_s_phandle,
 -                                    uint32_t *phandle,
 -                                    uint32_t *intc_phandles,
 -                                    uint32_t *aplic_phandles)
 +static void create_fdt_imsic(RISCVVirtState *s, const MemMapEntry *memmap,
 +                             uint32_t *phandle, uint32_t *intc_phandles,
 +                             uint32_t *msi_m_phandle, uint32_t *msi_s_phandle)
 +{
 +    *msi_m_phandle = (*phandle)++;
 +    *msi_s_phandle = (*phandle)++;
 +
 +    if (!kvm_enabled()) {
 +        /* M-level IMSIC node */
 +        create_fdt_one_imsic(s, memmap[VIRT_IMSIC_M].base, intc_phandles,
 +                             *msi_m_phandle, true, 0);
 +    }
 +
 +    /* S-level IMSIC node */
 +    create_fdt_one_imsic(s, memmap[VIRT_IMSIC_S].base, intc_phandles,
 +                         *msi_s_phandle, false,
 +                         imsic_num_bits(s->aia_guests + 1));
 +
 +}
 +
 +static void create_fdt_one_aplic(RISCVVirtState *s, int socket,
 +                                 unsigned long aplic_addr, uint32_t aplic_size,
 +                                 uint32_t msi_phandle,
 +                                 uint32_t *intc_phandles,
 +                                 uint32_t aplic_phandle,
 +                                 uint32_t aplic_child_phandle,
 +                                 bool m_mode)
  {
-+    uint32_t data = s->regs[IBEX_SPI_HOST_STATUS];
+     int cpu;
-     /* Empty the TX FIFO and assert TXEMPTY */
+     char *aplic_name;
-     fifo8_reset(&s->tx_fifo);
+     uint32_t *aplic_cells;
--    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXFULL_MASK;
+-    unsigned long aplic_addr;
--    s->regs[IBEX_SPI_HOST_STATUS] |= R_STATUS_TXEMPTY_MASK;
+     MachineState *ms = MACHINE(s);
-+    data = FIELD_DP32(data, STATUS, TXFULL, 0);
+-    uint32_t aplic_m_phandle, aplic_s_phandle;
-+    data = FIELD_DP32(data, STATUS, TXEMPTY, 1);
-+    s->regs[IBEX_SPI_HOST_STATUS] = data;
+-    aplic_m_phandle = (*phandle)++;
 -    aplic_s_phandle = (*phandle)++;
      aplic_cells = g_new0(uint32_t, s->soc[socket].num_harts * 2);
 -    /* M-level APLIC node */
      for (cpu = 0; cpu < s->soc[socket].num_harts; cpu++) {
          aplic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
 -        aplic_cells[cpu * 2 + 1] = cpu_to_be32(IRQ_M_EXT);
 +        aplic_cells[cpu * 2 + 1] = cpu_to_be32(m_mode ? IRQ_M_EXT : IRQ_S_EXT);
      }
 -    aplic_addr = memmap[VIRT_APLIC_M].base +
 -                 (memmap[VIRT_APLIC_M].size * socket);
 +
      aplic_name = g_strdup_printf("/soc/aplic@%lx", aplic_addr);
      qemu_fdt_add_subnode(ms->fdt, aplic_name);
      qemu_fdt_setprop_string(ms->fdt, aplic_name, "compatible", "riscv,aplic");
      qemu_fdt_setprop_cell(ms->fdt, aplic_name,
 -        "#interrupt-cells", FDT_APLIC_INT_CELLS);
 +                          "#interrupt-cells", FDT_APLIC_INT_CELLS);
      qemu_fdt_setprop(ms->fdt, aplic_name, "interrupt-controller", NULL, 0);
 +
      if (s->aia_type == VIRT_AIA_TYPE_APLIC) {
          qemu_fdt_setprop(ms->fdt, aplic_name, "interrupts-extended",
 -            aplic_cells, s->soc[socket].num_harts * sizeof(uint32_t) * 2);
 +                         aplic_cells,
 +                         s->soc[socket].num_harts * sizeof(uint32_t) * 2);
      } else {
 -        qemu_fdt_setprop_cell(ms->fdt, aplic_name, "msi-parent",
 -            msi_m_phandle);
 +        qemu_fdt_setprop_cell(ms->fdt, aplic_name, "msi-parent", msi_phandle);
      }
 +
      qemu_fdt_setprop_cells(ms->fdt, aplic_name, "reg",
 -        0x0, aplic_addr, 0x0, memmap[VIRT_APLIC_M].size);
 +                           0x0, aplic_addr, 0x0, aplic_size);
      qemu_fdt_setprop_cell(ms->fdt, aplic_name, "riscv,num-sources",
 -        VIRT_IRQCHIP_NUM_SOURCES);
 -    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "riscv,children",
 -        aplic_s_phandle);
 -    qemu_fdt_setprop_cells(ms->fdt, aplic_name, "riscv,delegate",
 -        aplic_s_phandle, 0x1, VIRT_IRQCHIP_NUM_SOURCES);
 +                          VIRT_IRQCHIP_NUM_SOURCES);
 +
 +    if (aplic_child_phandle) {
 +        qemu_fdt_setprop_cell(ms->fdt, aplic_name, "riscv,children",
 +                              aplic_child_phandle);
 +        qemu_fdt_setprop_cells(ms->fdt, aplic_name, "riscv,delegate",
 +                               aplic_child_phandle, 0x1,
 +                               VIRT_IRQCHIP_NUM_SOURCES);
 +    }
 +
      riscv_socket_fdt_write_id(ms, aplic_name, socket);
 -    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "phandle", aplic_m_phandle);
 +    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "phandle", aplic_phandle);
 +
      g_free(aplic_name);
 +    g_free(aplic_cells);
 +}
 -    /* S-level APLIC node */
 -    for (cpu = 0; cpu < s->soc[socket].num_harts; cpu++) {
 -        aplic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
 -        aplic_cells[cpu * 2 + 1] = cpu_to_be32(IRQ_S_EXT);
 +static void create_fdt_socket_aplic(RISCVVirtState *s,
 +                                    const MemMapEntry *memmap, int socket,
 +                                    uint32_t msi_m_phandle,
 +                                    uint32_t msi_s_phandle,
 +                                    uint32_t *phandle,
 +                                    uint32_t *intc_phandles,
 +                                    uint32_t *aplic_phandles)
 +{
 +    char *aplic_name;
 +    unsigned long aplic_addr;
 +    MachineState *ms = MACHINE(s);
 +    uint32_t aplic_m_phandle, aplic_s_phandle;
 +
 +    aplic_m_phandle = (*phandle)++;
 +    aplic_s_phandle = (*phandle)++;
 +
 +    if (!kvm_enabled()) {
 +        /* M-level APLIC node */
 +        aplic_addr = memmap[VIRT_APLIC_M].base +
 +                     (memmap[VIRT_APLIC_M].size * socket);
 +        create_fdt_one_aplic(s, socket, aplic_addr, memmap[VIRT_APLIC_M].size,
 +                             msi_m_phandle, intc_phandles,
 +                             aplic_m_phandle, aplic_s_phandle,
 +                             true);
      }
 +
 +    /* S-level APLIC node */
      aplic_addr = memmap[VIRT_APLIC_S].base +
                   (memmap[VIRT_APLIC_S].size * socket);
 +    create_fdt_one_aplic(s, socket, aplic_addr, memmap[VIRT_APLIC_S].size,
 +                         msi_s_phandle, intc_phandles,
 +                         aplic_s_phandle, 0,
 +                         false);
 +
      aplic_name = g_strdup_printf("/soc/aplic@%lx", aplic_addr);
 -    qemu_fdt_add_subnode(ms->fdt, aplic_name);
 -    qemu_fdt_setprop_string(ms->fdt, aplic_name, "compatible", "riscv,aplic");
 -    qemu_fdt_setprop_cell(ms->fdt, aplic_name,
 -        "#interrupt-cells", FDT_APLIC_INT_CELLS);
 -    qemu_fdt_setprop(ms->fdt, aplic_name, "interrupt-controller", NULL, 0);
 -    if (s->aia_type == VIRT_AIA_TYPE_APLIC) {
 -        qemu_fdt_setprop(ms->fdt, aplic_name, "interrupts-extended",
 -            aplic_cells, s->soc[socket].num_harts * sizeof(uint32_t) * 2);
 -    } else {
 -        qemu_fdt_setprop_cell(ms->fdt, aplic_name, "msi-parent",
 -            msi_s_phandle);
 -    }
 -    qemu_fdt_setprop_cells(ms->fdt, aplic_name, "reg",
 -        0x0, aplic_addr, 0x0, memmap[VIRT_APLIC_S].size);
 -    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "riscv,num-sources",
 -        VIRT_IRQCHIP_NUM_SOURCES);
 -    riscv_socket_fdt_write_id(ms, aplic_name, socket);
 -    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "phandle", aplic_s_phandle);
      if (!socket) {
          platform_bus_add_all_fdt_nodes(ms->fdt, aplic_name,
@@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aplic(RISCVVirtState *s,
      g_free(aplic_name);
 -    g_free(aplic_cells);
      aplic_phandles[socket] = aplic_s_phandle;
  }
- static void ibex_spi_host_reset(DeviceState *dev)
+@@ -XXX,XX +XXX,XX @@ static DeviceState *virt_create_aia(RISCVVirtAIAType aia_type, int aia_guests,
-@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_reset(DeviceState *dev)
+     int i;
-  */
+     hwaddr addr;
- static void ibex_spi_host_irq(IbexSPIHostState *s)
+     uint32_t guest_bits;
- {
+-    DeviceState *aplic_m;
--    bool error_en = s->regs[IBEX_SPI_HOST_INTR_ENABLE]
+-    bool msimode = (aia_type == VIRT_AIA_TYPE_APLIC_IMSIC) ? true : false;
--                    & R_INTR_ENABLE_ERROR_MASK;
++    DeviceState *aplic_s = NULL;
--    bool event_en = s->regs[IBEX_SPI_HOST_INTR_ENABLE]
++    DeviceState *aplic_m = NULL;
--                    & R_INTR_ENABLE_SPI_EVENT_MASK;
++    bool msimode = aia_type == VIRT_AIA_TYPE_APLIC_IMSIC;
--    bool err_pending = s->regs[IBEX_SPI_HOST_INTR_STATE]
--                        & R_INTR_STATE_ERROR_MASK;
+     if (msimode) {
--    bool status_pending = s->regs[IBEX_SPI_HOST_INTR_STATE]
+-        /* Per-socket M-level IMSICs */
--                        & R_INTR_STATE_SPI_EVENT_MASK;
+-        addr = memmap[VIRT_IMSIC_M].base + socket * VIRT_IMSIC_GROUP_MAX_SIZE;
-+    uint32_t intr_test_reg = s->regs[IBEX_SPI_HOST_INTR_TEST];
+-        for (i = 0; i < hart_count; i++) {
-+    uint32_t intr_en_reg = s->regs[IBEX_SPI_HOST_INTR_ENABLE];
+-            riscv_imsic_create(addr + i * IMSIC_HART_SIZE(0),
-+    uint32_t intr_state_reg = s->regs[IBEX_SPI_HOST_INTR_STATE];
+-                               base_hartid + i, true, 1,
-+
+-                               VIRT_IRQCHIP_NUM_MSIS);
-+    uint32_t err_en_reg = s->regs[IBEX_SPI_HOST_ERROR_ENABLE];
++        if (!kvm_enabled()) {
-+    uint32_t event_en_reg = s->regs[IBEX_SPI_HOST_EVENT_ENABLE];
++            /* Per-socket M-level IMSICs */
-+    uint32_t err_status_reg = s->regs[IBEX_SPI_HOST_ERROR_STATUS];
++            addr = memmap[VIRT_IMSIC_M].base +
-+    uint32_t status_reg = s->regs[IBEX_SPI_HOST_STATUS];
++                   socket * VIRT_IMSIC_GROUP_MAX_SIZE;
-+
++            for (i = 0; i < hart_count; i++) {
-+
++                riscv_imsic_create(addr + i * IMSIC_HART_SIZE(0),
-+    bool error_en = FIELD_EX32(intr_en_reg, INTR_ENABLE, ERROR);
++                                   base_hartid + i, true, 1,
-+    bool event_en = FIELD_EX32(intr_en_reg, INTR_ENABLE, SPI_EVENT);
++                                   VIRT_IRQCHIP_NUM_MSIS);
-+    bool err_pending = FIELD_EX32(intr_state_reg, INTR_STATE, ERROR);
++            }
 +    bool status_pending = FIELD_EX32(intr_state_reg, INTR_STATE, SPI_EVENT);
 +
      int err_irq = 0, event_irq = 0;
      /* Error IRQ enabled and Error IRQ Cleared */
      if (error_en && !err_pending) {
          /* Event enabled, Interrupt Test Error */
 -        if (s->regs[IBEX_SPI_HOST_INTR_TEST] & R_INTR_TEST_ERROR_MASK) {
 +        if (FIELD_EX32(intr_test_reg, INTR_TEST,  ERROR)) {
              err_irq = 1;
 -        } else if ((s->regs[IBEX_SPI_HOST_ERROR_ENABLE]
 -                    &  R_ERROR_ENABLE_CMDBUSY_MASK) &&
 -                    s->regs[IBEX_SPI_HOST_ERROR_STATUS]
 -                    & R_ERROR_STATUS_CMDBUSY_MASK) {
 +        } else if (FIELD_EX32(err_en_reg, ERROR_ENABLE,  CMDBUSY) &&
 +                   FIELD_EX32(err_status_reg, ERROR_STATUS,  CMDBUSY)) {
              /* Wrote to COMMAND when not READY */
              err_irq = 1;
 -        } else if ((s->regs[IBEX_SPI_HOST_ERROR_ENABLE]
 -                    &  R_ERROR_ENABLE_CMDINVAL_MASK) &&
 -                    s->regs[IBEX_SPI_HOST_ERROR_STATUS]
 -                    & R_ERROR_STATUS_CMDINVAL_MASK) {
 +        } else if (FIELD_EX32(err_en_reg, ERROR_ENABLE,  CMDINVAL)  &&
 +                   FIELD_EX32(err_status_reg, ERROR_STATUS,  CMDINVAL)) {
              /* Invalid command segment */
              err_irq = 1;
 -        } else if ((s->regs[IBEX_SPI_HOST_ERROR_ENABLE]
 -                    & R_ERROR_ENABLE_CSIDINVAL_MASK) &&
 -                    s->regs[IBEX_SPI_HOST_ERROR_STATUS]
 -                    & R_ERROR_STATUS_CSIDINVAL_MASK) {
 +        } else if (FIELD_EX32(err_en_reg, ERROR_ENABLE,  CSIDINVAL) &&
 +                   FIELD_EX32(err_status_reg, ERROR_STATUS,  CSIDINVAL)) {
              /* Invalid value for CSID */
              err_irq = 1;
          }
-@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_irq(IbexSPIHostState *s)
+         /* Per-socket S-level IMSICs */
-     /* Event IRQ Enabled and Event IRQ Cleared */
+@@ -XXX,XX +XXX,XX @@ static DeviceState *virt_create_aia(RISCVVirtAIAType aia_type, int aia_guests,
      if (event_en && !status_pending) {
 -        if (s->regs[IBEX_SPI_HOST_INTR_TEST] & R_INTR_TEST_SPI_EVENT_MASK) {
 +        if (FIELD_EX32(intr_test_reg, INTR_STATE,  SPI_EVENT)) {
              /* Event enabled, Interrupt Test Event */
              event_irq = 1;
 -        } else if ((s->regs[IBEX_SPI_HOST_EVENT_ENABLE]
 -                    & R_EVENT_ENABLE_READY_MASK) &&
 -                    (s->regs[IBEX_SPI_HOST_STATUS] & R_STATUS_READY_MASK)) {
 +        } else if (FIELD_EX32(event_en_reg, EVENT_ENABLE,  READY) &&
 +                   FIELD_EX32(status_reg, STATUS, READY)) {
              /* SPI Host ready for next command */
              event_irq = 1;
 -        } else if ((s->regs[IBEX_SPI_HOST_EVENT_ENABLE]
 -                    & R_EVENT_ENABLE_TXEMPTY_MASK) &&
 -                    (s->regs[IBEX_SPI_HOST_STATUS] & R_STATUS_TXEMPTY_MASK)) {
 +        } else if (FIELD_EX32(event_en_reg, EVENT_ENABLE,  TXEMPTY) &&
 +                   FIELD_EX32(status_reg, STATUS,  TXEMPTY)) {
              /* SPI TXEMPTY, TXFIFO drained */
              event_irq = 1;
 -        } else if ((s->regs[IBEX_SPI_HOST_EVENT_ENABLE]
 -                    & R_EVENT_ENABLE_RXFULL_MASK) &&
 -                    (s->regs[IBEX_SPI_HOST_STATUS] & R_STATUS_RXFULL_MASK)) {
 +        } else if (FIELD_EX32(event_en_reg, EVENT_ENABLE,  RXFULL) &&
 +                   FIELD_EX32(status_reg, STATUS,  RXFULL)) {
              /* SPI RXFULL, RXFIFO  full */
              event_irq = 1;
          }
-@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_irq(IbexSPIHostState *s)
+     }
- static void ibex_spi_host_transfer(IbexSPIHostState *s)
+-    /* Per-socket M-level APLIC */
- {
+-    aplic_m = riscv_aplic_create(
--    uint32_t rx, tx;
+-        memmap[VIRT_APLIC_M].base + socket * memmap[VIRT_APLIC_M].size,
-+    uint32_t rx, tx, data;
+-        memmap[VIRT_APLIC_M].size,
-     /* Get num of one byte transfers */
+-        (msimode) ? 0 : base_hartid,
--    uint8_t segment_len = ((s->regs[IBEX_SPI_HOST_COMMAND] & R_COMMAND_LEN_MASK)
+-        (msimode) ? 0 : hart_count,
--                          >> R_COMMAND_LEN_SHIFT);
+-        VIRT_IRQCHIP_NUM_SOURCES,
-+    uint8_t segment_len = FIELD_EX32(s->regs[IBEX_SPI_HOST_COMMAND],
+-        VIRT_IRQCHIP_NUM_PRIO_BITS,
-+                                     COMMAND,  LEN);
+-        msimode, true, NULL);
-+
+-
-     while (segment_len > 0) {
+-    if (aplic_m) {
-         if (fifo8_is_empty(&s->tx_fifo)) {
+-        /* Per-socket S-level APLIC */
-             /* Assert Stall */
+-        riscv_aplic_create(
-@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_transfer(IbexSPIHostState *s)
+-            memmap[VIRT_APLIC_S].base + socket * memmap[VIRT_APLIC_S].size,
-         --segment_len;
+-            memmap[VIRT_APLIC_S].size,
-     }
+-            (msimode) ? 0 : base_hartid,
+-            (msimode) ? 0 : hart_count,
-+    data = s->regs[IBEX_SPI_HOST_STATUS];
+-            VIRT_IRQCHIP_NUM_SOURCES,
-     /* Assert Ready */
+-            VIRT_IRQCHIP_NUM_PRIO_BITS,
--    s->regs[IBEX_SPI_HOST_STATUS] |= R_STATUS_READY_MASK;
+-            msimode, false, aplic_m);
-+    data = FIELD_DP32(data, STATUS, READY, 1);
++    if (!kvm_enabled()) {
-     /* Set RXQD */
++        /* Per-socket M-level APLIC */
--    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_RXQD_MASK;
++        aplic_m = riscv_aplic_create(memmap[VIRT_APLIC_M].base +
--    s->regs[IBEX_SPI_HOST_STATUS] |= (R_STATUS_RXQD_MASK
++                                     socket * memmap[VIRT_APLIC_M].size,
--                                    & div4_round_up(segment_len));
++                                     memmap[VIRT_APLIC_M].size,
-+    data = FIELD_DP32(data, STATUS, RXQD, div4_round_up(segment_len));
++                                     (msimode) ? 0 : base_hartid,
-     /* Set TXQD */
++                                     (msimode) ? 0 : hart_count,
--    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXQD_MASK;
++                                     VIRT_IRQCHIP_NUM_SOURCES,
--    s->regs[IBEX_SPI_HOST_STATUS] |= (fifo8_num_used(&s->tx_fifo) / 4)
++                                     VIRT_IRQCHIP_NUM_PRIO_BITS,
--                                    & R_STATUS_TXQD_MASK;
++                                     msimode, true, NULL);
-+    data = FIELD_DP32(data, STATUS, TXQD, fifo8_num_used(&s->tx_fifo) / 4);
+     }
-     /* Clear TXFULL */
--    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXFULL_MASK;
+-    return aplic_m;
--    /* Assert TXEMPTY and drop remaining bytes that exceed segment_len */
++    /* Per-socket S-level APLIC */
--    ibex_spi_txfifo_reset(s);
++    aplic_s = riscv_aplic_create(memmap[VIRT_APLIC_S].base +
-+    data = FIELD_DP32(data, STATUS, TXFULL, 0);
++                                 socket * memmap[VIRT_APLIC_S].size,
-     /* Reset RXEMPTY */
++                                 memmap[VIRT_APLIC_S].size,
--    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_RXEMPTY_MASK;
++                                 (msimode) ? 0 : base_hartid,
-+    data = FIELD_DP32(data, STATUS, RXEMPTY, 0);
++                                 (msimode) ? 0 : hart_count,
-+    /* Update register status */
++                                 VIRT_IRQCHIP_NUM_SOURCES,
-+    s->regs[IBEX_SPI_HOST_STATUS] = data;
++                                 VIRT_IRQCHIP_NUM_PRIO_BITS,
-+    /* Drop remaining bytes that exceed segment_len */
++                                 msimode, false, aplic_m);
-+    ibex_spi_txfifo_reset(s);
++
++    return kvm_enabled() ? aplic_s : aplic_m;
      ibex_spi_host_irq(s);
  }
-@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
- {
+ static void create_platform_bus(RISCVVirtState *s, DeviceState *irqchip)
      IbexSPIHostState *s = opaque;
      uint32_t val32 = val64;
 -    uint32_t shift_mask = 0xff;
 +    uint32_t shift_mask = 0xff, status = 0;
      uint8_t txqd_len;
      trace_ibex_spi_host_write(addr, size, val64);
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
          s->regs[addr] = val32;
          /* STALL, IP not enabled */
 -        if (!(s->regs[IBEX_SPI_HOST_CONTROL] & R_CONTROL_SPIEN_MASK)) {
 +        if (!(FIELD_EX32(s->regs[IBEX_SPI_HOST_CONTROL],
 +                         CONTROL, SPIEN))) {
              return;
          }
          /* SPI not ready, IRQ Error */
 -        if (!(s->regs[IBEX_SPI_HOST_STATUS] & R_STATUS_READY_MASK)) {
 +        if (!(FIELD_EX32(s->regs[IBEX_SPI_HOST_STATUS],
 +                         STATUS, READY))) {
              s->regs[IBEX_SPI_HOST_ERROR_STATUS] |= R_ERROR_STATUS_CMDBUSY_MASK;
              ibex_spi_host_irq(s);
              return;
          }
 +
          /* Assert Not Ready */
          s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_READY_MASK;
 -        if (((val32 & R_COMMAND_DIRECTION_MASK) >> R_COMMAND_DIRECTION_SHIFT)
 -            != BIDIRECTIONAL_TRANSFER) {
 -                qemu_log_mask(LOG_UNIMP,
 +        if (FIELD_EX32(val32, COMMAND, DIRECTION) != BIDIRECTIONAL_TRANSFER) {
 +            qemu_log_mask(LOG_UNIMP,
                            "%s: Rx Only/Tx Only are not supported\n", __func__);
          }
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
                  return;
              }
              /* Byte ordering is set by the IP */
 -            if ((s->regs[IBEX_SPI_HOST_STATUS] &
 -                R_STATUS_BYTEORDER_MASK) == 0) {
 +            status = s->regs[IBEX_SPI_HOST_STATUS];
 +            if (FIELD_EX32(status, STATUS, BYTEORDER) == 0) {
                  /* LE: LSB transmitted first (default for ibex processor) */
                  shift_mask = 0xff << (i * 8);
              } else {
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
              fifo8_push(&s->tx_fifo, (val32 & shift_mask) >> (i * 8));
          }
 -
 +        status = s->regs[IBEX_SPI_HOST_STATUS];
          /* Reset TXEMPTY */
 -        s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXEMPTY_MASK;
 +        status = FIELD_DP32(status, STATUS, TXEMPTY, 0);
          /* Update TXQD */
 -        txqd_len = (s->regs[IBEX_SPI_HOST_STATUS] &
 -                    R_STATUS_TXQD_MASK) >> R_STATUS_TXQD_SHIFT;
 +        txqd_len = FIELD_EX32(status, STATUS, TXQD);
          /* Partial bytes (size < 4) are padded, in words. */
          txqd_len += 1;
 -        s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXQD_MASK;
 -        s->regs[IBEX_SPI_HOST_STATUS] |= txqd_len;
 +        status = FIELD_DP32(status, STATUS, TXQD, txqd_len);
          /* Assert Ready */
 -        s->regs[IBEX_SPI_HOST_STATUS] |= R_STATUS_READY_MASK;
 +        status = FIELD_DP32(status, STATUS, READY, 1);
 +        /* Update register status */
 +        s->regs[IBEX_SPI_HOST_STATUS] = status;
          break;
      case IBEX_SPI_HOST_ERROR_ENABLE:
          s->regs[addr] = val32;
 --
-.37.3
+.41.0

-New patch
+[PULL v2 30/45] target/riscv: check the in-kernel irqchip support
+From: Yong-Xuan Wang <yongxuan.wang@sifive.com>
+We check the in-kernel irqchip support when using KVM acceleration.
+Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
+Reviewed-by: Jim Shu <jim.shu@sifive.com>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
+Message-ID: <20230727102439.22554-3-yongxuan.wang@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/kvm.c | 10 +++++++++-
+file changed, 9 insertions(+), 1 deletion(-)
+diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/kvm.c
++++ b/target/riscv/kvm.c
+@@ -XXX,XX +XXX,XX @@ int kvm_arch_init(MachineState *ms, KVMState *s)
+ int kvm_arch_irqchip_create(KVMState *s)
+ {
+-    return 0;
++    if (kvm_kernel_irqchip_split()) {
++        error_report("-machine kernel_irqchip=split is not supported on RISC-V.");
++        exit(1);
++    }
++
++    /*
++     * We can create the VAIA using the newer device control API.
++     */
++    return kvm_check_extension(s, KVM_CAP_DEVICE_CTRL);
+ }
+ int kvm_arch_process_async_events(CPUState *cs)
+--
+.41.0

-New patch
+[PULL v2 31/45] target/riscv: Create an KVM AIA irqchip
+From: Yong-Xuan Wang <yongxuan.wang@sifive.com>
+We create a vAIA chip by using the KVM_DEV_TYPE_RISCV_AIA and then set up
+the chip with the KVM_DEV_RISCV_AIA_GRP_* APIs.
+We also extend KVM accelerator to specify the KVM AIA mode. The "riscv-aia"
+parameter is passed along with --accel in QEMU command-line.
+) "riscv-aia=emul": IMSIC is emulated by hypervisor
+) "riscv-aia=hwaccel": use hardware guest IMSIC
+) "riscv-aia=auto": use the hardware guest IMSICs whenever available
+                     otherwise we fallback to software emulation.
+Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
+Reviewed-by: Jim Shu <jim.shu@sifive.com>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
+Message-ID: <20230727102439.22554-4-yongxuan.wang@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/kvm_riscv.h |   4 +
+ target/riscv/kvm.c       | 186 +++++++++++++++++++++++++++++++++++++++
+files changed, 190 insertions(+)
+diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/kvm_riscv.h
++++ b/target/riscv/kvm_riscv.h
+@@ -XXX,XX +XXX,XX @@
+ void kvm_riscv_init_user_properties(Object *cpu_obj);
+ void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
+ void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level);
++void kvm_riscv_aia_create(MachineState *machine, uint64_t group_shift,
++                          uint64_t aia_irq_num, uint64_t aia_msi_num,
++                          uint64_t aplic_base, uint64_t imsic_base,
++                          uint64_t guest_num);
+ #endif
+diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/kvm.c
++++ b/target/riscv/kvm.c
+@@ -XXX,XX +XXX,XX @@
+ #include "exec/address-spaces.h"
+ #include "hw/boards.h"
+ #include "hw/irq.h"
++#include "hw/intc/riscv_imsic.h"
+ #include "qemu/log.h"
+ #include "hw/loader.h"
+ #include "kvm_riscv.h"
+@@ -XXX,XX +XXX,XX @@
+ #include "chardev/char-fe.h"
+ #include "migration/migration.h"
+ #include "sysemu/runstate.h"
++#include "hw/riscv/numa.h"
+ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type,
+                                  uint64_t idx)
+@@ -XXX,XX +XXX,XX @@ bool kvm_arch_cpu_check_are_resettable(void)
+     return true;
+ }
++static int aia_mode;
++
++static const char *kvm_aia_mode_str(uint64_t mode)
++{
++    switch (mode) {
++    case KVM_DEV_RISCV_AIA_MODE_EMUL:
++        return "emul";
++    case KVM_DEV_RISCV_AIA_MODE_HWACCEL:
++        return "hwaccel";
++    case KVM_DEV_RISCV_AIA_MODE_AUTO:
++    default:
++        return "auto";
++    };
++}
++
++static char *riscv_get_kvm_aia(Object *obj, Error **errp)
++{
++    return g_strdup(kvm_aia_mode_str(aia_mode));
++}
++
++static void riscv_set_kvm_aia(Object *obj, const char *val, Error **errp)
++{
++    if (!strcmp(val, "emul")) {
++        aia_mode = KVM_DEV_RISCV_AIA_MODE_EMUL;
++    } else if (!strcmp(val, "hwaccel")) {
++        aia_mode = KVM_DEV_RISCV_AIA_MODE_HWACCEL;
++    } else if (!strcmp(val, "auto")) {
++        aia_mode = KVM_DEV_RISCV_AIA_MODE_AUTO;
++    } else {
++        error_setg(errp, "Invalid KVM AIA mode");
++        error_append_hint(errp, "Valid values are emul, hwaccel, and auto.\n");
++    }
++}
++
+ void kvm_arch_accel_class_init(ObjectClass *oc)
+ {
++    object_class_property_add_str(oc, "riscv-aia", riscv_get_kvm_aia,
++                                  riscv_set_kvm_aia);
++    object_class_property_set_description(oc, "riscv-aia",
++                                          "Set KVM AIA mode. Valid values are "
++                                          "emul, hwaccel, and auto. Default "
++                                          "is auto.");
++    object_property_set_default_str(object_class_property_find(oc, "riscv-aia"),
++                                    "auto");
++}
++
++void kvm_riscv_aia_create(MachineState *machine, uint64_t group_shift,
++                          uint64_t aia_irq_num, uint64_t aia_msi_num,
++                          uint64_t aplic_base, uint64_t imsic_base,
++                          uint64_t guest_num)
++{
++    int ret, i;
++    int aia_fd = -1;
++    uint64_t default_aia_mode;
++    uint64_t socket_count = riscv_socket_count(machine);
++    uint64_t max_hart_per_socket = 0;
++    uint64_t socket, base_hart, hart_count, socket_imsic_base, imsic_addr;
++    uint64_t socket_bits, hart_bits, guest_bits;
++
++    aia_fd = kvm_create_device(kvm_state, KVM_DEV_TYPE_RISCV_AIA, false);
++
++    if (aia_fd < 0) {
++        error_report("Unable to create in-kernel irqchip");
++        exit(1);
++    }
++
++    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
++                            KVM_DEV_RISCV_AIA_CONFIG_MODE,
++                            &default_aia_mode, false, NULL);
++    if (ret < 0) {
++        error_report("KVM AIA: failed to get current KVM AIA mode");
++        exit(1);
++    }
++    qemu_log("KVM AIA: default mode is %s\n",
++             kvm_aia_mode_str(default_aia_mode));
++
++    if (default_aia_mode != aia_mode) {
++        ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
++                                KVM_DEV_RISCV_AIA_CONFIG_MODE,
++                                &aia_mode, true, NULL);
++        if (ret < 0)
++            warn_report("KVM AIA: failed to set KVM AIA mode");
++        else
++            qemu_log("KVM AIA: set current mode to %s\n",
++                     kvm_aia_mode_str(aia_mode));
++    }
++
++    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
++                            KVM_DEV_RISCV_AIA_CONFIG_SRCS,
++                            &aia_irq_num, true, NULL);
++    if (ret < 0) {
++        error_report("KVM AIA: failed to set number of input irq lines");
++        exit(1);
++    }
++
++    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
++                            KVM_DEV_RISCV_AIA_CONFIG_IDS,
++                            &aia_msi_num, true, NULL);
++    if (ret < 0) {
++        error_report("KVM AIA: failed to set number of msi");
++        exit(1);
++    }
++
++    socket_bits = find_last_bit(&socket_count, BITS_PER_LONG) + 1;
++    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
++                            KVM_DEV_RISCV_AIA_CONFIG_GROUP_BITS,
++                            &socket_bits, true, NULL);
++    if (ret < 0) {
++        error_report("KVM AIA: failed to set group_bits");
++        exit(1);
++    }
++
++    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
++                            KVM_DEV_RISCV_AIA_CONFIG_GROUP_SHIFT,
++                            &group_shift, true, NULL);
++    if (ret < 0) {
++        error_report("KVM AIA: failed to set group_shift");
++        exit(1);
++    }
++
++    guest_bits = guest_num == 0 ? 0 :
++                 find_last_bit(&guest_num, BITS_PER_LONG) + 1;
++    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
++                            KVM_DEV_RISCV_AIA_CONFIG_GUEST_BITS,
++                            &guest_bits, true, NULL);
++    if (ret < 0) {
++        error_report("KVM AIA: failed to set guest_bits");
++        exit(1);
++    }
++
++    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_ADDR,
++                            KVM_DEV_RISCV_AIA_ADDR_APLIC,
++                            &aplic_base, true, NULL);
++    if (ret < 0) {
++        error_report("KVM AIA: failed to set the base address of APLIC");
++        exit(1);
++    }
++
++    for (socket = 0; socket < socket_count; socket++) {
++        socket_imsic_base = imsic_base + socket * (1U << group_shift);
++        hart_count = riscv_socket_hart_count(machine, socket);
++        base_hart = riscv_socket_first_hartid(machine, socket);
++
++        if (max_hart_per_socket < hart_count) {
++            max_hart_per_socket = hart_count;
++        }
++
++        for (i = 0; i < hart_count; i++) {
++            imsic_addr = socket_imsic_base + i * IMSIC_HART_SIZE(guest_bits);
++            ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_ADDR,
++                                    KVM_DEV_RISCV_AIA_ADDR_IMSIC(i + base_hart),
++                                    &imsic_addr, true, NULL);
++            if (ret < 0) {
++                error_report("KVM AIA: failed to set the IMSIC address for hart %d", i);
++                exit(1);
++            }
++        }
++    }
++
++    hart_bits = find_last_bit(&max_hart_per_socket, BITS_PER_LONG) + 1;
++    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
++                            KVM_DEV_RISCV_AIA_CONFIG_HART_BITS,
++                            &hart_bits, true, NULL);
++    if (ret < 0) {
++        error_report("KVM AIA: failed to set hart_bits");
++        exit(1);
++    }
++
++    if (kvm_has_gsi_routing()) {
++        for (uint64_t idx = 0; idx < aia_irq_num + 1; ++idx) {
++            /* KVM AIA only has one APLIC instance */
++            kvm_irqchip_add_irq_route(kvm_state, idx, 0, idx);
++        }
++        kvm_gsi_routing_allowed = true;
++        kvm_irqchip_commit_routes(kvm_state);
++    }
++
++    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CTRL,
++                            KVM_DEV_RISCV_AIA_CTRL_INIT,
++                            NULL, true, NULL);
++    if (ret < 0) {
++        error_report("KVM AIA: initialized fail");
++        exit(1);
++    }
++
++    kvm_msi_via_irqfd_allowed = kvm_irqfds_enabled();
+ }
+--
+.41.0

-New patch
+[PULL v2 32/45] target/riscv: update APLIC and IMSIC to support KVM AIA
+From: Yong-Xuan Wang <yongxuan.wang@sifive.com>
+KVM AIA can't emulate APLIC only. When "aia=aplic" parameter is passed,
+APLIC devices is emulated by QEMU. For "aia=aplic-imsic", remove the
+mmio operations of APLIC when using KVM AIA and send wired interrupt
+signal via KVM_IRQ_LINE API.
+After KVM AIA enabled, MSI messages are delivered by KVM_SIGNAL_MSI API
+when the IMSICs receive mmio write requests.
+Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
+Reviewed-by: Jim Shu <jim.shu@sifive.com>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
+Message-ID: <20230727102439.22554-5-yongxuan.wang@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ hw/intc/riscv_aplic.c | 56 ++++++++++++++++++++++++++++++-------------
+ hw/intc/riscv_imsic.c | 25 +++++++++++++++----
+files changed, 61 insertions(+), 20 deletions(-)
+diff --git a/hw/intc/riscv_aplic.c b/hw/intc/riscv_aplic.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/intc/riscv_aplic.c
++++ b/hw/intc/riscv_aplic.c
+@@ -XXX,XX +XXX,XX @@
+ #include "hw/irq.h"
+ #include "target/riscv/cpu.h"
+ #include "sysemu/sysemu.h"
++#include "sysemu/kvm.h"
+ #include "migration/vmstate.h"
+ #define APLIC_MAX_IDC                  (1UL << 14)
+@@ -XXX,XX +XXX,XX @@
+ #define APLIC_IDC_CLAIMI               0x1c
++/*
++ * KVM AIA only supports APLIC MSI, fallback to QEMU emulation if we want to use
++ * APLIC Wired.
++ */
++static bool is_kvm_aia(bool msimode)
++{
++    return kvm_irqchip_in_kernel() && msimode;
++}
++
+ static uint32_t riscv_aplic_read_input_word(RISCVAPLICState *aplic,
+                                             uint32_t word)
+ {
+@@ -XXX,XX +XXX,XX @@ static uint32_t riscv_aplic_idc_claimi(RISCVAPLICState *aplic, uint32_t idc)
+     return topi;
+ }
++static void riscv_kvm_aplic_request(void *opaque, int irq, int level)
++{
++    kvm_set_irq(kvm_state, irq, !!level);
++}
++
+ static void riscv_aplic_request(void *opaque, int irq, int level)
+ {
+     bool update = false;
+@@ -XXX,XX +XXX,XX @@ static void riscv_aplic_realize(DeviceState *dev, Error **errp)
+     uint32_t i;
+     RISCVAPLICState *aplic = RISCV_APLIC(dev);
+-    aplic->bitfield_words = (aplic->num_irqs + 31) >> 5;
+-    aplic->sourcecfg = g_new0(uint32_t, aplic->num_irqs);
+-    aplic->state = g_new0(uint32_t, aplic->num_irqs);
+-    aplic->target = g_new0(uint32_t, aplic->num_irqs);
+-    if (!aplic->msimode) {
+-        for (i = 0; i < aplic->num_irqs; i++) {
+-            aplic->target[i] = 1;
++    if (!is_kvm_aia(aplic->msimode)) {
++        aplic->bitfield_words = (aplic->num_irqs + 31) >> 5;
++        aplic->sourcecfg = g_new0(uint32_t, aplic->num_irqs);
++        aplic->state = g_new0(uint32_t, aplic->num_irqs);
++        aplic->target = g_new0(uint32_t, aplic->num_irqs);
++        if (!aplic->msimode) {
++            for (i = 0; i < aplic->num_irqs; i++) {
++                aplic->target[i] = 1;
++            }
+         }
+-    }
+-    aplic->idelivery = g_new0(uint32_t, aplic->num_harts);
+-    aplic->iforce = g_new0(uint32_t, aplic->num_harts);
+-    aplic->ithreshold = g_new0(uint32_t, aplic->num_harts);
++        aplic->idelivery = g_new0(uint32_t, aplic->num_harts);
++        aplic->iforce = g_new0(uint32_t, aplic->num_harts);
++        aplic->ithreshold = g_new0(uint32_t, aplic->num_harts);
+-    memory_region_init_io(&aplic->mmio, OBJECT(dev), &riscv_aplic_ops, aplic,
+-                          TYPE_RISCV_APLIC, aplic->aperture_size);
+-    sysbus_init_mmio(SYS_BUS_DEVICE(dev), &aplic->mmio);
++        memory_region_init_io(&aplic->mmio, OBJECT(dev), &riscv_aplic_ops,
++                              aplic, TYPE_RISCV_APLIC, aplic->aperture_size);
++        sysbus_init_mmio(SYS_BUS_DEVICE(dev), &aplic->mmio);
++    }
+     /*
+      * Only root APLICs have hardware IRQ lines. All non-root APLICs
+      * have IRQ lines delegated by their parent APLIC.
+      */
+     if (!aplic->parent) {
+-        qdev_init_gpio_in(dev, riscv_aplic_request, aplic->num_irqs);
++        if (is_kvm_aia(aplic->msimode)) {
++            qdev_init_gpio_in(dev, riscv_kvm_aplic_request, aplic->num_irqs);
++        } else {
++            qdev_init_gpio_in(dev, riscv_aplic_request, aplic->num_irqs);
++        }
+     }
+     /* Create output IRQ lines for non-MSI mode */
+@@ -XXX,XX +XXX,XX @@ DeviceState *riscv_aplic_create(hwaddr addr, hwaddr size,
+     qdev_prop_set_bit(dev, "mmode", mmode);
+     sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal);
+-    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, addr);
++
++    if (!is_kvm_aia(msimode)) {
++        sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, addr);
++    }
+     if (parent) {
+         riscv_aplic_add_child(parent, dev);
+diff --git a/hw/intc/riscv_imsic.c b/hw/intc/riscv_imsic.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/intc/riscv_imsic.c
++++ b/hw/intc/riscv_imsic.c
+@@ -XXX,XX +XXX,XX @@
+ #include "target/riscv/cpu.h"
+ #include "target/riscv/cpu_bits.h"
+ #include "sysemu/sysemu.h"
++#include "sysemu/kvm.h"
+ #include "migration/vmstate.h"
+ #define IMSIC_MMIO_PAGE_LE             0x00
+@@ -XXX,XX +XXX,XX @@ static void riscv_imsic_write(void *opaque, hwaddr addr, uint64_t value,
+         goto err;
+     }
++#if defined(CONFIG_KVM)
++    if (kvm_irqchip_in_kernel()) {
++        struct kvm_msi msi;
++
++        msi.address_lo = extract64(imsic->mmio.addr + addr, 0, 32);
++        msi.address_hi = extract64(imsic->mmio.addr + addr, 32, 32);
++        msi.data = le32_to_cpu(value);
++
++        kvm_vm_ioctl(kvm_state, KVM_SIGNAL_MSI, &msi);
++
++        return;
++    }
++#endif
++
+     /* Writes only supported for MSI little-endian registers */
+     page = addr >> IMSIC_MMIO_PAGE_SHIFT;
+     if ((addr & (IMSIC_MMIO_PAGE_SZ - 1)) == IMSIC_MMIO_PAGE_LE) {
+@@ -XXX,XX +XXX,XX @@ static void riscv_imsic_realize(DeviceState *dev, Error **errp)
+     CPUState *cpu = cpu_by_arch_id(imsic->hartid);
+     CPURISCVState *env = cpu ? cpu->env_ptr : NULL;
+-    imsic->num_eistate = imsic->num_pages * imsic->num_irqs;
+-    imsic->eidelivery = g_new0(uint32_t, imsic->num_pages);
+-    imsic->eithreshold = g_new0(uint32_t, imsic->num_pages);
+-    imsic->eistate = g_new0(uint32_t, imsic->num_eistate);
++    if (!kvm_irqchip_in_kernel()) {
++        imsic->num_eistate = imsic->num_pages * imsic->num_irqs;
++        imsic->eidelivery = g_new0(uint32_t, imsic->num_pages);
++        imsic->eithreshold = g_new0(uint32_t, imsic->num_pages);
++        imsic->eistate = g_new0(uint32_t, imsic->num_eistate);
++    }
+     memory_region_init_io(&imsic->mmio, OBJECT(dev), &riscv_imsic_ops,
+                           imsic, TYPE_RISCV_IMSIC,
+--
+.41.0

-New patch
+[PULL v2 33/45] target/riscv: select KVM AIA in riscv virt machine
+From: Yong-Xuan Wang <yongxuan.wang@sifive.com>
+Select KVM AIA when the host kernel has in-kernel AIA chip support.
+Since KVM AIA only has one APLIC instance, we map the QEMU APLIC
+devices to KVM APLIC.
+Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
+Reviewed-by: Jim Shu <jim.shu@sifive.com>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
+Message-ID: <20230727102439.22554-6-yongxuan.wang@sifive.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ hw/riscv/virt.c | 94 +++++++++++++++++++++++++++++++++----------------
+file changed, 63 insertions(+), 31 deletions(-)
+diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/riscv/virt.c
++++ b/hw/riscv/virt.c
+@@ -XXX,XX +XXX,XX @@
+ #include "hw/riscv/virt.h"
+ #include "hw/riscv/boot.h"
+ #include "hw/riscv/numa.h"
++#include "kvm_riscv.h"
+ #include "hw/intc/riscv_aclint.h"
+ #include "hw/intc/riscv_aplic.h"
+ #include "hw/intc/riscv_imsic.h"
+@@ -XXX,XX +XXX,XX @@
+ #error "Can't accommodate all IMSIC groups in address space"
+ #endif
++/* KVM AIA only supports APLIC MSI. APLIC Wired is always emulated by QEMU. */
++static bool virt_use_kvm_aia(RISCVVirtState *s)
++{
++    return kvm_irqchip_in_kernel() && s->aia_type == VIRT_AIA_TYPE_APLIC_IMSIC;
++}
++
+ static const MemMapEntry virt_memmap[] = {
+     [VIRT_DEBUG] =        {        0x0,         0x100 },
+     [VIRT_MROM] =         {     0x1000,        0xf000 },
+@@ -XXX,XX +XXX,XX @@ static void create_fdt_one_aplic(RISCVVirtState *s, int socket,
+                                  uint32_t *intc_phandles,
+                                  uint32_t aplic_phandle,
+                                  uint32_t aplic_child_phandle,
+-                                 bool m_mode)
++                                 bool m_mode, int num_harts)
+ {
+     int cpu;
+     char *aplic_name;
+     uint32_t *aplic_cells;
+     MachineState *ms = MACHINE(s);
+-    aplic_cells = g_new0(uint32_t, s->soc[socket].num_harts * 2);
++    aplic_cells = g_new0(uint32_t, num_harts * 2);
+-    for (cpu = 0; cpu < s->soc[socket].num_harts; cpu++) {
++    for (cpu = 0; cpu < num_harts; cpu++) {
+         aplic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
+         aplic_cells[cpu * 2 + 1] = cpu_to_be32(m_mode ? IRQ_M_EXT : IRQ_S_EXT);
+     }
+@@ -XXX,XX +XXX,XX @@ static void create_fdt_one_aplic(RISCVVirtState *s, int socket,
+     if (s->aia_type == VIRT_AIA_TYPE_APLIC) {
+         qemu_fdt_setprop(ms->fdt, aplic_name, "interrupts-extended",
+-                         aplic_cells,
+-                         s->soc[socket].num_harts * sizeof(uint32_t) * 2);
++                         aplic_cells, num_harts * sizeof(uint32_t) * 2);
+     } else {
+         qemu_fdt_setprop_cell(ms->fdt, aplic_name, "msi-parent", msi_phandle);
+     }
+@@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aplic(RISCVVirtState *s,
+                                     uint32_t msi_s_phandle,
+                                     uint32_t *phandle,
+                                     uint32_t *intc_phandles,
+-                                    uint32_t *aplic_phandles)
++                                    uint32_t *aplic_phandles,
++                                    int num_harts)
+ {
+     char *aplic_name;
+     unsigned long aplic_addr;
+@@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aplic(RISCVVirtState *s,
+         create_fdt_one_aplic(s, socket, aplic_addr, memmap[VIRT_APLIC_M].size,
+                              msi_m_phandle, intc_phandles,
+                              aplic_m_phandle, aplic_s_phandle,
+-                             true);
++                             true, num_harts);
+     }
+     /* S-level APLIC node */
+@@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aplic(RISCVVirtState *s,
+     create_fdt_one_aplic(s, socket, aplic_addr, memmap[VIRT_APLIC_S].size,
+                          msi_s_phandle, intc_phandles,
+                          aplic_s_phandle, 0,
+-                         false);
++                         false, num_harts);
+     aplic_name = g_strdup_printf("/soc/aplic@%lx", aplic_addr);
+@@ -XXX,XX +XXX,XX @@ static void create_fdt_sockets(RISCVVirtState *s, const MemMapEntry *memmap,
+         *msi_pcie_phandle = msi_s_phandle;
+     }
+-    phandle_pos = ms->smp.cpus;
+-    for (socket = (socket_count - 1); socket >= 0; socket--) {
+-        phandle_pos -= s->soc[socket].num_harts;
+-
+-        if (s->aia_type == VIRT_AIA_TYPE_NONE) {
+-            create_fdt_socket_plic(s, memmap, socket, phandle,
+-                &intc_phandles[phandle_pos], xplic_phandles);
+-        } else {
+-            create_fdt_socket_aplic(s, memmap, socket,
+-                msi_m_phandle, msi_s_phandle, phandle,
+-                &intc_phandles[phandle_pos], xplic_phandles);
++    /* KVM AIA only has one APLIC instance */
++    if (virt_use_kvm_aia(s)) {
++        create_fdt_socket_aplic(s, memmap, 0,
++                                msi_m_phandle, msi_s_phandle, phandle,
++                                &intc_phandles[0], xplic_phandles,
++                                ms->smp.cpus);
++    } else {
++        phandle_pos = ms->smp.cpus;
++        for (socket = (socket_count - 1); socket >= 0; socket--) {
++            phandle_pos -= s->soc[socket].num_harts;
++
++            if (s->aia_type == VIRT_AIA_TYPE_NONE) {
++                create_fdt_socket_plic(s, memmap, socket, phandle,
++                                       &intc_phandles[phandle_pos],
++                                       xplic_phandles);
++            } else {
++                create_fdt_socket_aplic(s, memmap, socket,
++                                        msi_m_phandle, msi_s_phandle, phandle,
++                                        &intc_phandles[phandle_pos],
++                                        xplic_phandles,
++                                        s->soc[socket].num_harts);
++            }
+         }
+     }
+     g_free(intc_phandles);
+-    for (socket = 0; socket < socket_count; socket++) {
+-        if (socket == 0) {
+-            *irq_mmio_phandle = xplic_phandles[socket];
+-            *irq_virtio_phandle = xplic_phandles[socket];
+-            *irq_pcie_phandle = xplic_phandles[socket];
+-        }
+-        if (socket == 1) {
+-            *irq_virtio_phandle = xplic_phandles[socket];
+-            *irq_pcie_phandle = xplic_phandles[socket];
+-        }
+-        if (socket == 2) {
+-            *irq_pcie_phandle = xplic_phandles[socket];
++    if (virt_use_kvm_aia(s)) {
++        *irq_mmio_phandle = xplic_phandles[0];
++        *irq_virtio_phandle = xplic_phandles[0];
++        *irq_pcie_phandle = xplic_phandles[0];
++    } else {
++        for (socket = 0; socket < socket_count; socket++) {
++            if (socket == 0) {
++                *irq_mmio_phandle = xplic_phandles[socket];
++                *irq_virtio_phandle = xplic_phandles[socket];
++                *irq_pcie_phandle = xplic_phandles[socket];
++            }
++            if (socket == 1) {
++                *irq_virtio_phandle = xplic_phandles[socket];
++                *irq_pcie_phandle = xplic_phandles[socket];
++            }
++            if (socket == 2) {
++                *irq_pcie_phandle = xplic_phandles[socket];
++            }
+         }
+     }
+@@ -XXX,XX +XXX,XX @@ static void virt_machine_init(MachineState *machine)
+         }
+     }
++    if (virt_use_kvm_aia(s)) {
++        kvm_riscv_aia_create(machine, IMSIC_MMIO_GROUP_MIN_SHIFT,
++                             VIRT_IRQCHIP_NUM_SOURCES, VIRT_IRQCHIP_NUM_MSIS,
++                             memmap[VIRT_APLIC_S].base,
++                             memmap[VIRT_IMSIC_S].base,
++                             s->aia_guests);
++    }
++
+     if (riscv_is_32bit(&s->soc[0])) {
+ #if HOST_LONG_BITS == 64
+         /* limit RAM size in a 32-bit system */
+--
+.41.0

-[PULL 06/10] hw/riscv: virt: Enable booting S-mode firmware from pflash
+[PULL v2 34/45] hw/riscv: virt: Fix riscv,pmu DT node path
-From: Sunil V L <sunilvl@ventanamicro.com>
+From: Conor Dooley <conor.dooley@microchip.com>
-To boot S-mode firmware payload like EDK2 from persistent
+On a dtb dumped from the virt machine, dt-validate complains:
-flash storage, qemu needs to pass the flash address as the
+soc: pmu: {'riscv,event-to-mhpmcounters': [[1, 1, 524281], [2, 2, 524284], [65561, 65561, 524280], [65563, 65563, 524280], [65569, 65569, 524280]], 'compatible': ['riscv,pmu']} should not be valid under {'type': 'object'}
-next_addr in fw_dynamic_info to the opensbi.
+        from schema $id: http://devicetree.org/schemas/simple-bus.yaml#
 That's pretty cryptic, but running the dtb back through dtc produces
 something a lot more reasonable:
 Warning (simple_bus_reg): /soc/pmu: missing or empty reg/ranges property
-When both -kernel and -pflash options are provided in command line,
+Moving the riscv,pmu node out of the soc bus solves the problem.
 the kernel (and initrd if -initrd) will be copied to fw_cfg table.
 The S-mode FW will load the kernel/initrd from fw_cfg table.
-If only pflash is given but not -kernel, then it is the job of
+Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
-of the S-mode firmware to locate and load the kernel.
+Acked-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
-In either case, update the kernel_entry with the flash address
+Message-ID: <20230727-groom-decline-2c57ce42841c@spud>
 so that the opensbi can jump to the entry point of the S-mode
 firmware.
 Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
 Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
 Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
 Message-Id: <20221004092351.18209-4-sunilvl@ventanamicro.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
- include/hw/riscv/boot.h |  1 +
+ hw/riscv/virt.c | 2 +-
- hw/riscv/boot.c         | 29 +++++++++++++++++++++++++++++
+file changed, 1 insertion(+), 1 deletion(-)
  hw/riscv/virt.c         | 18 +++++++++++++++++-
 files changed, 47 insertions(+), 1 deletion(-)
-diff --git a/include/hw/riscv/boot.h b/include/hw/riscv/boot.h
-index XXXXXXX..XXXXXXX 100644
---- a/include/hw/riscv/boot.h
-+++ b/include/hw/riscv/boot.h
-@@ -XXX,XX +XXX,XX @@ void riscv_rom_copy_firmware_info(MachineState *machine, hwaddr rom_base,
-                                   uint32_t reset_vec_size,
-                                   uint64_t kernel_entry);
- void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr);
-+void riscv_setup_firmware_boot(MachineState *machine);
- #endif /* RISCV_BOOT_H */
-diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c
-index XXXXXXX..XXXXXXX 100644
---- a/hw/riscv/boot.c
-+++ b/hw/riscv/boot.c
-@@ -XXX,XX +XXX,XX @@ void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr)
-         riscv_cpu->env.fdt_addr = fdt_addr;
-     }
- }
-+
-+void riscv_setup_firmware_boot(MachineState *machine)
-+{
-+    if (machine->kernel_filename) {
-+        FWCfgState *fw_cfg;
-+        fw_cfg = fw_cfg_find();
-+
-+        assert(fw_cfg);
-+        /*
-+         * Expose the kernel, the command line, and the initrd in fw_cfg.
-+         * We don't process them here at all, it's all left to the
-+         * firmware.
-+         */
-+        load_image_to_fw_cfg(fw_cfg,
-+                             FW_CFG_KERNEL_SIZE, FW_CFG_KERNEL_DATA,
-+                             machine->kernel_filename,
-+                             true);
-+        load_image_to_fw_cfg(fw_cfg,
-+                             FW_CFG_INITRD_SIZE, FW_CFG_INITRD_DATA,
-+                             machine->initrd_filename, false);
-+
-+        if (machine->kernel_cmdline) {
-+            fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_SIZE,
-+                           strlen(machine->kernel_cmdline) + 1);
-+            fw_cfg_add_string(fw_cfg, FW_CFG_CMDLINE_DATA,
-+                              machine->kernel_cmdline);
-+        }
-+    }
-+}
 diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/riscv/virt.c
 +++ b/hw/riscv/virt.c
-@@ -XXX,XX +XXX,XX @@ static void virt_machine_done(Notifier *notifier, void *data)
+@@ -XXX,XX +XXX,XX @@ static void create_fdt_pmu(RISCVVirtState *s)
-     s->fw_cfg = create_fw_cfg(machine);
+     MachineState *ms = MACHINE(s);
-     rom_set_fw(s->fw_cfg);
+     RISCVCPU hart = s->soc[0].harts[0];
--    if (machine->kernel_filename) {
+-    pmu_name = g_strdup_printf("/soc/pmu");
-+    if (drive_get(IF_PFLASH, 0, 1)) {
++    pmu_name = g_strdup_printf("/pmu");
-+        /*
+     qemu_fdt_add_subnode(ms->fdt, pmu_name);
-+         * S-mode FW like EDK2 will be kept in second plash (unit 1).
+     qemu_fdt_setprop_string(ms->fdt, pmu_name, "compatible", "riscv,pmu");
-+         * When both kernel, initrd and pflash options are provided in the
+     riscv_pmu_generate_fdt_node(ms->fdt, hart.cfg.pmu_num, pmu_name);
 +         * command line, the kernel and initrd will be copied to the fw_cfg
 +         * table and opensbi will jump to the flash address which is the
 +         * entry point of S-mode FW. It is the job of the S-mode FW to load
 +         * the kernel and initrd using fw_cfg table.
 +         *
 +         * If only pflash is given but not -kernel, then it is the job of
 +         * of the S-mode firmware to locate and load the kernel.
 +         * In either case, the next_addr for opensbi will be the flash address.
 +         */
 +        riscv_setup_firmware_boot(machine);
 +        kernel_entry = virt_memmap[VIRT_FLASH].base +
 +                       virt_memmap[VIRT_FLASH].size / 2;
 +    } else if (machine->kernel_filename) {
          kernel_start_addr = riscv_calc_kernel_start_addr(&s->soc[0],
                                                           firmware_end_addr);
 --
-.37.3
+.41.0

-New patch
+[PULL v2 35/45] target/riscv: Update CSR bits name for svadu extension
+From: Weiwei Li <liweiwei@iscas.ac.cn>
+The Svadu specification updated the name of the *envcfg bit from
+HADE to ADUE.
+Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn>
+Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn>
+Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Message-ID: <20230816141916.66898-1-liweiwei@iscas.ac.cn>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/cpu_bits.h   |  8 ++++----
+ target/riscv/cpu.c        |  4 ++--
+ target/riscv/cpu_helper.c |  6 +++---
+ target/riscv/csr.c        | 12 ++++++------
+files changed, 15 insertions(+), 15 deletions(-)
+diff --git a/target/riscv/cpu_bits.h b/target/riscv/cpu_bits.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu_bits.h
++++ b/target/riscv/cpu_bits.h
+@@ -XXX,XX +XXX,XX @@ typedef enum RISCVException {
+ #define MENVCFG_CBIE                       (3UL << 4)
+ #define MENVCFG_CBCFE                      BIT(6)
+ #define MENVCFG_CBZE                       BIT(7)
+-#define MENVCFG_HADE                       (1ULL << 61)
++#define MENVCFG_ADUE                       (1ULL << 61)
+ #define MENVCFG_PBMTE                      (1ULL << 62)
+ #define MENVCFG_STCE                       (1ULL << 63)
+ /* For RV32 */
+-#define MENVCFGH_HADE                      BIT(29)
++#define MENVCFGH_ADUE                      BIT(29)
+ #define MENVCFGH_PBMTE                     BIT(30)
+ #define MENVCFGH_STCE                      BIT(31)
+@@ -XXX,XX +XXX,XX @@ typedef enum RISCVException {
+ #define HENVCFG_CBIE                       MENVCFG_CBIE
+ #define HENVCFG_CBCFE                      MENVCFG_CBCFE
+ #define HENVCFG_CBZE                       MENVCFG_CBZE
+-#define HENVCFG_HADE                       MENVCFG_HADE
++#define HENVCFG_ADUE                       MENVCFG_ADUE
+ #define HENVCFG_PBMTE                      MENVCFG_PBMTE
+ #define HENVCFG_STCE                       MENVCFG_STCE
+ /* For RV32 */
+-#define HENVCFGH_HADE                       MENVCFGH_HADE
++#define HENVCFGH_ADUE                       MENVCFGH_ADUE
+ #define HENVCFGH_PBMTE                      MENVCFGH_PBMTE
+ #define HENVCFGH_STCE                       MENVCFGH_STCE
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu.c
++++ b/target/riscv/cpu.c
+@@ -XXX,XX +XXX,XX @@ static void riscv_cpu_reset_hold(Object *obj)
+     env->two_stage_lookup = false;
+     env->menvcfg = (cpu->cfg.ext_svpbmt ? MENVCFG_PBMTE : 0) |
+-                   (cpu->cfg.ext_svadu ? MENVCFG_HADE : 0);
++                   (cpu->cfg.ext_svadu ? MENVCFG_ADUE : 0);
+     env->henvcfg = (cpu->cfg.ext_svpbmt ? HENVCFG_PBMTE : 0) |
+-                   (cpu->cfg.ext_svadu ? HENVCFG_HADE : 0);
++                   (cpu->cfg.ext_svadu ? HENVCFG_ADUE : 0);
+     /* Initialized default priorities of local interrupts. */
+     for (i = 0; i < ARRAY_SIZE(env->miprio); i++) {
+diff --git a/target/riscv/cpu_helper.c b/target/riscv/cpu_helper.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu_helper.c
++++ b/target/riscv/cpu_helper.c
+@@ -XXX,XX +XXX,XX @@ static int get_physical_address(CPURISCVState *env, hwaddr *physical,
+     }
+     bool pbmte = env->menvcfg & MENVCFG_PBMTE;
+-    bool hade = env->menvcfg & MENVCFG_HADE;
++    bool adue = env->menvcfg & MENVCFG_ADUE;
+     if (first_stage && two_stage && env->virt_enabled) {
+         pbmte = pbmte && (env->henvcfg & HENVCFG_PBMTE);
+-        hade = hade && (env->henvcfg & HENVCFG_HADE);
++        adue = adue && (env->henvcfg & HENVCFG_ADUE);
+     }
+     int ptshift = (levels - 1) * ptidxbits;
+@@ -XXX,XX +XXX,XX @@ restart:
+     /* Page table updates need to be atomic with MTTCG enabled */
+     if (updated_pte != pte && !is_debug) {
+-        if (!hade) {
++        if (!adue) {
+             return TRANSLATE_FAIL;
+         }
+diff --git a/target/riscv/csr.c b/target/riscv/csr.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/csr.c
++++ b/target/riscv/csr.c
+@@ -XXX,XX +XXX,XX @@ static RISCVException write_menvcfg(CPURISCVState *env, int csrno,
+     if (riscv_cpu_mxl(env) == MXL_RV64) {
+         mask |= (cfg->ext_svpbmt ? MENVCFG_PBMTE : 0) |
+                 (cfg->ext_sstc ? MENVCFG_STCE : 0) |
+-                (cfg->ext_svadu ? MENVCFG_HADE : 0);
++                (cfg->ext_svadu ? MENVCFG_ADUE : 0);
+     }
+     env->menvcfg = (env->menvcfg & ~mask) | (val & mask);
+@@ -XXX,XX +XXX,XX @@ static RISCVException write_menvcfgh(CPURISCVState *env, int csrno,
+     const RISCVCPUConfig *cfg = riscv_cpu_cfg(env);
+     uint64_t mask = (cfg->ext_svpbmt ? MENVCFG_PBMTE : 0) |
+                     (cfg->ext_sstc ? MENVCFG_STCE : 0) |
+-                    (cfg->ext_svadu ? MENVCFG_HADE : 0);
++                    (cfg->ext_svadu ? MENVCFG_ADUE : 0);
+     uint64_t valh = (uint64_t)val << 32;
+     env->menvcfg = (env->menvcfg & ~mask) | (valh & mask);
+@@ -XXX,XX +XXX,XX @@ static RISCVException read_henvcfg(CPURISCVState *env, int csrno,
+      * henvcfg.stce is read_only 0 when menvcfg.stce = 0
+      * henvcfg.hade is read_only 0 when menvcfg.hade = 0
+      */
+-    *val = env->henvcfg & (~(HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_HADE) |
++    *val = env->henvcfg & (~(HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_ADUE) |
+                            env->menvcfg);
+     return RISCV_EXCP_NONE;
+ }
+@@ -XXX,XX +XXX,XX @@ static RISCVException write_henvcfg(CPURISCVState *env, int csrno,
+     }
+     if (riscv_cpu_mxl(env) == MXL_RV64) {
+-        mask |= env->menvcfg & (HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_HADE);
++        mask |= env->menvcfg & (HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_ADUE);
+     }
+     env->henvcfg = (env->henvcfg & ~mask) | (val & mask);
+@@ -XXX,XX +XXX,XX @@ static RISCVException read_henvcfgh(CPURISCVState *env, int csrno,
+         return ret;
+     }
+-    *val = (env->henvcfg & (~(HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_HADE) |
++    *val = (env->henvcfg & (~(HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_ADUE) |
+                             env->menvcfg)) >> 32;
+     return RISCV_EXCP_NONE;
+ }
+@@ -XXX,XX +XXX,XX @@ static RISCVException write_henvcfgh(CPURISCVState *env, int csrno,
+                                      target_ulong val)
+ {
+     uint64_t mask = env->menvcfg & (HENVCFG_PBMTE | HENVCFG_STCE |
+-                                    HENVCFG_HADE);
++                                    HENVCFG_ADUE);
+     uint64_t valh = (uint64_t)val << 32;
+     RISCVException ret;
+--
+.41.0

-New patch
+[PULL v2 36/45] target/riscv: fix satp_mode_finalize() when satp_mode.supported = 0
+From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+In the same emulated RISC-V host, the 'host' KVM CPU takes 4 times
+longer to boot than the 'rv64' KVM CPU.
+The reason is an unintended behavior of riscv_cpu_satp_mode_finalize()
+when satp_mode.supported = 0, i.e. when cpu_init() does not set
+satp_mode_max_supported(). satp_mode_max_from_map(map) does:
+- __builtin_clz(map)
+This means that, if satp_mode.supported = 0, satp_mode_supported_max
+wil be '31 - 32'. But this is C, so satp_mode_supported_max will gladly
+set it to UINT_MAX (4294967295). After that, if the user didn't set a
+satp_mode, set_satp_mode_default_map(cpu) will make
+cfg.satp_mode.map = cfg.satp_mode.supported
+So satp_mode.map = 0. And then satp_mode_map_max will be set to
+satp_mode_max_from_map(cpu->cfg.satp_mode.map), i.e. also UINT_MAX. The
+guard "satp_mode_map_max > satp_mode_supported_max" doesn't protect us
+here since both are UINT_MAX.
+And finally we have 2 loops:
+        for (int i = satp_mode_map_max - 1; i >= 0; --i) {
+Which are, in fact, 2 loops from UINT_MAX -1 to -1. This is where the
+extra delay when booting the 'host' CPU is coming from.
+Commit 43d1de32f8 already set a precedence for satp_mode.supported = 0
+in a different manner. We're doing the same here. If supported == 0,
+interpret as 'the CPU wants the OS to handle satp mode alone' and skip
+satp_mode_finalize().
+We'll also put a guard in satp_mode_max_from_map() to assert out if map
+is 0 since the function is not ready to deal with it.
+Cc: Alexandre Ghiti <alexghiti@rivosinc.com>
+Fixes: 6f23aaeb9b ("riscv: Allow user to set the satp mode")
+Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
+Message-ID: <20230817152903.694926-1-dbarboza@ventanamicro.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/cpu.c | 23 ++++++++++++++++++++---
+file changed, 20 insertions(+), 3 deletions(-)
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu.c
++++ b/target/riscv/cpu.c
+@@ -XXX,XX +XXX,XX @@ static uint8_t satp_mode_from_str(const char *satp_mode_str)
+ uint8_t satp_mode_max_from_map(uint32_t map)
+ {
++    /*
++     * 'map = 0' will make us return (31 - 32), which C will
++     * happily overflow to UINT_MAX. There's no good result to
++     * return if 'map = 0' (e.g. returning 0 will be ambiguous
++     * with the result for 'map = 1').
++     *
++     * Assert out if map = 0. Callers will have to deal with
++     * it outside of this function.
++     */
++    g_assert(map > 0);
++
+     /* map here has at least one bit set, so no problem with clz */
+     return 31 - __builtin_clz(map);
+ }
+@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
+ static void riscv_cpu_satp_mode_finalize(RISCVCPU *cpu, Error **errp)
+ {
+     bool rv32 = riscv_cpu_mxl(&cpu->env) == MXL_RV32;
+-    uint8_t satp_mode_map_max;
+-    uint8_t satp_mode_supported_max =
+-                        satp_mode_max_from_map(cpu->cfg.satp_mode.supported);
++    uint8_t satp_mode_map_max, satp_mode_supported_max;
++
++    /* The CPU wants the OS to decide which satp mode to use */
++    if (cpu->cfg.satp_mode.supported == 0) {
++        return;
++    }
++
++    satp_mode_supported_max =
++                    satp_mode_max_from_map(cpu->cfg.satp_mode.supported);
+     if (cpu->cfg.satp_mode.map == 0) {
+         if (cpu->cfg.satp_mode.init == 0) {
+--
+.41.0

-New patch
+[PULL v2 37/45] riscv: zicond: make non-experimental
+From: Vineet Gupta <vineetg@rivosinc.com>
+zicond is now codegen supported in both llvm and gcc.
+This change allows seamless enabling/testing of zicond in downstream
+projects. e.g. currently riscv-gnu-toolchain parses elf attributes
+to create a cmdline for qemu but fails short of enabling it because of
+the "x-" prefix.
+Signed-off-by: Vineet Gupta <vineetg@rivosinc.com>
+Message-ID: <20230808181715.436395-1-vineetg@rivosinc.com>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/cpu.c | 2 +-
+file changed, 1 insertion(+), 1 deletion(-)
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/cpu.c
++++ b/target/riscv/cpu.c
+@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
+     DEFINE_PROP_BOOL("zcf", RISCVCPU, cfg.ext_zcf, false),
+     DEFINE_PROP_BOOL("zcmp", RISCVCPU, cfg.ext_zcmp, false),
+     DEFINE_PROP_BOOL("zcmt", RISCVCPU, cfg.ext_zcmt, false),
++    DEFINE_PROP_BOOL("zicond", RISCVCPU, cfg.ext_zicond, false),
+     /* Vendor-specific custom extensions */
+     DEFINE_PROP_BOOL("xtheadba", RISCVCPU, cfg.ext_xtheadba, false),
+@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
+     DEFINE_PROP_BOOL("xventanacondops", RISCVCPU, cfg.ext_XVentanaCondOps, false),
+     /* These are experimental so mark with 'x-' */
+-    DEFINE_PROP_BOOL("x-zicond", RISCVCPU, cfg.ext_zicond, false),
+     /* ePMP 0.9.3 */
+     DEFINE_PROP_BOOL("x-epmp", RISCVCPU, cfg.epmp, false),
+--
+.41.0

-[PULL 05/10] hw/riscv: virt: Move create_fw_cfg() prior to loading kernel
+[PULL v2 38/45] hw/riscv/virt.c: fix non-KVM --enable-debug build
-From: Sunil V L <sunilvl@ventanamicro.com>
+From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
-To enable both -kernel and -pflash options, the fw_cfg needs to be
+A build with --enable-debug and without KVM will fail as follows:
 created prior to loading the kernel.
-Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
+/usr/bin/ld: libqemu-riscv64-softmmu.fa.p/hw_riscv_virt.c.o: in function `virt_machine_init':
 ./qemu/build/../hw/riscv/virt.c:1465: undefined reference to `kvm_riscv_aia_create'
 This happens because the code block with "if virt_use_kvm_aia(s)" isn't
 being ignored by the debug build, resulting in an undefined reference to
 a KVM only function.
 Add a 'kvm_enabled()' conditional together with virt_use_kvm_aia() will
 make the compiler crop the kvm_riscv_aia_create() call entirely from a
 non-KVM build. Note that adding the 'kvm_enabled()' conditional inside
 virt_use_kvm_aia() won't fix the build because this function would need
 to be inlined multiple times to make the compiler zero out the entire
 block.
 While we're at it, use kvm_enabled() in all instances where
 virt_use_kvm_aia() is checked to allow the compiler to elide these other
 kvm-only instances as well.
 Suggested-by: Richard Henderson <richard.henderson@linaro.org>
 Fixes: dbdb99948e ("target/riscv: select KVM AIA in riscv virt machine")
 Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
 Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
-Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
-Message-Id: <20221004092351.18209-3-sunilvl@ventanamicro.com>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
 Message-ID: <20230830133503.711138-2-dbarboza@ventanamicro.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
- hw/riscv/virt.c | 14 +++++++-------
+ hw/riscv/virt.c | 6 +++---
-file changed, 7 insertions(+), 7 deletions(-)
+file changed, 3 insertions(+), 3 deletions(-)
 diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/riscv/virt.c
 +++ b/hw/riscv/virt.c
-@@ -XXX,XX +XXX,XX @@ static void virt_machine_done(Notifier *notifier, void *data)
+@@ -XXX,XX +XXX,XX @@ static void create_fdt_sockets(RISCVVirtState *s, const MemMapEntry *memmap,
                                      RISCV64_BIOS_BIN, start_addr, NULL);
      }
-+    /*
+     /* KVM AIA only has one APLIC instance */
-+     * Init fw_cfg.  Must be done before riscv_load_fdt, otherwise the device
+-    if (virt_use_kvm_aia(s)) {
-+     * tree cannot be altered and we get FDT_ERR_NOSPACE.
++    if (kvm_enabled() && virt_use_kvm_aia(s)) {
-+     */
+         create_fdt_socket_aplic(s, memmap, 0,
-+    s->fw_cfg = create_fw_cfg(machine);
+                                 msi_m_phandle, msi_s_phandle, phandle,
-+    rom_set_fw(s->fw_cfg);
+                                 &intc_phandles[0], xplic_phandles,
-+
+@@ -XXX,XX +XXX,XX @@ static void create_fdt_sockets(RISCVVirtState *s, const MemMapEntry *memmap,
-     if (machine->kernel_filename) {
-         kernel_start_addr = riscv_calc_kernel_start_addr(&s->soc[0],
+     g_free(intc_phandles);
-                                                          firmware_end_addr);
-@@ -XXX,XX +XXX,XX @@ static void virt_machine_done(Notifier *notifier, void *data)
+-    if (virt_use_kvm_aia(s)) {
-         start_addr = virt_memmap[VIRT_FLASH].base;
++    if (kvm_enabled() && virt_use_kvm_aia(s)) {
          *irq_mmio_phandle = xplic_phandles[0];
          *irq_virtio_phandle = xplic_phandles[0];
          *irq_pcie_phandle = xplic_phandles[0];
@@ -XXX,XX +XXX,XX @@ static void virt_machine_init(MachineState *machine)
          }
      }
--    /*
+-    if (virt_use_kvm_aia(s)) {
--     * Init fw_cfg.  Must be done before riscv_load_fdt, otherwise the device
++    if (kvm_enabled() && virt_use_kvm_aia(s)) {
--     * tree cannot be altered and we get FDT_ERR_NOSPACE.
+         kvm_riscv_aia_create(machine, IMSIC_MMIO_GROUP_MIN_SHIFT,
--     */
+                              VIRT_IRQCHIP_NUM_SOURCES, VIRT_IRQCHIP_NUM_MSIS,
--    s->fw_cfg = create_fw_cfg(machine);
+                              memmap[VIRT_APLIC_S].base,
 -    rom_set_fw(s->fw_cfg);
 -
      /* Compute the fdt load address in dram */
      fdt_load_addr = riscv_load_fdt(memmap[VIRT_DRAM].base,
                                     machine->ram_size, machine->fdt);
 --
-.37.3
+.41.0

-New patch
+[PULL v2 39/45] hw/intc/riscv_aplic.c fix non-KVM --enable-debug build
+From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Commit 6df0b37e2ab breaks a --enable-debug build in a non-KVM
+environment with the following error:
+/usr/bin/ld: libqemu-riscv64-softmmu.fa.p/hw_intc_riscv_aplic.c.o: in function `riscv_kvm_aplic_request':
+./qemu/build/../hw/intc/riscv_aplic.c:486: undefined reference to `kvm_set_irq'
+collect2: error: ld returned 1 exit status
+This happens because the debug build will poke into the
+'if (is_kvm_aia(aplic->msimode))' block and fail to find a reference to
+the KVM only function riscv_kvm_aplic_request().
+There are multiple solutions to fix this. We'll go with the same
+solution from the previous patch, i.e. add a kvm_enabled() conditional
+to filter out the block. But there's a catch: riscv_kvm_aplic_request()
+is a local function that would end up being used if the compiler crops
+the block, and this won't work. Quoting Richard Henderson's explanation
+in [1]:
+"(...) the compiler won't eliminate entire unused functions with -O0"
+We'll solve it by moving riscv_kvm_aplic_request() to kvm.c and add its
+declaration in kvm_riscv.h, where all other KVM specific public
+functions are already declared. Other archs handles KVM specific code in
+this manner and we expect to do the same from now on.
+[1] https://lore.kernel.org/qemu-riscv/d2f1ad02-eb03-138f-9d08-db676deeed05@linaro.org/
+Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
+Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Message-ID: <20230830133503.711138-3-dbarboza@ventanamicro.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/kvm_riscv.h | 1 +
+ hw/intc/riscv_aplic.c    | 8 ++------
+ target/riscv/kvm.c       | 5 +++++
+files changed, 8 insertions(+), 6 deletions(-)
+diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/kvm_riscv.h
++++ b/target/riscv/kvm_riscv.h
+@@ -XXX,XX +XXX,XX @@ void kvm_riscv_aia_create(MachineState *machine, uint64_t group_shift,
+                           uint64_t aia_irq_num, uint64_t aia_msi_num,
+                           uint64_t aplic_base, uint64_t imsic_base,
+                           uint64_t guest_num);
++void riscv_kvm_aplic_request(void *opaque, int irq, int level);
+ #endif
+diff --git a/hw/intc/riscv_aplic.c b/hw/intc/riscv_aplic.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/intc/riscv_aplic.c
++++ b/hw/intc/riscv_aplic.c
+@@ -XXX,XX +XXX,XX @@
+ #include "target/riscv/cpu.h"
+ #include "sysemu/sysemu.h"
+ #include "sysemu/kvm.h"
++#include "kvm_riscv.h"
+ #include "migration/vmstate.h"
+ #define APLIC_MAX_IDC                  (1UL << 14)
+@@ -XXX,XX +XXX,XX @@ static uint32_t riscv_aplic_idc_claimi(RISCVAPLICState *aplic, uint32_t idc)
+     return topi;
+ }
+-static void riscv_kvm_aplic_request(void *opaque, int irq, int level)
+-{
+-    kvm_set_irq(kvm_state, irq, !!level);
+-}
+-
+ static void riscv_aplic_request(void *opaque, int irq, int level)
+ {
+     bool update = false;
+@@ -XXX,XX +XXX,XX @@ static void riscv_aplic_realize(DeviceState *dev, Error **errp)
+      * have IRQ lines delegated by their parent APLIC.
+      */
+     if (!aplic->parent) {
+-        if (is_kvm_aia(aplic->msimode)) {
++        if (kvm_enabled() && is_kvm_aia(aplic->msimode)) {
+             qdev_init_gpio_in(dev, riscv_kvm_aplic_request, aplic->num_irqs);
+         } else {
+             qdev_init_gpio_in(dev, riscv_aplic_request, aplic->num_irqs);
+diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/kvm.c
++++ b/target/riscv/kvm.c
+@@ -XXX,XX +XXX,XX @@
+ #include "sysemu/runstate.h"
+ #include "hw/riscv/numa.h"
++void riscv_kvm_aplic_request(void *opaque, int irq, int level)
++{
++    kvm_set_irq(kvm_state, irq, !!level);
++}
++
+ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type,
+                                  uint64_t idx)
+ {
+--
+.41.0

-New patch
+[PULL v2 40/45] linux-user/riscv: Add new extensions to hwprobe
+From: Robbin Ehn <rehn@rivosinc.com>
+This patch adds the new extensions in
+linux 6.5 to the hwprobe syscall.
+And fixes RVC check to OR with correct value.
+The previous variable contains 0 therefore it
+did work.
+Signed-off-by: Robbin Ehn <rehn@rivosinc.com>
+Acked-by: Richard Henderson <richard.henderson@linaro.org>
+Acked-by: Alistair Francis <alistair.francis@wdc.com>
+Message-ID: <bc82203b72d7efb30f1b4a8f9eb3d94699799dc8.camel@rivosinc.com>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ linux-user/syscall.c | 14 +++++++++++++-
+file changed, 13 insertions(+), 1 deletion(-)
+diff --git a/linux-user/syscall.c b/linux-user/syscall.c
+index XXXXXXX..XXXXXXX 100644
+--- a/linux-user/syscall.c
++++ b/linux-user/syscall.c
+@@ -XXX,XX +XXX,XX @@ static int do_getdents64(abi_long dirfd, abi_long arg2, abi_long count)
+ #define RISCV_HWPROBE_KEY_IMA_EXT_0     4
+ #define     RISCV_HWPROBE_IMA_FD       (1 << 0)
+ #define     RISCV_HWPROBE_IMA_C        (1 << 1)
++#define     RISCV_HWPROBE_IMA_V        (1 << 2)
++#define     RISCV_HWPROBE_EXT_ZBA      (1 << 3)
++#define     RISCV_HWPROBE_EXT_ZBB      (1 << 4)
++#define     RISCV_HWPROBE_EXT_ZBS      (1 << 5)
+ #define RISCV_HWPROBE_KEY_CPUPERF_0     5
+ #define     RISCV_HWPROBE_MISALIGNED_UNKNOWN     (0 << 0)
+@@ -XXX,XX +XXX,XX @@ static void risc_hwprobe_fill_pairs(CPURISCVState *env,
+                     riscv_has_ext(env, RVD) ?
+                     RISCV_HWPROBE_IMA_FD : 0;
+             value |= riscv_has_ext(env, RVC) ?
+-                     RISCV_HWPROBE_IMA_C : pair->value;
++                     RISCV_HWPROBE_IMA_C : 0;
++            value |= riscv_has_ext(env, RVV) ?
++                     RISCV_HWPROBE_IMA_V : 0;
++            value |= cfg->ext_zba ?
++                     RISCV_HWPROBE_EXT_ZBA : 0;
++            value |= cfg->ext_zbb ?
++                     RISCV_HWPROBE_EXT_ZBB : 0;
++            value |= cfg->ext_zbs ?
++                     RISCV_HWPROBE_EXT_ZBS : 0;
+             __put_user(value, &pair->value);
+             break;
+         case RISCV_HWPROBE_KEY_CPUPERF_0:
+--
+.41.0

-New patch
+[PULL v2 41/45] target/riscv: Use accelerated helper for AES64KS1I
+From: Ard Biesheuvel <ardb@kernel.org>
+Use the accelerated SubBytes/ShiftRows/AddRoundKey AES helper to
+implement the first half of the key schedule derivation. This does not
+actually involve shifting rows, so clone the same value into all four
+columns of the AES vector to counter that operation.
+Cc: Richard Henderson <richard.henderson@linaro.org>
+Cc: Philippe Mathieu-Daudé <philmd@linaro.org>
+Cc: Palmer Dabbelt <palmer@dabbelt.com>
+Cc: Alistair Francis <alistair.francis@wdc.com>
+Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Message-ID: <20230831154118.138727-1-ardb@kernel.org>
+Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
+---
+ target/riscv/crypto_helper.c | 17 +++++------------
+file changed, 5 insertions(+), 12 deletions(-)
+diff --git a/target/riscv/crypto_helper.c b/target/riscv/crypto_helper.c
+index XXXXXXX..XXXXXXX 100644
+--- a/target/riscv/crypto_helper.c
++++ b/target/riscv/crypto_helper.c
+@@ -XXX,XX +XXX,XX @@ target_ulong HELPER(aes64ks1i)(target_ulong rs1, target_ulong rnum)
+     uint8_t enc_rnum = rnum;
+     uint32_t temp = (RS1 >> 32) & 0xFFFFFFFF;
+-    uint8_t rcon_ = 0;
+-    target_ulong result;
++    AESState t, rc = {};
+     if (enc_rnum != 0xA) {
+         temp = ror32(temp, 8); /* Rotate right by 8 */
+-        rcon_ = round_consts[enc_rnum];
++        rc.w[0] = rc.w[1] = round_consts[enc_rnum];
+     }
+-    temp = ((uint32_t)AES_sbox[(temp >> 24) & 0xFF] << 24) |
+-           ((uint32_t)AES_sbox[(temp >> 16) & 0xFF] << 16) |
+-           ((uint32_t)AES_sbox[(temp >> 8) & 0xFF] << 8) |
+-           ((uint32_t)AES_sbox[(temp >> 0) & 0xFF] << 0);
++    t.w[0] = t.w[1] = t.w[2] = t.w[3] = temp;
++    aesenc_SB_SR_AK(&t, &t, &rc, false);
+-    temp ^= rcon_;
+-
+-    result = ((uint64_t)temp << 32) | temp;
+-
+-    return result;
++    return t.d[0];
+ }
+ target_ulong HELPER(aes64im)(target_ulong rs1)
+--
+.41.0

-[PULL 09/10] hw/intc: sifive_plic: change interrupt priority register to WARL field
+[PULL v2 42/45] target/riscv: Allocate itrigger timers only once
-From: Jim Shu <jim.shu@sifive.com>
+From: Akihiko Odaki <akihiko.odaki@daynix.com>
-PLIC spec [1] requires interrupt source priority registers are WARL
+riscv_trigger_init() had been called on reset events that can happen
-field and the number of supported priority is power-of-2 to simplify SW
+several times for a CPU and it allocated timers for itrigger. If old
-discovery.
+timers were present, they were simply overwritten by the new timers,
 resulting in a memory leak.
-Existing QEMU RISC-V machine (e.g. shakti_c) don't strictly follow PLIC
+Divide riscv_trigger_init() into two functions, namely
-spec, whose number of supported priority is not power-of-2. Just change
+riscv_trigger_realize() and riscv_trigger_reset() and call them in
-each bit of interrupt priority register to WARL field when the number of
+appropriate timing. The timer allocation will happen only once for a
-supported priority is power-of-2.
+CPU in riscv_trigger_realize().
-[1] https://github.com/riscv/riscv-plic-spec/blob/master/riscv-plic.adoc#interrupt-priorities
+Fixes: 5a4ae64cac ("target/riscv: Add itrigger support when icount is enabled")
+Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
-Signed-off-by: Jim Shu <jim.shu@sifive.com>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
-Reviewed-by: Clément Chigot <chigot@adacore.com>
+Reviewed-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
-Acked-by: Alistair Francis <alistair.francis@wdc.com>
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
-Message-Id: <20221003041440.2320-3-jim.shu@sifive.com>
+Message-ID: <20230818034059.9146-1-akihiko.odaki@daynix.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
- hw/intc/sifive_plic.c | 21 +++++++++++++++++++--
+ target/riscv/debug.h |  3 ++-
-file changed, 19 insertions(+), 2 deletions(-)
+ target/riscv/cpu.c   |  8 +++++++-
  target/riscv/debug.c | 15 ++++++++++++---
 files changed, 21 insertions(+), 5 deletions(-)
-diff --git a/hw/intc/sifive_plic.c b/hw/intc/sifive_plic.c
+diff --git a/target/riscv/debug.h b/target/riscv/debug.h
 index XXXXXXX..XXXXXXX 100644
---- a/hw/intc/sifive_plic.c
+--- a/target/riscv/debug.h
-+++ b/hw/intc/sifive_plic.c
++++ b/target/riscv/debug.h
-@@ -XXX,XX +XXX,XX @@ static void sifive_plic_write(void *opaque, hwaddr addr, uint64_t value,
+@@ -XXX,XX +XXX,XX @@ void riscv_cpu_debug_excp_handler(CPUState *cs);
-     if (addr_between(addr, plic->priority_base, plic->num_sources << 2)) {
+ bool riscv_cpu_debug_check_breakpoint(CPUState *cs);
-         uint32_t irq = ((addr - plic->priority_base) >> 2) + 1;
+ bool riscv_cpu_debug_check_watchpoint(CPUState *cs, CPUWatchpoint *wp);
--        if (value <= plic->num_priorities) {
+-void riscv_trigger_init(CPURISCVState *env);
-+        if (((plic->num_priorities + 1) & plic->num_priorities) == 0) {
++void riscv_trigger_realize(CPURISCVState *env);
-+            /*
++void riscv_trigger_reset_hold(CPURISCVState *env);
-+             * if "num_priorities + 1" is power-of-2, make each register bit of
-+             * interrupt priority WARL (Write-Any-Read-Legal). Just filter
+ bool riscv_itrigger_enabled(CPURISCVState *env);
-+             * out the access to unsupported priority bits.
+ void riscv_itrigger_update_priv(CPURISCVState *env);
-+             */
+diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
-+            plic->source_priority[irq] = value % (plic->num_priorities + 1);
+index XXXXXXX..XXXXXXX 100644
-+            sifive_plic_update(plic);
+--- a/target/riscv/cpu.c
-+        } else if (value <= plic->num_priorities) {
++++ b/target/riscv/cpu.c
-             plic->source_priority[irq] = value;
+@@ -XXX,XX +XXX,XX @@ static void riscv_cpu_reset_hold(Object *obj)
-             sifive_plic_update(plic);
-         }
+ #ifndef CONFIG_USER_ONLY
-@@ -XXX,XX +XXX,XX @@ static void sifive_plic_write(void *opaque, hwaddr addr, uint64_t value,
+     if (cpu->cfg.debug) {
-         uint32_t contextid = (addr & (plic->context_stride - 1));
+-        riscv_trigger_init(env);
++        riscv_trigger_reset_hold(env);
-         if (contextid == 0) {
+     }
--            if (value <= plic->num_priorities) {
-+            if (((plic->num_priorities + 1) & plic->num_priorities) == 0) {
+     if (kvm_enabled()) {
-+                /*
+@@ -XXX,XX +XXX,XX @@ static void riscv_cpu_realize(DeviceState *dev, Error **errp)
-+                 * if "num_priorities + 1" is power-of-2, each register bit of
-+                 * interrupt priority is WARL (Write-Any-Read-Legal). Just
+     riscv_cpu_register_gdb_regs_for_features(cs);
-+                 * filter out the access to unsupported priority bits.
-+                 */
++#ifndef CONFIG_USER_ONLY
-+                plic->target_priority[addrid] = value %
++    if (cpu->cfg.debug) {
-+                                                (plic->num_priorities + 1);
++        riscv_trigger_realize(&cpu->env);
-+                sifive_plic_update(plic);
++    }
-+            } else if (value <= plic->num_priorities) {
++#endif
-                 plic->target_priority[addrid] = value;
++
-                 sifive_plic_update(plic);
+     qemu_init_vcpu(cs);
-             }
+     cpu_reset(cs);
 diff --git a/target/riscv/debug.c b/target/riscv/debug.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/debug.c
 +++ b/target/riscv/debug.c
@@ -XXX,XX +XXX,XX @@ bool riscv_cpu_debug_check_watchpoint(CPUState *cs, CPUWatchpoint *wp)
      return false;
  }
 -void riscv_trigger_init(CPURISCVState *env)
 +void riscv_trigger_realize(CPURISCVState *env)
 +{
 +    int i;
 +
 +    for (i = 0; i < RV_MAX_TRIGGERS; i++) {
 +        env->itrigger_timer[i] = timer_new_ns(QEMU_CLOCK_VIRTUAL,
 +                                              riscv_itrigger_timer_cb, env);
 +    }
 +}
 +
 +void riscv_trigger_reset_hold(CPURISCVState *env)
  {
      target_ulong tdata1 = build_tdata1(env, TRIGGER_TYPE_AD_MATCH, 0, 0);
      int i;
@@ -XXX,XX +XXX,XX @@ void riscv_trigger_init(CPURISCVState *env)
          env->tdata3[i] = 0;
          env->cpu_breakpoint[i] = NULL;
          env->cpu_watchpoint[i] = NULL;
 -        env->itrigger_timer[i] = timer_new_ns(QEMU_CLOCK_VIRTUAL,
 -                                              riscv_itrigger_timer_cb, env);
 +        timer_del(env->itrigger_timer[i]);
      }
  }
 --
-.37.3
+.41.0

-[PULL 10/10] target/riscv: pmp: Fixup TLB size calculation
+[PULL v2 43/45] target/riscv/pmp.c: respect mseccfg.RLB for pmpaddrX changes
-From: Alistair Francis <alistair.francis@wdc.com>
+From: Leon Schuermann <leons@opentitan.org>
-Since commit 4047368938f6 "accel/tcg: Introduce tlb_set_page_full" we
+When the rule-lock bypass (RLB) bit is set in the mseccfg CSR, the PMP
-have been seeing this assert
+configuration lock bits must not apply. While this behavior is
 implemented for the pmpcfgX CSRs, this bit is not respected for
 changes to the pmpaddrX CSRs. This patch ensures that pmpaddrX CSR
 writes work even on locked regions when the global rule-lock bypass is
 enabled.
-    ../accel/tcg/cputlb.c:1294: tlb_set_page_with_attrs: Assertion `is_power_of_2(size)' failed.
+Signed-off-by: Leon Schuermann <leons@opentitan.org>
+Reviewed-by: Mayuresh Chitale <mchitale@ventanamicro.com>
-When running Tock on the OpenTitan machine.
+Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
+Message-ID: <20230829215046.1430463-1-leon@is.currently.online>
 The issue is that pmp_get_tlb_size() would return a TLB size that wasn't
 a power of 2. The size was also smaller then TARGET_PAGE_SIZE.
 This patch ensures that any TLB size less then TARGET_PAGE_SIZE is
 rounded down to 1 to ensure it's a valid size.
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
-Reviewed-by: LIU Zhiwei<zhiwei_liu@linux.alibaba.com>
-Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
-Message-id: 20221012011449.506928-1-alistair.francis@opensource.wdc.com
-Message-Id: <20221012011449.506928-1-alistair.francis@opensource.wdc.com>
 ---
- target/riscv/pmp.c | 12 ++++++++++++
+ target/riscv/pmp.c | 4 ++++
-file changed, 12 insertions(+)
+file changed, 4 insertions(+)
 diff --git a/target/riscv/pmp.c b/target/riscv/pmp.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/riscv/pmp.c
 +++ b/target/riscv/pmp.c
-@@ -XXX,XX +XXX,XX @@ bool pmp_is_range_in_tlb(CPURISCVState *env, hwaddr tlb_sa,
+@@ -XXX,XX +XXX,XX @@ static inline uint8_t pmp_get_a_field(uint8_t cfg)
-     }
+  */
+ static inline int pmp_is_locked(CPURISCVState *env, uint32_t pmp_index)
-     if (*tlb_size != 0) {
+ {
-+        /*
++    /* mseccfg.RLB is set */
-+         * At this point we have a tlb_size that is the smallest possible size
++    if (MSECCFG_RLB_ISSET(env)) {
-+         * That fits within a TARGET_PAGE_SIZE and the PMP region.
++        return 0;
-+         *
++    }
-+         * If the size is less then TARGET_PAGE_SIZE we drop the size to 1.
-+         * This means the result isn't cached in the TLB and is only used for
+     if (env->pmp_state.pmp[pmp_index].cfg_reg & PMP_LOCK) {
-+         * a single translation.
+         return 1;
 +         */
 +        if (*tlb_size < TARGET_PAGE_SIZE) {
 +            *tlb_size = 1;
 +        }
 +
          return true;
      }
 --
-.37.3
+.41.0

-[PULL 08/10] hw/intc: sifive_plic: fix hard-coded max priority level
+[PULL v2 44/45] target/riscv: Align the AIA model to v1.0 ratified spec
-From: Jim Shu <jim.shu@sifive.com>
+From: Tommy Wu <tommy.wu@sifive.com>
-The maximum priority level is hard-coded when writing to interrupt
+According to the new spec, when vsiselect has a reserved value, attempts
-priority register. However, when writing to priority threshold register,
+from M-mode or HS-mode to access vsireg, or from VS-mode to access
-the maximum priority level is from num_priorities Property which is
+sireg, should preferably raise an illegal instruction exception.
 configured by platform.
-Also change interrupt priority register to use num_priorities Property
+Signed-off-by: Tommy Wu <tommy.wu@sifive.com>
 in maximum priority level.
 Signed-off-by: Emmanuel Blot <emmanuel.blot@sifive.com>
 Signed-off-by: Jim Shu <jim.shu@sifive.com>
 Reviewed-by: Frank Chang <frank.chang@sifive.com>
-Acked-by: Alistair Francis <alistair.francis@wdc.com>
+Message-ID: <20230816061647.600672-1-tommy.wu@sifive.com>
 Message-Id: <20221003041440.2320-2-jim.shu@sifive.com>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
- hw/intc/sifive_plic.c | 6 ++++--
+ target/riscv/csr.c | 7 +++++--
-file changed, 4 insertions(+), 2 deletions(-)
+file changed, 5 insertions(+), 2 deletions(-)
-diff --git a/hw/intc/sifive_plic.c b/hw/intc/sifive_plic.c
+diff --git a/target/riscv/csr.c b/target/riscv/csr.c
 index XXXXXXX..XXXXXXX 100644
---- a/hw/intc/sifive_plic.c
+--- a/target/riscv/csr.c
-+++ b/hw/intc/sifive_plic.c
++++ b/target/riscv/csr.c
-@@ -XXX,XX +XXX,XX @@ static void sifive_plic_write(void *opaque, hwaddr addr, uint64_t value,
+@@ -XXX,XX +XXX,XX @@ static int rmw_iprio(target_ulong xlen,
-     if (addr_between(addr, plic->priority_base, plic->num_sources << 2)) {
+ static int rmw_xireg(CPURISCVState *env, int csrno, target_ulong *val,
-         uint32_t irq = ((addr - plic->priority_base) >> 2) + 1;
+                      target_ulong new_val, target_ulong wr_mask)
+ {
--        plic->source_priority[irq] = value & 7;
+-    bool virt;
--        sifive_plic_update(plic);
++    bool virt, isel_reserved;
-+        if (value <= plic->num_priorities) {
+     uint8_t *iprio;
-+            plic->source_priority[irq] = value;
+     int ret = -EINVAL;
-+            sifive_plic_update(plic);
+     target_ulong priv, isel, vgein;
-+        }
+@@ -XXX,XX +XXX,XX @@ static int rmw_xireg(CPURISCVState *env, int csrno, target_ulong *val,
-     } else if (addr_between(addr, plic->pending_base,
-                             plic->num_sources >> 3)) {
+     /* Decode register details from CSR number */
-         qemu_log_mask(LOG_GUEST_ERROR,
+     virt = false;
 +    isel_reserved = false;
      switch (csrno) {
      case CSR_MIREG:
          iprio = env->miprio;
@@ -XXX,XX +XXX,XX @@ static int rmw_xireg(CPURISCVState *env, int csrno, target_ulong *val,
                                                    riscv_cpu_mxl_bits(env)),
                                      val, new_val, wr_mask);
          }
 +    } else {
 +        isel_reserved = true;
      }
  done:
      if (ret) {
 -        return (env->virt_enabled && virt) ?
 +        return (env->virt_enabled && virt && !isel_reserved) ?
                 RISCV_EXCP_VIRT_INSTRUCTION_FAULT : RISCV_EXCP_ILLEGAL_INST;
      }
      return RISCV_EXCP_NONE;
 --
-.37.3
+.41.0

-[PULL 03/10] hw/ssi: ibex_spi: fixup/add rw1c functionality
+[PULL v2 45/45] target/riscv: don't read CSR in riscv_csrrw_do64
-From: Wilfred Mallawa <wilfred.mallawa@wdc.com>
+From: Nikita Shubin <n.shubin@yadro.com>
-This patch adds the `rw1c` functionality to the respective
+As per ISA:
 registers. The status fields are cleared when the respective
 field is set.
-Signed-off-by: Wilfred Mallawa <wilfred.mallawa@wdc.com>
+"For CSRRWI, if rd=x0, then the instruction shall not read the CSR and
 shall not cause any of the side effects that might occur on a CSR read."
 trans_csrrwi() and trans_csrrw() call do_csrw() if rd=x0, do_csrw() calls
 riscv_csrrw_do64(), via helper_csrw() passing NULL as *ret_value.
 Signed-off-by: Nikita Shubin <n.shubin@yadro.com>
 Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
-Message-Id: <20220930033241.206581-3-wilfred.mallawa@opensource.wdc.com>
+Message-ID: <20230808090914.17634-1-nikita.shubin@maquefel.me>
 Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
 ---
- include/hw/ssi/ibex_spi_host.h |  4 ++--
+ target/riscv/csr.c | 24 +++++++++++++++---------
- hw/ssi/ibex_spi_host.c         | 36 +++++++++++++++++++++++++++++++---
+file changed, 15 insertions(+), 9 deletions(-)
 files changed, 35 insertions(+), 5 deletions(-)
-diff --git a/include/hw/ssi/ibex_spi_host.h b/include/hw/ssi/ibex_spi_host.h
+diff --git a/target/riscv/csr.c b/target/riscv/csr.c
 index XXXXXXX..XXXXXXX 100644
---- a/include/hw/ssi/ibex_spi_host.h
+--- a/target/riscv/csr.c
-+++ b/include/hw/ssi/ibex_spi_host.h
++++ b/target/riscv/csr.c
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ static RISCVException riscv_csrrw_do64(CPURISCVState *env, int csrno,
-     OBJECT_CHECK(IbexSPIHostState, (obj), TYPE_IBEX_SPI_HOST)
+                                        target_ulong write_mask)
  /* SPI Registers */
 -#define IBEX_SPI_HOST_INTR_STATE         (0x00 / 4)  /* rw */
 +#define IBEX_SPI_HOST_INTR_STATE         (0x00 / 4)  /* rw1c */
  #define IBEX_SPI_HOST_INTR_ENABLE        (0x04 / 4)  /* rw */
  #define IBEX_SPI_HOST_INTR_TEST          (0x08 / 4)  /* wo */
  #define IBEX_SPI_HOST_ALERT_TEST         (0x0c / 4)  /* wo */
@@ -XXX,XX +XXX,XX @@
  #define IBEX_SPI_HOST_TXDATA             (0x28 / 4)
  #define IBEX_SPI_HOST_ERROR_ENABLE       (0x2c / 4)  /* rw */
 -#define IBEX_SPI_HOST_ERROR_STATUS       (0x30 / 4)  /* rw */
 +#define IBEX_SPI_HOST_ERROR_STATUS       (0x30 / 4)  /* rw1c */
  #define IBEX_SPI_HOST_EVENT_ENABLE       (0x34 / 4)  /* rw */
  /* FIFO Len in Bytes */
 diff --git a/hw/ssi/ibex_spi_host.c b/hw/ssi/ibex_spi_host.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/ssi/ibex_spi_host.c
 +++ b/hw/ssi/ibex_spi_host.c
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
  {
-     IbexSPIHostState *s = opaque;
+     RISCVException ret;
-     uint32_t val32 = val64;
+-    target_ulong old_value;
--    uint32_t shift_mask = 0xff, status = 0;
++    target_ulong old_value = 0;
-+    uint32_t shift_mask = 0xff, status = 0, data = 0;
-     uint8_t txqd_len;
+     /* execute combined read/write operation if it exists */
+     if (csr_ops[csrno].op) {
-     trace_ibex_spi_host_write(addr, size, val64);
+         return csr_ops[csrno].op(env, csrno, ret_value, new_value, write_mask);
-@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
+     }
-     switch (addr) {
+-    /* if no accessor exists then return failure */
-     /* Skipping any R/O registers */
+-    if (!csr_ops[csrno].read) {
--    case IBEX_SPI_HOST_INTR_STATE...IBEX_SPI_HOST_INTR_ENABLE:
+-        return RISCV_EXCP_ILLEGAL_INST;
-+    case IBEX_SPI_HOST_INTR_STATE:
+-    }
-+        /* rw1c status register */
+-    /* read old value */
-+        if (FIELD_EX32(val32, INTR_STATE, ERROR)) {
+-    ret = csr_ops[csrno].read(env, csrno, &old_value);
-+            data = FIELD_DP32(data, INTR_STATE, ERROR, 0);
+-    if (ret != RISCV_EXCP_NONE) {
 -        return ret;
 +    /*
 +     * ret_value == NULL means that rd=x0 and we're coming from helper_csrw()
 +     * and we can't throw side effects caused by CSR reads.
 +     */
 +    if (ret_value) {
 +        /* if no accessor exists then return failure */
 +        if (!csr_ops[csrno].read) {
 +            return RISCV_EXCP_ILLEGAL_INST;
 +        }
-+        if (FIELD_EX32(val32, INTR_STATE, SPI_EVENT)) {
++        /* read old value */
-+            data = FIELD_DP32(data, INTR_STATE, SPI_EVENT, 0);
++        ret = csr_ops[csrno].read(env, csrno, &old_value);
 +        if (ret != RISCV_EXCP_NONE) {
 +            return ret;
 +        }
-+        s->regs[addr] = data;
+     }
-+        break;
-+    case IBEX_SPI_HOST_INTR_ENABLE:
+     /* write value if writable and write mask set, otherwise drop writes */
          s->regs[addr] = val32;
          break;
      case IBEX_SPI_HOST_INTR_TEST:
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
       *  When an error occurs, the corresponding bit must be cleared
       *  here before issuing any further commands
       */
 -        s->regs[addr] = val32;
 +        status = s->regs[addr];
 +        /* rw1c status register */
 +        if (FIELD_EX32(val32, ERROR_STATUS, CMDBUSY)) {
 +            status = FIELD_DP32(status, ERROR_STATUS, CMDBUSY, 0);
 +        }
 +        if (FIELD_EX32(val32, ERROR_STATUS, OVERFLOW)) {
 +            status = FIELD_DP32(status, ERROR_STATUS, OVERFLOW, 0);
 +        }
 +        if (FIELD_EX32(val32, ERROR_STATUS, UNDERFLOW)) {
 +            status = FIELD_DP32(status, ERROR_STATUS, UNDERFLOW, 0);
 +        }
 +        if (FIELD_EX32(val32, ERROR_STATUS, CMDINVAL)) {
 +            status = FIELD_DP32(status, ERROR_STATUS, CMDINVAL, 0);
 +        }
 +        if (FIELD_EX32(val32, ERROR_STATUS, CSIDINVAL)) {
 +            status = FIELD_DP32(status, ERROR_STATUS, CSIDINVAL, 0);
 +        }
 +        if (FIELD_EX32(val32, ERROR_STATUS, ACCESSINVAL)) {
 +            status = FIELD_DP32(status, ERROR_STATUS, ACCESSINVAL, 0);
 +        }
 +        s->regs[addr] = status;
          break;
      case IBEX_SPI_HOST_EVENT_ENABLE:
      /* Controls which classes of SPI events raise an interrupt. */
 --
-.37.3
+.41.0

From: Alistair Francis <alistair.francis@wdc.com>

The following changes since commit 2ba341b3694cf3cff7b8a1df4cc765900d5c4f60:

Merge tag 'kraxel-20221013-pull-request' of https://gitlab.com/kraxel/qemu into staging (2022-10-13 13:55:53 -0400)

are available in the Git repository at:

https://github.com/alistair23/qemu.git tags/pull-riscv-to-apply-20221014

for you to fetch changes up to 47566421f029b0a489b63f8195b3ff944e017056:

target/riscv: pmp: Fixup TLB size calculation (2022-10-14 14:36:19 +1000)

----------------------------------------------------------------
Third RISC-V PR for QEMU 7.2

* Update qtest comment
* Fix coverity issue with Ibex SPI
* Move load_image_to_fw_cfg() to common location
* Enable booting S-mode firmware from pflash on virt machine
* Add disas support for vector instructions
* Priority level fixes for PLIC
* Fixup TLB size calculation when using PMP

----------------------------------------------------------------
Alistair Francis (1):
      target/riscv: pmp: Fixup TLB size calculation

Bin Meng (1):
      hw/riscv: Update comment for qtest check in riscv_find_firmware()

Jim Shu (2):
      hw/intc: sifive_plic: fix hard-coded max priority level
      hw/intc: sifive_plic: change interrupt priority register to WARL field

Sunil V L (3):
      hw/arm, loongarch: Move load_image_to_fw_cfg() to common location
      hw/riscv: virt: Move create_fw_cfg() prior to loading kernel
      hw/riscv: virt: Enable booting S-mode firmware from pflash

Wilfred Mallawa (2):
      hw/ssi: ibex_spi: fixup coverity issue
      hw/ssi: ibex_spi: fixup/add rw1c functionality

Yang Liu (1):
      disas/riscv.c: rvv: Add disas support for vector instructions

From: Bin Meng <bmeng.cn@gmail.com>

Since commit 4211fc553234 ("roms/opensbi: Remove ELF images"), the
comment for qtest check in riscv_find_firmware() is out of date.
Update it to reflect the latest status.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <BN7PR08MB435525C92550BAC5467BE672BF219@BN7PR08MB4355.namprd08.prod.outlook.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/riscv/boot.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/riscv/boot.c
+++ b/hw/riscv/boot.c
@@ -XXX,XX +XXX,XX @@ char *riscv_find_firmware(const char *firmware_filename)
     if (filename == NULL) {
         if (!qtest_enabled()) {
             /*
-             * We only ship plain binary bios images in the QEMU source.
-             * With Spike machine that uses ELF images as the default bios,
+             * We only ship OpenSBI binary bios images in the QEMU source.
+             * For machines that use images other than the default bios,
              * running QEMU test will complain hence let's suppress the error
              * report for QEMU testing.
              */
-- 
2.37.3

From: Wilfred Mallawa <wilfred.mallawa@wdc.com>

This patch addresses the coverity issues specified in [1],
as suggested, `FIELD_DP32()`/`FIELD_EX32()` macros have been
implemented to clean up the code.

[1] https://www.mail-archive.com/qemu-devel@nongnu.org/msg887713.html

Fixes: Coverity CID 1488107

Signed-off-by: Wilfred Mallawa <wilfred.mallawa@wdc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-Id: <20220930033241.206581-2-wilfred.mallawa@opensource.wdc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/ssi/ibex_spi_host.c | 132 +++++++++++++++++++++--------------------
 1 file changed, 68 insertions(+), 64 deletions(-)

diff --git a/hw/ssi/ibex_spi_host.c b/hw/ssi/ibex_spi_host.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/ssi/ibex_spi_host.c
+++ b/hw/ssi/ibex_spi_host.c
@@ -XXX,XX +XXX,XX @@ static inline uint8_t div4_round_up(uint8_t dividend)
 
 static void ibex_spi_rxfifo_reset(IbexSPIHostState *s)
 {
+    uint32_t data = s->regs[IBEX_SPI_HOST_STATUS];
     /* Empty the RX FIFO and assert RXEMPTY */
     fifo8_reset(&s->rx_fifo);
-    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_RXFULL_MASK;
-    s->regs[IBEX_SPI_HOST_STATUS] |= R_STATUS_RXEMPTY_MASK;
+    data = FIELD_DP32(data, STATUS, RXFULL, 0);
+    data = FIELD_DP32(data, STATUS, RXEMPTY, 1);
+    s->regs[IBEX_SPI_HOST_STATUS] = data;
 }
 
 static void ibex_spi_txfifo_reset(IbexSPIHostState *s)
 {
+    uint32_t data = s->regs[IBEX_SPI_HOST_STATUS];
     /* Empty the TX FIFO and assert TXEMPTY */
     fifo8_reset(&s->tx_fifo);
-    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXFULL_MASK;
-    s->regs[IBEX_SPI_HOST_STATUS] |= R_STATUS_TXEMPTY_MASK;
+    data = FIELD_DP32(data, STATUS, TXFULL, 0);
+    data = FIELD_DP32(data, STATUS, TXEMPTY, 1);
+    s->regs[IBEX_SPI_HOST_STATUS] = data;
 }
 
 static void ibex_spi_host_reset(DeviceState *dev)
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_reset(DeviceState *dev)
  */
 static void ibex_spi_host_irq(IbexSPIHostState *s)
 {
-    bool error_en = s->regs[IBEX_SPI_HOST_INTR_ENABLE]
-                    & R_INTR_ENABLE_ERROR_MASK;
-    bool event_en = s->regs[IBEX_SPI_HOST_INTR_ENABLE]
-                    & R_INTR_ENABLE_SPI_EVENT_MASK;
-    bool err_pending = s->regs[IBEX_SPI_HOST_INTR_STATE]
-                        & R_INTR_STATE_ERROR_MASK;
-    bool status_pending = s->regs[IBEX_SPI_HOST_INTR_STATE]
-                        & R_INTR_STATE_SPI_EVENT_MASK;
+    uint32_t intr_test_reg = s->regs[IBEX_SPI_HOST_INTR_TEST];
+    uint32_t intr_en_reg = s->regs[IBEX_SPI_HOST_INTR_ENABLE];
+    uint32_t intr_state_reg = s->regs[IBEX_SPI_HOST_INTR_STATE];
+
+    uint32_t err_en_reg = s->regs[IBEX_SPI_HOST_ERROR_ENABLE];
+    uint32_t event_en_reg = s->regs[IBEX_SPI_HOST_EVENT_ENABLE];
+    uint32_t err_status_reg = s->regs[IBEX_SPI_HOST_ERROR_STATUS];
+    uint32_t status_reg = s->regs[IBEX_SPI_HOST_STATUS];
+
+
+    bool error_en = FIELD_EX32(intr_en_reg, INTR_ENABLE, ERROR);
+    bool event_en = FIELD_EX32(intr_en_reg, INTR_ENABLE, SPI_EVENT);
+    bool err_pending = FIELD_EX32(intr_state_reg, INTR_STATE, ERROR);
+    bool status_pending = FIELD_EX32(intr_state_reg, INTR_STATE, SPI_EVENT);
+
     int err_irq = 0, event_irq = 0;
 
     /* Error IRQ enabled and Error IRQ Cleared */
     if (error_en && !err_pending) {
         /* Event enabled, Interrupt Test Error */
-        if (s->regs[IBEX_SPI_HOST_INTR_TEST] & R_INTR_TEST_ERROR_MASK) {
+        if (FIELD_EX32(intr_test_reg, INTR_TEST,  ERROR)) {
             err_irq = 1;
-        } else if ((s->regs[IBEX_SPI_HOST_ERROR_ENABLE]
-                    &  R_ERROR_ENABLE_CMDBUSY_MASK) &&
-                    s->regs[IBEX_SPI_HOST_ERROR_STATUS]
-                    & R_ERROR_STATUS_CMDBUSY_MASK) {
+        } else if (FIELD_EX32(err_en_reg, ERROR_ENABLE,  CMDBUSY) &&
+                   FIELD_EX32(err_status_reg, ERROR_STATUS,  CMDBUSY)) {
             /* Wrote to COMMAND when not READY */
             err_irq = 1;
-        } else if ((s->regs[IBEX_SPI_HOST_ERROR_ENABLE]
-                    &  R_ERROR_ENABLE_CMDINVAL_MASK) &&
-                    s->regs[IBEX_SPI_HOST_ERROR_STATUS]
-                    & R_ERROR_STATUS_CMDINVAL_MASK) {
+        } else if (FIELD_EX32(err_en_reg, ERROR_ENABLE,  CMDINVAL)  &&
+                   FIELD_EX32(err_status_reg, ERROR_STATUS,  CMDINVAL)) {
             /* Invalid command segment */
             err_irq = 1;
-        } else if ((s->regs[IBEX_SPI_HOST_ERROR_ENABLE]
-                    & R_ERROR_ENABLE_CSIDINVAL_MASK) &&
-                    s->regs[IBEX_SPI_HOST_ERROR_STATUS]
-                    & R_ERROR_STATUS_CSIDINVAL_MASK) {
+        } else if (FIELD_EX32(err_en_reg, ERROR_ENABLE,  CSIDINVAL) &&
+                   FIELD_EX32(err_status_reg, ERROR_STATUS,  CSIDINVAL)) {
             /* Invalid value for CSID */
             err_irq = 1;
         }
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_irq(IbexSPIHostState *s)
 
     /* Event IRQ Enabled and Event IRQ Cleared */
     if (event_en && !status_pending) {
-        if (s->regs[IBEX_SPI_HOST_INTR_TEST] & R_INTR_TEST_SPI_EVENT_MASK) {
+        if (FIELD_EX32(intr_test_reg, INTR_STATE,  SPI_EVENT)) {
             /* Event enabled, Interrupt Test Event */
             event_irq = 1;
-        } else if ((s->regs[IBEX_SPI_HOST_EVENT_ENABLE]
-                    & R_EVENT_ENABLE_READY_MASK) &&
-                    (s->regs[IBEX_SPI_HOST_STATUS] & R_STATUS_READY_MASK)) {
+        } else if (FIELD_EX32(event_en_reg, EVENT_ENABLE,  READY) &&
+                   FIELD_EX32(status_reg, STATUS, READY)) {
             /* SPI Host ready for next command */
             event_irq = 1;
-        } else if ((s->regs[IBEX_SPI_HOST_EVENT_ENABLE]
-                    & R_EVENT_ENABLE_TXEMPTY_MASK) &&
-                    (s->regs[IBEX_SPI_HOST_STATUS] & R_STATUS_TXEMPTY_MASK)) {
+        } else if (FIELD_EX32(event_en_reg, EVENT_ENABLE,  TXEMPTY) &&
+                   FIELD_EX32(status_reg, STATUS,  TXEMPTY)) {
             /* SPI TXEMPTY, TXFIFO drained */
             event_irq = 1;
-        } else if ((s->regs[IBEX_SPI_HOST_EVENT_ENABLE]
-                    & R_EVENT_ENABLE_RXFULL_MASK) &&
-                    (s->regs[IBEX_SPI_HOST_STATUS] & R_STATUS_RXFULL_MASK)) {
+        } else if (FIELD_EX32(event_en_reg, EVENT_ENABLE,  RXFULL) &&
+                   FIELD_EX32(status_reg, STATUS,  RXFULL)) {
             /* SPI RXFULL, RXFIFO  full */
             event_irq = 1;
         }
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_irq(IbexSPIHostState *s)
 
 static void ibex_spi_host_transfer(IbexSPIHostState *s)
 {
-    uint32_t rx, tx;
+    uint32_t rx, tx, data;
     /* Get num of one byte transfers */
-    uint8_t segment_len = ((s->regs[IBEX_SPI_HOST_COMMAND] & R_COMMAND_LEN_MASK)
-                          >> R_COMMAND_LEN_SHIFT);
+    uint8_t segment_len = FIELD_EX32(s->regs[IBEX_SPI_HOST_COMMAND],
+                                     COMMAND,  LEN);
+
     while (segment_len > 0) {
         if (fifo8_is_empty(&s->tx_fifo)) {
             /* Assert Stall */
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_transfer(IbexSPIHostState *s)
         --segment_len;
     }
 
+    data = s->regs[IBEX_SPI_HOST_STATUS];
     /* Assert Ready */
-    s->regs[IBEX_SPI_HOST_STATUS] |= R_STATUS_READY_MASK;
+    data = FIELD_DP32(data, STATUS, READY, 1);
     /* Set RXQD */
-    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_RXQD_MASK;
-    s->regs[IBEX_SPI_HOST_STATUS] |= (R_STATUS_RXQD_MASK
-                                    & div4_round_up(segment_len));
+    data = FIELD_DP32(data, STATUS, RXQD, div4_round_up(segment_len));
     /* Set TXQD */
-    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXQD_MASK;
-    s->regs[IBEX_SPI_HOST_STATUS] |= (fifo8_num_used(&s->tx_fifo) / 4)
-                                    & R_STATUS_TXQD_MASK;
+    data = FIELD_DP32(data, STATUS, TXQD, fifo8_num_used(&s->tx_fifo) / 4);
     /* Clear TXFULL */
-    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXFULL_MASK;
-    /* Assert TXEMPTY and drop remaining bytes that exceed segment_len */
-    ibex_spi_txfifo_reset(s);
+    data = FIELD_DP32(data, STATUS, TXFULL, 0);
     /* Reset RXEMPTY */
-    s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_RXEMPTY_MASK;
+    data = FIELD_DP32(data, STATUS, RXEMPTY, 0);
+    /* Update register status */
+    s->regs[IBEX_SPI_HOST_STATUS] = data;
+    /* Drop remaining bytes that exceed segment_len */
+    ibex_spi_txfifo_reset(s);
 
     ibex_spi_host_irq(s);
 }
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
 {
     IbexSPIHostState *s = opaque;
     uint32_t val32 = val64;
-    uint32_t shift_mask = 0xff;
+    uint32_t shift_mask = 0xff, status = 0;
     uint8_t txqd_len;
 
     trace_ibex_spi_host_write(addr, size, val64);
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
         s->regs[addr] = val32;
 
         /* STALL, IP not enabled */
-        if (!(s->regs[IBEX_SPI_HOST_CONTROL] & R_CONTROL_SPIEN_MASK)) {
+        if (!(FIELD_EX32(s->regs[IBEX_SPI_HOST_CONTROL],
+                         CONTROL, SPIEN))) {
             return;
         }
 
         /* SPI not ready, IRQ Error */
-        if (!(s->regs[IBEX_SPI_HOST_STATUS] & R_STATUS_READY_MASK)) {
+        if (!(FIELD_EX32(s->regs[IBEX_SPI_HOST_STATUS],
+                         STATUS, READY))) {
             s->regs[IBEX_SPI_HOST_ERROR_STATUS] |= R_ERROR_STATUS_CMDBUSY_MASK;
             ibex_spi_host_irq(s);
             return;
         }
+
         /* Assert Not Ready */
         s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_READY_MASK;
 
-        if (((val32 & R_COMMAND_DIRECTION_MASK) >> R_COMMAND_DIRECTION_SHIFT)
-            != BIDIRECTIONAL_TRANSFER) {
-                qemu_log_mask(LOG_UNIMP,
+        if (FIELD_EX32(val32, COMMAND, DIRECTION) != BIDIRECTIONAL_TRANSFER) {
+            qemu_log_mask(LOG_UNIMP,
                           "%s: Rx Only/Tx Only are not supported\n", __func__);
         }
 
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
                 return;
             }
             /* Byte ordering is set by the IP */
-            if ((s->regs[IBEX_SPI_HOST_STATUS] &
-                R_STATUS_BYTEORDER_MASK) == 0) {
+            status = s->regs[IBEX_SPI_HOST_STATUS];
+            if (FIELD_EX32(status, STATUS, BYTEORDER) == 0) {
                 /* LE: LSB transmitted first (default for ibex processor) */
                 shift_mask = 0xff << (i * 8);
             } else {
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
 
             fifo8_push(&s->tx_fifo, (val32 & shift_mask) >> (i * 8));
         }
-
+        status = s->regs[IBEX_SPI_HOST_STATUS];
         /* Reset TXEMPTY */
-        s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXEMPTY_MASK;
+        status = FIELD_DP32(status, STATUS, TXEMPTY, 0);
         /* Update TXQD */
-        txqd_len = (s->regs[IBEX_SPI_HOST_STATUS] &
-                    R_STATUS_TXQD_MASK) >> R_STATUS_TXQD_SHIFT;
+        txqd_len = FIELD_EX32(status, STATUS, TXQD);
         /* Partial bytes (size < 4) are padded, in words. */
         txqd_len += 1;
-        s->regs[IBEX_SPI_HOST_STATUS] &= ~R_STATUS_TXQD_MASK;
-        s->regs[IBEX_SPI_HOST_STATUS] |= txqd_len;
+        status = FIELD_DP32(status, STATUS, TXQD, txqd_len);
         /* Assert Ready */
-        s->regs[IBEX_SPI_HOST_STATUS] |= R_STATUS_READY_MASK;
+        status = FIELD_DP32(status, STATUS, READY, 1);
+        /* Update register status */
+        s->regs[IBEX_SPI_HOST_STATUS] = status;
         break;
     case IBEX_SPI_HOST_ERROR_ENABLE:
         s->regs[addr] = val32;
-- 
2.37.3

From: Wilfred Mallawa <wilfred.mallawa@wdc.com>

This patch adds the `rw1c` functionality to the respective
registers. The status fields are cleared when the respective
field is set.

Signed-off-by: Wilfred Mallawa <wilfred.mallawa@wdc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20220930033241.206581-3-wilfred.mallawa@opensource.wdc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 include/hw/ssi/ibex_spi_host.h |  4 ++--
 hw/ssi/ibex_spi_host.c         | 36 +++++++++++++++++++++++++++++++---
 2 files changed, 35 insertions(+), 5 deletions(-)

diff --git a/include/hw/ssi/ibex_spi_host.h b/include/hw/ssi/ibex_spi_host.h
index XXXXXXX..XXXXXXX 100644
--- a/include/hw/ssi/ibex_spi_host.h
+++ b/include/hw/ssi/ibex_spi_host.h
@@ -XXX,XX +XXX,XX @@
     OBJECT_CHECK(IbexSPIHostState, (obj), TYPE_IBEX_SPI_HOST)
 
 /* SPI Registers */
-#define IBEX_SPI_HOST_INTR_STATE         (0x00 / 4)  /* rw */
+#define IBEX_SPI_HOST_INTR_STATE         (0x00 / 4)  /* rw1c */
 #define IBEX_SPI_HOST_INTR_ENABLE        (0x04 / 4)  /* rw */
 #define IBEX_SPI_HOST_INTR_TEST          (0x08 / 4)  /* wo */
 #define IBEX_SPI_HOST_ALERT_TEST         (0x0c / 4)  /* wo */
@@ -XXX,XX +XXX,XX @@
 #define IBEX_SPI_HOST_TXDATA             (0x28 / 4)
 
 #define IBEX_SPI_HOST_ERROR_ENABLE       (0x2c / 4)  /* rw */
-#define IBEX_SPI_HOST_ERROR_STATUS       (0x30 / 4)  /* rw */
+#define IBEX_SPI_HOST_ERROR_STATUS       (0x30 / 4)  /* rw1c */
 #define IBEX_SPI_HOST_EVENT_ENABLE       (0x34 / 4)  /* rw */
 
 /* FIFO Len in Bytes */
diff --git a/hw/ssi/ibex_spi_host.c b/hw/ssi/ibex_spi_host.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/ssi/ibex_spi_host.c
+++ b/hw/ssi/ibex_spi_host.c
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
 {
     IbexSPIHostState *s = opaque;
     uint32_t val32 = val64;
-    uint32_t shift_mask = 0xff, status = 0;
+    uint32_t shift_mask = 0xff, status = 0, data = 0;
     uint8_t txqd_len;
 
     trace_ibex_spi_host_write(addr, size, val64);
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
 
     switch (addr) {
     /* Skipping any R/O registers */
-    case IBEX_SPI_HOST_INTR_STATE...IBEX_SPI_HOST_INTR_ENABLE:
+    case IBEX_SPI_HOST_INTR_STATE:
+        /* rw1c status register */
+        if (FIELD_EX32(val32, INTR_STATE, ERROR)) {
+            data = FIELD_DP32(data, INTR_STATE, ERROR, 0);
+        }
+        if (FIELD_EX32(val32, INTR_STATE, SPI_EVENT)) {
+            data = FIELD_DP32(data, INTR_STATE, SPI_EVENT, 0);
+        }
+        s->regs[addr] = data;
+        break;
+    case IBEX_SPI_HOST_INTR_ENABLE:
         s->regs[addr] = val32;
         break;
     case IBEX_SPI_HOST_INTR_TEST:
@@ -XXX,XX +XXX,XX @@ static void ibex_spi_host_write(void *opaque, hwaddr addr,
      *  When an error occurs, the corresponding bit must be cleared
      *  here before issuing any further commands
      */
-        s->regs[addr] = val32;
+        status = s->regs[addr];
+        /* rw1c status register */
+        if (FIELD_EX32(val32, ERROR_STATUS, CMDBUSY)) {
+            status = FIELD_DP32(status, ERROR_STATUS, CMDBUSY, 0);
+        }
+        if (FIELD_EX32(val32, ERROR_STATUS, OVERFLOW)) {
+            status = FIELD_DP32(status, ERROR_STATUS, OVERFLOW, 0);
+        }
+        if (FIELD_EX32(val32, ERROR_STATUS, UNDERFLOW)) {
+            status = FIELD_DP32(status, ERROR_STATUS, UNDERFLOW, 0);
+        }
+        if (FIELD_EX32(val32, ERROR_STATUS, CMDINVAL)) {
+            status = FIELD_DP32(status, ERROR_STATUS, CMDINVAL, 0);
+        }
+        if (FIELD_EX32(val32, ERROR_STATUS, CSIDINVAL)) {
+            status = FIELD_DP32(status, ERROR_STATUS, CSIDINVAL, 0);
+        }
+        if (FIELD_EX32(val32, ERROR_STATUS, ACCESSINVAL)) {
+            status = FIELD_DP32(status, ERROR_STATUS, ACCESSINVAL, 0);
+        }
+        s->regs[addr] = status;
         break;
     case IBEX_SPI_HOST_EVENT_ENABLE:
     /* Controls which classes of SPI events raise an interrupt. */
-- 
2.37.3

From: Sunil V L <sunilvl@ventanamicro.com>

load_image_to_fw_cfg() is duplicated by both arm and loongarch. The same
function will be required by riscv too. So, it's time to refactor and
move this function to a common path.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Song Gao <gaosong@loongson.cn>
Message-Id: <20221004092351.18209-2-sunilvl@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 include/hw/nvram/fw_cfg.h | 21 +++++++++++++++++
 hw/arm/boot.c             | 49 ---------------------------------------
 hw/loongarch/virt.c       | 33 --------------------------
 hw/nvram/fw_cfg.c         | 32 +++++++++++++++++++++++++
 4 files changed, 53 insertions(+), 82 deletions(-)

diff --git a/include/hw/nvram/fw_cfg.h b/include/hw/nvram/fw_cfg.h
index XXXXXXX..XXXXXXX 100644
--- a/include/hw/nvram/fw_cfg.h
+++ b/include/hw/nvram/fw_cfg.h
@@ -XXX,XX +XXX,XX @@ bool fw_cfg_dma_enabled(void *opaque);
  */
 const char *fw_cfg_arch_key_name(uint16_t key);
 
+/**
+ * load_image_to_fw_cfg() - Load an image file into an fw_cfg entry identified
+ *                          by key.
+ * @fw_cfg:         The firmware config instance to store the data in.
+ * @size_key:       The firmware config key to store the size of the loaded
+ *                  data under, with fw_cfg_add_i32().
+ * @data_key:       The firmware config key to store the loaded data under,
+ *                  with fw_cfg_add_bytes().
+ * @image_name:     The name of the image file to load. If it is NULL, the
+ *                  function returns without doing anything.
+ * @try_decompress: Whether the image should be decompressed (gunzipped) before
+ *                  adding it to fw_cfg. If decompression fails, the image is
+ *                  loaded as-is.
+ *
+ * In case of failure, the function prints an error message to stderr and the
+ * process exits with status 1.
+ */
+void load_image_to_fw_cfg(FWCfgState *fw_cfg, uint16_t size_key,
+                          uint16_t data_key, const char *image_name,
+                          bool try_decompress);
+
 #endif
diff --git a/hw/arm/boot.c b/hw/arm/boot.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/arm/boot.c
+++ b/hw/arm/boot.c
@@ -XXX,XX +XXX,XX @@ static void do_cpu_reset(void *opaque)
     }
 }
 
-/**
- * load_image_to_fw_cfg() - Load an image file into an fw_cfg entry identified
- *                          by key.
- * @fw_cfg:         The firmware config instance to store the data in.
- * @size_key:       The firmware config key to store the size of the loaded
- *                  data under, with fw_cfg_add_i32().
- * @data_key:       The firmware config key to store the loaded data under,
- *                  with fw_cfg_add_bytes().
- * @image_name:     The name of the image file to load. If it is NULL, the
- *                  function returns without doing anything.
- * @try_decompress: Whether the image should be decompressed (gunzipped) before
- *                  adding it to fw_cfg. If decompression fails, the image is
- *                  loaded as-is.
- *
- * In case of failure, the function prints an error message to stderr and the
- * process exits with status 1.
- */
-static void load_image_to_fw_cfg(FWCfgState *fw_cfg, uint16_t size_key,
-                                 uint16_t data_key, const char *image_name,
-                                 bool try_decompress)
-{
-    size_t size = -1;
-    uint8_t *data;
-
-    if (image_name == NULL) {
-        return;
-    }
-
-    if (try_decompress) {
-        size = load_image_gzipped_buffer(image_name,
-                                         LOAD_IMAGE_MAX_GUNZIP_BYTES, &data);
-    }
-
-    if (size == (size_t)-1) {
-        gchar *contents;
-        gsize length;
-
-        if (!g_file_get_contents(image_name, &contents, &length, NULL)) {
-            error_report("failed to load \"%s\"", image_name);
-            exit(1);
-        }
-        size = length;
-        data = (uint8_t *)contents;
-    }
-
-    fw_cfg_add_i32(fw_cfg, size_key, size);
-    fw_cfg_add_bytes(fw_cfg, data_key, data, size);
-}
-
 static int do_arm_linux_init(Object *obj, void *opaque)
 {
     if (object_dynamic_cast(obj, TYPE_ARM_LINUX_BOOT_IF)) {
diff --git a/hw/loongarch/virt.c b/hw/loongarch/virt.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/loongarch/virt.c
+++ b/hw/loongarch/virt.c
@@ -XXX,XX +XXX,XX @@ static void reset_load_elf(void *opaque)
     }
 }
 
-/* Load an image file into an fw_cfg entry identified by key. */
-static void load_image_to_fw_cfg(FWCfgState *fw_cfg, uint16_t size_key,
-                                 uint16_t data_key, const char *image_name,
-                                 bool try_decompress)
-{
-    size_t size = -1;
-    uint8_t *data;
-
-    if (image_name == NULL) {
-        return;
-    }
-
-    if (try_decompress) {
-        size = load_image_gzipped_buffer(image_name,
-                                         LOAD_IMAGE_MAX_GUNZIP_BYTES, &data);
-    }
-
-    if (size == (size_t)-1) {
-        gchar *contents;
-        gsize length;
-
-        if (!g_file_get_contents(image_name, &contents, &length, NULL)) {
-            error_report("failed to load \"%s\"", image_name);
-            exit(1);
-        }
-        size = length;
-        data = (uint8_t *)contents;
-    }
-
-    fw_cfg_add_i32(fw_cfg, size_key, size);
-    fw_cfg_add_bytes(fw_cfg, data_key, data, size);
-}
-
 static void fw_cfg_add_kernel_info(FWCfgState *fw_cfg)
 {
     /*
diff --git a/hw/nvram/fw_cfg.c b/hw/nvram/fw_cfg.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/nvram/fw_cfg.c
+++ b/hw/nvram/fw_cfg.c
@@ -XXX,XX +XXX,XX @@
 #include "qapi/error.h"
 #include "hw/acpi/aml-build.h"
 #include "hw/pci/pci_bus.h"
+#include "hw/loader.h"
 
 #define FW_CFG_FILE_SLOTS_DFLT 0x20
 
@@ -XXX,XX +XXX,XX @@ FWCfgState *fw_cfg_find(void)
     return FW_CFG(object_resolve_path_type("", TYPE_FW_CFG, NULL));
 }
 
+void load_image_to_fw_cfg(FWCfgState *fw_cfg, uint16_t size_key,
+                          uint16_t data_key, const char *image_name,
+                          bool try_decompress)
+{
+    size_t size = -1;
+    uint8_t *data;
+
+    if (image_name == NULL) {
+        return;
+    }
+
+    if (try_decompress) {
+        size = load_image_gzipped_buffer(image_name,
+                                         LOAD_IMAGE_MAX_GUNZIP_BYTES, &data);
+    }
+
+    if (size == (size_t)-1) {
+        gchar *contents;
+        gsize length;
+
+        if (!g_file_get_contents(image_name, &contents, &length, NULL)) {
+            error_report("failed to load \"%s\"", image_name);
+            exit(1);
+        }
+        size = length;
+        data = (uint8_t *)contents;
+    }
+
+    fw_cfg_add_i32(fw_cfg, size_key, size);
+    fw_cfg_add_bytes(fw_cfg, data_key, data, size);
+}
 
 static void fw_cfg_class_init(ObjectClass *klass, void *data)
 {
-- 
2.37.3

From: Sunil V L <sunilvl@ventanamicro.com>

To enable both -kernel and -pflash options, the fw_cfg needs to be
created prior to loading the kernel.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20221004092351.18209-3-sunilvl@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/riscv/virt.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -XXX,XX +XXX,XX @@ static void virt_machine_done(Notifier *notifier, void *data)
                                     RISCV64_BIOS_BIN, start_addr, NULL);
     }
 
+    /*
+     * Init fw_cfg.  Must be done before riscv_load_fdt, otherwise the device
+     * tree cannot be altered and we get FDT_ERR_NOSPACE.
+     */
+    s->fw_cfg = create_fw_cfg(machine);
+    rom_set_fw(s->fw_cfg);
+
     if (machine->kernel_filename) {
         kernel_start_addr = riscv_calc_kernel_start_addr(&s->soc[0],
                                                          firmware_end_addr);
@@ -XXX,XX +XXX,XX @@ static void virt_machine_done(Notifier *notifier, void *data)
         start_addr = virt_memmap[VIRT_FLASH].base;
     }
 
-    /*
-     * Init fw_cfg.  Must be done before riscv_load_fdt, otherwise the device
-     * tree cannot be altered and we get FDT_ERR_NOSPACE.
-     */
-    s->fw_cfg = create_fw_cfg(machine);
-    rom_set_fw(s->fw_cfg);
-
     /* Compute the fdt load address in dram */
     fdt_load_addr = riscv_load_fdt(memmap[VIRT_DRAM].base,
                                    machine->ram_size, machine->fdt);
-- 
2.37.3

From: Sunil V L <sunilvl@ventanamicro.com>

To boot S-mode firmware payload like EDK2 from persistent
flash storage, qemu needs to pass the flash address as the
next_addr in fw_dynamic_info to the opensbi.

When both -kernel and -pflash options are provided in command line,
the kernel (and initrd if -initrd) will be copied to fw_cfg table.
The S-mode FW will load the kernel/initrd from fw_cfg table.

If only pflash is given but not -kernel, then it is the job of
of the S-mode firmware to locate and load the kernel.

In either case, update the kernel_entry with the flash address
so that the opensbi can jump to the entry point of the S-mode
firmware.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20221004092351.18209-4-sunilvl@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 include/hw/riscv/boot.h |  1 +
 hw/riscv/boot.c         | 29 +++++++++++++++++++++++++++++
 hw/riscv/virt.c         | 18 +++++++++++++++++-
 3 files changed, 47 insertions(+), 1 deletion(-)

diff --git a/include/hw/riscv/boot.h b/include/hw/riscv/boot.h
index XXXXXXX..XXXXXXX 100644
--- a/include/hw/riscv/boot.h
+++ b/include/hw/riscv/boot.h
@@ -XXX,XX +XXX,XX @@ void riscv_rom_copy_firmware_info(MachineState *machine, hwaddr rom_base,
                                   uint32_t reset_vec_size,
                                   uint64_t kernel_entry);
 void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr);
+void riscv_setup_firmware_boot(MachineState *machine);
 
 #endif /* RISCV_BOOT_H */
diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/riscv/boot.c
+++ b/hw/riscv/boot.c
@@ -XXX,XX +XXX,XX @@ void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr)
         riscv_cpu->env.fdt_addr = fdt_addr;
     }
 }
+
+void riscv_setup_firmware_boot(MachineState *machine)
+{
+    if (machine->kernel_filename) {
+        FWCfgState *fw_cfg;
+        fw_cfg = fw_cfg_find();
+
+        assert(fw_cfg);
+        /*
+         * Expose the kernel, the command line, and the initrd in fw_cfg.
+         * We don't process them here at all, it's all left to the
+         * firmware.
+         */
+        load_image_to_fw_cfg(fw_cfg,
+                             FW_CFG_KERNEL_SIZE, FW_CFG_KERNEL_DATA,
+                             machine->kernel_filename,
+                             true);
+        load_image_to_fw_cfg(fw_cfg,
+                             FW_CFG_INITRD_SIZE, FW_CFG_INITRD_DATA,
+                             machine->initrd_filename, false);
+
+        if (machine->kernel_cmdline) {
+            fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_SIZE,
+                           strlen(machine->kernel_cmdline) + 1);
+            fw_cfg_add_string(fw_cfg, FW_CFG_CMDLINE_DATA,
+                              machine->kernel_cmdline);
+        }
+    }
+}
diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -XXX,XX +XXX,XX @@ static void virt_machine_done(Notifier *notifier, void *data)
     s->fw_cfg = create_fw_cfg(machine);
     rom_set_fw(s->fw_cfg);
 
-    if (machine->kernel_filename) {
+    if (drive_get(IF_PFLASH, 0, 1)) {
+        /*
+         * S-mode FW like EDK2 will be kept in second plash (unit 1).
+         * When both kernel, initrd and pflash options are provided in the
+         * command line, the kernel and initrd will be copied to the fw_cfg
+         * table and opensbi will jump to the flash address which is the
+         * entry point of S-mode FW. It is the job of the S-mode FW to load
+         * the kernel and initrd using fw_cfg table.
+         *
+         * If only pflash is given but not -kernel, then it is the job of
+         * of the S-mode firmware to locate and load the kernel.
+         * In either case, the next_addr for opensbi will be the flash address.
+         */
+        riscv_setup_firmware_boot(machine);
+        kernel_entry = virt_memmap[VIRT_FLASH].base +
+                       virt_memmap[VIRT_FLASH].size / 2;
+    } else if (machine->kernel_filename) {
         kernel_start_addr = riscv_calc_kernel_start_addr(&s->soc[0],
                                                          firmware_end_addr);
 
-- 
2.37.3

From: Yang Liu <liuyang22@iscas.ac.cn>

Tested with https://github.com/ksco/rvv-decoder-tests

Expected checkpatch errors for consistency and brevity reasons:

ERROR: line over 90 characters
ERROR: trailing statements should be on next line
ERROR: braces {} are necessary for all arms of this statement

Signed-off-by: Yang Liu <liuyang22@iscas.ac.cn>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20220928051842.16207-1-liuyang22@iscas.ac.cn>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 disas/riscv.c | 1432 ++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 1430 insertions(+), 2 deletions(-)

diff --git a/disas/riscv.c b/disas/riscv.c
index XXXXXXX..XXXXXXX 100644
--- a/disas/riscv.c
+++ b/disas/riscv.c
@@ -XXX,XX +XXX,XX @@ typedef enum {
     rv_codec_css_sqsp,
     rv_codec_k_bs,
     rv_codec_k_rnum,
+    rv_codec_v_r,
+    rv_codec_v_ldst,
+    rv_codec_v_i,
+    rv_codec_vsetvli,
+    rv_codec_vsetivli,
 } rv_codec;
 
 typedef enum {
@@ -XXX,XX +XXX,XX @@ typedef enum {
     rv_op_zip = 396,
     rv_op_xperm4 = 397,
     rv_op_xperm8 = 398,
+    rv_op_vle8_v = 399,
+    rv_op_vle16_v = 400,
+    rv_op_vle32_v = 401,
+    rv_op_vle64_v = 402,
+    rv_op_vse8_v = 403,
+    rv_op_vse16_v = 404,
+    rv_op_vse32_v = 405,
+    rv_op_vse64_v = 406,
+    rv_op_vlm_v = 407,
+    rv_op_vsm_v = 408,
+    rv_op_vlse8_v = 409,
+    rv_op_vlse16_v = 410,
+    rv_op_vlse32_v = 411,
+    rv_op_vlse64_v = 412,
+    rv_op_vsse8_v = 413,
+    rv_op_vsse16_v = 414,
+    rv_op_vsse32_v = 415,
+    rv_op_vsse64_v = 416,
+    rv_op_vluxei8_v = 417,
+    rv_op_vluxei16_v = 418,
+    rv_op_vluxei32_v = 419,
+    rv_op_vluxei64_v = 420,
+    rv_op_vloxei8_v = 421,
+    rv_op_vloxei16_v = 422,
+    rv_op_vloxei32_v = 423,
+    rv_op_vloxei64_v = 424,
+    rv_op_vsuxei8_v = 425,
+    rv_op_vsuxei16_v = 426,
+    rv_op_vsuxei32_v = 427,
+    rv_op_vsuxei64_v = 428,
+    rv_op_vsoxei8_v = 429,
+    rv_op_vsoxei16_v = 430,
+    rv_op_vsoxei32_v = 431,
+    rv_op_vsoxei64_v = 432,
+    rv_op_vle8ff_v = 433,
+    rv_op_vle16ff_v = 434,
+    rv_op_vle32ff_v = 435,
+    rv_op_vle64ff_v = 436,
+    rv_op_vl1re8_v = 437,
+    rv_op_vl1re16_v = 438,
+    rv_op_vl1re32_v = 439,
+    rv_op_vl1re64_v = 440,
+    rv_op_vl2re8_v = 441,
+    rv_op_vl2re16_v = 442,
+    rv_op_vl2re32_v = 443,
+    rv_op_vl2re64_v = 444,
+    rv_op_vl4re8_v = 445,
+    rv_op_vl4re16_v = 446,
+    rv_op_vl4re32_v = 447,
+    rv_op_vl4re64_v = 448,
+    rv_op_vl8re8_v = 449,
+    rv_op_vl8re16_v = 450,
+    rv_op_vl8re32_v = 451,
+    rv_op_vl8re64_v = 452,
+    rv_op_vs1r_v = 453,
+    rv_op_vs2r_v = 454,
+    rv_op_vs4r_v = 455,
+    rv_op_vs8r_v = 456,
+    rv_op_vadd_vv = 457,
+    rv_op_vadd_vx = 458,
+    rv_op_vadd_vi = 459,
+    rv_op_vsub_vv = 460,
+    rv_op_vsub_vx = 461,
+    rv_op_vrsub_vx = 462,
+    rv_op_vrsub_vi = 463,
+    rv_op_vwaddu_vv = 464,
+    rv_op_vwaddu_vx = 465,
+    rv_op_vwadd_vv = 466,
+    rv_op_vwadd_vx = 467,
+    rv_op_vwsubu_vv = 468,
+    rv_op_vwsubu_vx = 469,
+    rv_op_vwsub_vv = 470,
+    rv_op_vwsub_vx = 471,
+    rv_op_vwaddu_wv = 472,
+    rv_op_vwaddu_wx = 473,
+    rv_op_vwadd_wv = 474,
+    rv_op_vwadd_wx = 475,
+    rv_op_vwsubu_wv = 476,
+    rv_op_vwsubu_wx = 477,
+    rv_op_vwsub_wv = 478,
+    rv_op_vwsub_wx = 479,
+    rv_op_vadc_vvm = 480,
+    rv_op_vadc_vxm = 481,
+    rv_op_vadc_vim = 482,
+    rv_op_vmadc_vvm = 483,
+    rv_op_vmadc_vxm = 484,
+    rv_op_vmadc_vim = 485,
+    rv_op_vsbc_vvm = 486,
+    rv_op_vsbc_vxm = 487,
+    rv_op_vmsbc_vvm = 488,
+    rv_op_vmsbc_vxm = 489,
+    rv_op_vand_vv = 490,
+    rv_op_vand_vx = 491,
+    rv_op_vand_vi = 492,
+    rv_op_vor_vv = 493,
+    rv_op_vor_vx = 494,
+    rv_op_vor_vi = 495,
+    rv_op_vxor_vv = 496,
+    rv_op_vxor_vx = 497,
+    rv_op_vxor_vi = 498,
+    rv_op_vsll_vv = 499,
+    rv_op_vsll_vx = 500,
+    rv_op_vsll_vi = 501,
+    rv_op_vsrl_vv = 502,
+    rv_op_vsrl_vx = 503,
+    rv_op_vsrl_vi = 504,
+    rv_op_vsra_vv = 505,
+    rv_op_vsra_vx = 506,
+    rv_op_vsra_vi = 507,
+    rv_op_vnsrl_wv = 508,
+    rv_op_vnsrl_wx = 509,
+    rv_op_vnsrl_wi = 510,
+    rv_op_vnsra_wv = 511,
+    rv_op_vnsra_wx = 512,
+    rv_op_vnsra_wi = 513,
+    rv_op_vmseq_vv = 514,
+    rv_op_vmseq_vx = 515,
+    rv_op_vmseq_vi = 516,
+    rv_op_vmsne_vv = 517,
+    rv_op_vmsne_vx = 518,
+    rv_op_vmsne_vi = 519,
+    rv_op_vmsltu_vv = 520,
+    rv_op_vmsltu_vx = 521,
+    rv_op_vmslt_vv = 522,
+    rv_op_vmslt_vx = 523,
+    rv_op_vmsleu_vv = 524,
+    rv_op_vmsleu_vx = 525,
+    rv_op_vmsleu_vi = 526,
+    rv_op_vmsle_vv = 527,
+    rv_op_vmsle_vx = 528,
+    rv_op_vmsle_vi = 529,
+    rv_op_vmsgtu_vx = 530,
+    rv_op_vmsgtu_vi = 531,
+    rv_op_vmsgt_vx = 532,
+    rv_op_vmsgt_vi = 533,
+    rv_op_vminu_vv = 534,
+    rv_op_vminu_vx = 535,
+    rv_op_vmin_vv = 536,
+    rv_op_vmin_vx = 537,
+    rv_op_vmaxu_vv = 538,
+    rv_op_vmaxu_vx = 539,
+    rv_op_vmax_vv = 540,
+    rv_op_vmax_vx = 541,
+    rv_op_vmul_vv = 542,
+    rv_op_vmul_vx = 543,
+    rv_op_vmulh_vv = 544,
+    rv_op_vmulh_vx = 545,
+    rv_op_vmulhu_vv = 546,
+    rv_op_vmulhu_vx = 547,
+    rv_op_vmulhsu_vv = 548,
+    rv_op_vmulhsu_vx = 549,
+    rv_op_vdivu_vv = 550,
+    rv_op_vdivu_vx = 551,
+    rv_op_vdiv_vv = 552,
+    rv_op_vdiv_vx = 553,
+    rv_op_vremu_vv = 554,
+    rv_op_vremu_vx = 555,
+    rv_op_vrem_vv = 556,
+    rv_op_vrem_vx = 557,
+    rv_op_vwmulu_vv = 558,
+    rv_op_vwmulu_vx = 559,
+    rv_op_vwmulsu_vv = 560,
+    rv_op_vwmulsu_vx = 561,
+    rv_op_vwmul_vv = 562,
+    rv_op_vwmul_vx = 563,
+    rv_op_vmacc_vv = 564,
+    rv_op_vmacc_vx = 565,
+    rv_op_vnmsac_vv = 566,
+    rv_op_vnmsac_vx = 567,
+    rv_op_vmadd_vv = 568,
+    rv_op_vmadd_vx = 569,
+    rv_op_vnmsub_vv = 570,
+    rv_op_vnmsub_vx = 571,
+    rv_op_vwmaccu_vv = 572,
+    rv_op_vwmaccu_vx = 573,
+    rv_op_vwmacc_vv = 574,
+    rv_op_vwmacc_vx = 575,
+    rv_op_vwmaccsu_vv = 576,
+    rv_op_vwmaccsu_vx = 577,
+    rv_op_vwmaccus_vx = 578,
+    rv_op_vmv_v_v = 579,
+    rv_op_vmv_v_x = 580,
+    rv_op_vmv_v_i = 581,
+    rv_op_vmerge_vvm = 582,
+    rv_op_vmerge_vxm = 583,
+    rv_op_vmerge_vim = 584,
+    rv_op_vsaddu_vv = 585,
+    rv_op_vsaddu_vx = 586,
+    rv_op_vsaddu_vi = 587,
+    rv_op_vsadd_vv = 588,
+    rv_op_vsadd_vx = 589,
+    rv_op_vsadd_vi = 590,
+    rv_op_vssubu_vv = 591,
+    rv_op_vssubu_vx = 592,
+    rv_op_vssub_vv = 593,
+    rv_op_vssub_vx = 594,
+    rv_op_vaadd_vv = 595,
+    rv_op_vaadd_vx = 596,
+    rv_op_vaaddu_vv = 597,
+    rv_op_vaaddu_vx = 598,
+    rv_op_vasub_vv = 599,
+    rv_op_vasub_vx = 600,
+    rv_op_vasubu_vv = 601,
+    rv_op_vasubu_vx = 602,
+    rv_op_vsmul_vv = 603,
+    rv_op_vsmul_vx = 604,
+    rv_op_vssrl_vv = 605,
+    rv_op_vssrl_vx = 606,
+    rv_op_vssrl_vi = 607,
+    rv_op_vssra_vv = 608,
+    rv_op_vssra_vx = 609,
+    rv_op_vssra_vi = 610,
+    rv_op_vnclipu_wv = 611,
+    rv_op_vnclipu_wx = 612,
+    rv_op_vnclipu_wi = 613,
+    rv_op_vnclip_wv = 614,
+    rv_op_vnclip_wx = 615,
+    rv_op_vnclip_wi = 616,
+    rv_op_vfadd_vv = 617,
+    rv_op_vfadd_vf = 618,
+    rv_op_vfsub_vv = 619,
+    rv_op_vfsub_vf = 620,
+    rv_op_vfrsub_vf = 621,
+    rv_op_vfwadd_vv = 622,
+    rv_op_vfwadd_vf = 623,
+    rv_op_vfwadd_wv = 624,
+    rv_op_vfwadd_wf = 625,
+    rv_op_vfwsub_vv = 626,
+    rv_op_vfwsub_vf = 627,
+    rv_op_vfwsub_wv = 628,
+    rv_op_vfwsub_wf = 629,
+    rv_op_vfmul_vv = 630,
+    rv_op_vfmul_vf = 631,
+    rv_op_vfdiv_vv = 632,
+    rv_op_vfdiv_vf = 633,
+    rv_op_vfrdiv_vf = 634,
+    rv_op_vfwmul_vv = 635,
+    rv_op_vfwmul_vf = 636,
+    rv_op_vfmacc_vv = 637,
+    rv_op_vfmacc_vf = 638,
+    rv_op_vfnmacc_vv = 639,
+    rv_op_vfnmacc_vf = 640,
+    rv_op_vfmsac_vv = 641,
+    rv_op_vfmsac_vf = 642,
+    rv_op_vfnmsac_vv = 643,
+    rv_op_vfnmsac_vf = 644,
+    rv_op_vfmadd_vv = 645,
+    rv_op_vfmadd_vf = 646,
+    rv_op_vfnmadd_vv = 647,
+    rv_op_vfnmadd_vf = 648,
+    rv_op_vfmsub_vv = 649,
+    rv_op_vfmsub_vf = 650,
+    rv_op_vfnmsub_vv = 651,
+    rv_op_vfnmsub_vf = 652,
+    rv_op_vfwmacc_vv = 653,
+    rv_op_vfwmacc_vf = 654,
+    rv_op_vfwnmacc_vv = 655,
+    rv_op_vfwnmacc_vf = 656,
+    rv_op_vfwmsac_vv = 657,
+    rv_op_vfwmsac_vf = 658,
+    rv_op_vfwnmsac_vv = 659,
+    rv_op_vfwnmsac_vf = 660,
+    rv_op_vfsqrt_v = 661,
+    rv_op_vfrsqrt7_v = 662,
+    rv_op_vfrec7_v = 663,
+    rv_op_vfmin_vv = 664,
+    rv_op_vfmin_vf = 665,
+    rv_op_vfmax_vv = 666,
+    rv_op_vfmax_vf = 667,
+    rv_op_vfsgnj_vv = 668,
+    rv_op_vfsgnj_vf = 669,
+    rv_op_vfsgnjn_vv = 670,
+    rv_op_vfsgnjn_vf = 671,
+    rv_op_vfsgnjx_vv = 672,
+    rv_op_vfsgnjx_vf = 673,
+    rv_op_vfslide1up_vf = 674,
+    rv_op_vfslide1down_vf = 675,
+    rv_op_vmfeq_vv = 676,
+    rv_op_vmfeq_vf = 677,
+    rv_op_vmfne_vv = 678,
+    rv_op_vmfne_vf = 679,
+    rv_op_vmflt_vv = 680,
+    rv_op_vmflt_vf = 681,
+    rv_op_vmfle_vv = 682,
+    rv_op_vmfle_vf = 683,
+    rv_op_vmfgt_vf = 684,
+    rv_op_vmfge_vf = 685,
+    rv_op_vfclass_v = 686,
+    rv_op_vfmerge_vfm = 687,
+    rv_op_vfmv_v_f = 688,
+    rv_op_vfcvt_xu_f_v = 689,
+    rv_op_vfcvt_x_f_v = 690,
+    rv_op_vfcvt_f_xu_v = 691,
+    rv_op_vfcvt_f_x_v = 692,
+    rv_op_vfcvt_rtz_xu_f_v = 693,
+    rv_op_vfcvt_rtz_x_f_v = 694,
+    rv_op_vfwcvt_xu_f_v = 695,
+    rv_op_vfwcvt_x_f_v = 696,
+    rv_op_vfwcvt_f_xu_v = 697,
+    rv_op_vfwcvt_f_x_v = 698,
+    rv_op_vfwcvt_f_f_v = 699,
+    rv_op_vfwcvt_rtz_xu_f_v = 700,
+    rv_op_vfwcvt_rtz_x_f_v = 701,
+    rv_op_vfncvt_xu_f_w = 702,
+    rv_op_vfncvt_x_f_w = 703,
+    rv_op_vfncvt_f_xu_w = 704,
+    rv_op_vfncvt_f_x_w = 705,
+    rv_op_vfncvt_f_f_w = 706,
+    rv_op_vfncvt_rod_f_f_w = 707,
+    rv_op_vfncvt_rtz_xu_f_w = 708,
+    rv_op_vfncvt_rtz_x_f_w = 709,
+    rv_op_vredsum_vs = 710,
+    rv_op_vredand_vs = 711,
+    rv_op_vredor_vs = 712,
+    rv_op_vredxor_vs = 713,
+    rv_op_vredminu_vs = 714,
+    rv_op_vredmin_vs = 715,
+    rv_op_vredmaxu_vs = 716,
+    rv_op_vredmax_vs = 717,
+    rv_op_vwredsumu_vs = 718,
+    rv_op_vwredsum_vs = 719,
+    rv_op_vfredusum_vs = 720,
+    rv_op_vfredosum_vs = 721,
+    rv_op_vfredmin_vs = 722,
+    rv_op_vfredmax_vs = 723,
+    rv_op_vfwredusum_vs = 724,
+    rv_op_vfwredosum_vs = 725,
+    rv_op_vmand_mm = 726,
+    rv_op_vmnand_mm = 727,
+    rv_op_vmandn_mm = 728,
+    rv_op_vmxor_mm = 729,
+    rv_op_vmor_mm = 730,
+    rv_op_vmnor_mm = 731,
+    rv_op_vmorn_mm = 732,
+    rv_op_vmxnor_mm = 733,
+    rv_op_vcpop_m = 734,
+    rv_op_vfirst_m = 735,
+    rv_op_vmsbf_m = 736,
+    rv_op_vmsif_m = 737,
+    rv_op_vmsof_m = 738,
+    rv_op_viota_m = 739,
+    rv_op_vid_v = 740,
+    rv_op_vmv_x_s = 741,
+    rv_op_vmv_s_x = 742,
+    rv_op_vfmv_f_s = 743,
+    rv_op_vfmv_s_f = 744,
+    rv_op_vslideup_vx = 745,
+    rv_op_vslideup_vi = 746,
+    rv_op_vslide1up_vx = 747,
+    rv_op_vslidedown_vx = 748,
+    rv_op_vslidedown_vi = 749,
+    rv_op_vslide1down_vx = 750,
+    rv_op_vrgather_vv = 751,
+    rv_op_vrgatherei16_vv = 752,
+    rv_op_vrgather_vx = 753,
+    rv_op_vrgather_vi = 754,
+    rv_op_vcompress_vm = 755,
+    rv_op_vmv1r_v = 756,
+    rv_op_vmv2r_v = 757,
+    rv_op_vmv4r_v = 758,
+    rv_op_vmv8r_v = 759,
+    rv_op_vzext_vf2 = 760,
+    rv_op_vzext_vf4 = 761,
+    rv_op_vzext_vf8 = 762,
+    rv_op_vsext_vf2 = 763,
+    rv_op_vsext_vf4 = 764,
+    rv_op_vsext_vf8 = 765,
+    rv_op_vsetvli = 766,
+    rv_op_vsetivli = 767,
+    rv_op_vsetvl = 768,
 } rv_op;
 
 /* structures */
@@ -XXX,XX +XXX,XX @@ typedef struct {
     uint8_t   rl;
     uint8_t   bs;
     uint8_t   rnum;
+    uint8_t   vm;
+    uint32_t  vzimm;
 } rv_decode;
 
 typedef struct {
@@ -XXX,XX +XXX,XX @@ static const char rv_freg_name_sym[32][5] = {
     "fs8",  "fs9",  "fs10", "fs11", "ft8",  "ft9",  "ft10", "ft11",
 };
 
+static const char rv_vreg_name_sym[32][4] = {
+    "v0",  "v1",  "v2",  "v3",  "v4",  "v5",  "v6",  "v7",
+    "v8",  "v9",  "v10", "v11", "v12", "v13", "v14", "v15",
+    "v16", "v17", "v18", "v19", "v20", "v21", "v22", "v23",
+    "v24", "v25", "v26", "v27", "v28", "v29", "v30", "v31"
+};
+
 /* instruction formats */
 
 #define rv_fmt_none                   "O\t"
@@ -XXX,XX +XXX,XX @@ static const char rv_freg_name_sym[32][5] = {
 #define rv_fmt_rs2_offset             "O\t2,o"
 #define rv_fmt_rs1_rs2_bs             "O\t1,2,b"
 #define rv_fmt_rd_rs1_rnum            "O\t0,1,n"
+#define rv_fmt_ldst_vd_rs1_vm         "O\tD,(1)m"
+#define rv_fmt_ldst_vd_rs1_rs2_vm     "O\tD,(1),2m"
+#define rv_fmt_ldst_vd_rs1_vs2_vm     "O\tD,(1),Fm"
+#define rv_fmt_vd_vs2_vs1             "O\tD,F,E"
+#define rv_fmt_vd_vs2_vs1_vl          "O\tD,F,El"
+#define rv_fmt_vd_vs2_vs1_vm          "O\tD,F,Em"
+#define rv_fmt_vd_vs2_rs1_vl          "O\tD,F,1l"
+#define rv_fmt_vd_vs2_fs1_vl          "O\tD,F,4l"
+#define rv_fmt_vd_vs2_rs1_vm          "O\tD,F,1m"
+#define rv_fmt_vd_vs2_fs1_vm          "O\tD,F,4m"
+#define rv_fmt_vd_vs2_imm_vl          "O\tD,F,il"
+#define rv_fmt_vd_vs2_imm_vm          "O\tD,F,im"
+#define rv_fmt_vd_vs2_uimm_vm         "O\tD,F,um"
+#define rv_fmt_vd_vs1_vs2_vm          "O\tD,E,Fm"
+#define rv_fmt_vd_rs1_vs2_vm          "O\tD,1,Fm"
+#define rv_fmt_vd_fs1_vs2_vm          "O\tD,4,Fm"
+#define rv_fmt_vd_vs1                 "O\tD,E"
+#define rv_fmt_vd_rs1                 "O\tD,1"
+#define rv_fmt_vd_fs1                 "O\tD,4"
+#define rv_fmt_vd_imm                 "O\tD,i"
+#define rv_fmt_vd_vs2                 "O\tD,F"
+#define rv_fmt_vd_vs2_vm              "O\tD,Fm"
+#define rv_fmt_rd_vs2_vm              "O\t0,Fm"
+#define rv_fmt_rd_vs2                 "O\t0,F"
+#define rv_fmt_fd_vs2                 "O\t3,F"
+#define rv_fmt_vd_vm                  "O\tDm"
+#define rv_fmt_vsetvli                "O\t0,1,v"
+#define rv_fmt_vsetivli               "O\t0,u,v"
 
 /* pseudo-instruction constraints */
 
@@ -XXX,XX +XXX,XX @@ const rv_opcode_data opcode_data[] = {
     { "unzip", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0, 0 },
     { "zip", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0, 0 },
     { "xperm4", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 },
-    { "xperm8", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0, 0 }
+    { "xperm8", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0, 0 },
+    { "vle8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle8_v, rv_op_vle8_v, 0 },
+    { "vle16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle16_v, rv_op_vle16_v, 0 },
+    { "vle32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle32_v, rv_op_vle32_v, 0 },
+    { "vle64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle64_v, rv_op_vle64_v, 0 },
+    { "vse8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vse8_v, rv_op_vse8_v, 0 },
+    { "vse16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vse16_v, rv_op_vse16_v, 0 },
+    { "vse32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vse32_v, rv_op_vse32_v, 0 },
+    { "vse64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vse64_v, rv_op_vse64_v, 0 },
+    { "vlm.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vlm_v, rv_op_vlm_v, 0 },
+    { "vsm.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vsm_v, rv_op_vsm_v, 0 },
+    { "vlse8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vlse8_v, rv_op_vlse8_v, 0 },
+    { "vlse16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vlse16_v, rv_op_vlse16_v, 0 },
+    { "vlse32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vlse32_v, rv_op_vlse32_v, 0 },
+    { "vlse64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vlse64_v, rv_op_vlse64_v, 0 },
+    { "vsse8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vsse8_v, rv_op_vsse8_v, 0 },
+    { "vsse16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vsse16_v, rv_op_vsse16_v, 0 },
+    { "vsse32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vsse32_v, rv_op_vsse32_v, 0 },
+    { "vsse64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_rs2_vm, NULL, rv_op_vsse64_v, rv_op_vsse64_v, 0 },
+    { "vluxei8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vluxei8_v, rv_op_vluxei8_v, 0 },
+    { "vluxei16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vluxei16_v, rv_op_vluxei16_v, 0 },
+    { "vluxei32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vluxei32_v, rv_op_vluxei32_v, 0 },
+    { "vluxei64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vluxei64_v, rv_op_vluxei64_v, 0 },
+    { "vloxei8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vloxei8_v, rv_op_vloxei8_v, 0 },
+    { "vloxei16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vloxei16_v, rv_op_vloxei16_v, 0 },
+    { "vloxei32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vloxei32_v, rv_op_vloxei32_v, 0 },
+    { "vloxei64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vloxei64_v, rv_op_vloxei64_v, 0 },
+    { "vsuxei8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsuxei8_v, rv_op_vsuxei8_v, 0 },
+    { "vsuxei16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsuxei16_v, rv_op_vsuxei16_v, 0 },
+    { "vsuxei32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsuxei32_v, rv_op_vsuxei32_v, 0 },
+    { "vsuxei64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsuxei64_v, rv_op_vsuxei64_v, 0 },
+    { "vsoxei8.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsoxei8_v, rv_op_vsoxei8_v, 0 },
+    { "vsoxei16.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsoxei16_v, rv_op_vsoxei16_v, 0 },
+    { "vsoxei32.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsoxei32_v, rv_op_vsoxei32_v, 0 },
+    { "vsoxei64.v", rv_codec_v_r, rv_fmt_ldst_vd_rs1_vs2_vm, NULL, rv_op_vsoxei64_v, rv_op_vsoxei64_v, 0 },
+    { "vle8ff.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle8ff_v, rv_op_vle8ff_v, 0 },
+    { "vle16ff.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle16ff_v, rv_op_vle16ff_v, 0 },
+    { "vle32ff.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle32ff_v, rv_op_vle32ff_v, 0 },
+    { "vle64ff.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vle64ff_v, rv_op_vle64ff_v, 0 },
+    { "vl1re8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl1re8_v, rv_op_vl1re8_v, 0 },
+    { "vl1re16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl1re16_v, rv_op_vl1re16_v, 0 },
+    { "vl1re32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl1re32_v, rv_op_vl1re32_v, 0 },
+    { "vl1re64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl1re64_v, rv_op_vl1re64_v, 0 },
+    { "vl2re8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl2re8_v, rv_op_vl2re8_v, 0 },
+    { "vl2re16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl2re16_v, rv_op_vl2re16_v, 0 },
+    { "vl2re32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl2re32_v, rv_op_vl2re32_v, 0 },
+    { "vl2re64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl2re64_v, rv_op_vl2re64_v, 0 },
+    { "vl4re8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl4re8_v, rv_op_vl4re8_v, 0 },
+    { "vl4re16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl4re16_v, rv_op_vl4re16_v, 0 },
+    { "vl4re32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl4re32_v, rv_op_vl4re32_v, 0 },
+    { "vl4re64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl4re64_v, rv_op_vl4re64_v, 0 },
+    { "vl8re8.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl8re8_v, rv_op_vl8re8_v, 0 },
+    { "vl8re16.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl8re16_v, rv_op_vl8re16_v, 0 },
+    { "vl8re32.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl8re32_v, rv_op_vl8re32_v, 0 },
+    { "vl8re64.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vl8re64_v, rv_op_vl8re64_v, 0 },
+    { "vs1r.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vs1r_v, rv_op_vs1r_v, 0 },
+    { "vs2r.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vs2r_v, rv_op_vs2r_v, 0 },
+    { "vs4r.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vs4r_v, rv_op_vs4r_v, 0 },
+    { "vs8r.v", rv_codec_v_ldst, rv_fmt_ldst_vd_rs1_vm, NULL, rv_op_vs8r_v, rv_op_vs8r_v, 0 },
+    { "vadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vadd_vv, rv_op_vadd_vv, 0 },
+    { "vadd.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vadd_vx, rv_op_vadd_vx, 0 },
+    { "vadd.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vadd_vi, rv_op_vadd_vi, 0 },
+    { "vsub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsub_vv, rv_op_vsub_vv, 0 },
+    { "vsub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsub_vx, rv_op_vsub_vx, 0 },
+    { "vrsub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vrsub_vx, rv_op_vrsub_vx, 0 },
+    { "vrsub.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vrsub_vi, rv_op_vrsub_vi, 0 },
+    { "vwaddu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwaddu_vv, rv_op_vwaddu_vv, 0 },
+    { "vwaddu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwaddu_vx, rv_op_vwaddu_vx, 0 },
+    { "vwadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwadd_vv, rv_op_vwadd_vv, 0 },
+    { "vwadd.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwadd_vx, rv_op_vwadd_vx, 0 },
+    { "vwsubu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwsubu_vv, rv_op_vwsubu_vv, 0 },
+    { "vwsubu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwsubu_vx, rv_op_vwsubu_vx, 0 },
+    { "vwsub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwsub_vv, rv_op_vwsub_vv, 0 },
+    { "vwsub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwsub_vx, rv_op_vwsub_vx, 0 },
+    { "vwaddu.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwaddu_wv, rv_op_vwaddu_wv, 0 },
+    { "vwaddu.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwaddu_wx, rv_op_vwaddu_wx, 0 },
+    { "vwadd.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwadd_wv, rv_op_vwadd_wv, 0 },
+    { "vwadd.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwadd_wx, rv_op_vwadd_wx, 0 },
+    { "vwsubu.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwsubu_wv, rv_op_vwsubu_wv, 0 },
+    { "vwsubu.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwsubu_wx, rv_op_vwsubu_wx, 0 },
+    { "vwsub.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwsub_wv, rv_op_vwsub_wv, 0 },
+    { "vwsub.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwsub_wx, rv_op_vwsub_wx, 0 },
+    { "vadc.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vadc_vvm, rv_op_vadc_vvm, 0 },
+    { "vadc.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vadc_vxm, rv_op_vadc_vxm, 0 },
+    { "vadc.vim", rv_codec_v_i, rv_fmt_vd_vs2_imm_vl, NULL, rv_op_vadc_vim, rv_op_vadc_vim, 0 },
+    { "vmadc.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vmadc_vvm, rv_op_vmadc_vvm, 0 },
+    { "vmadc.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vmadc_vxm, rv_op_vmadc_vxm, 0 },
+    { "vmadc.vim", rv_codec_v_i, rv_fmt_vd_vs2_imm_vl, NULL, rv_op_vmadc_vim, rv_op_vmadc_vim, 0 },
+    { "vsbc.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vsbc_vvm, rv_op_vsbc_vvm, 0 },
+    { "vsbc.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vsbc_vxm, rv_op_vsbc_vxm, 0 },
+    { "vmsbc.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vmsbc_vvm, rv_op_vmsbc_vvm, 0 },
+    { "vmsbc.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vmsbc_vxm, rv_op_vmsbc_vxm, 0 },
+    { "vand.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vand_vv, rv_op_vand_vv, 0 },
+    { "vand.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vand_vx, rv_op_vand_vx, 0 },
+    { "vand.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vand_vi, rv_op_vand_vi, 0 },
+    { "vor.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vor_vv, rv_op_vor_vv, 0 },
+    { "vor.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vor_vx, rv_op_vor_vx, 0 },
+    { "vor.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vor_vi, rv_op_vor_vi, 0 },
+    { "vxor.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vxor_vv, rv_op_vxor_vv, 0 },
+    { "vxor.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vxor_vx, rv_op_vxor_vx, 0 },
+    { "vxor.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vxor_vi, rv_op_vxor_vi, 0 },
+    { "vsll.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsll_vv, rv_op_vsll_vv, 0 },
+    { "vsll.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsll_vx, rv_op_vsll_vx, 0 },
+    { "vsll.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vsll_vi, rv_op_vsll_vi, 0 },
+    { "vsrl.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsrl_vv, rv_op_vsrl_vv, 0 },
+    { "vsrl.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsrl_vx, rv_op_vsrl_vx, 0 },
+    { "vsrl.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vsrl_vi, rv_op_vsrl_vi, 0 },
+    { "vsra.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsra_vv, rv_op_vsra_vv, 0 },
+    { "vsra.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsra_vx, rv_op_vsra_vx, 0 },
+    { "vsra.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vsra_vi, rv_op_vsra_vi, 0 },
+    { "vnsrl.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vnsrl_wv, rv_op_vnsrl_wv, 0 },
+    { "vnsrl.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vnsrl_wx, rv_op_vnsrl_wx, 0 },
+    { "vnsrl.wi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vnsrl_wi, rv_op_vnsrl_wi, 0 },
+    { "vnsra.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vnsra_wv, rv_op_vnsra_wv, 0 },
+    { "vnsra.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vnsra_wx, rv_op_vnsra_wx, 0 },
+    { "vnsra.wi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vnsra_wi, rv_op_vnsra_wi, 0 },
+    { "vmseq.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmseq_vv, rv_op_vmseq_vv, 0 },
+    { "vmseq.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmseq_vx, rv_op_vmseq_vx, 0 },
+    { "vmseq.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmseq_vi, rv_op_vmseq_vi, 0 },
+    { "vmsne.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmsne_vv, rv_op_vmsne_vv, 0 },
+    { "vmsne.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsne_vx, rv_op_vmsne_vx, 0 },
+    { "vmsne.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsne_vi, rv_op_vmsne_vi, 0 },
+    { "vmsltu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmsltu_vv, rv_op_vmsltu_vv, 0 },
+    { "vmsltu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsltu_vx, rv_op_vmsltu_vx, 0 },
+    { "vmslt.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmslt_vv, rv_op_vmslt_vv, 0 },
+    { "vmslt.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmslt_vx, rv_op_vmslt_vx, 0 },
+    { "vmsleu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmsleu_vv, rv_op_vmsleu_vv, 0 },
+    { "vmsleu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsleu_vx, rv_op_vmsleu_vx, 0 },
+    { "vmsleu.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsleu_vi, rv_op_vmsleu_vi, 0 },
+    { "vmsle.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmsle_vv, rv_op_vmsle_vv, 0 },
+    { "vmsle.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsle_vx, rv_op_vmsle_vx, 0 },
+    { "vmsle.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsle_vi, rv_op_vmsle_vi, 0 },
+    { "vmsgtu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsgtu_vx, rv_op_vmsgtu_vx, 0 },
+    { "vmsgtu.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsgtu_vi, rv_op_vmsgtu_vi, 0 },
+    { "vmsgt.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmsgt_vx, rv_op_vmsgt_vx, 0 },
+    { "vmsgt.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vmsgt_vi, rv_op_vmsgt_vi, 0 },
+    { "vminu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vminu_vv, rv_op_vminu_vv, 0 },
+    { "vminu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vminu_vx, rv_op_vminu_vx, 0 },
+    { "vmin.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmin_vv, rv_op_vmin_vv, 0 },
+    { "vmin.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmin_vx, rv_op_vmin_vx, 0 },
+    { "vmaxu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmaxu_vv, rv_op_vmaxu_vv, 0 },
+    { "vmaxu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmaxu_vx, rv_op_vmaxu_vx, 0 },
+    { "vmax.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmax_vv, rv_op_vmax_vv, 0 },
+    { "vmax.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmax_vx, rv_op_vmax_vx, 0 },
+    { "vmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmul_vv, rv_op_vmul_vv, 0 },
+    { "vmul.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmul_vx, rv_op_vmul_vx, 0 },
+    { "vmulh.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmulh_vv, rv_op_vmulh_vv, 0 },
+    { "vmulh.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmulh_vx, rv_op_vmulh_vx, 0 },
+    { "vmulhu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmulhu_vv, rv_op_vmulhu_vv, 0 },
+    { "vmulhu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmulhu_vx, rv_op_vmulhu_vx, 0 },
+    { "vmulhsu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmulhsu_vv, rv_op_vmulhsu_vv, 0 },
+    { "vmulhsu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vmulhsu_vx, rv_op_vmulhsu_vx, 0 },
+    { "vdivu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vdivu_vv, rv_op_vdivu_vv, 0 },
+    { "vdivu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vdivu_vx, rv_op_vdivu_vx, 0 },
+    { "vdiv.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vdiv_vv, rv_op_vdiv_vv, 0 },
+    { "vdiv.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vdiv_vx, rv_op_vdiv_vx, 0 },
+    { "vremu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vremu_vv, rv_op_vremu_vv, 0 },
+    { "vremu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vremu_vx, rv_op_vremu_vx, 0 },
+    { "vrem.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vrem_vv, rv_op_vrem_vv, 0 },
+    { "vrem.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vrem_vx, rv_op_vrem_vx, 0 },
+    { "vwmulu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwmulu_vv, rv_op_vwmulu_vv, 0 },
+    { "vwmulu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwmulu_vx, rv_op_vwmulu_vx, 0 },
+    { "vwmulsu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwmulsu_vv, rv_op_vwmulsu_vv, 0 },
+    { "vwmulsu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwmulsu_vx, rv_op_vwmulsu_vx, 0 },
+    { "vwmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwmul_vv, rv_op_vwmul_vv, 0 },
+    { "vwmul.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vwmul_vx, rv_op_vwmul_vx, 0 },
+    { "vmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vmacc_vv, rv_op_vmacc_vv, 0 },
+    { "vmacc.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vmacc_vx, rv_op_vmacc_vx, 0 },
+    { "vnmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vnmsac_vv, rv_op_vnmsac_vv, 0 },
+    { "vnmsac.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vnmsac_vx, rv_op_vnmsac_vx, 0 },
+    { "vmadd.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vmadd_vv, rv_op_vmadd_vv, 0 },
+    { "vmadd.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vmadd_vx, rv_op_vmadd_vx, 0 },
+    { "vnmsub.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vnmsub_vv, rv_op_vnmsub_vv, 0 },
+    { "vnmsub.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vnmsub_vx, rv_op_vnmsub_vx, 0 },
+    { "vwmaccu.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vwmaccu_vv, rv_op_vwmaccu_vv, 0 },
+    { "vwmaccu.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vwmaccu_vx, rv_op_vwmaccu_vx, 0 },
+    { "vwmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vwmacc_vv, rv_op_vwmacc_vv, 0 },
+    { "vwmacc.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vwmacc_vx, rv_op_vwmacc_vx, 0 },
+    { "vwmaccsu.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vwmaccsu_vv, rv_op_vwmaccsu_vv, 0 },
+    { "vwmaccsu.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vwmaccsu_vx, rv_op_vwmaccsu_vx, 0 },
+    { "vwmaccus.vx", rv_codec_v_r, rv_fmt_vd_rs1_vs2_vm, NULL, rv_op_vwmaccus_vx, rv_op_vwmaccus_vx, 0 },
+    { "vmv.v.v", rv_codec_v_r, rv_fmt_vd_vs1, NULL, rv_op_vmv_v_v, rv_op_vmv_v_v, 0 },
+    { "vmv.v.x", rv_codec_v_r, rv_fmt_vd_rs1, NULL, rv_op_vmv_v_x, rv_op_vmv_v_x, 0 },
+    { "vmv.v.i", rv_codec_v_i, rv_fmt_vd_imm, NULL, rv_op_vmv_v_i, rv_op_vmv_v_i, 0 },
+    { "vmerge.vvm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vl, NULL, rv_op_vmerge_vvm, rv_op_vmerge_vvm, 0 },
+    { "vmerge.vxm", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vl, NULL, rv_op_vmerge_vxm, rv_op_vmerge_vxm, 0 },
+    { "vmerge.vim", rv_codec_v_i, rv_fmt_vd_vs2_imm_vl, NULL, rv_op_vmerge_vim, rv_op_vmerge_vim, 0 },
+    { "vsaddu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsaddu_vv, rv_op_vsaddu_vv, 0 },
+    { "vsaddu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsaddu_vx, rv_op_vsaddu_vx, 0 },
+    { "vsaddu.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vsaddu_vi, rv_op_vsaddu_vi, 0 },
+    { "vsadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsadd_vv, rv_op_vsadd_vv, 0 },
+    { "vsadd.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsadd_vx, rv_op_vsadd_vx, 0 },
+    { "vsadd.vi", rv_codec_v_i, rv_fmt_vd_vs2_imm_vm, NULL, rv_op_vsadd_vi, rv_op_vsadd_vi, 0 },
+    { "vssubu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vssubu_vv, rv_op_vssubu_vv, 0 },
+    { "vssubu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vssubu_vx, rv_op_vssubu_vx, 0 },
+    { "vssub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vssub_vv, rv_op_vssub_vv, 0 },
+    { "vssub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vssub_vx, rv_op_vssub_vx, 0 },
+    { "vaadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vaadd_vv, rv_op_vaadd_vv, 0 },
+    { "vaadd.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vaadd_vx, rv_op_vaadd_vx, 0 },
+    { "vaaddu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vaaddu_vv, rv_op_vaaddu_vv, 0 },
+    { "vaaddu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vaaddu_vx, rv_op_vaaddu_vx, 0 },
+    { "vasub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vasub_vv, rv_op_vasub_vv, 0 },
+    { "vasub.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vasub_vx, rv_op_vasub_vx, 0 },
+    { "vasubu.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vasubu_vv, rv_op_vasubu_vv, 0 },
+    { "vasubu.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vasubu_vx, rv_op_vasubu_vx, 0 },
+    { "vsmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vsmul_vv, rv_op_vsmul_vv, 0 },
+    { "vsmul.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vsmul_vx, rv_op_vsmul_vx, 0 },
+    { "vssrl.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vssrl_vv, rv_op_vssrl_vv, 0 },
+    { "vssrl.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vssrl_vx, rv_op_vssrl_vx, 0 },
+    { "vssrl.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vssrl_vi, rv_op_vssrl_vi, 0 },
+    { "vssra.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vssra_vv, rv_op_vssra_vv, 0 },
+    { "vssra.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vssra_vx, rv_op_vssra_vx, 0 },
+    { "vssra.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vssra_vi, rv_op_vssra_vi, 0 },
+    { "vnclipu.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vnclipu_wv, rv_op_vnclipu_wv, 0 },
+    { "vnclipu.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vnclipu_wx, rv_op_vnclipu_wx, 0 },
+    { "vnclipu.wi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vnclipu_wi, rv_op_vnclipu_wi, 0 },
+    { "vnclip.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vnclip_wv, rv_op_vnclip_wv, 0 },
+    { "vnclip.wx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vnclip_wx, rv_op_vnclip_wx, 0 },
+    { "vnclip.wi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vnclip_wi, rv_op_vnclip_wi, 0 },
+    { "vfadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfadd_vv, rv_op_vfadd_vv, 0 },
+    { "vfadd.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfadd_vf, rv_op_vfadd_vf, 0 },
+    { "vfsub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfsub_vv, rv_op_vfsub_vv, 0 },
+    { "vfsub.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfsub_vf, rv_op_vfsub_vf, 0 },
+    { "vfrsub.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfrsub_vf, rv_op_vfrsub_vf, 0 },
+    { "vfwadd.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwadd_vv, rv_op_vfwadd_vv, 0 },
+    { "vfwadd.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwadd_vf, rv_op_vfwadd_vf, 0 },
+    { "vfwadd.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwadd_wv, rv_op_vfwadd_wv, 0 },
+    { "vfwadd.wf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwadd_wf, rv_op_vfwadd_wf, 0 },
+    { "vfwsub.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwsub_vv, rv_op_vfwsub_vv, 0 },
+    { "vfwsub.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwsub_vf, rv_op_vfwsub_vf, 0 },
+    { "vfwsub.wv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwsub_wv, rv_op_vfwsub_wv, 0 },
+    { "vfwsub.wf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwsub_wf, rv_op_vfwsub_wf, 0 },
+    { "vfmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfmul_vv, rv_op_vfmul_vv, 0 },
+    { "vfmul.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfmul_vf, rv_op_vfmul_vf, 0 },
+    { "vfdiv.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfdiv_vv, rv_op_vfdiv_vv, 0 },
+    { "vfdiv.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfdiv_vf, rv_op_vfdiv_vf, 0 },
+    { "vfrdiv.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfrdiv_vf, rv_op_vfrdiv_vf, 0 },
+    { "vfwmul.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwmul_vv, rv_op_vfwmul_vv, 0 },
+    { "vfwmul.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfwmul_vf, rv_op_vfwmul_vf, 0 },
+    { "vfmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfmacc_vv, rv_op_vfmacc_vv, 0 },
+    { "vfmacc.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfmacc_vf, rv_op_vfmacc_vf, 0 },
+    { "vfnmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfnmacc_vv, rv_op_vfnmacc_vv, 0 },
+    { "vfnmacc.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfnmacc_vf, rv_op_vfnmacc_vf, 0 },
+    { "vfmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfmsac_vv, rv_op_vfmsac_vv, 0 },
+    { "vfmsac.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfmsac_vf, rv_op_vfmsac_vf, 0 },
+    { "vfnmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfnmsac_vv, rv_op_vfnmsac_vv, 0 },
+    { "vfnmsac.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfnmsac_vf, rv_op_vfnmsac_vf, 0 },
+    { "vfmadd.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfmadd_vv, rv_op_vfmadd_vv, 0 },
+    { "vfmadd.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfmadd_vf, rv_op_vfmadd_vf, 0 },
+    { "vfnmadd.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfnmadd_vv, rv_op_vfnmadd_vv, 0 },
+    { "vfnmadd.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfnmadd_vf, rv_op_vfnmadd_vf, 0 },
+    { "vfmsub.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfmsub_vv, rv_op_vfmsub_vv, 0 },
+    { "vfmsub.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfmsub_vf, rv_op_vfmsub_vf, 0 },
+    { "vfnmsub.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfnmsub_vv, rv_op_vfnmsub_vv, 0 },
+    { "vfnmsub.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfnmsub_vf, rv_op_vfnmsub_vf, 0 },
+    { "vfwmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfwmacc_vv, rv_op_vfwmacc_vv, 0 },
+    { "vfwmacc.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfwmacc_vf, rv_op_vfwmacc_vf, 0 },
+    { "vfwnmacc.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfwnmacc_vv, rv_op_vfwnmacc_vv, 0 },
+    { "vfwnmacc.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfwnmacc_vf, rv_op_vfwnmacc_vf, 0 },
+    { "vfwmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfwmsac_vv, rv_op_vfwmsac_vv, 0 },
+    { "vfwmsac.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfwmsac_vf, rv_op_vfwmsac_vf, 0 },
+    { "vfwnmsac.vv", rv_codec_v_r, rv_fmt_vd_vs1_vs2_vm, NULL, rv_op_vfwnmsac_vv, rv_op_vfwnmsac_vv, 0 },
+    { "vfwnmsac.vf", rv_codec_v_r, rv_fmt_vd_fs1_vs2_vm, NULL, rv_op_vfwnmsac_vf, rv_op_vfwnmsac_vf, 0 },
+    { "vfsqrt.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vfsqrt_v, rv_op_vfsqrt_v, 0 },
+    { "vfrsqrt7.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vfrsqrt7_v, rv_op_vfrsqrt7_v, 0 },
+    { "vfrec7.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vfrec7_v, rv_op_vfrec7_v, 0 },
+    { "vfmin.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfmin_vv, rv_op_vfmin_vv, 0 },
+    { "vfmin.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfmin_vf, rv_op_vfmin_vf, 0 },
+    { "vfmax.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfmax_vv, rv_op_vfmax_vv, 0 },
+    { "vfmax.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfmax_vf, rv_op_vfmax_vf, 0 },
+    { "vfsgnj.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfsgnj_vv, rv_op_vfsgnj_vv, 0 },
+    { "vfsgnj.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfsgnj_vf, rv_op_vfsgnj_vf, 0 },
+    { "vfsgnjn.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfsgnjn_vv, rv_op_vfsgnjn_vv, 0 },
+    { "vfsgnjn.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfsgnjn_vf, rv_op_vfsgnjn_vf, 0 },
+    { "vfsgnjx.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfsgnjx_vv, rv_op_vfsgnjx_vv, 0 },
+    { "vfsgnjx.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfsgnjx_vf, rv_op_vfsgnjx_vf, 0 },
+    { "vfslide1up.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfslide1up_vf, rv_op_vfslide1up_vf, 0 },
+    { "vfslide1down.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vfslide1down_vf, rv_op_vfslide1down_vf, 0 },
+    { "vmfeq.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmfeq_vv, rv_op_vmfeq_vv, 0 },
+    { "vmfeq.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfeq_vf, rv_op_vmfeq_vf, 0 },
+    { "vmfne.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmfne_vv, rv_op_vmfne_vv, 0 },
+    { "vmfne.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfne_vf, rv_op_vmfne_vf, 0 },
+    { "vmflt.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmflt_vv, rv_op_vmflt_vv, 0 },
+    { "vmflt.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmflt_vf, rv_op_vmflt_vf, 0 },
+    { "vmfle.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmfle_vv, rv_op_vmfle_vv, 0 },
+    { "vmfle.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfle_vf, rv_op_vmfle_vf, 0 },
+    { "vmfgt.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfgt_vf, rv_op_vmfgt_vf, 0 },
+    { "vmfge.vf", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vm, NULL, rv_op_vmfge_vf, rv_op_vmfge_vf, 0 },
+    { "vfclass.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfclass_v, rv_op_vfclass_v, 0 },
+    { "vfmerge.vfm", rv_codec_v_r, rv_fmt_vd_vs2_fs1_vl, NULL, rv_op_vfmerge_vfm, rv_op_vfmerge_vfm, 0 },
+    { "vfmv.v.f", rv_codec_v_r, rv_fmt_vd_fs1, NULL, rv_op_vfmv_v_f, rv_op_vfmv_v_f, 0 },
+    { "vfcvt.xu.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_xu_f_v, rv_op_vfcvt_xu_f_v, 0 },
+    { "vfcvt.x.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_x_f_v, rv_op_vfcvt_x_f_v, 0 },
+    { "vfcvt.f.xu.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_f_xu_v, rv_op_vfcvt_f_xu_v, 0 },
+    { "vfcvt.f.x.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_f_x_v, rv_op_vfcvt_f_x_v, 0 },
+    { "vfcvt.rtz.xu.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_rtz_xu_f_v, rv_op_vfcvt_rtz_xu_f_v, 0 },
+    { "vfcvt.rtz.x.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfcvt_rtz_x_f_v, rv_op_vfcvt_rtz_x_f_v, 0 },
+    { "vfwcvt.xu.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_xu_f_v, rv_op_vfwcvt_xu_f_v, 0 },
+    { "vfwcvt.x.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_x_f_v, rv_op_vfwcvt_x_f_v, 0 },
+    { "vfwcvt.f.xu.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_f_xu_v, rv_op_vfwcvt_f_xu_v, 0 },
+    { "vfwcvt.f.x.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_f_x_v, rv_op_vfwcvt_f_x_v, 0 },
+    { "vfwcvt.f.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_f_f_v, rv_op_vfwcvt_f_f_v, 0 },
+    { "vfwcvt.rtz.xu.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_rtz_xu_f_v, rv_op_vfwcvt_rtz_xu_f_v, 0 },
+    { "vfwcvt.rtz.x.f.v", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfwcvt_rtz_x_f_v, rv_op_vfwcvt_rtz_x_f_v, 0 },
+    { "vfncvt.xu.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_xu_f_w, rv_op_vfncvt_xu_f_w, 0 },
+    { "vfncvt.x.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_x_f_w, rv_op_vfncvt_x_f_w, 0 },
+    { "vfncvt.f.xu.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_f_xu_w, rv_op_vfncvt_f_xu_w, 0 },
+    { "vfncvt.f.x.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_f_x_w, rv_op_vfncvt_f_x_w, 0 },
+    { "vfncvt.f.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_f_f_w, rv_op_vfncvt_f_f_w, 0 },
+    { "vfncvt.rod.f.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_rod_f_f_w, rv_op_vfncvt_rod_f_f_w, 0 },
+    { "vfncvt.rtz.xu.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_rtz_xu_f_w, rv_op_vfncvt_rtz_xu_f_w, 0 },
+    { "vfncvt.rtz.x.f.w", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vfncvt_rtz_x_f_w, rv_op_vfncvt_rtz_x_f_w, 0 },
+    { "vredsum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredsum_vs, rv_op_vredsum_vs, 0 },
+    { "vredand.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredand_vs, rv_op_vredand_vs, 0 },
+    { "vredor.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredor_vs, rv_op_vredor_vs, 0 },
+    { "vredxor.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredxor_vs, rv_op_vredxor_vs, 0 },
+    { "vredminu.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredminu_vs, rv_op_vredminu_vs, 0 },
+    { "vredmin.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredmin_vs, rv_op_vredmin_vs, 0 },
+    { "vredmaxu.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredmaxu_vs, rv_op_vredmaxu_vs, 0 },
+    { "vredmax.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vredmax_vs, rv_op_vredmax_vs, 0 },
+    { "vwredsumu.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwredsumu_vs, rv_op_vwredsumu_vs, 0 },
+    { "vwredsum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vwredsum_vs, rv_op_vwredsum_vs, 0 },
+    { "vfredusum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfredusum_vs, rv_op_vfredusum_vs, 0 },
+    { "vfredosum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfredosum_vs, rv_op_vfredosum_vs, 0 },
+    { "vfredmin.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfredmin_vs, rv_op_vfredmin_vs, 0 },
+    { "vfredmax.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfredmax_vs, rv_op_vfredmax_vs, 0 },
+    { "vfwredusum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwredusum_vs, rv_op_vfwredusum_vs, 0 },
+    { "vfwredosum.vs", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vfwredosum_vs, rv_op_vfwredosum_vs, 0 },
+    { "vmand.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmand_mm, rv_op_vmand_mm, 0 },
+    { "vmnand.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmnand_mm, rv_op_vmnand_mm, 0 },
+    { "vmandn.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmandn_mm, rv_op_vmandn_mm, 0 },
+    { "vmxor.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmxor_mm, rv_op_vmxor_mm, 0 },
+    { "vmor.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmor_mm, rv_op_vmor_mm, 0 },
+    { "vmnor.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmnor_mm, rv_op_vmnor_mm, 0 },
+    { "vmorn.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmorn_mm, rv_op_vmorn_mm, 0 },
+    { "vmxnor.mm", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vmxnor_mm, rv_op_vmxnor_mm, 0 },
+    { "vcpop.m", rv_codec_v_r, rv_fmt_rd_vs2_vm, NULL, rv_op_vcpop_m, rv_op_vcpop_m, 0 },
+    { "vfirst.m", rv_codec_v_r, rv_fmt_rd_vs2_vm, NULL, rv_op_vfirst_m, rv_op_vfirst_m, 0 },
+    { "vmsbf.m", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vmsbf_m, rv_op_vmsbf_m, 0 },
+    { "vmsif.m", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vmsif_m, rv_op_vmsif_m, 0 },
+    { "vmsof.m", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vmsof_m, rv_op_vmsof_m, 0 },
+    { "viota.m", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_viota_m, rv_op_viota_m, 0 },
+    { "vid.v", rv_codec_v_r, rv_fmt_vd_vm, NULL, rv_op_vid_v, rv_op_vid_v, 0 },
+    { "vmv.x.s", rv_codec_v_r, rv_fmt_rd_vs2, NULL, rv_op_vmv_x_s, rv_op_vmv_x_s, 0 },
+    { "vmv.s.x", rv_codec_v_r, rv_fmt_vd_rs1, NULL, rv_op_vmv_s_x, rv_op_vmv_s_x, 0 },
+    { "vfmv.f.s", rv_codec_v_r, rv_fmt_fd_vs2, NULL, rv_op_vfmv_f_s, rv_op_vfmv_f_s, 0 },
+    { "vfmv.s.f", rv_codec_v_r, rv_fmt_vd_fs1, NULL, rv_op_vfmv_s_f, rv_op_vfmv_s_f, 0 },
+    { "vslideup.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vslideup_vx, rv_op_vslideup_vx, 0 },
+    { "vslideup.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vslideup_vi, rv_op_vslideup_vi, 0 },
+    { "vslide1up.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vslide1up_vx, rv_op_vslide1up_vx, 0 },
+    { "vslidedown.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vslidedown_vx, rv_op_vslidedown_vx, 0 },
+    { "vslidedown.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vslidedown_vi, rv_op_vslidedown_vi, 0 },
+    { "vslide1down.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vslide1down_vx, rv_op_vslide1down_vx, 0 },
+    { "vrgather.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vrgather_vv, rv_op_vrgather_vv, 0 },
+    { "vrgatherei16.vv", rv_codec_v_r, rv_fmt_vd_vs2_vs1_vm, NULL, rv_op_vrgatherei16_vv, rv_op_vrgatherei16_vv, 0 },
+    { "vrgather.vx", rv_codec_v_r, rv_fmt_vd_vs2_rs1_vm, NULL, rv_op_vrgather_vx, rv_op_vrgather_vx, 0 },
+    { "vrgather.vi", rv_codec_v_i, rv_fmt_vd_vs2_uimm_vm, NULL, rv_op_vrgather_vi, rv_op_vrgather_vi, 0 },
+    { "vcompress.vm", rv_codec_v_r, rv_fmt_vd_vs2_vs1, NULL, rv_op_vcompress_vm, rv_op_vcompress_vm, 0 },
+    { "vmv1r.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vmv1r_v, rv_op_vmv1r_v, 0 },
+    { "vmv2r.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vmv2r_v, rv_op_vmv2r_v, 0 },
+    { "vmv4r.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vmv4r_v, rv_op_vmv4r_v, 0 },
+    { "vmv8r.v", rv_codec_v_r, rv_fmt_vd_vs2, NULL, rv_op_vmv8r_v, rv_op_vmv8r_v, 0 },
+    { "vzext.vf2", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vzext_vf2, rv_op_vzext_vf2, 0 },
+    { "vzext.vf4", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vzext_vf4, rv_op_vzext_vf4, 0 },
+    { "vzext.vf8", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vzext_vf8, rv_op_vzext_vf8, 0 },
+    { "vsext.vf2", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vsext_vf2, rv_op_vsext_vf2, 0 },
+    { "vsext.vf4", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vsext_vf4, rv_op_vsext_vf4, 0 },
+    { "vsext.vf8", rv_codec_v_r, rv_fmt_vd_vs2_vm, NULL, rv_op_vsext_vf8, rv_op_vsext_vf8, 0 },
+    { "vsetvli", rv_codec_vsetvli, rv_fmt_vsetvli, NULL, rv_op_vsetvli, rv_op_vsetvli, 0 },
+    { "vsetivli", rv_codec_vsetivli, rv_fmt_vsetivli, NULL, rv_op_vsetivli, rv_op_vsetivli, 0 },
+    { "vsetvl", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, rv_op_vsetvl, rv_op_vsetvl, 0 }
 };
 
 /* CSR names */
@@ -XXX,XX +XXX,XX @@ static const char *csr_name(int csrno)
     case 0x0003: return "fcsr";
     case 0x0004: return "uie";
     case 0x0005: return "utvec";
+    case 0x0008: return "vstart";
+    case 0x0009: return "vxsat";
+    case 0x000a: return "vxrm";
+    case 0x000f: return "vcsr";
     case 0x0015: return "seed";
     case 0x0040: return "uscratch";
     case 0x0041: return "uepc";
@@ -XXX,XX +XXX,XX @@ static const char *csr_name(int csrno)
     case 0x0c00: return "cycle";
     case 0x0c01: return "time";
     case 0x0c02: return "instret";
+    case 0x0c20: return "vl";
+    case 0x0c21: return "vtype";
+    case 0x0c22: return "vlenb";
     case 0x0c80: return "cycleh";
     case 0x0c81: return "timeh";
     case 0x0c82: return "instreth";
@@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa)
             break;
         case 1:
             switch (((inst >> 12) & 0b111)) {
+            case 0:
+                switch (((inst >> 20) & 0b111111111111)) {
+                case 40: op = rv_op_vl1re8_v; break;
+                case 552: op = rv_op_vl2re8_v; break;
+                case 1576: op = rv_op_vl4re8_v; break;
+                case 3624: op = rv_op_vl8re8_v; break;
+                }
+                switch (((inst >> 26) & 0b111)) {
+                case 0:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: op = rv_op_vle8_v; break;
+                    case 11: op = rv_op_vlm_v; break;
+                    case 16: op = rv_op_vle8ff_v; break;
+                    }
+                    break;
+                case 1: op = rv_op_vluxei8_v; break;
+                case 2: op = rv_op_vlse8_v; break;
+                case 3: op = rv_op_vloxei8_v; break;
+                }
+                break;
             case 2: op = rv_op_flw; break;
             case 3: op = rv_op_fld; break;
             case 4: op = rv_op_flq; break;
+            case 5:
+                switch (((inst >> 20) & 0b111111111111)) {
+                case 40: op = rv_op_vl1re16_v; break;
+                case 552: op = rv_op_vl2re16_v; break;
+                case 1576: op = rv_op_vl4re16_v; break;
+                case 3624: op = rv_op_vl8re16_v; break;
+                }
+                switch (((inst >> 26) & 0b111)) {
+                case 0:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: op = rv_op_vle16_v; break;
+                    case 16: op = rv_op_vle16ff_v; break;
+                    }
+                    break;
+                case 1: op = rv_op_vluxei16_v; break;
+                case 2: op = rv_op_vlse16_v; break;
+                case 3: op = rv_op_vloxei16_v; break;
+                }
+                break;
+            case 6:
+                switch (((inst >> 20) & 0b111111111111)) {
+                case 40: op = rv_op_vl1re32_v; break;
+                case 552: op = rv_op_vl2re32_v; break;
+                case 1576: op = rv_op_vl4re32_v; break;
+                case 3624: op = rv_op_vl8re32_v; break;
+                }
+                switch (((inst >> 26) & 0b111)) {
+                case 0:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: op = rv_op_vle32_v; break;
+                    case 16: op = rv_op_vle32ff_v; break;
+                    }
+                    break;
+                case 1: op = rv_op_vluxei32_v; break;
+                case 2: op = rv_op_vlse32_v; break;
+                case 3: op = rv_op_vloxei32_v; break;
+                }
+                break;
+            case 7:
+                switch (((inst >> 20) & 0b111111111111)) {
+                case 40: op = rv_op_vl1re64_v; break;
+                case 552: op = rv_op_vl2re64_v; break;
+                case 1576: op = rv_op_vl4re64_v; break;
+                case 3624: op = rv_op_vl8re64_v; break;
+                }
+                switch (((inst >> 26) & 0b111)) {
+                case 0:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: op = rv_op_vle64_v; break;
+                    case 16: op = rv_op_vle64ff_v; break;
+                    }
+                    break;
+                case 1: op = rv_op_vluxei64_v; break;
+                case 2: op = rv_op_vlse64_v; break;
+                case 3: op = rv_op_vloxei64_v; break;
+                }
+                break;
             }
             break;
         case 3:
@@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa)
             break;
         case 9:
             switch (((inst >> 12) & 0b111)) {
+            case 0:
+                switch (((inst >> 20) & 0b111111111111)) {
+                case 40: op = rv_op_vs1r_v; break;
+                case 552: op = rv_op_vs2r_v; break;
+                case 1576: op = rv_op_vs4r_v; break;
+                case 3624: op = rv_op_vs8r_v; break;
+                }
+                switch (((inst >> 26) & 0b111)) {
+                case 0:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: op = rv_op_vse8_v; break;
+                    case 11: op = rv_op_vsm_v; break;
+                    }
+                    break;
+                case 1: op = rv_op_vsuxei8_v; break;
+                case 2: op = rv_op_vsse8_v; break;
+                case 3: op = rv_op_vsoxei8_v; break;
+                }
+                break;
             case 2: op = rv_op_fsw; break;
             case 3: op = rv_op_fsd; break;
             case 4: op = rv_op_fsq; break;
+            case 5:
+                switch (((inst >> 26) & 0b111)) {
+                case 0:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: op = rv_op_vse16_v; break;
+                    }
+                    break;
+                case 1: op = rv_op_vsuxei16_v; break;
+                case 2: op = rv_op_vsse16_v; break;
+                case 3: op = rv_op_vsoxei16_v; break;
+                }
+                break;
+            case 6:
+                switch (((inst >> 26) & 0b111)) {
+                case 0:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: op = rv_op_vse32_v; break;
+                    }
+                    break;
+                case 1: op = rv_op_vsuxei32_v; break;
+                case 2: op = rv_op_vsse32_v; break;
+                case 3: op = rv_op_vsoxei32_v; break;
+                }
+                break;
+            case 7:
+                switch (((inst >> 26) & 0b111)) {
+                case 0:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: op = rv_op_vse64_v; break;
+                    }
+                    break;
+                case 1: op = rv_op_vsuxei64_v; break;
+                case 2: op = rv_op_vsse64_v; break;
+                case 3: op = rv_op_vsoxei64_v; break;
+                }
+                break;
             }
             break;
         case 11:
@@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa)
                 break;
             }
             break;
+        case 21:
+            switch (((inst >> 12) & 0b111)) {
+            case 0:
+                switch (((inst >> 26) & 0b111111)) {
+                case 0: op = rv_op_vadd_vv; break;
+                case 2: op = rv_op_vsub_vv; break;
+                case 4: op = rv_op_vminu_vv; break;
+                case 5: op = rv_op_vmin_vv; break;
+                case 6: op = rv_op_vmaxu_vv; break;
+                case 7: op = rv_op_vmax_vv; break;
+                case 9: op = rv_op_vand_vv; break;
+                case 10: op = rv_op_vor_vv; break;
+                case 11: op = rv_op_vxor_vv; break;
+                case 12: op = rv_op_vrgather_vv; break;
+                case 14: op = rv_op_vrgatherei16_vv; break;
+                case 16: if (((inst >> 25) & 1) == 0) op = rv_op_vadc_vvm; break;
+                case 17: op = rv_op_vmadc_vvm; break;
+                case 18: if (((inst >> 25) & 1) == 0) op = rv_op_vsbc_vvm; break;
+                case 19: op = rv_op_vmsbc_vvm; break;
+                case 23:
+                    if (((inst >> 20) & 0b111111) == 32)
+                        op = rv_op_vmv_v_v;
+                    else if (((inst >> 25) & 1) == 0)
+                        op = rv_op_vmerge_vvm;
+                    break;
+                case 24: op = rv_op_vmseq_vv; break;
+                case 25: op = rv_op_vmsne_vv; break;
+                case 26: op = rv_op_vmsltu_vv; break;
+                case 27: op = rv_op_vmslt_vv; break;
+                case 28: op = rv_op_vmsleu_vv; break;
+                case 29: op = rv_op_vmsle_vv; break;
+                case 32: op = rv_op_vsaddu_vv; break;
+                case 33: op = rv_op_vsadd_vv; break;
+                case 34: op = rv_op_vssubu_vv; break;
+                case 35: op = rv_op_vssub_vv; break;
+                case 37: op = rv_op_vsll_vv; break;
+                case 39: op = rv_op_vsmul_vv; break;
+                case 40: op = rv_op_vsrl_vv; break;
+                case 41: op = rv_op_vsra_vv; break;
+                case 42: op = rv_op_vssrl_vv; break;
+                case 43: op = rv_op_vssra_vv; break;
+                case 44: op = rv_op_vnsrl_wv; break;
+                case 45: op = rv_op_vnsra_wv; break;
+                case 46: op = rv_op_vnclipu_wv; break;
+                case 47: op = rv_op_vnclip_wv; break;
+                case 48: op = rv_op_vwredsumu_vs; break;
+                case 49: op = rv_op_vwredsum_vs; break;
+                }
+                break;
+            case 1:
+                switch (((inst >> 26) & 0b111111)) {
+                case 0: op = rv_op_vfadd_vv; break;
+                case 1: op = rv_op_vfredusum_vs; break;
+                case 2: op = rv_op_vfsub_vv; break;
+                case 3: op = rv_op_vfredosum_vs; break;
+                case 4: op = rv_op_vfmin_vv; break;
+                case 5: op = rv_op_vfredmin_vs; break;
+                case 6: op = rv_op_vfmax_vv; break;
+                case 7: op = rv_op_vfredmax_vs; break;
+                case 8: op = rv_op_vfsgnj_vv; break;
+                case 9: op = rv_op_vfsgnjn_vv; break;
+                case 10: op = rv_op_vfsgnjx_vv; break;
+                case 16:
+                    switch (((inst >> 15) & 0b11111)) {
+                    case 0: if ((inst >> 25) & 1) op = rv_op_vfmv_f_s; break;
+                    }
+                    break;
+                case 18:
+                    switch (((inst >> 15) & 0b11111)) {
+                    case 0: op = rv_op_vfcvt_xu_f_v; break;
+                    case 1: op = rv_op_vfcvt_x_f_v; break;
+                    case 2: op = rv_op_vfcvt_f_xu_v; break;
+                    case 3: op = rv_op_vfcvt_f_x_v; break;
+                    case 6: op = rv_op_vfcvt_rtz_xu_f_v; break;
+                    case 7: op = rv_op_vfcvt_rtz_x_f_v; break;
+                    case 8: op = rv_op_vfwcvt_xu_f_v; break;
+                    case 9: op = rv_op_vfwcvt_x_f_v; break;
+                    case 10: op = rv_op_vfwcvt_f_xu_v; break;
+                    case 11: op = rv_op_vfwcvt_f_x_v; break;
+                    case 12: op = rv_op_vfwcvt_f_f_v; break;
+                    case 14: op = rv_op_vfwcvt_rtz_xu_f_v; break;
+                    case 15: op = rv_op_vfwcvt_rtz_x_f_v; break;
+                    case 16: op = rv_op_vfncvt_xu_f_w; break;
+                    case 17: op = rv_op_vfncvt_x_f_w; break;
+                    case 18: op = rv_op_vfncvt_f_xu_w; break;
+                    case 19: op = rv_op_vfncvt_f_x_w; break;
+                    case 20: op = rv_op_vfncvt_f_f_w; break;
+                    case 21: op = rv_op_vfncvt_rod_f_f_w; break;
+                    case 22: op = rv_op_vfncvt_rtz_xu_f_w; break;
+                    case 23: op = rv_op_vfncvt_rtz_x_f_w; break;
+                    }
+                    break;
+                case 19:
+                    switch (((inst >> 15) & 0b11111)) {
+                    case 0: op = rv_op_vfsqrt_v; break;
+                    case 4: op = rv_op_vfrsqrt7_v; break;
+                    case 5: op = rv_op_vfrec7_v; break;
+                    case 16: op = rv_op_vfclass_v; break;
+                    }
+                    break;
+                case 24: op = rv_op_vmfeq_vv; break;
+                case 25: op = rv_op_vmfle_vv; break;
+                case 27: op = rv_op_vmflt_vv; break;
+                case 28: op = rv_op_vmfne_vv; break;
+                case 32: op = rv_op_vfdiv_vv; break;
+                case 36: op = rv_op_vfmul_vv; break;
+                case 40: op = rv_op_vfmadd_vv; break;
+                case 41: op = rv_op_vfnmadd_vv; break;
+                case 42: op = rv_op_vfmsub_vv; break;
+                case 43: op = rv_op_vfnmsub_vv; break;
+                case 44: op = rv_op_vfmacc_vv; break;
+                case 45: op = rv_op_vfnmacc_vv; break;
+                case 46: op = rv_op_vfmsac_vv; break;
+                case 47: op = rv_op_vfnmsac_vv; break;
+                case 48: op = rv_op_vfwadd_vv; break;
+                case 49: op = rv_op_vfwredusum_vs; break;
+                case 50: op = rv_op_vfwsub_vv; break;
+                case 51: op = rv_op_vfwredosum_vs; break;
+                case 52: op = rv_op_vfwadd_wv; break;
+                case 54: op = rv_op_vfwsub_wv; break;
+                case 56: op = rv_op_vfwmul_vv; break;
+                case 60: op = rv_op_vfwmacc_vv; break;
+                case 61: op = rv_op_vfwnmacc_vv; break;
+                case 62: op = rv_op_vfwmsac_vv; break;
+                case 63: op = rv_op_vfwnmsac_vv; break;
+                }
+                break;
+            case 2:
+                switch (((inst >> 26) & 0b111111)) {
+                case 0: op = rv_op_vredsum_vs; break;
+                case 1: op = rv_op_vredand_vs; break;
+                case 2: op = rv_op_vredor_vs; break;
+                case 3: op = rv_op_vredxor_vs; break;
+                case 4: op = rv_op_vredminu_vs; break;
+                case 5: op = rv_op_vredmin_vs; break;
+                case 6: op = rv_op_vredmaxu_vs; break;
+                case 7: op = rv_op_vredmax_vs; break;
+                case 8: op = rv_op_vaaddu_vv; break;
+                case 9: op = rv_op_vaadd_vv; break;
+                case 10: op = rv_op_vasubu_vv; break;
+                case 11: op = rv_op_vasub_vv; break;
+                case 16:
+                    switch (((inst >> 15) & 0b11111)) {
+                    case 0: if ((inst >> 25) & 1) op = rv_op_vmv_x_s; break;
+                    case 16: op = rv_op_vcpop_m; break;
+                    case 17: op = rv_op_vfirst_m; break;
+                    }
+                    break;
+                case 18:
+                    switch (((inst >> 15) & 0b11111)) {
+                    case 2: op = rv_op_vzext_vf8; break;
+                    case 3: op = rv_op_vsext_vf8; break;
+                    case 4: op = rv_op_vzext_vf4; break;
+                    case 5: op = rv_op_vsext_vf4; break;
+                    case 6: op = rv_op_vzext_vf2; break;
+                    case 7: op = rv_op_vsext_vf2; break;
+                    }
+                    break;
+                case 20:
+                    switch (((inst >> 15) & 0b11111)) {
+                    case 1: op = rv_op_vmsbf_m;  break;
+                    case 2: op = rv_op_vmsof_m; break;
+                    case 3: op = rv_op_vmsif_m; break;
+                    case 16: op = rv_op_viota_m; break;
+                    case 17: if (((inst >> 20) & 0b11111) == 0) op = rv_op_vid_v; break;
+                    }
+                    break;
+                case 23: if ((inst >> 25) & 1) op = rv_op_vcompress_vm; break;
+                case 24: if ((inst >> 25) & 1) op = rv_op_vmandn_mm; break;
+                case 25: if ((inst >> 25) & 1) op = rv_op_vmand_mm; break;
+                case 26: if ((inst >> 25) & 1) op = rv_op_vmor_mm; break;
+                case 27: if ((inst >> 25) & 1) op = rv_op_vmxor_mm; break;
+                case 28: if ((inst >> 25) & 1) op = rv_op_vmorn_mm; break;
+                case 29: if ((inst >> 25) & 1) op = rv_op_vmnand_mm; break;
+                case 30: if ((inst >> 25) & 1) op = rv_op_vmnor_mm; break;
+                case 31: if ((inst >> 25) & 1) op = rv_op_vmxnor_mm; break;
+                case 32: op = rv_op_vdivu_vv; break;
+                case 33: op = rv_op_vdiv_vv; break;
+                case 34: op = rv_op_vremu_vv; break;
+                case 35: op = rv_op_vrem_vv; break;
+                case 36: op = rv_op_vmulhu_vv; break;
+                case 37: op = rv_op_vmul_vv; break;
+                case 38: op = rv_op_vmulhsu_vv; break;
+                case 39: op = rv_op_vmulh_vv; break;
+                case 41: op = rv_op_vmadd_vv; break;
+                case 43: op = rv_op_vnmsub_vv; break;
+                case 45: op = rv_op_vmacc_vv; break;
+                case 47: op = rv_op_vnmsac_vv; break;
+                case 48: op = rv_op_vwaddu_vv; break;
+                case 49: op = rv_op_vwadd_vv; break;
+                case 50: op = rv_op_vwsubu_vv; break;
+                case 51: op = rv_op_vwsub_vv; break;
+                case 52: op = rv_op_vwaddu_wv; break;
+                case 53: op = rv_op_vwadd_wv; break;
+                case 54: op = rv_op_vwsubu_wv; break;
+                case 55: op = rv_op_vwsub_wv; break;
+                case 56: op = rv_op_vwmulu_vv; break;
+                case 58: op = rv_op_vwmulsu_vv; break;
+                case 59: op = rv_op_vwmul_vv; break;
+                case 60: op = rv_op_vwmaccu_vv; break;
+                case 61: op = rv_op_vwmacc_vv; break;
+                case 63: op = rv_op_vwmaccsu_vv; break;
+                }
+                break;
+            case 3:
+                switch (((inst >> 26) & 0b111111)) {
+                case 0: op = rv_op_vadd_vi; break;
+                case 3: op = rv_op_vrsub_vi; break;
+                case 9: op = rv_op_vand_vi; break;
+                case 10: op = rv_op_vor_vi; break;
+                case 11: op = rv_op_vxor_vi; break;
+                case 12: op = rv_op_vrgather_vi; break;
+                case 14: op = rv_op_vslideup_vi; break;
+                case 15: op = rv_op_vslidedown_vi; break;
+                case 16: if (((inst >> 25) & 1) == 0) op = rv_op_vadc_vim; break;
+                case 17: op = rv_op_vmadc_vim; break;
+                case 23:
+                    if (((inst >> 20) & 0b111111) == 32)
+                        op = rv_op_vmv_v_i;
+                    else if (((inst >> 25) & 1) == 0)
+                        op = rv_op_vmerge_vim;
+                    break;
+                case 24: op = rv_op_vmseq_vi; break;
+                case 25: op = rv_op_vmsne_vi; break;
+                case 28: op = rv_op_vmsleu_vi; break;
+                case 29: op = rv_op_vmsle_vi; break;
+                case 30: op = rv_op_vmsgtu_vi; break;
+                case 31: op = rv_op_vmsgt_vi; break;
+                case 32: op = rv_op_vsaddu_vi; break;
+                case 33: op = rv_op_vsadd_vi; break;
+                case 37: op = rv_op_vsll_vi; break;
+                case 39:
+                    switch (((inst >> 15) & 0b11111)) {
+                    case 0: op = rv_op_vmv1r_v; break;
+                    case 1: op = rv_op_vmv2r_v; break;
+                    case 3: op = rv_op_vmv4r_v; break;
+                    case 7: op = rv_op_vmv8r_v; break;
+                    }
+                    break;
+                case 40: op = rv_op_vsrl_vi; break;
+                case 41: op = rv_op_vsra_vi; break;
+                case 42: op = rv_op_vssrl_vi; break;
+                case 43: op = rv_op_vssra_vi; break;
+                case 44: op = rv_op_vnsrl_wi; break;
+                case 45: op = rv_op_vnsra_wi; break;
+                case 46: op = rv_op_vnclipu_wi; break;
+                case 47: op = rv_op_vnclip_wi; break;
+                }
+                break;
+            case 4:
+                switch (((inst >> 26) & 0b111111)) {
+                case 0: op = rv_op_vadd_vx; break;
+                case 2: op = rv_op_vsub_vx; break;
+                case 3: op = rv_op_vrsub_vx; break;
+                case 4: op = rv_op_vminu_vx; break;
+                case 5: op = rv_op_vmin_vx; break;
+                case 6: op = rv_op_vmaxu_vx; break;
+                case 7: op = rv_op_vmax_vx; break;
+                case 9: op = rv_op_vand_vx; break;
+                case 10: op = rv_op_vor_vx; break;
+                case 11: op = rv_op_vxor_vx; break;
+                case 12: op = rv_op_vrgather_vx; break;
+                case 14: op = rv_op_vslideup_vx; break;
+                case 15: op = rv_op_vslidedown_vx; break;
+                case 16: if (((inst >> 25) & 1) == 0) op = rv_op_vadc_vxm; break;
+                case 17: op = rv_op_vmadc_vxm; break;
+                case 18: if (((inst >> 25) & 1) == 0) op = rv_op_vsbc_vxm; break;
+                case 19: op = rv_op_vmsbc_vxm; break;
+                case 23:
+                    if (((inst >> 20) & 0b111111) == 32)
+                        op = rv_op_vmv_v_x;
+                    else if (((inst >> 25) & 1) == 0)
+                        op = rv_op_vmerge_vxm;
+                    break;
+                case 24: op = rv_op_vmseq_vx; break;
+                case 25: op = rv_op_vmsne_vx; break;
+                case 26: op = rv_op_vmsltu_vx; break;
+                case 27: op = rv_op_vmslt_vx; break;
+                case 28: op = rv_op_vmsleu_vx; break;
+                case 29: op = rv_op_vmsle_vx; break;
+                case 30: op = rv_op_vmsgtu_vx; break;
+                case 31: op = rv_op_vmsgt_vx; break;
+                case 32: op = rv_op_vsaddu_vx; break;
+                case 33: op = rv_op_vsadd_vx; break;
+                case 34: op = rv_op_vssubu_vx; break;
+                case 35: op = rv_op_vssub_vx; break;
+                case 37: op = rv_op_vsll_vx; break;
+                case 39: op = rv_op_vsmul_vx; break;
+                case 40: op = rv_op_vsrl_vx; break;
+                case 41: op = rv_op_vsra_vx; break;
+                case 42: op = rv_op_vssrl_vx; break;
+                case 43: op = rv_op_vssra_vx; break;
+                case 44: op = rv_op_vnsrl_wx; break;
+                case 45: op = rv_op_vnsra_wx; break;
+                case 46: op = rv_op_vnclipu_wx; break;
+                case 47: op = rv_op_vnclip_wx; break;
+                }
+                break;
+            case 5:
+                switch (((inst >> 26) & 0b111111)) {
+                case 0: op = rv_op_vfadd_vf; break;
+                case 2: op = rv_op_vfsub_vf; break;
+                case 4: op = rv_op_vfmin_vf; break;
+                case 6: op = rv_op_vfmax_vf; break;
+                case 8: op = rv_op_vfsgnj_vf; break;
+                case 9: op = rv_op_vfsgnjn_vf; break;
+                case 10: op = rv_op_vfsgnjx_vf; break;
+                case 14: op = rv_op_vfslide1up_vf; break;
+                case 15: op = rv_op_vfslide1down_vf; break;
+                case 16:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: if ((inst >> 25) & 1) op = rv_op_vfmv_s_f; break;
+                    }
+                    break;
+                case 23:
+                    if (((inst >> 25) & 1) == 0)
+                        op = rv_op_vfmerge_vfm;
+                    else if (((inst >> 20) & 0b111111) == 32)
+                        op = rv_op_vfmv_v_f;
+                    break;
+                case 24: op = rv_op_vmfeq_vf; break;
+                case 25: op = rv_op_vmfle_vf; break;
+                case 27: op = rv_op_vmflt_vf; break;
+                case 28: op = rv_op_vmfne_vf; break;
+                case 29: op = rv_op_vmfgt_vf; break;
+                case 31: op = rv_op_vmfge_vf; break;
+                case 32: op = rv_op_vfdiv_vf; break;
+                case 33: op = rv_op_vfrdiv_vf; break;
+                case 36: op = rv_op_vfmul_vf; break;
+                case 39: op = rv_op_vfrsub_vf; break;
+                case 40: op = rv_op_vfmadd_vf; break;
+                case 41: op = rv_op_vfnmadd_vf; break;
+                case 42: op = rv_op_vfmsub_vf; break;
+                case 43: op = rv_op_vfnmsub_vf; break;
+                case 44: op = rv_op_vfmacc_vf; break;
+                case 45: op = rv_op_vfnmacc_vf; break;
+                case 46: op = rv_op_vfmsac_vf; break;
+                case 47: op = rv_op_vfnmsac_vf; break;
+                case 48: op = rv_op_vfwadd_vf; break;
+                case 50: op = rv_op_vfwsub_vf; break;
+                case 52: op = rv_op_vfwadd_wf; break;
+                case 54: op = rv_op_vfwsub_wf; break;
+                case 56: op = rv_op_vfwmul_vf; break;
+                case 60: op = rv_op_vfwmacc_vf; break;
+                case 61: op = rv_op_vfwnmacc_vf; break;
+                case 62: op = rv_op_vfwmsac_vf; break;
+                case 63: op = rv_op_vfwnmsac_vf; break;
+                }
+                break;
+            case 6:
+                switch (((inst >> 26) & 0b111111)) {
+                case 8: op = rv_op_vaaddu_vx; break;
+                case 9: op = rv_op_vaadd_vx; break;
+                case 10: op = rv_op_vasubu_vx; break;
+                case 11: op = rv_op_vasub_vx; break;
+                case 14: op = rv_op_vslide1up_vx; break;
+                case 15: op = rv_op_vslide1down_vx; break;
+                case 16:
+                    switch (((inst >> 20) & 0b11111)) {
+                    case 0: if ((inst >> 25) & 1) op = rv_op_vmv_s_x; break;
+                    }
+                    break;
+                case 32: op = rv_op_vdivu_vx; break;
+                case 33: op = rv_op_vdiv_vx; break;
+                case 34: op = rv_op_vremu_vx; break;
+                case 35: op = rv_op_vrem_vx; break;
+                case 36: op = rv_op_vmulhu_vx; break;
+                case 37: op = rv_op_vmul_vx; break;
+                case 38: op = rv_op_vmulhsu_vx; break;
+                case 39: op = rv_op_vmulh_vx; break;
+                case 41: op = rv_op_vmadd_vx; break;
+                case 43: op = rv_op_vnmsub_vx; break;
+                case 45: op = rv_op_vmacc_vx; break;
+                case 47: op = rv_op_vnmsac_vx; break;
+                case 48: op = rv_op_vwaddu_vx; break;
+                case 49: op = rv_op_vwadd_vx; break;
+                case 50: op = rv_op_vwsubu_vx; break;
+                case 51: op = rv_op_vwsub_vx; break;
+                case 52: op = rv_op_vwaddu_wx; break;
+                case 53: op = rv_op_vwadd_wx; break;
+                case 54: op = rv_op_vwsubu_wx; break;
+                case 55: op = rv_op_vwsub_wx; break;
+                case 56: op = rv_op_vwmulu_vx; break;
+                case 58: op = rv_op_vwmulsu_vx; break;
+                case 59: op = rv_op_vwmul_vx; break;
+                case 60: op = rv_op_vwmaccu_vx; break;
+                case 61: op = rv_op_vwmacc_vx; break;
+                case 62: op = rv_op_vwmaccus_vx; break;
+                case 63: op = rv_op_vwmaccsu_vx; break;
+                }
+                break;
+            case 7:
+                if (((inst >> 31) & 1) == 0) {
+                    op = rv_op_vsetvli;
+                } else if ((inst >> 30) & 1) {
+                    op = rv_op_vsetivli;
+                } else if (((inst >> 25) & 0b11111) == 0) {
+                    op = rv_op_vsetvl;
+                }
+                break;
+            }
+            break;
         case 22:
             switch (((inst >> 12) & 0b111)) {
             case 0: op = rv_op_addid; break;
@@ -XXX,XX +XXX,XX @@ static uint32_t operand_cimmq(rv_inst inst)
         ((inst << 57) >> 62) << 6;
 }
 
+static uint32_t operand_vimm(rv_inst inst)
+{
+    return (int64_t)(inst << 44) >> 59;
+}
+
+static uint32_t operand_vzimm11(rv_inst inst)
+{
+    return (inst << 33) >> 53;
+}
+
+static uint32_t operand_vzimm10(rv_inst inst)
+{
+    return (inst << 34) >> 54;
+}
+
 static uint32_t operand_bs(rv_inst inst)
 {
     return (inst << 32) >> 62;
@@ -XXX,XX +XXX,XX @@ static uint32_t operand_rnum(rv_inst inst)
     return (inst << 40) >> 60;
 }
 
+static uint32_t operand_vm(rv_inst inst)
+{
+    return (inst << 38) >> 63;
+}
+
 /* decode operands */
 
 static void decode_inst_operands(rv_decode *dec, rv_isa isa)
@@ -XXX,XX +XXX,XX @@ static void decode_inst_operands(rv_decode *dec, rv_isa isa)
         dec->rs1 = operand_rs1(inst);
         dec->rnum = operand_rnum(inst);
         break;
+    case rv_codec_v_r:
+        dec->rd = operand_rd(inst);
+        dec->rs1 = operand_rs1(inst);
+        dec->rs2 = operand_rs2(inst);
+        dec->vm = operand_vm(inst);
+        break;
+    case rv_codec_v_ldst:
+        dec->rd = operand_rd(inst);
+        dec->rs1 = operand_rs1(inst);
+        dec->vm = operand_vm(inst);
+        break;
+    case rv_codec_v_i:
+        dec->rd = operand_rd(inst);
+        dec->rs2 = operand_rs2(inst);
+        dec->imm = operand_vimm(inst);
+        dec->vm = operand_vm(inst);
+        break;
+    case rv_codec_vsetvli:
+        dec->rd = operand_rd(inst);
+        dec->rs1 = operand_rs1(inst);
+        dec->vzimm = operand_vzimm11(inst);
+        break;
+    case rv_codec_vsetivli:
+        dec->rd = operand_rd(inst);
+        dec->imm = operand_vimm(inst);
+        dec->vzimm = operand_vzimm10(inst);
+        break;
     };
 }
 
@@ -XXX,XX +XXX,XX @@ static void format_inst(char *buf, size_t buflen, size_t tab, rv_decode *dec)
             snprintf(tmp, sizeof(tmp), "%d", dec->imm);
             append(buf, tmp, buflen);
             break;
+        case 'u':
+            snprintf(tmp, sizeof(tmp), "%u", ((uint32_t)dec->imm & 0b11111));
+            append(buf, tmp, buflen);
+            break;
         case 'o':
             snprintf(tmp, sizeof(tmp), "%d", dec->imm);
             append(buf, tmp, buflen);
@@ -XXX,XX +XXX,XX @@ static void format_inst(char *buf, size_t buflen, size_t tab, rv_decode *dec)
                 append(buf, ".rl", buflen);
             }
             break;
+        case 'l':
+            append(buf, ",v0", buflen);
+            break;
+        case 'm':
+            if (dec->vm == 0) {
+                append(buf, ",v0.t", buflen);
+            }
+            break;
+        case 'D':
+            append(buf, rv_vreg_name_sym[dec->rd], buflen);
+            break;
+        case 'E':
+            append(buf, rv_vreg_name_sym[dec->rs1], buflen);
+            break;
+        case 'F':
+            append(buf, rv_vreg_name_sym[dec->rs2], buflen);
+            break;
+        case 'G':
+            append(buf, rv_vreg_name_sym[dec->rs3], buflen);
+            break;
+        case 'v': {
+            char nbuf[32] = {0};
+            const int sew = 1 << (((dec->vzimm >> 3) & 0b111) + 3);
+            sprintf(nbuf, "%d", sew);
+            const int lmul = dec->vzimm & 0b11;
+            const int flmul = (dec->vzimm >> 2) & 1;
+            const char *vta = (dec->vzimm >> 6) & 1 ? "ta" : "tu";
+            const char *vma = (dec->vzimm >> 7) & 1 ? "ma" : "mu";
+            append(buf, "e", buflen);
+            append(buf, nbuf, buflen);
+            append(buf, ",m", buflen);
+            if (flmul) {
+                switch (lmul) {
+                case 3:
+                    sprintf(nbuf, "f2");
+                    break;
+                case 2:
+                    sprintf(nbuf, "f4");
+                    break;
+                case 1:
+                    sprintf(nbuf, "f8");
+                break;
+                }
+                append(buf, nbuf, buflen);
+            } else {
+                sprintf(nbuf, "%d", 1 << lmul);
+                append(buf, nbuf, buflen);
+            }
+            append(buf, ",", buflen);
+            append(buf, vta, buflen);
+            append(buf, ",", buflen);
+            append(buf, vma, buflen);
+            break;
+        }
         default:
             break;
         }
@@ -XXX,XX +XXX,XX @@ disasm_inst(char *buf, size_t buflen, rv_isa isa, uint64_t pc, rv_inst inst)
     decode_inst_operands(&dec, isa);
     decode_inst_decompress(&dec, isa);
     decode_inst_lift_pseudo(&dec);
-    format_inst(buf, buflen, 16, &dec);
+    format_inst(buf, buflen, 24, &dec);
 }
 
 #define INST_FMT_2 "%04" PRIx64 "              "
-- 
2.37.3

From: Jim Shu <jim.shu@sifive.com>

The maximum priority level is hard-coded when writing to interrupt
priority register. However, when writing to priority threshold register,
the maximum priority level is from num_priorities Property which is
configured by platform.

Also change interrupt priority register to use num_priorities Property
in maximum priority level.

Signed-off-by: Emmanuel Blot <emmanuel.blot@sifive.com>
Signed-off-by: Jim Shu <jim.shu@sifive.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20221003041440.2320-2-jim.shu@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/intc/sifive_plic.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

From: Jim Shu <jim.shu@sifive.com>

PLIC spec [1] requires interrupt source priority registers are WARL
field and the number of supported priority is power-of-2 to simplify SW
discovery.

Existing QEMU RISC-V machine (e.g. shakti_c) don't strictly follow PLIC
spec, whose number of supported priority is not power-of-2. Just change
each bit of interrupt priority register to WARL field when the number of
supported priority is power-of-2.

[1] https://github.com/riscv/riscv-plic-spec/blob/master/riscv-plic.adoc#interrupt-priorities

Signed-off-by: Jim Shu <jim.shu@sifive.com>
Reviewed-by: Clément Chigot <chigot@adacore.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20221003041440.2320-3-jim.shu@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/intc/sifive_plic.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

diff --git a/hw/intc/sifive_plic.c b/hw/intc/sifive_plic.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/sifive_plic.c
+++ b/hw/intc/sifive_plic.c
@@ -XXX,XX +XXX,XX @@ static void sifive_plic_write(void *opaque, hwaddr addr, uint64_t value,
     if (addr_between(addr, plic->priority_base, plic->num_sources << 2)) {
         uint32_t irq = ((addr - plic->priority_base) >> 2) + 1;
 
-        if (value <= plic->num_priorities) {
+        if (((plic->num_priorities + 1) & plic->num_priorities) == 0) {
+            /*
+             * if "num_priorities + 1" is power-of-2, make each register bit of
+             * interrupt priority WARL (Write-Any-Read-Legal). Just filter
+             * out the access to unsupported priority bits.
+             */
+            plic->source_priority[irq] = value % (plic->num_priorities + 1);
+            sifive_plic_update(plic);
+        } else if (value <= plic->num_priorities) {
             plic->source_priority[irq] = value;
             sifive_plic_update(plic);
         }
@@ -XXX,XX +XXX,XX @@ static void sifive_plic_write(void *opaque, hwaddr addr, uint64_t value,
         uint32_t contextid = (addr & (plic->context_stride - 1));
 
         if (contextid == 0) {
-            if (value <= plic->num_priorities) {
+            if (((plic->num_priorities + 1) & plic->num_priorities) == 0) {
+                /*
+                 * if "num_priorities + 1" is power-of-2, each register bit of
+                 * interrupt priority is WARL (Write-Any-Read-Legal). Just
+                 * filter out the access to unsupported priority bits.
+                 */
+                plic->target_priority[addrid] = value %
+                                                (plic->num_priorities + 1);
+                sifive_plic_update(plic);
+            } else if (value <= plic->num_priorities) {
                 plic->target_priority[addrid] = value;
                 sifive_plic_update(plic);
             }
-- 
2.37.3

From: Alistair Francis <alistair.francis@wdc.com>

Since commit 4047368938f6 "accel/tcg: Introduce tlb_set_page_full" we
have been seeing this assert

../accel/tcg/cputlb.c:1294: tlb_set_page_with_attrs: Assertion `is_power_of_2(size)' failed.

When running Tock on the OpenTitan machine.

The issue is that pmp_get_tlb_size() would return a TLB size that wasn't
a power of 2. The size was also smaller then TARGET_PAGE_SIZE.

This patch ensures that any TLB size less then TARGET_PAGE_SIZE is
rounded down to 1 to ensure it's a valid size.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: LIU Zhiwei<zhiwei_liu@linux.alibaba.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20221012011449.506928-1-alistair.francis@opensource.wdc.com
Message-Id: <20221012011449.506928-1-alistair.francis@opensource.wdc.com>
---
 target/riscv/pmp.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/target/riscv/pmp.c b/target/riscv/pmp.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/pmp.c
+++ b/target/riscv/pmp.c
@@ -XXX,XX +XXX,XX @@ bool pmp_is_range_in_tlb(CPURISCVState *env, hwaddr tlb_sa,
     }
 
     if (*tlb_size != 0) {
+        /*
+         * At this point we have a tlb_size that is the smallest possible size
+         * That fits within a TARGET_PAGE_SIZE and the PMP region.
+         *
+         * If the size is less then TARGET_PAGE_SIZE we drop the size to 1.
+         * This means the result isn't cached in the TLB and is only used for
+         * a single translation.
+         */
+        if (*tlb_size < TARGET_PAGE_SIZE) {
+            *tlb_size = 1;
+        }
+
         return true;
     }
 
-- 
2.37.3

The following changes since commit c5ea91da443b458352c1b629b490ee6631775cb4:

Merge tag 'pull-trivial-patches' of https://gitlab.com/mjt0k/qemu into staging (2023-09-08 10:06:25 -0400)

are available in the Git repository at:

https://github.com/alistair23/qemu.git tags/pull-riscv-to-apply-20230911

for you to fetch changes up to e7a03409f29e2da59297d55afbaec98c96e43e3a:

target/riscv: don't read CSR in riscv_csrrw_do64 (2023-09-11 11:45:55 +1000)

----------------------------------------------------------------
First RISC-V PR for 8.2

* Remove 'host' CPU from TCG
 * riscv_htif Fixup printing on big endian hosts
 * Add zmmul isa string
 * Add smepmp isa string
 * Fix page_check_range use in fault-only-first
 * Use existing lookup tables for MixColumns
 * Add RISC-V vector cryptographic instruction set support
 * Implement WARL behaviour for mcountinhibit/mcounteren
 * Add Zihintntl extension ISA string to DTS
 * Fix zfa fleq.d and fltq.d
 * Fix upper/lower mtime write calculation
 * Make rtc variable names consistent
 * Use abi type for linux-user target_ucontext
 * Add RISC-V KVM AIA Support
 * Fix riscv,pmu DT node path in the virt machine
 * Update CSR bits name for svadu extension
 * Mark zicond non-experimental
 * Fix satp_mode_finalize() when satp_mode.supported = 0
 * Fix non-KVM --enable-debug build
 * Add new extensions to hwprobe
 * Use accelerated helper for AES64KS1I
 * Allocate itrigger timers only once
 * Respect mseccfg.RLB for pmpaddrX changes
 * Align the AIA model to v1.0 ratified spec
 * Don't read the CSR in riscv_csrrw_do64

----------------------------------------------------------------
Akihiko Odaki (1):
      target/riscv: Allocate itrigger timers only once

Ard Biesheuvel (2):
      target/riscv: Use existing lookup tables for MixColumns
      target/riscv: Use accelerated helper for AES64KS1I

Conor Dooley (1):
      hw/riscv: virt: Fix riscv,pmu DT node path

Daniel Henrique Barboza (6):
      target/riscv/cpu.c: do not run 'host' CPU with TCG
      target/riscv/cpu.c: add zmmul isa string
      target/riscv/cpu.c: add smepmp isa string
      target/riscv: fix satp_mode_finalize() when satp_mode.supported = 0
      hw/riscv/virt.c: fix non-KVM --enable-debug build
      hw/intc/riscv_aplic.c fix non-KVM --enable-debug build

Dickon Hood (2):
      target/riscv: Refactor translation of vector-widening instruction
      target/riscv: Add Zvbb ISA extension support

Jason Chien (3):
      target/riscv: Add Zihintntl extension ISA string to DTS
      hw/intc: Fix upper/lower mtime write calculation
      hw/intc: Make rtc variable names consistent

Kiran Ostrolenk (4):
      target/riscv: Refactor some of the generic vector functionality
      target/riscv: Refactor vector-vector translation macro
      target/riscv: Refactor some of the generic vector functionality
      target/riscv: Add Zvknh ISA extension support

LIU Zhiwei (3):
      target/riscv: Fix page_check_range use in fault-only-first
      target/riscv: Fix zfa fleq.d and fltq.d
      linux-user/riscv: Use abi type for target_ucontext

Lawrence Hunter (2):
      target/riscv: Add Zvbc ISA extension support
      target/riscv: Add Zvksh ISA extension support

Leon Schuermann (1):
      target/riscv/pmp.c: respect mseccfg.RLB for pmpaddrX changes

Max Chou (3):
      crypto: Create sm4_subword
      crypto: Add SM4 constant parameter CK
      target/riscv: Add Zvksed ISA extension support

Nazar Kazakov (4):
      target/riscv: Remove redundant "cpu_vl == 0" checks
      target/riscv: Move vector translation checks
      target/riscv: Add Zvkned ISA extension support
      target/riscv: Add Zvkg ISA extension support

Nikita Shubin (1):
      target/riscv: don't read CSR in riscv_csrrw_do64

Rob Bradford (1):
      target/riscv: Implement WARL behaviour for mcountinhibit/mcounteren

Robbin Ehn (1):
      linux-user/riscv: Add new extensions to hwprobe

Thomas Huth (2):
      hw/char/riscv_htif: Fix printing of console characters on big endian hosts
      hw/char/riscv_htif: Fix the console syscall on big endian hosts

Tommy Wu (1):
      target/riscv: Align the AIA model to v1.0 ratified spec

Vineet Gupta (1):
      riscv: zicond: make non-experimental

Weiwei Li (1):
      target/riscv: Update CSR bits name for svadu extension

Yong-Xuan Wang (5):
      target/riscv: support the AIA device emulation with KVM enabled
      target/riscv: check the in-kernel irqchip support
      target/riscv: Create an KVM AIA irqchip
      target/riscv: update APLIC and IMSIC to support KVM AIA
      target/riscv: select KVM AIA in riscv virt machine

From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>

The 'host' CPU is available in a CONFIG_KVM build and it's currently
available for all accels, but is a KVM only CPU. This means that in a
RISC-V KVM capable host we can do things like this:

$ ./build/qemu-system-riscv64 -M virt,accel=tcg -cpu host --nographic
qemu-system-riscv64: H extension requires priv spec 1.12.0

This CPU does not have a priv spec because we don't filter its extensions
via priv spec. We shouldn't be reaching riscv_cpu_realize_tcg() at all
with the 'host' CPU.

We don't have a way to filter the 'host' CPU out of the available CPU
options (-cpu help) if the build includes both KVM and TCG. What we can
do is to error out during riscv_cpu_realize_tcg() if the user chooses
the 'host' CPU with accel=tcg:

$ ./build/qemu-system-riscv64 -M virt,accel=tcg -cpu host --nographic
qemu-system-riscv64: 'host' CPU is not compatible with TCG acceleration

Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230721133411.474105-1-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static void riscv_cpu_realize_tcg(DeviceState *dev, Error **errp)
     CPURISCVState *env = &cpu->env;
     Error *local_err = NULL;
 
+    if (object_dynamic_cast(OBJECT(dev), TYPE_RISCV_CPU_HOST)) {
+        error_setg(errp, "'host' CPU is not compatible with TCG acceleration");
+        return;
+    }
+
     riscv_cpu_validate_misa_mxl(cpu, &local_err);
     if (local_err != NULL) {
         error_propagate(errp, local_err);
-- 
2.41.0

From: Thomas Huth <thuth@redhat.com>

The character that should be printed is stored in the 64 bit "payload"
variable. The code currently tries to print it by taking the address
of the variable and passing this pointer to qemu_chr_fe_write(). However,
this only works on little endian hosts where the least significant bits
are stored on the lowest address. To do this in a portable way, we have
to store the value in an uint8_t variable instead.

Fixes: 5033606780 ("RISC-V HTIF Console")
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230721094720.902454-2-thuth@redhat.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/char/riscv_htif.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/hw/char/riscv_htif.c b/hw/char/riscv_htif.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/char/riscv_htif.c
+++ b/hw/char/riscv_htif.c
@@ -XXX,XX +XXX,XX @@ static void htif_handle_tohost_write(HTIFState *s, uint64_t val_written)
             s->tohost = 0; /* clear to indicate we read */
             return;
         } else if (cmd == HTIF_CONSOLE_CMD_PUTC) {
-            qemu_chr_fe_write(&s->chr, (uint8_t *)&payload, 1);
+            uint8_t ch = (uint8_t)payload;
+            qemu_chr_fe_write(&s->chr, &ch, 1);
             resp = 0x100 | (uint8_t)payload;
         } else {
             qemu_log("HTIF device %d: unknown command\n", device);
-- 
2.41.0

From: Thomas Huth <thuth@redhat.com>

Values that have been read via cpu_physical_memory_read() from the
guest's memory have to be swapped in case the host endianess differs
from the guest.

Fixes: a6e13e31d5 ("riscv_htif: Support console output via proxy syscall")
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Message-Id: <20230721094720.902454-3-thuth@redhat.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/char/riscv_htif.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/hw/char/riscv_htif.c b/hw/char/riscv_htif.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/char/riscv_htif.c
+++ b/hw/char/riscv_htif.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/timer.h"
 #include "qemu/error-report.h"
 #include "exec/address-spaces.h"
+#include "exec/tswap.h"
 #include "sysemu/dma.h"
 
 #define RISCV_DEBUG_HTIF 0
@@ -XXX,XX +XXX,XX @@ static void htif_handle_tohost_write(HTIFState *s, uint64_t val_written)
             } else {
                 uint64_t syscall[8];
                 cpu_physical_memory_read(payload, syscall, sizeof(syscall));
-                if (syscall[0] == PK_SYS_WRITE &&
-                    syscall[1] == HTIF_DEV_CONSOLE &&
-                    syscall[3] == HTIF_CONSOLE_CMD_PUTC) {
+                if (tswap64(syscall[0]) == PK_SYS_WRITE &&
+                    tswap64(syscall[1]) == HTIF_DEV_CONSOLE &&
+                    tswap64(syscall[3]) == HTIF_CONSOLE_CMD_PUTC) {
                     uint8_t ch;
-                    cpu_physical_memory_read(syscall[2], &ch, 1);
+                    cpu_physical_memory_read(tswap64(syscall[2]), &ch, 1);
                     qemu_chr_fe_write(&s->chr, &ch, 1);
                     resp = 0x100 | (uint8_t)payload;
                 } else {
-- 
2.41.0

From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>

zmmul was promoted from experimental to ratified in commit 6d00ffad4e95.
Add a riscv,isa string for it.

Fixes: 6d00ffad4e95 ("target/riscv: move zmmul out of the experimental properties")
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20230720132424.371132-2-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
     ISA_EXT_DATA_ENTRY(zicsr, PRIV_VERSION_1_10_0, ext_icsr),
     ISA_EXT_DATA_ENTRY(zifencei, PRIV_VERSION_1_10_0, ext_ifencei),
     ISA_EXT_DATA_ENTRY(zihintpause, PRIV_VERSION_1_10_0, ext_zihintpause),
+    ISA_EXT_DATA_ENTRY(zmmul, PRIV_VERSION_1_12_0, ext_zmmul),
     ISA_EXT_DATA_ENTRY(zawrs, PRIV_VERSION_1_12_0, ext_zawrs),
     ISA_EXT_DATA_ENTRY(zfa, PRIV_VERSION_1_12_0, ext_zfa),
     ISA_EXT_DATA_ENTRY(zfbfmin, PRIV_VERSION_1_12_0, ext_zfbfmin),
-- 
2.41.0

From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>

The cpu->cfg.epmp extension is still experimental, but it already has a
'smepmp' riscv,isa string. Add it.

Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20230720132424.371132-3-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
     ISA_EXT_DATA_ENTRY(zhinx, PRIV_VERSION_1_12_0, ext_zhinx),
     ISA_EXT_DATA_ENTRY(zhinxmin, PRIV_VERSION_1_12_0, ext_zhinxmin),
     ISA_EXT_DATA_ENTRY(smaia, PRIV_VERSION_1_12_0, ext_smaia),
+    ISA_EXT_DATA_ENTRY(smepmp, PRIV_VERSION_1_12_0, epmp),
     ISA_EXT_DATA_ENTRY(smstateen, PRIV_VERSION_1_12_0, ext_smstateen),
     ISA_EXT_DATA_ENTRY(ssaia, PRIV_VERSION_1_12_0, ext_ssaia),
     ISA_EXT_DATA_ENTRY(sscofpmf, PRIV_VERSION_1_12_0, ext_sscofpmf),
-- 
2.41.0

From: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>

Commit bef6f008b98(accel/tcg: Return bool from page_check_range) converts
integer return value to bool type. However, it wrongly converted the use
of the API in riscv fault-only-first, where page_check_range < = 0, should
be converted to !page_check_range.

Signed-off-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20230729031618.821-1-zhiwei_liu@linux.alibaba.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/vector_helper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/vector_helper.c
+++ b/target/riscv/vector_helper.c
@@ -XXX,XX +XXX,XX @@ vext_ldff(void *vd, void *v0, target_ulong base,
                                          cpu_mmu_index(env, false));
                 if (host) {
 #ifdef CONFIG_USER_ONLY
-                    if (page_check_range(addr, offset, PAGE_READ)) {
+                    if (!page_check_range(addr, offset, PAGE_READ)) {
                         vl = i;
                         goto ProbeSuccess;
                     }
-- 
2.41.0

From: Ard Biesheuvel <ardb@kernel.org>

The AES MixColumns and InvMixColumns operations are relatively
expensive 4x4 matrix multiplications in GF(2^8), which is why C
implementations usually rely on precomputed lookup tables rather than
performing the calculations on demand.

Given that we already carry those tables in QEMU, we can just grab the
right value in the implementation of the RISC-V AES32 instructions. Note
that the tables in question are permuted according to the respective
Sbox, so we can omit the Sbox lookup as well in this case.

Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Philippe Mathieu-Daudé <philmd@linaro.org>
Cc: Zewen Ye <lustrew@foxmail.com>
Cc: Weiwei Li <liweiwei@iscas.ac.cn>
Cc: Junqiang Wang <wangjunqiang@iscas.ac.cn>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20230731084043.1791984-1-ardb@kernel.org>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 include/crypto/aes.h         |  7 +++++++
 crypto/aes.c                 |  4 ++--
 target/riscv/crypto_helper.c | 34 ++++------------------------------
 3 files changed, 13 insertions(+), 32 deletions(-)

diff --git a/include/crypto/aes.h b/include/crypto/aes.h
index XXXXXXX..XXXXXXX 100644
--- a/include/crypto/aes.h
+++ b/include/crypto/aes.h
@@ -XXX,XX +XXX,XX @@ void AES_decrypt(const unsigned char *in, unsigned char *out,
 extern const uint8_t AES_sbox[256];
 extern const uint8_t AES_isbox[256];
 
+/*
+AES_Te0[x] = S [x].[02, 01, 01, 03];
+AES_Td0[x] = Si[x].[0e, 09, 0d, 0b];
+*/
+
+extern const uint32_t AES_Te0[256], AES_Td0[256];
+
 #endif
diff --git a/crypto/aes.c b/crypto/aes.c
index XXXXXXX..XXXXXXX 100644
--- a/crypto/aes.c
+++ b/crypto/aes.c
@@ -XXX,XX +XXX,XX @@ AES_Td3[x] = Si[x].[09, 0d, 0b, 0e];
 AES_Td4[x] = Si[x].[01, 01, 01, 01];
 */
 
-static const uint32_t AES_Te0[256] = {
+const uint32_t AES_Te0[256] = {
     0xc66363a5U, 0xf87c7c84U, 0xee777799U, 0xf67b7b8dU,
     0xfff2f20dU, 0xd66b6bbdU, 0xde6f6fb1U, 0x91c5c554U,
     0x60303050U, 0x02010103U, 0xce6767a9U, 0x562b2b7dU,
@@ -XXX,XX +XXX,XX @@ static const uint32_t AES_Te4[256] = {
     0xb0b0b0b0U, 0x54545454U, 0xbbbbbbbbU, 0x16161616U,
 };
 
-static const uint32_t AES_Td0[256] = {
+const uint32_t AES_Td0[256] = {
     0x51f4a750U, 0x7e416553U, 0x1a17a4c3U, 0x3a275e96U,
     0x3bab6bcbU, 0x1f9d45f1U, 0xacfa58abU, 0x4be30393U,
     0x2030fa55U, 0xad766df6U, 0x88cc7691U, 0xf5024c25U,
diff --git a/target/riscv/crypto_helper.c b/target/riscv/crypto_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/crypto_helper.c
+++ b/target/riscv/crypto_helper.c
@@ -XXX,XX +XXX,XX @@
 #include "crypto/aes-round.h"
 #include "crypto/sm4.h"
 
-#define AES_XTIME(a) \
-    ((a << 1) ^ ((a & 0x80) ? 0x1b : 0))
-
-#define AES_GFMUL(a, b) (( \
-    (((b) & 0x1) ? (a) : 0) ^ \
-    (((b) & 0x2) ? AES_XTIME(a) : 0) ^ \
-    (((b) & 0x4) ? AES_XTIME(AES_XTIME(a)) : 0) ^ \
-    (((b) & 0x8) ? AES_XTIME(AES_XTIME(AES_XTIME(a))) : 0)) & 0xFF)
-
-static inline uint32_t aes_mixcolumn_byte(uint8_t x, bool fwd)
-{
-    uint32_t u;
-
-    if (fwd) {
-        u = (AES_GFMUL(x, 3) << 24) | (x << 16) | (x << 8) |
-            (AES_GFMUL(x, 2) << 0);
-    } else {
-        u = (AES_GFMUL(x, 0xb) << 24) | (AES_GFMUL(x, 0xd) << 16) |
-            (AES_GFMUL(x, 0x9) << 8) | (AES_GFMUL(x, 0xe) << 0);
-    }
-    return u;
-}
-
 #define sext32_xlen(x) (target_ulong)(int32_t)(x)
 
 static inline target_ulong aes32_operation(target_ulong shamt,
@@ -XXX,XX +XXX,XX @@ static inline target_ulong aes32_operation(target_ulong shamt,
                                            bool enc, bool mix)
 {
     uint8_t si = rs2 >> shamt;
-    uint8_t so;
     uint32_t mixed;
     target_ulong res;
 
     if (enc) {
-        so = AES_sbox[si];
         if (mix) {
-            mixed = aes_mixcolumn_byte(so, true);
+            mixed = be32_to_cpu(AES_Te0[si]);
         } else {
-            mixed = so;
+            mixed = AES_sbox[si];
         }
     } else {
-        so = AES_isbox[si];
         if (mix) {
-            mixed = aes_mixcolumn_byte(so, false);
+            mixed = be32_to_cpu(AES_Td0[si]);
         } else {
-            mixed = so;
+            mixed = AES_isbox[si];
         }
     }
     mixed = rol32(mixed, shamt);
-- 
2.41.0

From: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>

Take some functions/macros out of `vector_helper` and put them in a new
module called `vector_internals`. This ensures they can be used by both
vector and vector-crypto helpers (latter implemented in proceeding
commits).

Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
Signed-off-by: Max Chou <max.chou@sifive.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <20230711165917.2629866-2-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/vector_internals.h | 182 +++++++++++++++++++++++++++++
 target/riscv/vector_helper.c    | 201 +-------------------------------
 target/riscv/vector_internals.c |  81 +++++++++++++
 target/riscv/meson.build        |   1 +
 4 files changed, 265 insertions(+), 200 deletions(-)
 create mode 100644 target/riscv/vector_internals.h
 create mode 100644 target/riscv/vector_internals.c

diff --git a/target/riscv/vector_internals.h b/target/riscv/vector_internals.h
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/target/riscv/vector_internals.h
@@ -XXX,XX +XXX,XX @@
+/*
+ * RISC-V Vector Extension Internals
+ *
+ * Copyright (c) 2020 T-Head Semiconductor Co., Ltd. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef TARGET_RISCV_VECTOR_INTERNALS_H
+#define TARGET_RISCV_VECTOR_INTERNALS_H
+
+#include "qemu/osdep.h"
+#include "qemu/bitops.h"
+#include "cpu.h"
+#include "tcg/tcg-gvec-desc.h"
+#include "internals.h"
+
+static inline uint32_t vext_nf(uint32_t desc)
+{
+    return FIELD_EX32(simd_data(desc), VDATA, NF);
+}
+
+/*
+ * Note that vector data is stored in host-endian 64-bit chunks,
+ * so addressing units smaller than that needs a host-endian fixup.
+ */
+#if HOST_BIG_ENDIAN
+#define H1(x)   ((x) ^ 7)
+#define H1_2(x) ((x) ^ 6)
+#define H1_4(x) ((x) ^ 4)
+#define H2(x)   ((x) ^ 3)
+#define H4(x)   ((x) ^ 1)
+#define H8(x)   ((x))
+#else
+#define H1(x)   (x)
+#define H1_2(x) (x)
+#define H1_4(x) (x)
+#define H2(x)   (x)
+#define H4(x)   (x)
+#define H8(x)   (x)
+#endif
+
+/*
+ * Encode LMUL to lmul as following:
+ *     LMUL    vlmul    lmul
+ *      1       000       0
+ *      2       001       1
+ *      4       010       2
+ *      8       011       3
+ *      -       100       -
+ *     1/8      101      -3
+ *     1/4      110      -2
+ *     1/2      111      -1
+ */
+static inline int32_t vext_lmul(uint32_t desc)
+{
+    return sextract32(FIELD_EX32(simd_data(desc), VDATA, LMUL), 0, 3);
+}
+
+static inline uint32_t vext_vm(uint32_t desc)
+{
+    return FIELD_EX32(simd_data(desc), VDATA, VM);
+}
+
+static inline uint32_t vext_vma(uint32_t desc)
+{
+    return FIELD_EX32(simd_data(desc), VDATA, VMA);
+}
+
+static inline uint32_t vext_vta(uint32_t desc)
+{
+    return FIELD_EX32(simd_data(desc), VDATA, VTA);
+}
+
+static inline uint32_t vext_vta_all_1s(uint32_t desc)
+{
+    return FIELD_EX32(simd_data(desc), VDATA, VTA_ALL_1S);
+}
+
+/*
+ * Earlier designs (pre-0.9) had a varying number of bits
+ * per mask value (MLEN). In the 0.9 design, MLEN=1.
+ * (Section 4.5)
+ */
+static inline int vext_elem_mask(void *v0, int index)
+{
+    int idx = index / 64;
+    int pos = index  % 64;
+    return (((uint64_t *)v0)[idx] >> pos) & 1;
+}
+
+/*
+ * Get number of total elements, including prestart, body and tail elements.
+ * Note that when LMUL < 1, the tail includes the elements past VLMAX that
+ * are held in the same vector register.
+ */
+static inline uint32_t vext_get_total_elems(CPURISCVState *env, uint32_t desc,
+                                            uint32_t esz)
+{
+    uint32_t vlenb = simd_maxsz(desc);
+    uint32_t sew = 1 << FIELD_EX64(env->vtype, VTYPE, VSEW);
+    int8_t emul = ctzl(esz) - ctzl(sew) + vext_lmul(desc) < 0 ? 0 :
+                  ctzl(esz) - ctzl(sew) + vext_lmul(desc);
+    return (vlenb << emul) / esz;
+}
+
+/* set agnostic elements to 1s */
+void vext_set_elems_1s(void *base, uint32_t is_agnostic, uint32_t cnt,
+                       uint32_t tot);
+
+/* expand macro args before macro */
+#define RVVCALL(macro, ...)  macro(__VA_ARGS__)
+
+/* (TD, T1, T2, TX1, TX2) */
+#define OP_UUU_B uint8_t, uint8_t, uint8_t, uint8_t, uint8_t
+#define OP_UUU_H uint16_t, uint16_t, uint16_t, uint16_t, uint16_t
+#define OP_UUU_W uint32_t, uint32_t, uint32_t, uint32_t, uint32_t
+#define OP_UUU_D uint64_t, uint64_t, uint64_t, uint64_t, uint64_t
+
+/* operation of two vector elements */
+typedef void opivv2_fn(void *vd, void *vs1, void *vs2, int i);
+
+#define OPIVV2(NAME, TD, T1, T2, TX1, TX2, HD, HS1, HS2, OP)    \
+static void do_##NAME(void *vd, void *vs1, void *vs2, int i)    \
+{                                                               \
+    TX1 s1 = *((T1 *)vs1 + HS1(i));                             \
+    TX2 s2 = *((T2 *)vs2 + HS2(i));                             \
+    *((TD *)vd + HD(i)) = OP(s2, s1);                           \
+}
+
+void do_vext_vv(void *vd, void *v0, void *vs1, void *vs2,
+                CPURISCVState *env, uint32_t desc,
+                opivv2_fn *fn, uint32_t esz);
+
+/* generate the helpers for OPIVV */
+#define GEN_VEXT_VV(NAME, ESZ)                            \
+void HELPER(NAME)(void *vd, void *v0, void *vs1,          \
+                  void *vs2, CPURISCVState *env,          \
+                  uint32_t desc)                          \
+{                                                         \
+    do_vext_vv(vd, v0, vs1, vs2, env, desc,               \
+               do_##NAME, ESZ);                           \
+}
+
+typedef void opivx2_fn(void *vd, target_long s1, void *vs2, int i);
+
+/*
+ * (T1)s1 gives the real operator type.
+ * (TX1)(T1)s1 expands the operator type of widen or narrow operations.
+ */
+#define OPIVX2(NAME, TD, T1, T2, TX1, TX2, HD, HS2, OP)             \
+static void do_##NAME(void *vd, target_long s1, void *vs2, int i)   \
+{                                                                   \
+    TX2 s2 = *((T2 *)vs2 + HS2(i));                                 \
+    *((TD *)vd + HD(i)) = OP(s2, (TX1)(T1)s1);                      \
+}
+
+void do_vext_vx(void *vd, void *v0, target_long s1, void *vs2,
+                CPURISCVState *env, uint32_t desc,
+                opivx2_fn fn, uint32_t esz);
+
+/* generate the helpers for OPIVX */
+#define GEN_VEXT_VX(NAME, ESZ)                            \
+void HELPER(NAME)(void *vd, void *v0, target_ulong s1,    \
+                  void *vs2, CPURISCVState *env,          \
+                  uint32_t desc)                          \
+{                                                         \
+    do_vext_vx(vd, v0, s1, vs2, env, desc,                \
+               do_##NAME, ESZ);                           \
+}
+
+#endif /* TARGET_RISCV_VECTOR_INTERNALS_H */
diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/vector_helper.c
+++ b/target/riscv/vector_helper.c
@@ -XXX,XX +XXX,XX @@
 #include "fpu/softfloat.h"
 #include "tcg/tcg-gvec-desc.h"
 #include "internals.h"
+#include "vector_internals.h"
 #include <math.h>
 
 target_ulong HELPER(vsetvl)(CPURISCVState *env, target_ulong s1,
@@ -XXX,XX +XXX,XX @@ target_ulong HELPER(vsetvl)(CPURISCVState *env, target_ulong s1,
     return vl;
 }
 
-/*
- * Note that vector data is stored in host-endian 64-bit chunks,
- * so addressing units smaller than that needs a host-endian fixup.
- */
-#if HOST_BIG_ENDIAN
-#define H1(x)   ((x) ^ 7)
-#define H1_2(x) ((x) ^ 6)
-#define H1_4(x) ((x) ^ 4)
-#define H2(x)   ((x) ^ 3)
-#define H4(x)   ((x) ^ 1)
-#define H8(x)   ((x))
-#else
-#define H1(x)   (x)
-#define H1_2(x) (x)
-#define H1_4(x) (x)
-#define H2(x)   (x)
-#define H4(x)   (x)
-#define H8(x)   (x)
-#endif
-
-static inline uint32_t vext_nf(uint32_t desc)
-{
-    return FIELD_EX32(simd_data(desc), VDATA, NF);
-}
-
-static inline uint32_t vext_vm(uint32_t desc)
-{
-    return FIELD_EX32(simd_data(desc), VDATA, VM);
-}
-
-/*
- * Encode LMUL to lmul as following:
- *     LMUL    vlmul    lmul
- *      1       000       0
- *      2       001       1
- *      4       010       2
- *      8       011       3
- *      -       100       -
- *     1/8      101      -3
- *     1/4      110      -2
- *     1/2      111      -1
- */
-static inline int32_t vext_lmul(uint32_t desc)
-{
-    return sextract32(FIELD_EX32(simd_data(desc), VDATA, LMUL), 0, 3);
-}
-
-static inline uint32_t vext_vta(uint32_t desc)
-{
-    return FIELD_EX32(simd_data(desc), VDATA, VTA);
-}
-
-static inline uint32_t vext_vma(uint32_t desc)
-{
-    return FIELD_EX32(simd_data(desc), VDATA, VMA);
-}
-
-static inline uint32_t vext_vta_all_1s(uint32_t desc)
-{
-    return FIELD_EX32(simd_data(desc), VDATA, VTA_ALL_1S);
-}
-
 /*
  * Get the maximum number of elements can be operated.
  *
@@ -XXX,XX +XXX,XX @@ static inline uint32_t vext_max_elems(uint32_t desc, uint32_t log2_esz)
     return scale < 0 ? vlenb >> -scale : vlenb << scale;
 }
 
-/*
- * Get number of total elements, including prestart, body and tail elements.
- * Note that when LMUL < 1, the tail includes the elements past VLMAX that
- * are held in the same vector register.
- */
-static inline uint32_t vext_get_total_elems(CPURISCVState *env, uint32_t desc,
-                                            uint32_t esz)
-{
-    uint32_t vlenb = simd_maxsz(desc);
-    uint32_t sew = 1 << FIELD_EX64(env->vtype, VTYPE, VSEW);
-    int8_t emul = ctzl(esz) - ctzl(sew) + vext_lmul(desc) < 0 ? 0 :
-                  ctzl(esz) - ctzl(sew) + vext_lmul(desc);
-    return (vlenb << emul) / esz;
-}
-
 static inline target_ulong adjust_addr(CPURISCVState *env, target_ulong addr)
 {
     return (addr & ~env->cur_pmmask) | env->cur_pmbase;
@@ -XXX,XX +XXX,XX @@ static void probe_pages(CPURISCVState *env, target_ulong addr,
     }
 }
 
-/* set agnostic elements to 1s */
-static void vext_set_elems_1s(void *base, uint32_t is_agnostic, uint32_t cnt,
-                              uint32_t tot)
-{
-    if (is_agnostic == 0) {
-        /* policy undisturbed */
-        return;
-    }
-    if (tot - cnt == 0) {
-        return;
-    }
-    memset(base + cnt, -1, tot - cnt);
-}
-
 static inline void vext_set_elem_mask(void *v0, int index,
                                       uint8_t value)
 {
@@ -XXX,XX +XXX,XX @@ static inline void vext_set_elem_mask(void *v0, int index,
     ((uint64_t *)v0)[idx] = deposit64(old, pos, 1, value);
 }
 
-/*
- * Earlier designs (pre-0.9) had a varying number of bits
- * per mask value (MLEN). In the 0.9 design, MLEN=1.
- * (Section 4.5)
- */
-static inline int vext_elem_mask(void *v0, int index)
-{
-    int idx = index / 64;
-    int pos = index  % 64;
-    return (((uint64_t *)v0)[idx] >> pos) & 1;
-}
-
 /* elements operations for load and store */
 typedef void vext_ldst_elem_fn(CPURISCVState *env, abi_ptr addr,
                                uint32_t idx, void *vd, uintptr_t retaddr);
@@ -XXX,XX +XXX,XX @@ GEN_VEXT_ST_WHOLE(vs8r_v, int8_t, ste_b)
  * Vector Integer Arithmetic Instructions
  */
 
-/* expand macro args before macro */
-#define RVVCALL(macro, ...)  macro(__VA_ARGS__)
-
 /* (TD, T1, T2, TX1, TX2) */
 #define OP_SSS_B int8_t, int8_t, int8_t, int8_t, int8_t
 #define OP_SSS_H int16_t, int16_t, int16_t, int16_t, int16_t
 #define OP_SSS_W int32_t, int32_t, int32_t, int32_t, int32_t
 #define OP_SSS_D int64_t, int64_t, int64_t, int64_t, int64_t
-#define OP_UUU_B uint8_t, uint8_t, uint8_t, uint8_t, uint8_t
-#define OP_UUU_H uint16_t, uint16_t, uint16_t, uint16_t, uint16_t
-#define OP_UUU_W uint32_t, uint32_t, uint32_t, uint32_t, uint32_t
-#define OP_UUU_D uint64_t, uint64_t, uint64_t, uint64_t, uint64_t
 #define OP_SUS_B int8_t, uint8_t, int8_t, uint8_t, int8_t
 #define OP_SUS_H int16_t, uint16_t, int16_t, uint16_t, int16_t
 #define OP_SUS_W int32_t, uint32_t, int32_t, uint32_t, int32_t
@@ -XXX,XX +XXX,XX @@ GEN_VEXT_ST_WHOLE(vs8r_v, int8_t, ste_b)
 #define NOP_UUU_H uint16_t, uint16_t, uint32_t, uint16_t, uint32_t
 #define NOP_UUU_W uint32_t, uint32_t, uint64_t, uint32_t, uint64_t
 
-/* operation of two vector elements */
-typedef void opivv2_fn(void *vd, void *vs1, void *vs2, int i);
-
-#define OPIVV2(NAME, TD, T1, T2, TX1, TX2, HD, HS1, HS2, OP)    \
-static void do_##NAME(void *vd, void *vs1, void *vs2, int i)    \
-{                                                               \
-    TX1 s1 = *((T1 *)vs1 + HS1(i));                             \
-    TX2 s2 = *((T2 *)vs2 + HS2(i));                             \
-    *((TD *)vd + HD(i)) = OP(s2, s1);                           \
-}
 #define DO_SUB(N, M) (N - M)
 #define DO_RSUB(N, M) (M - N)
 
@@ -XXX,XX +XXX,XX @@ RVVCALL(OPIVV2, vsub_vv_h, OP_SSS_H, H2, H2, H2, DO_SUB)
 RVVCALL(OPIVV2, vsub_vv_w, OP_SSS_W, H4, H4, H4, DO_SUB)
 RVVCALL(OPIVV2, vsub_vv_d, OP_SSS_D, H8, H8, H8, DO_SUB)
 
-static void do_vext_vv(void *vd, void *v0, void *vs1, void *vs2,
-                       CPURISCVState *env, uint32_t desc,
-                       opivv2_fn *fn, uint32_t esz)
-{
-    uint32_t vm = vext_vm(desc);
-    uint32_t vl = env->vl;
-    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
-    uint32_t vta = vext_vta(desc);
-    uint32_t vma = vext_vma(desc);
-    uint32_t i;
-
-    for (i = env->vstart; i < vl; i++) {
-        if (!vm && !vext_elem_mask(v0, i)) {
-            /* set masked-off elements to 1s */
-            vext_set_elems_1s(vd, vma, i * esz, (i + 1) * esz);
-            continue;
-        }
-        fn(vd, vs1, vs2, i);
-    }
-    env->vstart = 0;
-    /* set tail elements to 1s */
-    vext_set_elems_1s(vd, vta, vl * esz, total_elems * esz);
-}
-
-/* generate the helpers for OPIVV */
-#define GEN_VEXT_VV(NAME, ESZ)                            \
-void HELPER(NAME)(void *vd, void *v0, void *vs1,          \
-                  void *vs2, CPURISCVState *env,          \
-                  uint32_t desc)                          \
-{                                                         \
-    do_vext_vv(vd, v0, vs1, vs2, env, desc,               \
-               do_##NAME, ESZ);                           \
-}
-
 GEN_VEXT_VV(vadd_vv_b, 1)
 GEN_VEXT_VV(vadd_vv_h, 2)
 GEN_VEXT_VV(vadd_vv_w, 4)
@@ -XXX,XX +XXX,XX @@ GEN_VEXT_VV(vsub_vv_h, 2)
 GEN_VEXT_VV(vsub_vv_w, 4)
 GEN_VEXT_VV(vsub_vv_d, 8)
 
-typedef void opivx2_fn(void *vd, target_long s1, void *vs2, int i);
-
-/*
- * (T1)s1 gives the real operator type.
- * (TX1)(T1)s1 expands the operator type of widen or narrow operations.
- */
-#define OPIVX2(NAME, TD, T1, T2, TX1, TX2, HD, HS2, OP)             \
-static void do_##NAME(void *vd, target_long s1, void *vs2, int i)   \
-{                                                                   \
-    TX2 s2 = *((T2 *)vs2 + HS2(i));                                 \
-    *((TD *)vd + HD(i)) = OP(s2, (TX1)(T1)s1);                      \
-}
 
 RVVCALL(OPIVX2, vadd_vx_b, OP_SSS_B, H1, H1, DO_ADD)
 RVVCALL(OPIVX2, vadd_vx_h, OP_SSS_H, H2, H2, DO_ADD)
@@ -XXX,XX +XXX,XX @@ RVVCALL(OPIVX2, vrsub_vx_h, OP_SSS_H, H2, H2, DO_RSUB)
 RVVCALL(OPIVX2, vrsub_vx_w, OP_SSS_W, H4, H4, DO_RSUB)
 RVVCALL(OPIVX2, vrsub_vx_d, OP_SSS_D, H8, H8, DO_RSUB)
 
-static void do_vext_vx(void *vd, void *v0, target_long s1, void *vs2,
-                       CPURISCVState *env, uint32_t desc,
-                       opivx2_fn fn, uint32_t esz)
-{
-    uint32_t vm = vext_vm(desc);
-    uint32_t vl = env->vl;
-    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
-    uint32_t vta = vext_vta(desc);
-    uint32_t vma = vext_vma(desc);
-    uint32_t i;
-
-    for (i = env->vstart; i < vl; i++) {
-        if (!vm && !vext_elem_mask(v0, i)) {
-            /* set masked-off elements to 1s */
-            vext_set_elems_1s(vd, vma, i * esz, (i + 1) * esz);
-            continue;
-        }
-        fn(vd, s1, vs2, i);
-    }
-    env->vstart = 0;
-    /* set tail elements to 1s */
-    vext_set_elems_1s(vd, vta, vl * esz, total_elems * esz);
-}
-
-/* generate the helpers for OPIVX */
-#define GEN_VEXT_VX(NAME, ESZ)                            \
-void HELPER(NAME)(void *vd, void *v0, target_ulong s1,    \
-                  void *vs2, CPURISCVState *env,          \
-                  uint32_t desc)                          \
-{                                                         \
-    do_vext_vx(vd, v0, s1, vs2, env, desc,                \
-               do_##NAME, ESZ);                           \
-}
-
 GEN_VEXT_VX(vadd_vx_b, 1)
 GEN_VEXT_VX(vadd_vx_h, 2)
 GEN_VEXT_VX(vadd_vx_w, 4)
diff --git a/target/riscv/vector_internals.c b/target/riscv/vector_internals.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/target/riscv/vector_internals.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * RISC-V Vector Extension Internals
+ *
+ * Copyright (c) 2020 T-Head Semiconductor Co., Ltd. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "vector_internals.h"
+
+/* set agnostic elements to 1s */
+void vext_set_elems_1s(void *base, uint32_t is_agnostic, uint32_t cnt,
+                       uint32_t tot)
+{
+    if (is_agnostic == 0) {
+        /* policy undisturbed */
+        return;
+    }
+    if (tot - cnt == 0) {
+        return ;
+    }
+    memset(base + cnt, -1, tot - cnt);
+}
+
+void do_vext_vv(void *vd, void *v0, void *vs1, void *vs2,
+                CPURISCVState *env, uint32_t desc,
+                opivv2_fn *fn, uint32_t esz)
+{
+    uint32_t vm = vext_vm(desc);
+    uint32_t vl = env->vl;
+    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
+    uint32_t vta = vext_vta(desc);
+    uint32_t vma = vext_vma(desc);
+    uint32_t i;
+
+    for (i = env->vstart; i < vl; i++) {
+        if (!vm && !vext_elem_mask(v0, i)) {
+            /* set masked-off elements to 1s */
+            vext_set_elems_1s(vd, vma, i * esz, (i + 1) * esz);
+            continue;
+        }
+        fn(vd, vs1, vs2, i);
+    }
+    env->vstart = 0;
+    /* set tail elements to 1s */
+    vext_set_elems_1s(vd, vta, vl * esz, total_elems * esz);
+}
+
+void do_vext_vx(void *vd, void *v0, target_long s1, void *vs2,
+                CPURISCVState *env, uint32_t desc,
+                opivx2_fn fn, uint32_t esz)
+{
+    uint32_t vm = vext_vm(desc);
+    uint32_t vl = env->vl;
+    uint32_t total_elems = vext_get_total_elems(env, desc, esz);
+    uint32_t vta = vext_vta(desc);
+    uint32_t vma = vext_vma(desc);
+    uint32_t i;
+
+    for (i = env->vstart; i < vl; i++) {
+        if (!vm && !vext_elem_mask(v0, i)) {
+            /* set masked-off elements to 1s */
+            vext_set_elems_1s(vd, vma, i * esz, (i + 1) * esz);
+            continue;
+        }
+        fn(vd, s1, vs2, i);
+    }
+    env->vstart = 0;
+    /* set tail elements to 1s */
+    vext_set_elems_1s(vd, vta, vl * esz, total_elems * esz);
+}
diff --git a/target/riscv/meson.build b/target/riscv/meson.build
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/meson.build
+++ b/target/riscv/meson.build
@@ -XXX,XX +XXX,XX @@ riscv_ss.add(files(
   'gdbstub.c',
   'op_helper.c',
   'vector_helper.c',
+  'vector_internals.c',
   'bitmanip_helper.c',
   'translate.c',
   'm128_helper.c',
-- 
2.41.0

From: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>

Refactor the non SEW-specific stuff out of `GEN_OPIVV_TRANS` into
function `opivv_trans` (similar to `opivi_trans`). `opivv_trans` will be
used in proceeding vector-crypto commits.

Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
Signed-off-by: Max Chou <max.chou@sifive.com>
Message-ID: <20230711165917.2629866-3-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/insn_trans/trans_rvv.c.inc | 62 +++++++++++++------------
 1 file changed, 32 insertions(+), 30 deletions(-)

From: Nazar Kazakov <nazar.kazakov@codethink.co.uk>

Remove the redundant "vl == 0" check which is already included within the  vstart >= vl check, when vl == 0.

Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
Signed-off-by: Max Chou <max.chou@sifive.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <20230711165917.2629866-4-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/insn_trans/trans_rvv.c.inc | 31 +------------------------
 1 file changed, 1 insertion(+), 30 deletions(-)

diff --git a/target/riscv/insn_trans/trans_rvv.c.inc b/target/riscv/insn_trans/trans_rvv.c.inc
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/insn_trans/trans_rvv.c.inc
+++ b/target/riscv/insn_trans/trans_rvv.c.inc
@@ -XXX,XX +XXX,XX @@ static bool ldst_us_trans(uint32_t vd, uint32_t rs1, uint32_t data,
     TCGv_i32 desc;
 
     TCGLabel *over = gen_new_label();
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     dest = tcg_temp_new_ptr();
@@ -XXX,XX +XXX,XX @@ static bool ldst_stride_trans(uint32_t vd, uint32_t rs1, uint32_t rs2,
     TCGv_i32 desc;
 
     TCGLabel *over = gen_new_label();
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     dest = tcg_temp_new_ptr();
@@ -XXX,XX +XXX,XX @@ static bool ldst_index_trans(uint32_t vd, uint32_t rs1, uint32_t vs2,
     TCGv_i32 desc;
 
     TCGLabel *over = gen_new_label();
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     dest = tcg_temp_new_ptr();
@@ -XXX,XX +XXX,XX @@ static bool ldff_trans(uint32_t vd, uint32_t rs1, uint32_t data,
     TCGv_i32 desc;
 
     TCGLabel *over = gen_new_label();
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     dest = tcg_temp_new_ptr();
@@ -XXX,XX +XXX,XX @@ do_opivv_gvec(DisasContext *s, arg_rmrr *a, GVecGen3Fn *gvec_fn,
         return false;
     }
 
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     if (a->vm && s->vl_eq_vlmax && !(s->vta && s->lmul < 0)) {
@@ -XXX,XX +XXX,XX @@ static bool opivx_trans(uint32_t vd, uint32_t rs1, uint32_t vs2, uint32_t vm,
     uint32_t data = 0;
 
     TCGLabel *over = gen_new_label();
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     dest = tcg_temp_new_ptr();
@@ -XXX,XX +XXX,XX @@ static bool opivi_trans(uint32_t vd, uint32_t imm, uint32_t vs2, uint32_t vm,
     uint32_t data = 0;
 
     TCGLabel *over = gen_new_label();
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     dest = tcg_temp_new_ptr();
@@ -XXX,XX +XXX,XX @@ static bool do_opivv_widen(DisasContext *s, arg_rmrr *a,
     if (checkfn(s, a)) {
         uint32_t data = 0;
         TCGLabel *over = gen_new_label();
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
         data = FIELD_DP32(data, VDATA, VM, a->vm);
@@ -XXX,XX +XXX,XX @@ static bool do_opiwv_widen(DisasContext *s, arg_rmrr *a,
     if (opiwv_widen_check(s, a)) {
         uint32_t data = 0;
         TCGLabel *over = gen_new_label();
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
         data = FIELD_DP32(data, VDATA, VM, a->vm);
@@ -XXX,XX +XXX,XX @@ static bool opivv_trans(uint32_t vd, uint32_t vs1, uint32_t vs2, uint32_t vm,
 {
     uint32_t data = 0;
     TCGLabel *over = gen_new_label();
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     data = FIELD_DP32(data, VDATA, VM, vm);
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
             gen_helper_##NAME##_w,                                 \
         };                                                         \
         TCGLabel *over = gen_new_label();                          \
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
                                                                    \
         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
@@ -XXX,XX +XXX,XX @@ static bool trans_vmv_v_v(DisasContext *s, arg_vmv_v_v *a)
                 gen_helper_vmv_v_v_w, gen_helper_vmv_v_v_d,
             };
             TCGLabel *over = gen_new_label();
-            tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
             tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
             tcg_gen_gvec_2_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, a->rs1),
@@ -XXX,XX +XXX,XX @@ static bool trans_vmv_v_x(DisasContext *s, arg_vmv_v_x *a)
         vext_check_ss(s, a->rd, 0, 1)) {
         TCGv s1;
         TCGLabel *over = gen_new_label();
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
         s1 = get_gpr(s, a->rs1, EXT_SIGN);
@@ -XXX,XX +XXX,XX @@ static bool trans_vmv_v_i(DisasContext *s, arg_vmv_v_i *a)
                 gen_helper_vmv_v_x_w, gen_helper_vmv_v_x_d,
             };
             TCGLabel *over = gen_new_label();
-            tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
             tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
             s1 = tcg_constant_i64(simm);
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
         };                                                         \
         TCGLabel *over = gen_new_label();                          \
         gen_set_rm(s, RISCV_FRM_DYN);                              \
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
                                                                    \
         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
@@ -XXX,XX +XXX,XX @@ static bool opfvf_trans(uint32_t vd, uint32_t rs1, uint32_t vs2,
     TCGv_i64 t1;
 
     TCGLabel *over = gen_new_label();
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     dest = tcg_temp_new_ptr();
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)           \
         };                                                       \
         TCGLabel *over = gen_new_label();                        \
         gen_set_rm(s, RISCV_FRM_DYN);                            \
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);        \
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);\
                                                                  \
         data = FIELD_DP32(data, VDATA, VM, a->vm);               \
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmrr *a)             \
         };                                                         \
         TCGLabel *over = gen_new_label();                          \
         gen_set_rm(s, RISCV_FRM_DYN);                              \
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
                                                                    \
         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
@@ -XXX,XX +XXX,XX @@ static bool do_opfv(DisasContext *s, arg_rmr *a,
         uint32_t data = 0;
         TCGLabel *over = gen_new_label();
         gen_set_rm_chkfrm(s, rm);
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
         data = FIELD_DP32(data, VDATA, VM, a->vm);
@@ -XXX,XX +XXX,XX @@ static bool trans_vfmv_v_f(DisasContext *s, arg_vfmv_v_f *a)
                 gen_helper_vmv_v_x_d,
             };
             TCGLabel *over = gen_new_label();
-            tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
             tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
             t1 = tcg_temp_new_i64();
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a)              \
         };                                                         \
         TCGLabel *over = gen_new_label();                          \
         gen_set_rm_chkfrm(s, FRM);                                 \
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
                                                                    \
         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a)              \
         };                                                         \
         TCGLabel *over = gen_new_label();                          \
         gen_set_rm(s, RISCV_FRM_DYN);                              \
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
                                                                    \
         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a)              \
         };                                                         \
         TCGLabel *over = gen_new_label();                          \
         gen_set_rm_chkfrm(s, FRM);                                 \
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
                                                                    \
         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a)              \
         };                                                         \
         TCGLabel *over = gen_new_label();                          \
         gen_set_rm_chkfrm(s, FRM);                                 \
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
                                                                    \
         data = FIELD_DP32(data, VDATA, VM, a->vm);                 \
@@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_r *a)                \
         uint32_t data = 0;                                         \
         gen_helper_gvec_4_ptr *fn = gen_helper_##NAME;             \
         TCGLabel *over = gen_new_label();                          \
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);          \
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over); \
                                                                    \
         data = FIELD_DP32(data, VDATA, LMUL, s->lmul);             \
@@ -XXX,XX +XXX,XX @@ static bool trans_vid_v(DisasContext *s, arg_vid_v *a)
         require_vm(a->vm, a->rd)) {
         uint32_t data = 0;
         TCGLabel *over = gen_new_label();
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
         data = FIELD_DP32(data, VDATA, VM, a->vm);
@@ -XXX,XX +XXX,XX @@ static bool trans_vmv_s_x(DisasContext *s, arg_vmv_s_x *a)
         TCGv s1;
         TCGLabel *over = gen_new_label();
 
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
         t1 = tcg_temp_new_i64();
@@ -XXX,XX +XXX,XX @@ static bool trans_vfmv_s_f(DisasContext *s, arg_vfmv_s_f *a)
         TCGv_i64 t1;
         TCGLabel *over = gen_new_label();
 
-        /* if vl == 0 or vstart >= vl, skip vector register write back */
-        tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
+        /* if vstart >= vl, skip vector register write back */
         tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
         /* NaN-box f[rs1] */
@@ -XXX,XX +XXX,XX @@ static bool int_ext_op(DisasContext *s, arg_rmr *a, uint8_t seq)
     uint32_t data = 0;
     gen_helper_gvec_3_ptr *fn;
     TCGLabel *over = gen_new_label();
-    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_vl, 0, over);
     tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
 
     static gen_helper_gvec_3_ptr * const fns[6][4] = {
-- 
2.41.0

From: Lawrence Hunter <lawrence.hunter@codethink.co.uk>

This commit adds support for the Zvbc vector-crypto extension, which
consists of the following instructions:

* vclmulh.[vx,vv]
* vclmul.[vx,vv]

Translation functions are defined in
`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
`target/riscv/vcrypto_helper.c`.

Co-authored-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Co-authored-by: Max Chou <max.chou@sifive.com>
Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
Signed-off-by: Max Chou <max.chou@sifive.com>
[max.chou@sifive.com: Exposed x-zvbc property]
Message-ID: <20230711165917.2629866-5-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu_cfg.h                   |  1 +
 target/riscv/helper.h                    |  6 +++
 target/riscv/insn32.decode               |  6 +++
 target/riscv/cpu.c                       |  9 ++++
 target/riscv/translate.c                 |  1 +
 target/riscv/vcrypto_helper.c            | 59 ++++++++++++++++++++++
 target/riscv/insn_trans/trans_rvvk.c.inc | 62 ++++++++++++++++++++++++
 target/riscv/meson.build                 |  3 +-
 8 files changed, 146 insertions(+), 1 deletion(-)
 create mode 100644 target/riscv/vcrypto_helper.c
 create mode 100644 target/riscv/insn_trans/trans_rvvk.c.inc

From: Nazar Kazakov <nazar.kazakov@codethink.co.uk>

Move the checks out of `do_opiv{v,x,i}_gvec{,_shift}` functions
and into the corresponding macros. This enables the functions to be
reused in proceeding commits without check duplication.

Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
Signed-off-by: Max Chou <max.chou@sifive.com>
Message-ID: <20230711165917.2629866-6-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/insn_trans/trans_rvv.c.inc | 28 +++++++++++--------------
 1 file changed, 12 insertions(+), 16 deletions(-)

From: Dickon Hood <dickon.hood@codethink.co.uk>

Zvbb (implemented in later commit) has a widening instruction, which
requires an extra check on the enabled extensions.  Refactor
GEN_OPIVX_WIDEN_TRANS() to take a check function to avoid reimplementing
it.

Signed-off-by: Dickon Hood <dickon.hood@codethink.co.uk>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
Signed-off-by: Max Chou <max.chou@sifive.com>
Message-ID: <20230711165917.2629866-7-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/insn_trans/trans_rvv.c.inc | 52 +++++++++++--------------
 1 file changed, 23 insertions(+), 29 deletions(-)

From: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>

Move some macros out of `vector_helper` and into `vector_internals`.
This ensures they can be used by both vector and vector-crypto helpers
(latter implemented in proceeding commits).

Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
Signed-off-by: Max Chou <max.chou@sifive.com>
Message-ID: <20230711165917.2629866-8-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/vector_internals.h | 46 +++++++++++++++++++++++++++++++++
 target/riscv/vector_helper.c    | 42 ------------------------------
 2 files changed, 46 insertions(+), 42 deletions(-)

diff --git a/target/riscv/vector_internals.h b/target/riscv/vector_internals.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/vector_internals.h
+++ b/target/riscv/vector_internals.h
@@ -XXX,XX +XXX,XX @@ void vext_set_elems_1s(void *base, uint32_t is_agnostic, uint32_t cnt,
 /* expand macro args before macro */
 #define RVVCALL(macro, ...)  macro(__VA_ARGS__)
 
+/* (TD, T2, TX2) */
+#define OP_UU_B uint8_t, uint8_t, uint8_t
+#define OP_UU_H uint16_t, uint16_t, uint16_t
+#define OP_UU_W uint32_t, uint32_t, uint32_t
+#define OP_UU_D uint64_t, uint64_t, uint64_t
+
 /* (TD, T1, T2, TX1, TX2) */
 #define OP_UUU_B uint8_t, uint8_t, uint8_t, uint8_t, uint8_t
 #define OP_UUU_H uint16_t, uint16_t, uint16_t, uint16_t, uint16_t
 #define OP_UUU_W uint32_t, uint32_t, uint32_t, uint32_t, uint32_t
 #define OP_UUU_D uint64_t, uint64_t, uint64_t, uint64_t, uint64_t
 
+#define OPIVV1(NAME, TD, T2, TX2, HD, HS2, OP)         \
+static void do_##NAME(void *vd, void *vs2, int i)      \
+{                                                      \
+    TX2 s2 = *((T2 *)vs2 + HS2(i));                    \
+    *((TD *)vd + HD(i)) = OP(s2);                      \
+}
+
+#define GEN_VEXT_V(NAME, ESZ)                          \
+void HELPER(NAME)(void *vd, void *v0, void *vs2,       \
+                  CPURISCVState *env, uint32_t desc)   \
+{                                                      \
+    uint32_t vm = vext_vm(desc);                       \
+    uint32_t vl = env->vl;                             \
+    uint32_t total_elems =                             \
+        vext_get_total_elems(env, desc, ESZ);          \
+    uint32_t vta = vext_vta(desc);                     \
+    uint32_t vma = vext_vma(desc);                     \
+    uint32_t i;                                        \
+                                                       \
+    for (i = env->vstart; i < vl; i++) {               \
+        if (!vm && !vext_elem_mask(v0, i)) {           \
+            /* set masked-off elements to 1s */        \
+            vext_set_elems_1s(vd, vma, i * ESZ,        \
+                              (i + 1) * ESZ);          \
+            continue;                                  \
+        }                                              \
+        do_##NAME(vd, vs2, i);                         \
+    }                                                  \
+    env->vstart = 0;                                   \
+    /* set tail elements to 1s */                      \
+    vext_set_elems_1s(vd, vta, vl * ESZ,               \
+                      total_elems * ESZ);              \
+}
+
 /* operation of two vector elements */
 typedef void opivv2_fn(void *vd, void *vs1, void *vs2, int i);
 
@@ -XXX,XX +XXX,XX @@ void HELPER(NAME)(void *vd, void *v0, target_ulong s1,    \
                do_##NAME, ESZ);                           \
 }
 
+/* Three of the widening shortening macros: */
+/* (TD, T1, T2, TX1, TX2) */
+#define WOP_UUU_B uint16_t, uint8_t, uint8_t, uint16_t, uint16_t
+#define WOP_UUU_H uint32_t, uint16_t, uint16_t, uint32_t, uint32_t
+#define WOP_UUU_W uint64_t, uint32_t, uint32_t, uint64_t, uint64_t
+
 #endif /* TARGET_RISCV_VECTOR_INTERNALS_H */
diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/vector_helper.c
+++ b/target/riscv/vector_helper.c
@@ -XXX,XX +XXX,XX @@ GEN_VEXT_ST_WHOLE(vs8r_v, int8_t, ste_b)
 #define OP_SUS_H int16_t, uint16_t, int16_t, uint16_t, int16_t
 #define OP_SUS_W int32_t, uint32_t, int32_t, uint32_t, int32_t
 #define OP_SUS_D int64_t, uint64_t, int64_t, uint64_t, int64_t
-#define WOP_UUU_B uint16_t, uint8_t, uint8_t, uint16_t, uint16_t
-#define WOP_UUU_H uint32_t, uint16_t, uint16_t, uint32_t, uint32_t
-#define WOP_UUU_W uint64_t, uint32_t, uint32_t, uint64_t, uint64_t
 #define WOP_SSS_B int16_t, int8_t, int8_t, int16_t, int16_t
 #define WOP_SSS_H int32_t, int16_t, int16_t, int32_t, int32_t
 #define WOP_SSS_W int64_t, int32_t, int32_t, int64_t, int64_t
@@ -XXX,XX +XXX,XX @@ GEN_VEXT_VF(vfwnmsac_vf_h, 4)
 GEN_VEXT_VF(vfwnmsac_vf_w, 8)
 
 /* Vector Floating-Point Square-Root Instruction */
-/* (TD, T2, TX2) */
-#define OP_UU_H uint16_t, uint16_t, uint16_t
-#define OP_UU_W uint32_t, uint32_t, uint32_t
-#define OP_UU_D uint64_t, uint64_t, uint64_t
-
 #define OPFVV1(NAME, TD, T2, TX2, HD, HS2, OP)         \
 static void do_##NAME(void *vd, void *vs2, int i,      \
                       CPURISCVState *env)              \
@@ -XXX,XX +XXX,XX @@ GEN_VEXT_CMP_VF(vmfge_vf_w, uint32_t, H4, vmfge32)
 GEN_VEXT_CMP_VF(vmfge_vf_d, uint64_t, H8, vmfge64)
 
 /* Vector Floating-Point Classify Instruction */
-#define OPIVV1(NAME, TD, T2, TX2, HD, HS2, OP)         \
-static void do_##NAME(void *vd, void *vs2, int i)      \
-{                                                      \
-    TX2 s2 = *((T2 *)vs2 + HS2(i));                    \
-    *((TD *)vd + HD(i)) = OP(s2);                      \
-}
-
-#define GEN_VEXT_V(NAME, ESZ)                          \
-void HELPER(NAME)(void *vd, void *v0, void *vs2,       \
-                  CPURISCVState *env, uint32_t desc)   \
-{                                                      \
-    uint32_t vm = vext_vm(desc);                       \
-    uint32_t vl = env->vl;                             \
-    uint32_t total_elems =                             \
-        vext_get_total_elems(env, desc, ESZ);          \
-    uint32_t vta = vext_vta(desc);                     \
-    uint32_t vma = vext_vma(desc);                     \
-    uint32_t i;                                        \
-                                                       \
-    for (i = env->vstart; i < vl; i++) {               \
-        if (!vm && !vext_elem_mask(v0, i)) {           \
-            /* set masked-off elements to 1s */        \
-            vext_set_elems_1s(vd, vma, i * ESZ,        \
-                              (i + 1) * ESZ);          \
-            continue;                                  \
-        }                                              \
-        do_##NAME(vd, vs2, i);                         \
-    }                                                  \
-    env->vstart = 0;                                   \
-    /* set tail elements to 1s */                      \
-    vext_set_elems_1s(vd, vta, vl * ESZ,               \
-                      total_elems * ESZ);              \
-}
-
 target_ulong fclass_h(uint64_t frs1)
 {
     float16 f = frs1;
-- 
2.41.0

From: Dickon Hood <dickon.hood@codethink.co.uk>

This commit adds support for the Zvbb vector-crypto extension, which
consists of the following instructions:

* vrol.[vv,vx]
* vror.[vv,vx,vi]
* vbrev8.v
* vrev8.v
* vandn.[vv,vx]
* vbrev.v
* vclz.v
* vctz.v
* vcpop.v
* vwsll.[vv,vx,vi]

Translation functions are defined in
`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
`target/riscv/vcrypto_helper.c`.

Co-authored-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Co-authored-by: William Salmon <will.salmon@codethink.co.uk>
Co-authored-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
[max.chou@sifive.com: Fix imm mode of vror.vi]
Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Signed-off-by: William Salmon <will.salmon@codethink.co.uk>
Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
Signed-off-by: Dickon Hood <dickon.hood@codethink.co.uk>
Signed-off-by: Max Chou <max.chou@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
[max.chou@sifive.com: Exposed x-zvbb property]
Message-ID: <20230711165917.2629866-9-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu_cfg.h                   |   1 +
 target/riscv/helper.h                    |  62 +++++++++
 target/riscv/insn32.decode               |  20 +++
 target/riscv/cpu.c                       |  12 ++
 target/riscv/vcrypto_helper.c            | 138 +++++++++++++++++++
 target/riscv/insn_trans/trans_rvvk.c.inc | 164 +++++++++++++++++++++++
 6 files changed, 397 insertions(+)

diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu_cfg.h
+++ b/target/riscv/cpu_cfg.h
@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
     bool ext_zve32f;
     bool ext_zve64f;
     bool ext_zve64d;
+    bool ext_zvbb;
     bool ext_zvbc;
     bool ext_zmmul;
     bool ext_zvfbfmin;
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -XXX,XX +XXX,XX @@ DEF_HELPER_6(vclmul_vv, void, ptr, ptr, ptr, ptr, env, i32)
 DEF_HELPER_6(vclmul_vx, void, ptr, ptr, tl, ptr, env, i32)
 DEF_HELPER_6(vclmulh_vv, void, ptr, ptr, ptr, ptr, env, i32)
 DEF_HELPER_6(vclmulh_vx, void, ptr, ptr, tl, ptr, env, i32)
+
+DEF_HELPER_6(vror_vv_b, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vror_vv_h, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vror_vv_w, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vror_vv_d, void, ptr, ptr, ptr, ptr, env, i32)
+
+DEF_HELPER_6(vror_vx_b, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vror_vx_h, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vror_vx_w, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vror_vx_d, void, ptr, ptr, tl, ptr, env, i32)
+
+DEF_HELPER_6(vrol_vv_b, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vrol_vv_h, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vrol_vv_w, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vrol_vv_d, void, ptr, ptr, ptr, ptr, env, i32)
+
+DEF_HELPER_6(vrol_vx_b, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vrol_vx_h, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vrol_vx_w, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vrol_vx_d, void, ptr, ptr, tl, ptr, env, i32)
+
+DEF_HELPER_5(vrev8_v_b, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vrev8_v_h, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vrev8_v_w, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vrev8_v_d, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vbrev8_v_b, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vbrev8_v_h, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vbrev8_v_w, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vbrev8_v_d, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vbrev_v_b, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vbrev_v_h, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vbrev_v_w, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vbrev_v_d, void, ptr, ptr, ptr, env, i32)
+
+DEF_HELPER_5(vclz_v_b, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vclz_v_h, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vclz_v_w, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vclz_v_d, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vctz_v_b, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vctz_v_h, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vctz_v_w, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vctz_v_d, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vcpop_v_b, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vcpop_v_h, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vcpop_v_w, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vcpop_v_d, void, ptr, ptr, ptr, env, i32)
+
+DEF_HELPER_6(vwsll_vv_b, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vwsll_vv_h, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vwsll_vv_w, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vwsll_vx_b, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vwsll_vx_h, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vwsll_vx_w, void, ptr, ptr, tl, ptr, env, i32)
+
+DEF_HELPER_6(vandn_vv_b, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vandn_vv_h, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vandn_vv_w, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vandn_vv_d, void, ptr, ptr, ptr, ptr, env, i32)
+DEF_HELPER_6(vandn_vx_b, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vandn_vx_h, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vandn_vx_w, void, ptr, ptr, tl, ptr, env, i32)
+DEF_HELPER_6(vandn_vx_d, void, ptr, ptr, tl, ptr, env, i32)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -XXX,XX +XXX,XX @@
 %imm_u    12:s20                 !function=ex_shift_12
 %imm_bs   30:2                   !function=ex_shift_3
 %imm_rnum 20:4
+%imm_z6   26:1 15:5
 
 # Argument sets:
 &empty
@@ -XXX,XX +XXX,XX @@
 @r_vm    ...... vm:1 ..... ..... ... ..... ....... &rmrr %rs2 %rs1 %rd
 @r_vm_1  ...... . ..... ..... ... ..... .......    &rmrr vm=1 %rs2 %rs1 %rd
 @r_vm_0  ...... . ..... ..... ... ..... .......    &rmrr vm=0 %rs2 %rs1 %rd
+@r2_zimm6  ..... . vm:1 ..... ..... ... ..... .......  &rmrr %rs2 rs1=%imm_z6 %rd
 @r2_zimm11 . zimm:11  ..... ... ..... ....... %rs1 %rd
 @r2_zimm10 .. zimm:10  ..... ... ..... ....... %rs1 %rd
 @r2_s    .......   ..... ..... ... ..... ....... %rs2 %rs1
@@ -XXX,XX +XXX,XX @@ vclmul_vv   001100 . ..... ..... 010 ..... 1010111 @r_vm
 vclmul_vx   001100 . ..... ..... 110 ..... 1010111 @r_vm
 vclmulh_vv  001101 . ..... ..... 010 ..... 1010111 @r_vm
 vclmulh_vx  001101 . ..... ..... 110 ..... 1010111 @r_vm
+
+# *** Zvbb vector crypto extension ***
+vrol_vv     010101 . ..... ..... 000 ..... 1010111 @r_vm
+vrol_vx     010101 . ..... ..... 100 ..... 1010111 @r_vm
+vror_vv     010100 . ..... ..... 000 ..... 1010111 @r_vm
+vror_vx     010100 . ..... ..... 100 ..... 1010111 @r_vm
+vror_vi     01010. . ..... ..... 011 ..... 1010111 @r2_zimm6
+vbrev8_v    010010 . ..... 01000 010 ..... 1010111 @r2_vm
+vrev8_v     010010 . ..... 01001 010 ..... 1010111 @r2_vm
+vandn_vv    000001 . ..... ..... 000 ..... 1010111 @r_vm
+vandn_vx    000001 . ..... ..... 100 ..... 1010111 @r_vm
+vbrev_v     010010 . ..... 01010 010 ..... 1010111 @r2_vm
+vclz_v      010010 . ..... 01100 010 ..... 1010111 @r2_vm
+vctz_v      010010 . ..... 01101 010 ..... 1010111 @r2_vm
+vcpop_v     010010 . ..... 01110 010 ..... 1010111 @r2_vm
+vwsll_vv    110101 . ..... ..... 000 ..... 1010111 @r_vm
+vwsll_vx    110101 . ..... ..... 100 ..... 1010111 @r_vm
+vwsll_vi    110101 . ..... ..... 011 ..... 1010111 @r_vm
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
     ISA_EXT_DATA_ENTRY(zksed, PRIV_VERSION_1_12_0, ext_zksed),
     ISA_EXT_DATA_ENTRY(zksh, PRIV_VERSION_1_12_0, ext_zksh),
     ISA_EXT_DATA_ENTRY(zkt, PRIV_VERSION_1_12_0, ext_zkt),
+    ISA_EXT_DATA_ENTRY(zvbb, PRIV_VERSION_1_12_0, ext_zvbb),
     ISA_EXT_DATA_ENTRY(zvbc, PRIV_VERSION_1_12_0, ext_zvbc),
     ISA_EXT_DATA_ENTRY(zve32f, PRIV_VERSION_1_10_0, ext_zve32f),
     ISA_EXT_DATA_ENTRY(zve64f, PRIV_VERSION_1_10_0, ext_zve64f),
@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
         return;
     }
 
+    /*
+     * In principle Zve*x would also suffice here, were they supported
+     * in qemu
+     */
+    if (cpu->cfg.ext_zvbb && !cpu->cfg.ext_zve32f) {
+        error_setg(errp,
+                   "Vector crypto extensions require V or Zve* extensions");
+        return;
+    }
+
     if (cpu->cfg.ext_zvbc && !cpu->cfg.ext_zve64f) {
         error_setg(errp, "Zvbc extension requires V or Zve64{f,d} extensions");
         return;
@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
     DEFINE_PROP_BOOL("x-zvfbfwma", RISCVCPU, cfg.ext_zvfbfwma, false),
 
     /* Vector cryptography extensions */
+    DEFINE_PROP_BOOL("x-zvbb", RISCVCPU, cfg.ext_zvbb, false),
     DEFINE_PROP_BOOL("x-zvbc", RISCVCPU, cfg.ext_zvbc, false),
 
     DEFINE_PROP_END_OF_LIST(),
diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/vcrypto_helper.c
+++ b/target/riscv/vcrypto_helper.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/osdep.h"
 #include "qemu/host-utils.h"
 #include "qemu/bitops.h"
+#include "qemu/bswap.h"
 #include "cpu.h"
 #include "exec/memop.h"
 #include "exec/exec-all.h"
@@ -XXX,XX +XXX,XX @@ RVVCALL(OPIVV2, vclmulh_vv, OP_UUU_D, H8, H8, H8, clmulh64)
 GEN_VEXT_VV(vclmulh_vv, 8)
 RVVCALL(OPIVX2, vclmulh_vx, OP_UUU_D, H8, H8, clmulh64)
 GEN_VEXT_VX(vclmulh_vx, 8)
+
+RVVCALL(OPIVV2, vror_vv_b, OP_UUU_B, H1, H1, H1, ror8)
+RVVCALL(OPIVV2, vror_vv_h, OP_UUU_H, H2, H2, H2, ror16)
+RVVCALL(OPIVV2, vror_vv_w, OP_UUU_W, H4, H4, H4, ror32)
+RVVCALL(OPIVV2, vror_vv_d, OP_UUU_D, H8, H8, H8, ror64)
+GEN_VEXT_VV(vror_vv_b, 1)
+GEN_VEXT_VV(vror_vv_h, 2)
+GEN_VEXT_VV(vror_vv_w, 4)
+GEN_VEXT_VV(vror_vv_d, 8)
+
+RVVCALL(OPIVX2, vror_vx_b, OP_UUU_B, H1, H1, ror8)
+RVVCALL(OPIVX2, vror_vx_h, OP_UUU_H, H2, H2, ror16)
+RVVCALL(OPIVX2, vror_vx_w, OP_UUU_W, H4, H4, ror32)
+RVVCALL(OPIVX2, vror_vx_d, OP_UUU_D, H8, H8, ror64)
+GEN_VEXT_VX(vror_vx_b, 1)
+GEN_VEXT_VX(vror_vx_h, 2)
+GEN_VEXT_VX(vror_vx_w, 4)
+GEN_VEXT_VX(vror_vx_d, 8)
+
+RVVCALL(OPIVV2, vrol_vv_b, OP_UUU_B, H1, H1, H1, rol8)
+RVVCALL(OPIVV2, vrol_vv_h, OP_UUU_H, H2, H2, H2, rol16)
+RVVCALL(OPIVV2, vrol_vv_w, OP_UUU_W, H4, H4, H4, rol32)
+RVVCALL(OPIVV2, vrol_vv_d, OP_UUU_D, H8, H8, H8, rol64)
+GEN_VEXT_VV(vrol_vv_b, 1)
+GEN_VEXT_VV(vrol_vv_h, 2)
+GEN_VEXT_VV(vrol_vv_w, 4)
+GEN_VEXT_VV(vrol_vv_d, 8)
+
+RVVCALL(OPIVX2, vrol_vx_b, OP_UUU_B, H1, H1, rol8)
+RVVCALL(OPIVX2, vrol_vx_h, OP_UUU_H, H2, H2, rol16)
+RVVCALL(OPIVX2, vrol_vx_w, OP_UUU_W, H4, H4, rol32)
+RVVCALL(OPIVX2, vrol_vx_d, OP_UUU_D, H8, H8, rol64)
+GEN_VEXT_VX(vrol_vx_b, 1)
+GEN_VEXT_VX(vrol_vx_h, 2)
+GEN_VEXT_VX(vrol_vx_w, 4)
+GEN_VEXT_VX(vrol_vx_d, 8)
+
+static uint64_t brev8(uint64_t val)
+{
+    val = ((val & 0x5555555555555555ull) << 1) |
+          ((val & 0xAAAAAAAAAAAAAAAAull) >> 1);
+    val = ((val & 0x3333333333333333ull) << 2) |
+          ((val & 0xCCCCCCCCCCCCCCCCull) >> 2);
+    val = ((val & 0x0F0F0F0F0F0F0F0Full) << 4) |
+          ((val & 0xF0F0F0F0F0F0F0F0ull) >> 4);
+
+    return val;
+}
+
+RVVCALL(OPIVV1, vbrev8_v_b, OP_UU_B, H1, H1, brev8)
+RVVCALL(OPIVV1, vbrev8_v_h, OP_UU_H, H2, H2, brev8)
+RVVCALL(OPIVV1, vbrev8_v_w, OP_UU_W, H4, H4, brev8)
+RVVCALL(OPIVV1, vbrev8_v_d, OP_UU_D, H8, H8, brev8)
+GEN_VEXT_V(vbrev8_v_b, 1)
+GEN_VEXT_V(vbrev8_v_h, 2)
+GEN_VEXT_V(vbrev8_v_w, 4)
+GEN_VEXT_V(vbrev8_v_d, 8)
+
+#define DO_IDENTITY(a) (a)
+RVVCALL(OPIVV1, vrev8_v_b, OP_UU_B, H1, H1, DO_IDENTITY)
+RVVCALL(OPIVV1, vrev8_v_h, OP_UU_H, H2, H2, bswap16)
+RVVCALL(OPIVV1, vrev8_v_w, OP_UU_W, H4, H4, bswap32)
+RVVCALL(OPIVV1, vrev8_v_d, OP_UU_D, H8, H8, bswap64)
+GEN_VEXT_V(vrev8_v_b, 1)
+GEN_VEXT_V(vrev8_v_h, 2)
+GEN_VEXT_V(vrev8_v_w, 4)
+GEN_VEXT_V(vrev8_v_d, 8)
+
+#define DO_ANDN(a, b) ((a) & ~(b))
+RVVCALL(OPIVV2, vandn_vv_b, OP_UUU_B, H1, H1, H1, DO_ANDN)
+RVVCALL(OPIVV2, vandn_vv_h, OP_UUU_H, H2, H2, H2, DO_ANDN)
+RVVCALL(OPIVV2, vandn_vv_w, OP_UUU_W, H4, H4, H4, DO_ANDN)
+RVVCALL(OPIVV2, vandn_vv_d, OP_UUU_D, H8, H8, H8, DO_ANDN)
+GEN_VEXT_VV(vandn_vv_b, 1)
+GEN_VEXT_VV(vandn_vv_h, 2)
+GEN_VEXT_VV(vandn_vv_w, 4)
+GEN_VEXT_VV(vandn_vv_d, 8)
+
+RVVCALL(OPIVX2, vandn_vx_b, OP_UUU_B, H1, H1, DO_ANDN)
+RVVCALL(OPIVX2, vandn_vx_h, OP_UUU_H, H2, H2, DO_ANDN)
+RVVCALL(OPIVX2, vandn_vx_w, OP_UUU_W, H4, H4, DO_ANDN)
+RVVCALL(OPIVX2, vandn_vx_d, OP_UUU_D, H8, H8, DO_ANDN)
+GEN_VEXT_VX(vandn_vx_b, 1)
+GEN_VEXT_VX(vandn_vx_h, 2)
+GEN_VEXT_VX(vandn_vx_w, 4)
+GEN_VEXT_VX(vandn_vx_d, 8)
+
+RVVCALL(OPIVV1, vbrev_v_b, OP_UU_B, H1, H1, revbit8)
+RVVCALL(OPIVV1, vbrev_v_h, OP_UU_H, H2, H2, revbit16)
+RVVCALL(OPIVV1, vbrev_v_w, OP_UU_W, H4, H4, revbit32)
+RVVCALL(OPIVV1, vbrev_v_d, OP_UU_D, H8, H8, revbit64)
+GEN_VEXT_V(vbrev_v_b, 1)
+GEN_VEXT_V(vbrev_v_h, 2)
+GEN_VEXT_V(vbrev_v_w, 4)
+GEN_VEXT_V(vbrev_v_d, 8)
+
+RVVCALL(OPIVV1, vclz_v_b, OP_UU_B, H1, H1, clz8)
+RVVCALL(OPIVV1, vclz_v_h, OP_UU_H, H2, H2, clz16)
+RVVCALL(OPIVV1, vclz_v_w, OP_UU_W, H4, H4, clz32)
+RVVCALL(OPIVV1, vclz_v_d, OP_UU_D, H8, H8, clz64)
+GEN_VEXT_V(vclz_v_b, 1)
+GEN_VEXT_V(vclz_v_h, 2)
+GEN_VEXT_V(vclz_v_w, 4)
+GEN_VEXT_V(vclz_v_d, 8)
+
+RVVCALL(OPIVV1, vctz_v_b, OP_UU_B, H1, H1, ctz8)
+RVVCALL(OPIVV1, vctz_v_h, OP_UU_H, H2, H2, ctz16)
+RVVCALL(OPIVV1, vctz_v_w, OP_UU_W, H4, H4, ctz32)
+RVVCALL(OPIVV1, vctz_v_d, OP_UU_D, H8, H8, ctz64)
+GEN_VEXT_V(vctz_v_b, 1)
+GEN_VEXT_V(vctz_v_h, 2)
+GEN_VEXT_V(vctz_v_w, 4)
+GEN_VEXT_V(vctz_v_d, 8)
+
+RVVCALL(OPIVV1, vcpop_v_b, OP_UU_B, H1, H1, ctpop8)
+RVVCALL(OPIVV1, vcpop_v_h, OP_UU_H, H2, H2, ctpop16)
+RVVCALL(OPIVV1, vcpop_v_w, OP_UU_W, H4, H4, ctpop32)
+RVVCALL(OPIVV1, vcpop_v_d, OP_UU_D, H8, H8, ctpop64)
+GEN_VEXT_V(vcpop_v_b, 1)
+GEN_VEXT_V(vcpop_v_h, 2)
+GEN_VEXT_V(vcpop_v_w, 4)
+GEN_VEXT_V(vcpop_v_d, 8)
+
+#define DO_SLL(N, M) (N << (M & (sizeof(N) * 8 - 1)))
+RVVCALL(OPIVV2, vwsll_vv_b, WOP_UUU_B, H2, H1, H1, DO_SLL)
+RVVCALL(OPIVV2, vwsll_vv_h, WOP_UUU_H, H4, H2, H2, DO_SLL)
+RVVCALL(OPIVV2, vwsll_vv_w, WOP_UUU_W, H8, H4, H4, DO_SLL)
+GEN_VEXT_VV(vwsll_vv_b, 2)
+GEN_VEXT_VV(vwsll_vv_h, 4)
+GEN_VEXT_VV(vwsll_vv_w, 8)
+
+RVVCALL(OPIVX2, vwsll_vx_b, WOP_UUU_B, H2, H1, DO_SLL)
+RVVCALL(OPIVX2, vwsll_vx_h, WOP_UUU_H, H4, H2, DO_SLL)
+RVVCALL(OPIVX2, vwsll_vx_w, WOP_UUU_W, H8, H4, DO_SLL)
+GEN_VEXT_VX(vwsll_vx_b, 2)
+GEN_VEXT_VX(vwsll_vx_h, 4)
+GEN_VEXT_VX(vwsll_vx_w, 8)
diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/insn_trans/trans_rvvk.c.inc
+++ b/target/riscv/insn_trans/trans_rvvk.c.inc
@@ -XXX,XX +XXX,XX @@ static bool vclmul_vx_check(DisasContext *s, arg_rmrr *a)
 
 GEN_VX_MASKED_TRANS(vclmul_vx, vclmul_vx_check)
 GEN_VX_MASKED_TRANS(vclmulh_vx, vclmul_vx_check)
+
+/*
+ * Zvbb
+ */
+
+#define GEN_OPIVI_GVEC_TRANS_CHECK(NAME, IMM_MODE, OPIVX, SUF, CHECK)   \
+    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)              \
+    {                                                                   \
+        if (CHECK(s, a)) {                                              \
+            static gen_helper_opivx *const fns[4] = {                   \
+                gen_helper_##OPIVX##_b,                                 \
+                gen_helper_##OPIVX##_h,                                 \
+                gen_helper_##OPIVX##_w,                                 \
+                gen_helper_##OPIVX##_d,                                 \
+            };                                                          \
+            return do_opivi_gvec(s, a, tcg_gen_gvec_##SUF, fns[s->sew], \
+                                 IMM_MODE);                             \
+        }                                                               \
+        return false;                                                   \
+    }
+
+#define GEN_OPIVV_GVEC_TRANS_CHECK(NAME, SUF, CHECK)                     \
+    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)               \
+    {                                                                    \
+        if (CHECK(s, a)) {                                               \
+            static gen_helper_gvec_4_ptr *const fns[4] = {               \
+                gen_helper_##NAME##_b,                                   \
+                gen_helper_##NAME##_h,                                   \
+                gen_helper_##NAME##_w,                                   \
+                gen_helper_##NAME##_d,                                   \
+            };                                                           \
+            return do_opivv_gvec(s, a, tcg_gen_gvec_##SUF, fns[s->sew]); \
+        }                                                                \
+        return false;                                                    \
+    }
+
+#define GEN_OPIVX_GVEC_SHIFT_TRANS_CHECK(NAME, SUF, CHECK)       \
+    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)       \
+    {                                                            \
+        if (CHECK(s, a)) {                                       \
+            static gen_helper_opivx *const fns[4] = {            \
+                gen_helper_##NAME##_b,                           \
+                gen_helper_##NAME##_h,                           \
+                gen_helper_##NAME##_w,                           \
+                gen_helper_##NAME##_d,                           \
+            };                                                   \
+            return do_opivx_gvec_shift(s, a, tcg_gen_gvec_##SUF, \
+                                       fns[s->sew]);             \
+        }                                                        \
+        return false;                                            \
+    }
+
+static bool zvbb_vv_check(DisasContext *s, arg_rmrr *a)
+{
+    return opivv_check(s, a) && s->cfg_ptr->ext_zvbb == true;
+}
+
+static bool zvbb_vx_check(DisasContext *s, arg_rmrr *a)
+{
+    return opivx_check(s, a) && s->cfg_ptr->ext_zvbb == true;
+}
+
+/* vrol.v[vx] */
+GEN_OPIVV_GVEC_TRANS_CHECK(vrol_vv, rotlv, zvbb_vv_check)
+GEN_OPIVX_GVEC_SHIFT_TRANS_CHECK(vrol_vx, rotls, zvbb_vx_check)
+
+/* vror.v[vxi] */
+GEN_OPIVV_GVEC_TRANS_CHECK(vror_vv, rotrv, zvbb_vv_check)
+GEN_OPIVX_GVEC_SHIFT_TRANS_CHECK(vror_vx, rotrs, zvbb_vx_check)
+GEN_OPIVI_GVEC_TRANS_CHECK(vror_vi, IMM_TRUNC_SEW, vror_vx, rotri, zvbb_vx_check)
+
+#define GEN_OPIVX_GVEC_TRANS_CHECK(NAME, SUF, CHECK)                     \
+    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)               \
+    {                                                                    \
+        if (CHECK(s, a)) {                                               \
+            static gen_helper_opivx *const fns[4] = {                    \
+                gen_helper_##NAME##_b,                                   \
+                gen_helper_##NAME##_h,                                   \
+                gen_helper_##NAME##_w,                                   \
+                gen_helper_##NAME##_d,                                   \
+            };                                                           \
+            return do_opivx_gvec(s, a, tcg_gen_gvec_##SUF, fns[s->sew]); \
+        }                                                                \
+        return false;                                                    \
+    }
+
+/* vandn.v[vx] */
+GEN_OPIVV_GVEC_TRANS_CHECK(vandn_vv, andc, zvbb_vv_check)
+GEN_OPIVX_GVEC_TRANS_CHECK(vandn_vx, andcs, zvbb_vx_check)
+
+#define GEN_OPIV_TRANS(NAME, CHECK)                                        \
+    static bool trans_##NAME(DisasContext *s, arg_rmr *a)                  \
+    {                                                                      \
+        if (CHECK(s, a)) {                                                 \
+            uint32_t data = 0;                                             \
+            static gen_helper_gvec_3_ptr *const fns[4] = {                 \
+                gen_helper_##NAME##_b,                                     \
+                gen_helper_##NAME##_h,                                     \
+                gen_helper_##NAME##_w,                                     \
+                gen_helper_##NAME##_d,                                     \
+            };                                                             \
+            TCGLabel *over = gen_new_label();                              \
+            tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);     \
+                                                                           \
+            data = FIELD_DP32(data, VDATA, VM, a->vm);                     \
+            data = FIELD_DP32(data, VDATA, LMUL, s->lmul);                 \
+            data = FIELD_DP32(data, VDATA, VTA, s->vta);                   \
+            data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s); \
+            data = FIELD_DP32(data, VDATA, VMA, s->vma);                   \
+            tcg_gen_gvec_3_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, 0),         \
+                               vreg_ofs(s, a->rs2), cpu_env,               \
+                               s->cfg_ptr->vlen / 8, s->cfg_ptr->vlen / 8, \
+                               data, fns[s->sew]);                         \
+            mark_vs_dirty(s);                                              \
+            gen_set_label(over);                                           \
+            return true;                                                   \
+        }                                                                  \
+        return false;                                                      \
+    }
+
+static bool zvbb_opiv_check(DisasContext *s, arg_rmr *a)
+{
+    return s->cfg_ptr->ext_zvbb == true &&
+           require_rvv(s) &&
+           vext_check_isa_ill(s) &&
+           vext_check_ss(s, a->rd, a->rs2, a->vm);
+}
+
+GEN_OPIV_TRANS(vbrev8_v, zvbb_opiv_check)
+GEN_OPIV_TRANS(vrev8_v, zvbb_opiv_check)
+GEN_OPIV_TRANS(vbrev_v, zvbb_opiv_check)
+GEN_OPIV_TRANS(vclz_v, zvbb_opiv_check)
+GEN_OPIV_TRANS(vctz_v, zvbb_opiv_check)
+GEN_OPIV_TRANS(vcpop_v, zvbb_opiv_check)
+
+static bool vwsll_vv_check(DisasContext *s, arg_rmrr *a)
+{
+    return s->cfg_ptr->ext_zvbb && opivv_widen_check(s, a);
+}
+
+static bool vwsll_vx_check(DisasContext *s, arg_rmrr *a)
+{
+    return s->cfg_ptr->ext_zvbb && opivx_widen_check(s, a);
+}
+
+/* OPIVI without GVEC IR */
+#define GEN_OPIVI_WIDEN_TRANS(NAME, IMM_MODE, OPIVX, CHECK)                  \
+    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)                   \
+    {                                                                        \
+        if (CHECK(s, a)) {                                                   \
+            static gen_helper_opivx *const fns[3] = {                        \
+                gen_helper_##OPIVX##_b,                                      \
+                gen_helper_##OPIVX##_h,                                      \
+                gen_helper_##OPIVX##_w,                                      \
+            };                                                               \
+            return opivi_trans(a->rd, a->rs1, a->rs2, a->vm, fns[s->sew], s, \
+                               IMM_MODE);                                    \
+        }                                                                    \
+        return false;                                                        \
+    }
+
+GEN_OPIVV_WIDEN_TRANS(vwsll_vv, vwsll_vv_check)
+GEN_OPIVX_WIDEN_TRANS(vwsll_vx, vwsll_vx_check)
+GEN_OPIVI_WIDEN_TRANS(vwsll_vi, IMM_ZX, vwsll_vx, vwsll_vx_check)
-- 
2.41.0

From: Nazar Kazakov <nazar.kazakov@codethink.co.uk>

This commit adds support for the Zvkned vector-crypto extension, which
consists of the following instructions:

* vaesef.[vv,vs]
* vaesdf.[vv,vs]
* vaesdm.[vv,vs]
* vaesz.vs
* vaesem.[vv,vs]
* vaeskf1.vi
* vaeskf2.vi

Translation functions are defined in
`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
`target/riscv/vcrypto_helper.c`.

Co-authored-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
Co-authored-by: William Salmon <will.salmon@codethink.co.uk>
[max.chou@sifive.com: Replaced vstart checking by TCG op]
Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
Signed-off-by: William Salmon <will.salmon@codethink.co.uk>
Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Signed-off-by: Max Chou <max.chou@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
[max.chou@sifive.com: Imported aes-round.h and exposed x-zvkned
property]
[max.chou@sifive.com: Fixed endian issues and replaced the vstart & vl
egs checking by helper function]
[max.chou@sifive.com: Replaced bswap32 calls in aes key expanding]
Message-ID: <20230711165917.2629866-10-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu_cfg.h                   |   1 +
 target/riscv/helper.h                    |  14 ++
 target/riscv/insn32.decode               |  14 ++
 target/riscv/cpu.c                       |   4 +-
 target/riscv/vcrypto_helper.c            | 202 +++++++++++++++++++++++
 target/riscv/insn_trans/trans_rvvk.c.inc | 147 +++++++++++++++++
 6 files changed, 381 insertions(+), 1 deletion(-)

diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu_cfg.h
+++ b/target/riscv/cpu_cfg.h
@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
     bool ext_zve64d;
     bool ext_zvbb;
     bool ext_zvbc;
+    bool ext_zvkned;
     bool ext_zmmul;
     bool ext_zvfbfmin;
     bool ext_zvfbfwma;
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -XXX,XX +XXX,XX @@ DEF_HELPER_6(vandn_vx_b, void, ptr, ptr, tl, ptr, env, i32)
 DEF_HELPER_6(vandn_vx_h, void, ptr, ptr, tl, ptr, env, i32)
 DEF_HELPER_6(vandn_vx_w, void, ptr, ptr, tl, ptr, env, i32)
 DEF_HELPER_6(vandn_vx_d, void, ptr, ptr, tl, ptr, env, i32)
+
+DEF_HELPER_2(egs_check, void, i32, env)
+
+DEF_HELPER_4(vaesef_vv, void, ptr, ptr, env, i32)
+DEF_HELPER_4(vaesef_vs, void, ptr, ptr, env, i32)
+DEF_HELPER_4(vaesdf_vv, void, ptr, ptr, env, i32)
+DEF_HELPER_4(vaesdf_vs, void, ptr, ptr, env, i32)
+DEF_HELPER_4(vaesem_vv, void, ptr, ptr, env, i32)
+DEF_HELPER_4(vaesem_vs, void, ptr, ptr, env, i32)
+DEF_HELPER_4(vaesdm_vv, void, ptr, ptr, env, i32)
+DEF_HELPER_4(vaesdm_vs, void, ptr, ptr, env, i32)
+DEF_HELPER_4(vaesz_vs, void, ptr, ptr, env, i32)
+DEF_HELPER_5(vaeskf1_vi, void, ptr, ptr, i32, env, i32)
+DEF_HELPER_5(vaeskf2_vi, void, ptr, ptr, i32, env, i32)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -XXX,XX +XXX,XX @@
 @r_rm    .......   ..... ..... ... ..... ....... %rs2 %rs1 %rm %rd
 @r2_rm   .......   ..... ..... ... ..... ....... %rs1 %rm %rd
 @r2      .......   ..... ..... ... ..... ....... &r2 %rs1 %rd
+@r2_vm_1 ...... . ..... ..... ... ..... ....... &rmr vm=1 %rs2 %rd
 @r2_nfvm ... ... vm:1 ..... ..... ... ..... ....... &r2nfvm %nf %rs1 %rd
 @r2_vm   ...... vm:1 ..... ..... ... ..... ....... &rmr %rs2 %rd
 @r1_vm   ...... vm:1 ..... ..... ... ..... ....... %rd
@@ -XXX,XX +XXX,XX @@ vcpop_v     010010 . ..... 01110 010 ..... 1010111 @r2_vm
 vwsll_vv    110101 . ..... ..... 000 ..... 1010111 @r_vm
 vwsll_vx    110101 . ..... ..... 100 ..... 1010111 @r_vm
 vwsll_vi    110101 . ..... ..... 011 ..... 1010111 @r_vm
+
+# *** Zvkned vector crypto extension ***
+vaesef_vv   101000 1 ..... 00011 010 ..... 1110111 @r2_vm_1
+vaesef_vs   101001 1 ..... 00011 010 ..... 1110111 @r2_vm_1
+vaesdf_vv   101000 1 ..... 00001 010 ..... 1110111 @r2_vm_1
+vaesdf_vs   101001 1 ..... 00001 010 ..... 1110111 @r2_vm_1
+vaesem_vv   101000 1 ..... 00010 010 ..... 1110111 @r2_vm_1
+vaesem_vs   101001 1 ..... 00010 010 ..... 1110111 @r2_vm_1
+vaesdm_vv   101000 1 ..... 00000 010 ..... 1110111 @r2_vm_1
+vaesdm_vs   101001 1 ..... 00000 010 ..... 1110111 @r2_vm_1
+vaesz_vs    101001 1 ..... 00111 010 ..... 1110111 @r2_vm_1
+vaeskf1_vi  100010 1 ..... ..... 010 ..... 1110111 @r_vm_1
+vaeskf2_vi  101010 1 ..... ..... 010 ..... 1110111 @r_vm_1
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
     ISA_EXT_DATA_ENTRY(zvfbfwma, PRIV_VERSION_1_12_0, ext_zvfbfwma),
     ISA_EXT_DATA_ENTRY(zvfh, PRIV_VERSION_1_12_0, ext_zvfh),
     ISA_EXT_DATA_ENTRY(zvfhmin, PRIV_VERSION_1_12_0, ext_zvfhmin),
+    ISA_EXT_DATA_ENTRY(zvkned, PRIV_VERSION_1_12_0, ext_zvkned),
     ISA_EXT_DATA_ENTRY(zhinx, PRIV_VERSION_1_12_0, ext_zhinx),
     ISA_EXT_DATA_ENTRY(zhinxmin, PRIV_VERSION_1_12_0, ext_zhinxmin),
     ISA_EXT_DATA_ENTRY(smaia, PRIV_VERSION_1_12_0, ext_smaia),
@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
      * In principle Zve*x would also suffice here, were they supported
      * in qemu
      */
-    if (cpu->cfg.ext_zvbb && !cpu->cfg.ext_zve32f) {
+    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned) && !cpu->cfg.ext_zve32f) {
         error_setg(errp,
                    "Vector crypto extensions require V or Zve* extensions");
         return;
@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
     /* Vector cryptography extensions */
     DEFINE_PROP_BOOL("x-zvbb", RISCVCPU, cfg.ext_zvbb, false),
     DEFINE_PROP_BOOL("x-zvbc", RISCVCPU, cfg.ext_zvbc, false),
+    DEFINE_PROP_BOOL("x-zvkned", RISCVCPU, cfg.ext_zvkned, false),
 
     DEFINE_PROP_END_OF_LIST(),
 };
diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/vcrypto_helper.c
+++ b/target/riscv/vcrypto_helper.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/bitops.h"
 #include "qemu/bswap.h"
 #include "cpu.h"
+#include "crypto/aes.h"
+#include "crypto/aes-round.h"
 #include "exec/memop.h"
 #include "exec/exec-all.h"
 #include "exec/helper-proto.h"
@@ -XXX,XX +XXX,XX @@ RVVCALL(OPIVX2, vwsll_vx_w, WOP_UUU_W, H8, H4, DO_SLL)
 GEN_VEXT_VX(vwsll_vx_b, 2)
 GEN_VEXT_VX(vwsll_vx_h, 4)
 GEN_VEXT_VX(vwsll_vx_w, 8)
+
+void HELPER(egs_check)(uint32_t egs, CPURISCVState *env)
+{
+    uint32_t vl = env->vl;
+    uint32_t vstart = env->vstart;
+
+    if (vl % egs != 0 || vstart % egs != 0) {
+        riscv_raise_exception(env, RISCV_EXCP_ILLEGAL_INST, GETPC());
+    }
+}
+
+static inline void xor_round_key(AESState *round_state, AESState *round_key)
+{
+    round_state->v = round_state->v ^ round_key->v;
+}
+
+#define GEN_ZVKNED_HELPER_VV(NAME, ...)                                   \
+    void HELPER(NAME)(void *vd, void *vs2, CPURISCVState *env,            \
+                      uint32_t desc)                                      \
+    {                                                                     \
+        uint32_t vl = env->vl;                                            \
+        uint32_t total_elems = vext_get_total_elems(env, desc, 4);        \
+        uint32_t vta = vext_vta(desc);                                    \
+                                                                          \
+        for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {        \
+            AESState round_key;                                           \
+            round_key.d[0] = *((uint64_t *)vs2 + H8(i * 2 + 0));          \
+            round_key.d[1] = *((uint64_t *)vs2 + H8(i * 2 + 1));          \
+            AESState round_state;                                         \
+            round_state.d[0] = *((uint64_t *)vd + H8(i * 2 + 0));         \
+            round_state.d[1] = *((uint64_t *)vd + H8(i * 2 + 1));         \
+            __VA_ARGS__;                                                  \
+            *((uint64_t *)vd + H8(i * 2 + 0)) = round_state.d[0];         \
+            *((uint64_t *)vd + H8(i * 2 + 1)) = round_state.d[1];         \
+        }                                                                 \
+        env->vstart = 0;                                                  \
+        /* set tail elements to 1s */                                     \
+        vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);              \
+    }
+
+#define GEN_ZVKNED_HELPER_VS(NAME, ...)                                   \
+    void HELPER(NAME)(void *vd, void *vs2, CPURISCVState *env,            \
+                      uint32_t desc)                                      \
+    {                                                                     \
+        uint32_t vl = env->vl;                                            \
+        uint32_t total_elems = vext_get_total_elems(env, desc, 4);        \
+        uint32_t vta = vext_vta(desc);                                    \
+                                                                          \
+        for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {        \
+            AESState round_key;                                           \
+            round_key.d[0] = *((uint64_t *)vs2 + H8(0));                  \
+            round_key.d[1] = *((uint64_t *)vs2 + H8(1));                  \
+            AESState round_state;                                         \
+            round_state.d[0] = *((uint64_t *)vd + H8(i * 2 + 0));         \
+            round_state.d[1] = *((uint64_t *)vd + H8(i * 2 + 1));         \
+            __VA_ARGS__;                                                  \
+            *((uint64_t *)vd + H8(i * 2 + 0)) = round_state.d[0];         \
+            *((uint64_t *)vd + H8(i * 2 + 1)) = round_state.d[1];         \
+        }                                                                 \
+        env->vstart = 0;                                                  \
+        /* set tail elements to 1s */                                     \
+        vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);              \
+    }
+
+GEN_ZVKNED_HELPER_VV(vaesef_vv, aesenc_SB_SR_AK(&round_state,
+                                                &round_state,
+                                                &round_key,
+                                                false);)
+GEN_ZVKNED_HELPER_VS(vaesef_vs, aesenc_SB_SR_AK(&round_state,
+                                                &round_state,
+                                                &round_key,
+                                                false);)
+GEN_ZVKNED_HELPER_VV(vaesdf_vv, aesdec_ISB_ISR_AK(&round_state,
+                                                  &round_state,
+                                                  &round_key,
+                                                  false);)
+GEN_ZVKNED_HELPER_VS(vaesdf_vs, aesdec_ISB_ISR_AK(&round_state,
+                                                  &round_state,
+                                                  &round_key,
+                                                  false);)
+GEN_ZVKNED_HELPER_VV(vaesem_vv, aesenc_SB_SR_MC_AK(&round_state,
+                                                   &round_state,
+                                                   &round_key,
+                                                   false);)
+GEN_ZVKNED_HELPER_VS(vaesem_vs, aesenc_SB_SR_MC_AK(&round_state,
+                                                   &round_state,
+                                                   &round_key,
+                                                   false);)
+GEN_ZVKNED_HELPER_VV(vaesdm_vv, aesdec_ISB_ISR_AK_IMC(&round_state,
+                                                      &round_state,
+                                                      &round_key,
+                                                      false);)
+GEN_ZVKNED_HELPER_VS(vaesdm_vs, aesdec_ISB_ISR_AK_IMC(&round_state,
+                                                      &round_state,
+                                                      &round_key,
+                                                      false);)
+GEN_ZVKNED_HELPER_VS(vaesz_vs, xor_round_key(&round_state, &round_key);)
+
+void HELPER(vaeskf1_vi)(void *vd_vptr, void *vs2_vptr, uint32_t uimm,
+                        CPURISCVState *env, uint32_t desc)
+{
+    uint32_t *vd = vd_vptr;
+    uint32_t *vs2 = vs2_vptr;
+    uint32_t vl = env->vl;
+    uint32_t total_elems = vext_get_total_elems(env, desc, 4);
+    uint32_t vta = vext_vta(desc);
+
+    uimm &= 0b1111;
+    if (uimm > 10 || uimm == 0) {
+        uimm ^= 0b1000;
+    }
+
+    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
+        uint32_t rk[8], tmp;
+        static const uint32_t rcon[] = {
+            0x00000001, 0x00000002, 0x00000004, 0x00000008, 0x00000010,
+            0x00000020, 0x00000040, 0x00000080, 0x0000001B, 0x00000036,
+        };
+
+        rk[0] = vs2[i * 4 + H4(0)];
+        rk[1] = vs2[i * 4 + H4(1)];
+        rk[2] = vs2[i * 4 + H4(2)];
+        rk[3] = vs2[i * 4 + H4(3)];
+        tmp = ror32(rk[3], 8);
+
+        rk[4] = rk[0] ^ (((uint32_t)AES_sbox[(tmp >> 24) & 0xff] << 24) |
+                         ((uint32_t)AES_sbox[(tmp >> 16) & 0xff] << 16) |
+                         ((uint32_t)AES_sbox[(tmp >> 8) & 0xff] << 8) |
+                         ((uint32_t)AES_sbox[(tmp >> 0) & 0xff] << 0))
+                      ^ rcon[uimm - 1];
+        rk[5] = rk[1] ^ rk[4];
+        rk[6] = rk[2] ^ rk[5];
+        rk[7] = rk[3] ^ rk[6];
+
+        vd[i * 4 + H4(0)] = rk[4];
+        vd[i * 4 + H4(1)] = rk[5];
+        vd[i * 4 + H4(2)] = rk[6];
+        vd[i * 4 + H4(3)] = rk[7];
+    }
+    env->vstart = 0;
+    /* set tail elements to 1s */
+    vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);
+}
+
+void HELPER(vaeskf2_vi)(void *vd_vptr, void *vs2_vptr, uint32_t uimm,
+                        CPURISCVState *env, uint32_t desc)
+{
+    uint32_t *vd = vd_vptr;
+    uint32_t *vs2 = vs2_vptr;
+    uint32_t vl = env->vl;
+    uint32_t total_elems = vext_get_total_elems(env, desc, 4);
+    uint32_t vta = vext_vta(desc);
+
+    uimm &= 0b1111;
+    if (uimm > 14 || uimm < 2) {
+        uimm ^= 0b1000;
+    }
+
+    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
+        uint32_t rk[12], tmp;
+        static const uint32_t rcon[] = {
+            0x00000001, 0x00000002, 0x00000004, 0x00000008, 0x00000010,
+            0x00000020, 0x00000040, 0x00000080, 0x0000001B, 0x00000036,
+        };
+
+        rk[0] = vd[i * 4 + H4(0)];
+        rk[1] = vd[i * 4 + H4(1)];
+        rk[2] = vd[i * 4 + H4(2)];
+        rk[3] = vd[i * 4 + H4(3)];
+        rk[4] = vs2[i * 4 + H4(0)];
+        rk[5] = vs2[i * 4 + H4(1)];
+        rk[6] = vs2[i * 4 + H4(2)];
+        rk[7] = vs2[i * 4 + H4(3)];
+
+        if (uimm % 2 == 0) {
+            tmp = ror32(rk[7], 8);
+            rk[8] = rk[0] ^ (((uint32_t)AES_sbox[(tmp >> 24) & 0xff] << 24) |
+                             ((uint32_t)AES_sbox[(tmp >> 16) & 0xff] << 16) |
+                             ((uint32_t)AES_sbox[(tmp >> 8) & 0xff] << 8) |
+                             ((uint32_t)AES_sbox[(tmp >> 0) & 0xff] << 0))
+                          ^ rcon[(uimm - 1) / 2];
+        } else {
+            rk[8] = rk[0] ^ (((uint32_t)AES_sbox[(rk[7] >> 24) & 0xff] << 24) |
+                             ((uint32_t)AES_sbox[(rk[7] >> 16) & 0xff] << 16) |
+                             ((uint32_t)AES_sbox[(rk[7] >> 8) & 0xff] << 8) |
+                             ((uint32_t)AES_sbox[(rk[7] >> 0) & 0xff] << 0));
+        }
+        rk[9] = rk[1] ^ rk[8];
+        rk[10] = rk[2] ^ rk[9];
+        rk[11] = rk[3] ^ rk[10];
+
+        vd[i * 4 + H4(0)] = rk[8];
+        vd[i * 4 + H4(1)] = rk[9];
+        vd[i * 4 + H4(2)] = rk[10];
+        vd[i * 4 + H4(3)] = rk[11];
+    }
+    env->vstart = 0;
+    /* set tail elements to 1s */
+    vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);
+}
diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/insn_trans/trans_rvvk.c.inc
+++ b/target/riscv/insn_trans/trans_rvvk.c.inc
@@ -XXX,XX +XXX,XX @@ static bool vwsll_vx_check(DisasContext *s, arg_rmrr *a)
 GEN_OPIVV_WIDEN_TRANS(vwsll_vv, vwsll_vv_check)
 GEN_OPIVX_WIDEN_TRANS(vwsll_vx, vwsll_vx_check)
 GEN_OPIVI_WIDEN_TRANS(vwsll_vi, IMM_ZX, vwsll_vx, vwsll_vx_check)
+
+/*
+ * Zvkned
+ */
+
+#define ZVKNED_EGS 4
+
+#define GEN_V_UNMASKED_TRANS(NAME, CHECK, EGS)                                \
+    static bool trans_##NAME(DisasContext *s, arg_##NAME *a)                  \
+    {                                                                         \
+        if (CHECK(s, a)) {                                                    \
+            TCGv_ptr rd_v, rs2_v;                                             \
+            TCGv_i32 desc, egs;                                               \
+            uint32_t data = 0;                                                \
+            TCGLabel *over = gen_new_label();                                 \
+                                                                              \
+            if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {                      \
+                /* save opcode for unwinding in case we throw an exception */ \
+                decode_save_opc(s);                                           \
+                egs = tcg_constant_i32(EGS);                                  \
+                gen_helper_egs_check(egs, cpu_env);                           \
+                tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);    \
+            }                                                                 \
+                                                                              \
+            data = FIELD_DP32(data, VDATA, VM, a->vm);                        \
+            data = FIELD_DP32(data, VDATA, LMUL, s->lmul);                    \
+            data = FIELD_DP32(data, VDATA, VTA, s->vta);                      \
+            data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);    \
+            data = FIELD_DP32(data, VDATA, VMA, s->vma);                      \
+            rd_v = tcg_temp_new_ptr();                                        \
+            rs2_v = tcg_temp_new_ptr();                                       \
+            desc = tcg_constant_i32(                                          \
+                simd_desc(s->cfg_ptr->vlen / 8, s->cfg_ptr->vlen / 8, data)); \
+            tcg_gen_addi_ptr(rd_v, cpu_env, vreg_ofs(s, a->rd));              \
+            tcg_gen_addi_ptr(rs2_v, cpu_env, vreg_ofs(s, a->rs2));            \
+            gen_helper_##NAME(rd_v, rs2_v, cpu_env, desc);                    \
+            mark_vs_dirty(s);                                                 \
+            gen_set_label(over);                                              \
+            return true;                                                      \
+        }                                                                     \
+        return false;                                                         \
+    }
+
+static bool vaes_check_vv(DisasContext *s, arg_rmr *a)
+{
+    int egw_bytes = ZVKNED_EGS << s->sew;
+    return s->cfg_ptr->ext_zvkned == true &&
+           require_rvv(s) &&
+           vext_check_isa_ill(s) &&
+           MAXSZ(s) >= egw_bytes &&
+           require_align(a->rd, s->lmul) &&
+           require_align(a->rs2, s->lmul) &&
+           s->sew == MO_32;
+}
+
+static bool vaes_check_overlap(DisasContext *s, int vd, int vs2)
+{
+    int8_t op_size = s->lmul <= 0 ? 1 : 1 << s->lmul;
+    return !is_overlapped(vd, op_size, vs2, 1);
+}
+
+static bool vaes_check_vs(DisasContext *s, arg_rmr *a)
+{
+    int egw_bytes = ZVKNED_EGS << s->sew;
+    return vaes_check_overlap(s, a->rd, a->rs2) &&
+           MAXSZ(s) >= egw_bytes &&
+           s->cfg_ptr->ext_zvkned == true &&
+           require_rvv(s) &&
+           vext_check_isa_ill(s) &&
+           require_align(a->rd, s->lmul) &&
+           s->sew == MO_32;
+}
+
+GEN_V_UNMASKED_TRANS(vaesef_vv, vaes_check_vv, ZVKNED_EGS)
+GEN_V_UNMASKED_TRANS(vaesef_vs, vaes_check_vs, ZVKNED_EGS)
+GEN_V_UNMASKED_TRANS(vaesdf_vv, vaes_check_vv, ZVKNED_EGS)
+GEN_V_UNMASKED_TRANS(vaesdf_vs, vaes_check_vs, ZVKNED_EGS)
+GEN_V_UNMASKED_TRANS(vaesdm_vv, vaes_check_vv, ZVKNED_EGS)
+GEN_V_UNMASKED_TRANS(vaesdm_vs, vaes_check_vs, ZVKNED_EGS)
+GEN_V_UNMASKED_TRANS(vaesz_vs, vaes_check_vs, ZVKNED_EGS)
+GEN_V_UNMASKED_TRANS(vaesem_vv, vaes_check_vv, ZVKNED_EGS)
+GEN_V_UNMASKED_TRANS(vaesem_vs, vaes_check_vs, ZVKNED_EGS)
+
+#define GEN_VI_UNMASKED_TRANS(NAME, CHECK, EGS)                               \
+    static bool trans_##NAME(DisasContext *s, arg_##NAME *a)                  \
+    {                                                                         \
+        if (CHECK(s, a)) {                                                    \
+            TCGv_ptr rd_v, rs2_v;                                             \
+            TCGv_i32 uimm_v, desc, egs;                                       \
+            uint32_t data = 0;                                                \
+            TCGLabel *over = gen_new_label();                                 \
+                                                                              \
+            if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {                      \
+                /* save opcode for unwinding in case we throw an exception */ \
+                decode_save_opc(s);                                           \
+                egs = tcg_constant_i32(EGS);                                  \
+                gen_helper_egs_check(egs, cpu_env);                           \
+                tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);    \
+            }                                                                 \
+                                                                              \
+            data = FIELD_DP32(data, VDATA, VM, a->vm);                        \
+            data = FIELD_DP32(data, VDATA, LMUL, s->lmul);                    \
+            data = FIELD_DP32(data, VDATA, VTA, s->vta);                      \
+            data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);    \
+            data = FIELD_DP32(data, VDATA, VMA, s->vma);                      \
+                                                                              \
+            rd_v = tcg_temp_new_ptr();                                        \
+            rs2_v = tcg_temp_new_ptr();                                       \
+            uimm_v = tcg_constant_i32(a->rs1);                                \
+            desc = tcg_constant_i32(                                          \
+                simd_desc(s->cfg_ptr->vlen / 8, s->cfg_ptr->vlen / 8, data)); \
+            tcg_gen_addi_ptr(rd_v, cpu_env, vreg_ofs(s, a->rd));              \
+            tcg_gen_addi_ptr(rs2_v, cpu_env, vreg_ofs(s, a->rs2));            \
+            gen_helper_##NAME(rd_v, rs2_v, uimm_v, cpu_env, desc);            \
+            mark_vs_dirty(s);                                                 \
+            gen_set_label(over);                                              \
+            return true;                                                      \
+        }                                                                     \
+        return false;                                                         \
+    }
+
+static bool vaeskf1_check(DisasContext *s, arg_vaeskf1_vi *a)
+{
+    int egw_bytes = ZVKNED_EGS << s->sew;
+    return s->cfg_ptr->ext_zvkned == true &&
+           require_rvv(s) &&
+           vext_check_isa_ill(s) &&
+           MAXSZ(s) >= egw_bytes &&
+           s->sew == MO_32 &&
+           require_align(a->rd, s->lmul) &&
+           require_align(a->rs2, s->lmul);
+}
+
+static bool vaeskf2_check(DisasContext *s, arg_vaeskf2_vi *a)
+{
+    int egw_bytes = ZVKNED_EGS << s->sew;
+    return s->cfg_ptr->ext_zvkned == true &&
+           require_rvv(s) &&
+           vext_check_isa_ill(s) &&
+           MAXSZ(s) >= egw_bytes &&
+           s->sew == MO_32 &&
+           require_align(a->rd, s->lmul) &&
+           require_align(a->rs2, s->lmul);
+}
+
+GEN_VI_UNMASKED_TRANS(vaeskf1_vi, vaeskf1_check, ZVKNED_EGS)
+GEN_VI_UNMASKED_TRANS(vaeskf2_vi, vaeskf2_check, ZVKNED_EGS)
-- 
2.41.0

From: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>

This commit adds support for the Zvknh vector-crypto extension, which
consists of the following instructions:

* vsha2ms.vv
* vsha2c[hl].vv

Translation functions are defined in
`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
`target/riscv/vcrypto_helper.c`.

Co-authored-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Co-authored-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
[max.chou@sifive.com: Replaced vstart checking by TCG op]
Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
Signed-off-by: Max Chou <max.chou@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
[max.chou@sifive.com: Exposed x-zvknha & x-zvknhb properties]
[max.chou@sifive.com: Replaced SEW selection to happened during
translation]
Message-ID: <20230711165917.2629866-11-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu_cfg.h                   |   2 +
 target/riscv/helper.h                    |   6 +
 target/riscv/insn32.decode               |   5 +
 target/riscv/cpu.c                       |  13 +-
 target/riscv/vcrypto_helper.c            | 238 +++++++++++++++++++++++
 target/riscv/insn_trans/trans_rvvk.c.inc | 129 ++++++++++++
 6 files changed, 390 insertions(+), 3 deletions(-)

diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu_cfg.h
+++ b/target/riscv/cpu_cfg.h
@@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig {
     bool ext_zvbb;
     bool ext_zvbc;
     bool ext_zvkned;
+    bool ext_zvknha;
+    bool ext_zvknhb;
     bool ext_zmmul;
     bool ext_zvfbfmin;
     bool ext_zvfbfwma;
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -XXX,XX +XXX,XX @@ DEF_HELPER_4(vaesdm_vs, void, ptr, ptr, env, i32)
 DEF_HELPER_4(vaesz_vs, void, ptr, ptr, env, i32)
 DEF_HELPER_5(vaeskf1_vi, void, ptr, ptr, i32, env, i32)
 DEF_HELPER_5(vaeskf2_vi, void, ptr, ptr, i32, env, i32)
+
+DEF_HELPER_5(vsha2ms_vv, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vsha2ch32_vv, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vsha2ch64_vv, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vsha2cl32_vv, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vsha2cl64_vv, void, ptr, ptr, ptr, env, i32)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -XXX,XX +XXX,XX @@ vaesdm_vs   101001 1 ..... 00000 010 ..... 1110111 @r2_vm_1
 vaesz_vs    101001 1 ..... 00111 010 ..... 1110111 @r2_vm_1
 vaeskf1_vi  100010 1 ..... ..... 010 ..... 1110111 @r_vm_1
 vaeskf2_vi  101010 1 ..... ..... 010 ..... 1110111 @r_vm_1
+
+# *** Zvknh vector crypto extension ***
+vsha2ms_vv  101101 1 ..... ..... 010 ..... 1110111 @r_vm_1
+vsha2ch_vv  101110 1 ..... ..... 010 ..... 1110111 @r_vm_1
+vsha2cl_vv  101111 1 ..... ..... 010 ..... 1110111 @r_vm_1
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static const struct isa_ext_data isa_edata_arr[] = {
     ISA_EXT_DATA_ENTRY(zvfh, PRIV_VERSION_1_12_0, ext_zvfh),
     ISA_EXT_DATA_ENTRY(zvfhmin, PRIV_VERSION_1_12_0, ext_zvfhmin),
     ISA_EXT_DATA_ENTRY(zvkned, PRIV_VERSION_1_12_0, ext_zvkned),
+    ISA_EXT_DATA_ENTRY(zvknha, PRIV_VERSION_1_12_0, ext_zvknha),
+    ISA_EXT_DATA_ENTRY(zvknhb, PRIV_VERSION_1_12_0, ext_zvknhb),
     ISA_EXT_DATA_ENTRY(zhinx, PRIV_VERSION_1_12_0, ext_zhinx),
     ISA_EXT_DATA_ENTRY(zhinxmin, PRIV_VERSION_1_12_0, ext_zhinxmin),
     ISA_EXT_DATA_ENTRY(smaia, PRIV_VERSION_1_12_0, ext_smaia),
@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
      * In principle Zve*x would also suffice here, were they supported
      * in qemu
      */
-    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned) && !cpu->cfg.ext_zve32f) {
+    if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned || cpu->cfg.ext_zvknha) &&
+        !cpu->cfg.ext_zve32f) {
         error_setg(errp,
                    "Vector crypto extensions require V or Zve* extensions");
         return;
     }
 
-    if (cpu->cfg.ext_zvbc && !cpu->cfg.ext_zve64f) {
-        error_setg(errp, "Zvbc extension requires V or Zve64{f,d} extensions");
+    if ((cpu->cfg.ext_zvbc || cpu->cfg.ext_zvknhb) && !cpu->cfg.ext_zve64f) {
+        error_setg(
+            errp,
+            "Zvbc and Zvknhb extensions require V or Zve64{f,d} extensions");
         return;
     }
 
@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
     DEFINE_PROP_BOOL("x-zvbb", RISCVCPU, cfg.ext_zvbb, false),
     DEFINE_PROP_BOOL("x-zvbc", RISCVCPU, cfg.ext_zvbc, false),
     DEFINE_PROP_BOOL("x-zvkned", RISCVCPU, cfg.ext_zvkned, false),
+    DEFINE_PROP_BOOL("x-zvknha", RISCVCPU, cfg.ext_zvknha, false),
+    DEFINE_PROP_BOOL("x-zvknhb", RISCVCPU, cfg.ext_zvknhb, false),
 
     DEFINE_PROP_END_OF_LIST(),
 };
diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/vcrypto_helper.c
+++ b/target/riscv/vcrypto_helper.c
@@ -XXX,XX +XXX,XX @@ void HELPER(vaeskf2_vi)(void *vd_vptr, void *vs2_vptr, uint32_t uimm,
     /* set tail elements to 1s */
     vext_set_elems_1s(vd, vta, vl * 4, total_elems * 4);
 }
+
+static inline uint32_t sig0_sha256(uint32_t x)
+{
+    return ror32(x, 7) ^ ror32(x, 18) ^ (x >> 3);
+}
+
+static inline uint32_t sig1_sha256(uint32_t x)
+{
+    return ror32(x, 17) ^ ror32(x, 19) ^ (x >> 10);
+}
+
+static inline uint64_t sig0_sha512(uint64_t x)
+{
+    return ror64(x, 1) ^ ror64(x, 8) ^ (x >> 7);
+}
+
+static inline uint64_t sig1_sha512(uint64_t x)
+{
+    return ror64(x, 19) ^ ror64(x, 61) ^ (x >> 6);
+}
+
+static inline void vsha2ms_e32(uint32_t *vd, uint32_t *vs1, uint32_t *vs2)
+{
+    uint32_t res[4];
+    res[0] = sig1_sha256(vs1[H4(2)]) + vs2[H4(1)] + sig0_sha256(vd[H4(1)]) +
+             vd[H4(0)];
+    res[1] = sig1_sha256(vs1[H4(3)]) + vs2[H4(2)] + sig0_sha256(vd[H4(2)]) +
+             vd[H4(1)];
+    res[2] =
+        sig1_sha256(res[0]) + vs2[H4(3)] + sig0_sha256(vd[H4(3)]) + vd[H4(2)];
+    res[3] =
+        sig1_sha256(res[1]) + vs1[H4(0)] + sig0_sha256(vs2[H4(0)]) + vd[H4(3)];
+    vd[H4(3)] = res[3];
+    vd[H4(2)] = res[2];
+    vd[H4(1)] = res[1];
+    vd[H4(0)] = res[0];
+}
+
+static inline void vsha2ms_e64(uint64_t *vd, uint64_t *vs1, uint64_t *vs2)
+{
+    uint64_t res[4];
+    res[0] = sig1_sha512(vs1[2]) + vs2[1] + sig0_sha512(vd[1]) + vd[0];
+    res[1] = sig1_sha512(vs1[3]) + vs2[2] + sig0_sha512(vd[2]) + vd[1];
+    res[2] = sig1_sha512(res[0]) + vs2[3] + sig0_sha512(vd[3]) + vd[2];
+    res[3] = sig1_sha512(res[1]) + vs1[0] + sig0_sha512(vs2[0]) + vd[3];
+    vd[3] = res[3];
+    vd[2] = res[2];
+    vd[1] = res[1];
+    vd[0] = res[0];
+}
+
+void HELPER(vsha2ms_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
+                        uint32_t desc)
+{
+    uint32_t sew = FIELD_EX64(env->vtype, VTYPE, VSEW);
+    uint32_t esz = sew == MO_32 ? 4 : 8;
+    uint32_t total_elems;
+    uint32_t vta = vext_vta(desc);
+
+    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
+        if (sew == MO_32) {
+            vsha2ms_e32(((uint32_t *)vd) + i * 4, ((uint32_t *)vs1) + i * 4,
+                        ((uint32_t *)vs2) + i * 4);
+        } else {
+            /* If not 32 then SEW should be 64 */
+            vsha2ms_e64(((uint64_t *)vd) + i * 4, ((uint64_t *)vs1) + i * 4,
+                        ((uint64_t *)vs2) + i * 4);
+        }
+    }
+    /* set tail elements to 1s */
+    total_elems = vext_get_total_elems(env, desc, esz);
+    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
+    env->vstart = 0;
+}
+
+static inline uint64_t sum0_64(uint64_t x)
+{
+    return ror64(x, 28) ^ ror64(x, 34) ^ ror64(x, 39);
+}
+
+static inline uint32_t sum0_32(uint32_t x)
+{
+    return ror32(x, 2) ^ ror32(x, 13) ^ ror32(x, 22);
+}
+
+static inline uint64_t sum1_64(uint64_t x)
+{
+    return ror64(x, 14) ^ ror64(x, 18) ^ ror64(x, 41);
+}
+
+static inline uint32_t sum1_32(uint32_t x)
+{
+    return ror32(x, 6) ^ ror32(x, 11) ^ ror32(x, 25);
+}
+
+#define ch(x, y, z) ((x & y) ^ ((~x) & z))
+
+#define maj(x, y, z) ((x & y) ^ (x & z) ^ (y & z))
+
+static void vsha2c_64(uint64_t *vs2, uint64_t *vd, uint64_t *vs1)
+{
+    uint64_t a = vs2[3], b = vs2[2], e = vs2[1], f = vs2[0];
+    uint64_t c = vd[3], d = vd[2], g = vd[1], h = vd[0];
+    uint64_t W0 = vs1[0], W1 = vs1[1];
+    uint64_t T1 = h + sum1_64(e) + ch(e, f, g) + W0;
+    uint64_t T2 = sum0_64(a) + maj(a, b, c);
+
+    h = g;
+    g = f;
+    f = e;
+    e = d + T1;
+    d = c;
+    c = b;
+    b = a;
+    a = T1 + T2;
+
+    T1 = h + sum1_64(e) + ch(e, f, g) + W1;
+    T2 = sum0_64(a) + maj(a, b, c);
+    h = g;
+    g = f;
+    f = e;
+    e = d + T1;
+    d = c;
+    c = b;
+    b = a;
+    a = T1 + T2;
+
+    vd[0] = f;
+    vd[1] = e;
+    vd[2] = b;
+    vd[3] = a;
+}
+
+static void vsha2c_32(uint32_t *vs2, uint32_t *vd, uint32_t *vs1)
+{
+    uint32_t a = vs2[H4(3)], b = vs2[H4(2)], e = vs2[H4(1)], f = vs2[H4(0)];
+    uint32_t c = vd[H4(3)], d = vd[H4(2)], g = vd[H4(1)], h = vd[H4(0)];
+    uint32_t W0 = vs1[H4(0)], W1 = vs1[H4(1)];
+    uint32_t T1 = h + sum1_32(e) + ch(e, f, g) + W0;
+    uint32_t T2 = sum0_32(a) + maj(a, b, c);
+
+    h = g;
+    g = f;
+    f = e;
+    e = d + T1;
+    d = c;
+    c = b;
+    b = a;
+    a = T1 + T2;
+
+    T1 = h + sum1_32(e) + ch(e, f, g) + W1;
+    T2 = sum0_32(a) + maj(a, b, c);
+    h = g;
+    g = f;
+    f = e;
+    e = d + T1;
+    d = c;
+    c = b;
+    b = a;
+    a = T1 + T2;
+
+    vd[H4(0)] = f;
+    vd[H4(1)] = e;
+    vd[H4(2)] = b;
+    vd[H4(3)] = a;
+}
+
+void HELPER(vsha2ch32_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
+                          uint32_t desc)
+{
+    const uint32_t esz = 4;
+    uint32_t total_elems;
+    uint32_t vta = vext_vta(desc);
+
+    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
+        vsha2c_32(((uint32_t *)vs2) + 4 * i, ((uint32_t *)vd) + 4 * i,
+                  ((uint32_t *)vs1) + 4 * i + 2);
+    }
+
+    /* set tail elements to 1s */
+    total_elems = vext_get_total_elems(env, desc, esz);
+    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
+    env->vstart = 0;
+}
+
+void HELPER(vsha2ch64_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
+                          uint32_t desc)
+{
+    const uint32_t esz = 8;
+    uint32_t total_elems;
+    uint32_t vta = vext_vta(desc);
+
+    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
+        vsha2c_64(((uint64_t *)vs2) + 4 * i, ((uint64_t *)vd) + 4 * i,
+                  ((uint64_t *)vs1) + 4 * i + 2);
+    }
+
+    /* set tail elements to 1s */
+    total_elems = vext_get_total_elems(env, desc, esz);
+    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
+    env->vstart = 0;
+}
+
+void HELPER(vsha2cl32_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
+                          uint32_t desc)
+{
+    const uint32_t esz = 4;
+    uint32_t total_elems;
+    uint32_t vta = vext_vta(desc);
+
+    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
+        vsha2c_32(((uint32_t *)vs2) + 4 * i, ((uint32_t *)vd) + 4 * i,
+                  (((uint32_t *)vs1) + 4 * i));
+    }
+
+    /* set tail elements to 1s */
+    total_elems = vext_get_total_elems(env, desc, esz);
+    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
+    env->vstart = 0;
+}
+
+void HELPER(vsha2cl64_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
+                          uint32_t desc)
+{
+    uint32_t esz = 8;
+    uint32_t total_elems;
+    uint32_t vta = vext_vta(desc);
+
+    for (uint32_t i = env->vstart / 4; i < env->vl / 4; i++) {
+        vsha2c_64(((uint64_t *)vs2) + 4 * i, ((uint64_t *)vd) + 4 * i,
+                  (((uint64_t *)vs1) + 4 * i));
+    }
+
+    /* set tail elements to 1s */
+    total_elems = vext_get_total_elems(env, desc, esz);
+    vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
+    env->vstart = 0;
+}
diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/insn_trans/trans_rvvk.c.inc
+++ b/target/riscv/insn_trans/trans_rvvk.c.inc
@@ -XXX,XX +XXX,XX @@ static bool vaeskf2_check(DisasContext *s, arg_vaeskf2_vi *a)
 
 GEN_VI_UNMASKED_TRANS(vaeskf1_vi, vaeskf1_check, ZVKNED_EGS)
 GEN_VI_UNMASKED_TRANS(vaeskf2_vi, vaeskf2_check, ZVKNED_EGS)
+
+/*
+ * Zvknh
+ */
+
+#define ZVKNH_EGS 4
+
+#define GEN_VV_UNMASKED_TRANS(NAME, CHECK, EGS)                               \
+    static bool trans_##NAME(DisasContext *s, arg_rmrr *a)                    \
+    {                                                                         \
+        if (CHECK(s, a)) {                                                    \
+            uint32_t data = 0;                                                \
+            TCGLabel *over = gen_new_label();                                 \
+            TCGv_i32 egs;                                                     \
+                                                                              \
+            if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {                      \
+                /* save opcode for unwinding in case we throw an exception */ \
+                decode_save_opc(s);                                           \
+                egs = tcg_constant_i32(EGS);                                  \
+                gen_helper_egs_check(egs, cpu_env);                           \
+                tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);    \
+            }                                                                 \
+                                                                              \
+            data = FIELD_DP32(data, VDATA, VM, a->vm);                        \
+            data = FIELD_DP32(data, VDATA, LMUL, s->lmul);                    \
+            data = FIELD_DP32(data, VDATA, VTA, s->vta);                      \
+            data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);    \
+            data = FIELD_DP32(data, VDATA, VMA, s->vma);                      \
+                                                                              \
+            tcg_gen_gvec_3_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, a->rs1),       \
+                               vreg_ofs(s, a->rs2), cpu_env,                  \
+                               s->cfg_ptr->vlen / 8, s->cfg_ptr->vlen / 8,    \
+                               data, gen_helper_##NAME);                      \
+                                                                              \
+            mark_vs_dirty(s);                                                 \
+            gen_set_label(over);                                              \
+            return true;                                                      \
+        }                                                                     \
+        return false;                                                         \
+    }
+
+static bool vsha_check_sew(DisasContext *s)
+{
+    return (s->cfg_ptr->ext_zvknha == true && s->sew == MO_32) ||
+           (s->cfg_ptr->ext_zvknhb == true &&
+            (s->sew == MO_32 || s->sew == MO_64));
+}
+
+static bool vsha_check(DisasContext *s, arg_rmrr *a)
+{
+    int egw_bytes = ZVKNH_EGS << s->sew;
+    int mult = 1 << MAX(s->lmul, 0);
+    return opivv_check(s, a) &&
+           vsha_check_sew(s) &&
+           MAXSZ(s) >= egw_bytes &&
+           !is_overlapped(a->rd, mult, a->rs1, mult) &&
+           !is_overlapped(a->rd, mult, a->rs2, mult) &&
+           s->lmul >= 0;
+}
+
+GEN_VV_UNMASKED_TRANS(vsha2ms_vv, vsha_check, ZVKNH_EGS)
+
+static bool trans_vsha2cl_vv(DisasContext *s, arg_rmrr *a)
+{
+    if (vsha_check(s, a)) {
+        uint32_t data = 0;
+        TCGLabel *over = gen_new_label();
+        TCGv_i32 egs;
+
+        if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {
+            /* save opcode for unwinding in case we throw an exception */
+            decode_save_opc(s);
+            egs = tcg_constant_i32(ZVKNH_EGS);
+            gen_helper_egs_check(egs, cpu_env);
+            tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+        }
+
+        data = FIELD_DP32(data, VDATA, VM, a->vm);
+        data = FIELD_DP32(data, VDATA, LMUL, s->lmul);
+        data = FIELD_DP32(data, VDATA, VTA, s->vta);
+        data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);
+        data = FIELD_DP32(data, VDATA, VMA, s->vma);
+
+        tcg_gen_gvec_3_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, a->rs1),
+            vreg_ofs(s, a->rs2), cpu_env, s->cfg_ptr->vlen / 8,
+            s->cfg_ptr->vlen / 8, data,
+            s->sew == MO_32 ?
+                gen_helper_vsha2cl32_vv : gen_helper_vsha2cl64_vv);
+
+        mark_vs_dirty(s);
+        gen_set_label(over);
+        return true;
+    }
+    return false;
+}
+
+static bool trans_vsha2ch_vv(DisasContext *s, arg_rmrr *a)
+{
+    if (vsha_check(s, a)) {
+        uint32_t data = 0;
+        TCGLabel *over = gen_new_label();
+        TCGv_i32 egs;
+
+        if (!s->vstart_eq_zero || !s->vl_eq_vlmax) {
+            /* save opcode for unwinding in case we throw an exception */
+            decode_save_opc(s);
+            egs = tcg_constant_i32(ZVKNH_EGS);
+            gen_helper_egs_check(egs, cpu_env);
+            tcg_gen_brcond_tl(TCG_COND_GEU, cpu_vstart, cpu_vl, over);
+        }
+
+        data = FIELD_DP32(data, VDATA, VM, a->vm);
+        data = FIELD_DP32(data, VDATA, LMUL, s->lmul);
+        data = FIELD_DP32(data, VDATA, VTA, s->vta);
+        data = FIELD_DP32(data, VDATA, VTA_ALL_1S, s->cfg_vta_all_1s);
+        data = FIELD_DP32(data, VDATA, VMA, s->vma);
+
+        tcg_gen_gvec_3_ptr(vreg_ofs(s, a->rd), vreg_ofs(s, a->rs1),
+            vreg_ofs(s, a->rs2), cpu_env, s->cfg_ptr->vlen / 8,
+            s->cfg_ptr->vlen / 8, data,
+            s->sew == MO_32 ?
+                gen_helper_vsha2ch32_vv : gen_helper_vsha2ch64_vv);
+
+        mark_vs_dirty(s);
+        gen_set_label(over);
+        return true;
+    }
+    return false;
+}
-- 
2.41.0

From: Lawrence Hunter <lawrence.hunter@codethink.co.uk>

This commit adds support for the Zvksh vector-crypto extension, which
consists of the following instructions:

* vsm3me.vv
* vsm3c.vi

Translation functions are defined in
`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
`target/riscv/vcrypto_helper.c`.

Co-authored-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
[max.chou@sifive.com: Replaced vstart checking by TCG op]
Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
Signed-off-by: Max Chou <max.chou@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
[max.chou@sifive.com: Exposed x-zvksh property]
Message-ID: <20230711165917.2629866-12-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu_cfg.h                   |   1 +
 target/riscv/helper.h                    |   3 +
 target/riscv/insn32.decode               |   4 +
 target/riscv/cpu.c                       |   6 +-
 target/riscv/vcrypto_helper.c            | 134 +++++++++++++++++++++++
 target/riscv/insn_trans/trans_rvvk.c.inc |  31 ++++++
 6 files changed, 177 insertions(+), 2 deletions(-)

From: Nazar Kazakov <nazar.kazakov@codethink.co.uk>

This commit adds support for the Zvkg vector-crypto extension, which
consists of the following instructions:

* vgmul.vv
* vghsh.vv

Translation functions are defined in
`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
`target/riscv/vcrypto_helper.c`.

Co-authored-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
[max.chou@sifive.com: Replaced vstart checking by TCG op]
Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
Signed-off-by: Nazar Kazakov <nazar.kazakov@codethink.co.uk>
Signed-off-by: Max Chou <max.chou@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
[max.chou@sifive.com: Exposed x-zvkg property]
[max.chou@sifive.com: Replaced uint by int for cross win32 build]
Message-ID: <20230711165917.2629866-13-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu_cfg.h                   |  1 +
 target/riscv/helper.h                    |  3 +
 target/riscv/insn32.decode               |  4 ++
 target/riscv/cpu.c                       |  6 +-
 target/riscv/vcrypto_helper.c            | 72 ++++++++++++++++++++++++
 target/riscv/insn_trans/trans_rvvk.c.inc | 30 ++++++++++
 6 files changed, 114 insertions(+), 2 deletions(-)

From: Max Chou <max.chou@sifive.com>

Allows sharing of sm4_subword between different targets.

Signed-off-by: Max Chou <max.chou@sifive.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Max Chou <max.chou@sifive.com>
Message-ID: <20230711165917.2629866-14-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 include/crypto/sm4.h           |  8 ++++++++
 target/arm/tcg/crypto_helper.c | 10 ++--------
 2 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/include/crypto/sm4.h b/include/crypto/sm4.h
index XXXXXXX..XXXXXXX 100644
--- a/include/crypto/sm4.h
+++ b/include/crypto/sm4.h
@@ -XXX,XX +XXX,XX @@
 
 extern const uint8_t sm4_sbox[256];
 
+static inline uint32_t sm4_subword(uint32_t word)
+{
+    return sm4_sbox[word & 0xff] |
+           sm4_sbox[(word >> 8) & 0xff] << 8 |
+           sm4_sbox[(word >> 16) & 0xff] << 16 |
+           sm4_sbox[(word >> 24) & 0xff] << 24;
+}
+
 #endif
diff --git a/target/arm/tcg/crypto_helper.c b/target/arm/tcg/crypto_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/tcg/crypto_helper.c
+++ b/target/arm/tcg/crypto_helper.c
@@ -XXX,XX +XXX,XX @@ static void do_crypto_sm4e(uint64_t *rd, uint64_t *rn, uint64_t *rm)
             CR_ST_WORD(d, (i + 3) % 4) ^
             CR_ST_WORD(n, i);
 
-        t = sm4_sbox[t & 0xff] |
-            sm4_sbox[(t >> 8) & 0xff] << 8 |
-            sm4_sbox[(t >> 16) & 0xff] << 16 |
-            sm4_sbox[(t >> 24) & 0xff] << 24;
+        t = sm4_subword(t);
 
         CR_ST_WORD(d, i) ^= t ^ rol32(t, 2) ^ rol32(t, 10) ^ rol32(t, 18) ^
                             rol32(t, 24);
@@ -XXX,XX +XXX,XX @@ static void do_crypto_sm4ekey(uint64_t *rd, uint64_t *rn, uint64_t *rm)
             CR_ST_WORD(d, (i + 3) % 4) ^
             CR_ST_WORD(m, i);
 
-        t = sm4_sbox[t & 0xff] |
-            sm4_sbox[(t >> 8) & 0xff] << 8 |
-            sm4_sbox[(t >> 16) & 0xff] << 16 |
-            sm4_sbox[(t >> 24) & 0xff] << 24;
+        t = sm4_subword(t);
 
         CR_ST_WORD(d, i) ^= t ^ rol32(t, 13) ^ rol32(t, 23);
     }
-- 
2.41.0

From: Max Chou <max.chou@sifive.com>

Adds sm4_ck constant for use in sm4 cryptography across different targets.

Signed-off-by: Max Chou <max.chou@sifive.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Signed-off-by: Max Chou <max.chou@sifive.com>
Message-ID: <20230711165917.2629866-15-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 include/crypto/sm4.h |  1 +
 crypto/sm4.c         | 10 ++++++++++
 2 files changed, 11 insertions(+)

diff --git a/include/crypto/sm4.h b/include/crypto/sm4.h
index XXXXXXX..XXXXXXX 100644
--- a/include/crypto/sm4.h
+++ b/include/crypto/sm4.h
@@ -XXX,XX +XXX,XX @@
 #define QEMU_SM4_H
 
 extern const uint8_t sm4_sbox[256];
+extern const uint32_t sm4_ck[32];
 
 static inline uint32_t sm4_subword(uint32_t word)
 {
diff --git a/crypto/sm4.c b/crypto/sm4.c
index XXXXXXX..XXXXXXX 100644
--- a/crypto/sm4.c
+++ b/crypto/sm4.c
@@ -XXX,XX +XXX,XX @@ uint8_t const sm4_sbox[] = {
     0x79, 0xee, 0x5f, 0x3e, 0xd7, 0xcb, 0x39, 0x48,
 };
 
+uint32_t const sm4_ck[] = {
+    0x00070e15, 0x1c232a31, 0x383f464d, 0x545b6269,
+    0x70777e85, 0x8c939aa1, 0xa8afb6bd, 0xc4cbd2d9,
+    0xe0e7eef5, 0xfc030a11, 0x181f262d, 0x343b4249,
+    0x50575e65, 0x6c737a81, 0x888f969d, 0xa4abb2b9,
+    0xc0c7ced5, 0xdce3eaf1, 0xf8ff060d, 0x141b2229,
+    0x30373e45, 0x4c535a61, 0x686f767d, 0x848b9299,
+    0xa0a7aeb5, 0xbcc3cad1, 0xd8dfe6ed, 0xf4fb0209,
+    0x10171e25, 0x2c333a41, 0x484f565d, 0x646b7279
+};
-- 
2.41.0

From: Max Chou <max.chou@sifive.com>

This commit adds support for the Zvksed vector-crypto extension, which
consists of the following instructions:

* vsm4k.vi
* vsm4r.[vv,vs]

Translation functions are defined in
`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
`target/riscv/vcrypto_helper.c`.

Signed-off-by: Max Chou <max.chou@sifive.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
[lawrence.hunter@codethink.co.uk: Moved SM4 functions from
crypto_helper.c to vcrypto_helper.c]
[nazar.kazakov@codethink.co.uk: Added alignment checks, refactored code to
use macros, and minor style changes]
Signed-off-by: Max Chou <max.chou@sifive.com>
Message-ID: <20230711165917.2629866-16-max.chou@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu_cfg.h                   |   1 +
 target/riscv/helper.h                    |   4 +
 target/riscv/insn32.decode               |   5 +
 target/riscv/cpu.c                       |   5 +-
 target/riscv/vcrypto_helper.c            | 127 +++++++++++++++++++++++
 target/riscv/insn_trans/trans_rvvk.c.inc |  43 ++++++++
 6 files changed, 184 insertions(+), 1 deletion(-)

From: Rob Bradford <rbradford@rivosinc.com>

These are WARL fields - zero out the bits for unavailable counters and
special case the TM bit in mcountinhibit which is hardwired to zero.
This patch achieves this by modifying the value written so that any use
of the field will see the correctly masked bits.

Tested by modifying OpenSBI to write max value to these CSRs and upon
subsequent read the appropriate number of bits for number of PMUs is
enabled and the TM bit is zero in mcountinhibit.

Signed-off-by: Rob Bradford <rbradford@rivosinc.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Message-ID: <20230802124906.24197-1-rbradford@rivosinc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/csr.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/target/riscv/csr.c b/target/riscv/csr.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/csr.c
+++ b/target/riscv/csr.c
@@ -XXX,XX +XXX,XX @@ static RISCVException write_mcountinhibit(CPURISCVState *env, int csrno,
 {
     int cidx;
     PMUCTRState *counter;
+    RISCVCPU *cpu = env_archcpu(env);
 
-    env->mcountinhibit = val;
+    /* WARL register - disable unavailable counters; TM bit is always 0 */
+    env->mcountinhibit =
+        val & (cpu->pmu_avail_ctrs | COUNTEREN_CY | COUNTEREN_IR);
 
     /* Check if any other counter is also monitoring cycles/instructions */
     for (cidx = 0; cidx < RV_MAX_MHPMCOUNTERS; cidx++) {
@@ -XXX,XX +XXX,XX @@ static RISCVException read_mcounteren(CPURISCVState *env, int csrno,
 static RISCVException write_mcounteren(CPURISCVState *env, int csrno,
                                        target_ulong val)
 {
-    env->mcounteren = val;
+    RISCVCPU *cpu = env_archcpu(env);
+
+    /* WARL register - disable unavailable counters */
+    env->mcounteren = val & (cpu->pmu_avail_ctrs | COUNTEREN_CY | COUNTEREN_TM |
+                             COUNTEREN_IR);
     return RISCV_EXCP_NONE;
 }
 
-- 
2.41.0

From: Jason Chien <jason.chien@sifive.com>

RVA23 Profiles states:
The RVA23 profiles are intended to be used for 64-bit application
processors that will run rich OS stacks from standard binary OS
distributions and with a substantial number of third-party binary user
applications that will be supported over a considerable length of time
in the field.

The chapter 4 of the unprivileged spec introduces the Zihintntl extension
and Zihintntl is a mandatory extension presented in RVA23 Profiles, whose
purpose is to enable application and operating system portability across
different implementations. Thus the DTS should contain the Zihintntl ISA
string in order to pass to software.

The unprivileged spec states:
Like any HINTs, these instructions may be freely ignored. Hence, although
they are described in terms of cache-based memory hierarchies, they do not
mandate the provision of caches.

These instructions are encoded with non-used opcode, e.g. ADD x0, x0, x2,
which QEMU already supports, and QEMU does not emulate cache. Therefore
these instructions can be considered as a no-op, and we only need to add
a new property for the Zihintntl extension.

Reviewed-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Jason Chien <jason.chien@sifive.com>
Message-ID: <20230726074049.19505-2-jason.chien@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu_cfg.h | 1 +
 target/riscv/cpu.c     | 2 ++
 2 files changed, 3 insertions(+)

From: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>

Commit a47842d ("riscv: Add support for the Zfa extension") implemented the zfa extension.
However, it has some typos for fleq.d and fltq.d. Both of them misused the fltq.s
helper function.

Fixes: a47842d ("riscv: Add support for the Zfa extension")
Signed-off-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Weiwei Li <liweiwei@iscas.ac.cn>
Message-ID: <20230728003906.768-1-zhiwei_liu@linux.alibaba.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/insn_trans/trans_rvzfa.c.inc | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/target/riscv/insn_trans/trans_rvzfa.c.inc b/target/riscv/insn_trans/trans_rvzfa.c.inc
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/insn_trans/trans_rvzfa.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfa.c.inc
@@ -XXX,XX +XXX,XX @@ bool trans_fleq_d(DisasContext *ctx, arg_fleq_d *a)
     TCGv_i64 src1 = get_fpr_hs(ctx, a->rs1);
     TCGv_i64 src2 = get_fpr_hs(ctx, a->rs2);
 
-    gen_helper_fltq_s(dest, cpu_env, src1, src2);
+    gen_helper_fleq_d(dest, cpu_env, src1, src2);
     gen_set_gpr(ctx, a->rd, dest);
     return true;
 }
@@ -XXX,XX +XXX,XX @@ bool trans_fltq_d(DisasContext *ctx, arg_fltq_d *a)
     TCGv_i64 src1 = get_fpr_hs(ctx, a->rs1);
     TCGv_i64 src2 = get_fpr_hs(ctx, a->rs2);
 
-    gen_helper_fltq_s(dest, cpu_env, src1, src2);
+    gen_helper_fltq_d(dest, cpu_env, src1, src2);
     gen_set_gpr(ctx, a->rd, dest);
     return true;
 }
-- 
2.41.0

From: Jason Chien <jason.chien@sifive.com>

When writing the upper mtime, we should keep the original lower mtime
whose value is given by cpu_riscv_read_rtc() instead of
cpu_riscv_read_rtc_raw(). The same logic applies to writes to lower mtime.

Signed-off-by: Jason Chien <jason.chien@sifive.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <20230728082502.26439-1-jason.chien@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/intc/riscv_aclint.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/hw/intc/riscv_aclint.c b/hw/intc/riscv_aclint.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/riscv_aclint.c
+++ b/hw/intc/riscv_aclint.c
@@ -XXX,XX +XXX,XX @@ static void riscv_aclint_mtimer_write(void *opaque, hwaddr addr,
         return;
     } else if (addr == mtimer->time_base || addr == mtimer->time_base + 4) {
         uint64_t rtc_r = cpu_riscv_read_rtc_raw(mtimer->timebase_freq);
+        uint64_t rtc = cpu_riscv_read_rtc(mtimer);
 
         if (addr == mtimer->time_base) {
             if (size == 4) {
                 /* time_lo for RV32/RV64 */
-                mtimer->time_delta = ((rtc_r & ~0xFFFFFFFFULL) | value) - rtc_r;
+                mtimer->time_delta = ((rtc & ~0xFFFFFFFFULL) | value) - rtc_r;
             } else {
                 /* time for RV64 */
                 mtimer->time_delta = value - rtc_r;
@@ -XXX,XX +XXX,XX @@ static void riscv_aclint_mtimer_write(void *opaque, hwaddr addr,
         } else {
             if (size == 4) {
                 /* time_hi for RV32/RV64 */
-                mtimer->time_delta = (value << 32 | (rtc_r & 0xFFFFFFFF)) - rtc_r;
+                mtimer->time_delta = (value << 32 | (rtc & 0xFFFFFFFF)) - rtc_r;
             } else {
                 qemu_log_mask(LOG_GUEST_ERROR,
                               "aclint-mtimer: invalid time_hi write: %08x",
-- 
2.41.0

From: Jason Chien <jason.chien@sifive.com>

The variables whose values are given by cpu_riscv_read_rtc() should be named
"rtc". The variables whose value are given by cpu_riscv_read_rtc_raw()
should be named "rtc_r".

Signed-off-by: Jason Chien <jason.chien@sifive.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <20230728082502.26439-2-jason.chien@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/intc/riscv_aclint.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/hw/intc/riscv_aclint.c b/hw/intc/riscv_aclint.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/riscv_aclint.c
+++ b/hw/intc/riscv_aclint.c
@@ -XXX,XX +XXX,XX @@ static void riscv_aclint_mtimer_write_timecmp(RISCVAclintMTimerState *mtimer,
     uint64_t next;
     uint64_t diff;
 
-    uint64_t rtc_r = cpu_riscv_read_rtc(mtimer);
+    uint64_t rtc = cpu_riscv_read_rtc(mtimer);
 
     /* Compute the relative hartid w.r.t the socket */
     hartid = hartid - mtimer->hartid_base;
 
     mtimer->timecmp[hartid] = value;
-    if (mtimer->timecmp[hartid] <= rtc_r) {
+    if (mtimer->timecmp[hartid] <= rtc) {
         /*
          * If we're setting an MTIMECMP value in the "past",
          * immediately raise the timer interrupt
@@ -XXX,XX +XXX,XX @@ static void riscv_aclint_mtimer_write_timecmp(RISCVAclintMTimerState *mtimer,
 
     /* otherwise, set up the future timer interrupt */
     qemu_irq_lower(mtimer->timer_irqs[hartid]);
-    diff = mtimer->timecmp[hartid] - rtc_r;
+    diff = mtimer->timecmp[hartid] - rtc;
     /* back to ns (note args switched in muldiv64) */
     uint64_t ns_diff = muldiv64(diff, NANOSECONDS_PER_SECOND, timebase_freq);
 
-- 
2.41.0

From: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>

We should not use types dependend on host arch for target_ucontext.
This bug is found when run rv32 applications.

Signed-off-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20230811055438.1945-1-zhiwei_liu@linux.alibaba.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 linux-user/riscv/signal.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/linux-user/riscv/signal.c b/linux-user/riscv/signal.c
index XXXXXXX..XXXXXXX 100644
--- a/linux-user/riscv/signal.c
+++ b/linux-user/riscv/signal.c
@@ -XXX,XX +XXX,XX @@ struct target_sigcontext {
 }; /* cf. riscv-linux:arch/riscv/include/uapi/asm/ptrace.h */
 
 struct target_ucontext {
-    unsigned long uc_flags;
-    struct target_ucontext *uc_link;
+    abi_ulong uc_flags;
+    abi_ptr uc_link;
     target_stack_t uc_stack;
     target_sigset_t uc_sigmask;
     uint8_t   __unused[1024 / 8 - sizeof(target_sigset_t)];
-- 
2.41.0

From: Yong-Xuan Wang <yongxuan.wang@sifive.com>

In this patch, we create the APLIC and IMSIC FDT helper functions and
remove M mode AIA devices when using KVM acceleration.

Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
Reviewed-by: Jim Shu <jim.shu@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230727102439.22554-2-yongxuan.wang@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/riscv/virt.c | 290 +++++++++++++++++++++++-------------------------
 1 file changed, 137 insertions(+), 153 deletions(-)

diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -XXX,XX +XXX,XX @@ static uint32_t imsic_num_bits(uint32_t count)
     return ret;
 }
 
-static void create_fdt_imsic(RISCVVirtState *s, const MemMapEntry *memmap,
-                             uint32_t *phandle, uint32_t *intc_phandles,
-                             uint32_t *msi_m_phandle, uint32_t *msi_s_phandle)
+static void create_fdt_one_imsic(RISCVVirtState *s, hwaddr base_addr,
+                                 uint32_t *intc_phandles, uint32_t msi_phandle,
+                                 bool m_mode, uint32_t imsic_guest_bits)
 {
     int cpu, socket;
     char *imsic_name;
     MachineState *ms = MACHINE(s);
     int socket_count = riscv_socket_count(ms);
-    uint32_t imsic_max_hart_per_socket, imsic_guest_bits;
+    uint32_t imsic_max_hart_per_socket;
     uint32_t *imsic_cells, *imsic_regs, imsic_addr, imsic_size;
 
-    *msi_m_phandle = (*phandle)++;
-    *msi_s_phandle = (*phandle)++;
     imsic_cells = g_new0(uint32_t, ms->smp.cpus * 2);
     imsic_regs = g_new0(uint32_t, socket_count * 4);
 
-    /* M-level IMSIC node */
     for (cpu = 0; cpu < ms->smp.cpus; cpu++) {
         imsic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
-        imsic_cells[cpu * 2 + 1] = cpu_to_be32(IRQ_M_EXT);
+        imsic_cells[cpu * 2 + 1] = cpu_to_be32(m_mode ? IRQ_M_EXT : IRQ_S_EXT);
     }
-    imsic_max_hart_per_socket = 0;
-    for (socket = 0; socket < socket_count; socket++) {
-        imsic_addr = memmap[VIRT_IMSIC_M].base +
-                     socket * VIRT_IMSIC_GROUP_MAX_SIZE;
-        imsic_size = IMSIC_HART_SIZE(0) * s->soc[socket].num_harts;
-        imsic_regs[socket * 4 + 0] = 0;
-        imsic_regs[socket * 4 + 1] = cpu_to_be32(imsic_addr);
-        imsic_regs[socket * 4 + 2] = 0;
-        imsic_regs[socket * 4 + 3] = cpu_to_be32(imsic_size);
-        if (imsic_max_hart_per_socket < s->soc[socket].num_harts) {
-            imsic_max_hart_per_socket = s->soc[socket].num_harts;
-        }
-    }
-    imsic_name = g_strdup_printf("/soc/imsics@%lx",
-        (unsigned long)memmap[VIRT_IMSIC_M].base);
-    qemu_fdt_add_subnode(ms->fdt, imsic_name);
-    qemu_fdt_setprop_string(ms->fdt, imsic_name, "compatible",
-        "riscv,imsics");
-    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "#interrupt-cells",
-        FDT_IMSIC_INT_CELLS);
-    qemu_fdt_setprop(ms->fdt, imsic_name, "interrupt-controller",
-        NULL, 0);
-    qemu_fdt_setprop(ms->fdt, imsic_name, "msi-controller",
-        NULL, 0);
-    qemu_fdt_setprop(ms->fdt, imsic_name, "interrupts-extended",
-        imsic_cells, ms->smp.cpus * sizeof(uint32_t) * 2);
-    qemu_fdt_setprop(ms->fdt, imsic_name, "reg", imsic_regs,
-        socket_count * sizeof(uint32_t) * 4);
-    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,num-ids",
-        VIRT_IRQCHIP_NUM_MSIS);
-    if (socket_count > 1) {
-        qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,hart-index-bits",
-            imsic_num_bits(imsic_max_hart_per_socket));
-        qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,group-index-bits",
-            imsic_num_bits(socket_count));
-        qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,group-index-shift",
-            IMSIC_MMIO_GROUP_MIN_SHIFT);
-    }
-    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "phandle", *msi_m_phandle);
-
-    g_free(imsic_name);
 
-    /* S-level IMSIC node */
-    for (cpu = 0; cpu < ms->smp.cpus; cpu++) {
-        imsic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
-        imsic_cells[cpu * 2 + 1] = cpu_to_be32(IRQ_S_EXT);
-    }
-    imsic_guest_bits = imsic_num_bits(s->aia_guests + 1);
     imsic_max_hart_per_socket = 0;
     for (socket = 0; socket < socket_count; socket++) {
-        imsic_addr = memmap[VIRT_IMSIC_S].base +
-                     socket * VIRT_IMSIC_GROUP_MAX_SIZE;
+        imsic_addr = base_addr + socket * VIRT_IMSIC_GROUP_MAX_SIZE;
         imsic_size = IMSIC_HART_SIZE(imsic_guest_bits) *
                      s->soc[socket].num_harts;
         imsic_regs[socket * 4 + 0] = 0;
@@ -XXX,XX +XXX,XX @@ static void create_fdt_imsic(RISCVVirtState *s, const MemMapEntry *memmap,
             imsic_max_hart_per_socket = s->soc[socket].num_harts;
         }
     }
-    imsic_name = g_strdup_printf("/soc/imsics@%lx",
-        (unsigned long)memmap[VIRT_IMSIC_S].base);
+
+    imsic_name = g_strdup_printf("/soc/imsics@%lx", (unsigned long)base_addr);
     qemu_fdt_add_subnode(ms->fdt, imsic_name);
-    qemu_fdt_setprop_string(ms->fdt, imsic_name, "compatible",
-        "riscv,imsics");
+    qemu_fdt_setprop_string(ms->fdt, imsic_name, "compatible", "riscv,imsics");
     qemu_fdt_setprop_cell(ms->fdt, imsic_name, "#interrupt-cells",
-        FDT_IMSIC_INT_CELLS);
-    qemu_fdt_setprop(ms->fdt, imsic_name, "interrupt-controller",
-        NULL, 0);
-    qemu_fdt_setprop(ms->fdt, imsic_name, "msi-controller",
-        NULL, 0);
+                          FDT_IMSIC_INT_CELLS);
+    qemu_fdt_setprop(ms->fdt, imsic_name, "interrupt-controller", NULL, 0);
+    qemu_fdt_setprop(ms->fdt, imsic_name, "msi-controller", NULL, 0);
     qemu_fdt_setprop(ms->fdt, imsic_name, "interrupts-extended",
-        imsic_cells, ms->smp.cpus * sizeof(uint32_t) * 2);
+                     imsic_cells, ms->smp.cpus * sizeof(uint32_t) * 2);
     qemu_fdt_setprop(ms->fdt, imsic_name, "reg", imsic_regs,
-        socket_count * sizeof(uint32_t) * 4);
+                     socket_count * sizeof(uint32_t) * 4);
     qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,num-ids",
-        VIRT_IRQCHIP_NUM_MSIS);
+                     VIRT_IRQCHIP_NUM_MSIS);
+
     if (imsic_guest_bits) {
         qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,guest-index-bits",
-            imsic_guest_bits);
+                              imsic_guest_bits);
     }
+
     if (socket_count > 1) {
         qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,hart-index-bits",
-            imsic_num_bits(imsic_max_hart_per_socket));
+                              imsic_num_bits(imsic_max_hart_per_socket));
         qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,group-index-bits",
-            imsic_num_bits(socket_count));
+                              imsic_num_bits(socket_count));
         qemu_fdt_setprop_cell(ms->fdt, imsic_name, "riscv,group-index-shift",
-            IMSIC_MMIO_GROUP_MIN_SHIFT);
+                              IMSIC_MMIO_GROUP_MIN_SHIFT);
     }
-    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "phandle", *msi_s_phandle);
-    g_free(imsic_name);
+    qemu_fdt_setprop_cell(ms->fdt, imsic_name, "phandle", msi_phandle);
 
+    g_free(imsic_name);
     g_free(imsic_regs);
     g_free(imsic_cells);
 }
 
-static void create_fdt_socket_aplic(RISCVVirtState *s,
-                                    const MemMapEntry *memmap, int socket,
-                                    uint32_t msi_m_phandle,
-                                    uint32_t msi_s_phandle,
-                                    uint32_t *phandle,
-                                    uint32_t *intc_phandles,
-                                    uint32_t *aplic_phandles)
+static void create_fdt_imsic(RISCVVirtState *s, const MemMapEntry *memmap,
+                             uint32_t *phandle, uint32_t *intc_phandles,
+                             uint32_t *msi_m_phandle, uint32_t *msi_s_phandle)
+{
+    *msi_m_phandle = (*phandle)++;
+    *msi_s_phandle = (*phandle)++;
+
+    if (!kvm_enabled()) {
+        /* M-level IMSIC node */
+        create_fdt_one_imsic(s, memmap[VIRT_IMSIC_M].base, intc_phandles,
+                             *msi_m_phandle, true, 0);
+    }
+
+    /* S-level IMSIC node */
+    create_fdt_one_imsic(s, memmap[VIRT_IMSIC_S].base, intc_phandles,
+                         *msi_s_phandle, false,
+                         imsic_num_bits(s->aia_guests + 1));
+
+}
+
+static void create_fdt_one_aplic(RISCVVirtState *s, int socket,
+                                 unsigned long aplic_addr, uint32_t aplic_size,
+                                 uint32_t msi_phandle,
+                                 uint32_t *intc_phandles,
+                                 uint32_t aplic_phandle,
+                                 uint32_t aplic_child_phandle,
+                                 bool m_mode)
 {
     int cpu;
     char *aplic_name;
     uint32_t *aplic_cells;
-    unsigned long aplic_addr;
     MachineState *ms = MACHINE(s);
-    uint32_t aplic_m_phandle, aplic_s_phandle;
 
-    aplic_m_phandle = (*phandle)++;
-    aplic_s_phandle = (*phandle)++;
     aplic_cells = g_new0(uint32_t, s->soc[socket].num_harts * 2);
 
-    /* M-level APLIC node */
     for (cpu = 0; cpu < s->soc[socket].num_harts; cpu++) {
         aplic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
-        aplic_cells[cpu * 2 + 1] = cpu_to_be32(IRQ_M_EXT);
+        aplic_cells[cpu * 2 + 1] = cpu_to_be32(m_mode ? IRQ_M_EXT : IRQ_S_EXT);
     }
-    aplic_addr = memmap[VIRT_APLIC_M].base +
-                 (memmap[VIRT_APLIC_M].size * socket);
+
     aplic_name = g_strdup_printf("/soc/aplic@%lx", aplic_addr);
     qemu_fdt_add_subnode(ms->fdt, aplic_name);
     qemu_fdt_setprop_string(ms->fdt, aplic_name, "compatible", "riscv,aplic");
     qemu_fdt_setprop_cell(ms->fdt, aplic_name,
-        "#interrupt-cells", FDT_APLIC_INT_CELLS);
+                          "#interrupt-cells", FDT_APLIC_INT_CELLS);
     qemu_fdt_setprop(ms->fdt, aplic_name, "interrupt-controller", NULL, 0);
+
     if (s->aia_type == VIRT_AIA_TYPE_APLIC) {
         qemu_fdt_setprop(ms->fdt, aplic_name, "interrupts-extended",
-            aplic_cells, s->soc[socket].num_harts * sizeof(uint32_t) * 2);
+                         aplic_cells,
+                         s->soc[socket].num_harts * sizeof(uint32_t) * 2);
     } else {
-        qemu_fdt_setprop_cell(ms->fdt, aplic_name, "msi-parent",
-            msi_m_phandle);
+        qemu_fdt_setprop_cell(ms->fdt, aplic_name, "msi-parent", msi_phandle);
     }
+
     qemu_fdt_setprop_cells(ms->fdt, aplic_name, "reg",
-        0x0, aplic_addr, 0x0, memmap[VIRT_APLIC_M].size);
+                           0x0, aplic_addr, 0x0, aplic_size);
     qemu_fdt_setprop_cell(ms->fdt, aplic_name, "riscv,num-sources",
-        VIRT_IRQCHIP_NUM_SOURCES);
-    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "riscv,children",
-        aplic_s_phandle);
-    qemu_fdt_setprop_cells(ms->fdt, aplic_name, "riscv,delegate",
-        aplic_s_phandle, 0x1, VIRT_IRQCHIP_NUM_SOURCES);
+                          VIRT_IRQCHIP_NUM_SOURCES);
+
+    if (aplic_child_phandle) {
+        qemu_fdt_setprop_cell(ms->fdt, aplic_name, "riscv,children",
+                              aplic_child_phandle);
+        qemu_fdt_setprop_cells(ms->fdt, aplic_name, "riscv,delegate",
+                               aplic_child_phandle, 0x1,
+                               VIRT_IRQCHIP_NUM_SOURCES);
+    }
+
     riscv_socket_fdt_write_id(ms, aplic_name, socket);
-    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "phandle", aplic_m_phandle);
+    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "phandle", aplic_phandle);
+
     g_free(aplic_name);
+    g_free(aplic_cells);
+}
 
-    /* S-level APLIC node */
-    for (cpu = 0; cpu < s->soc[socket].num_harts; cpu++) {
-        aplic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
-        aplic_cells[cpu * 2 + 1] = cpu_to_be32(IRQ_S_EXT);
+static void create_fdt_socket_aplic(RISCVVirtState *s,
+                                    const MemMapEntry *memmap, int socket,
+                                    uint32_t msi_m_phandle,
+                                    uint32_t msi_s_phandle,
+                                    uint32_t *phandle,
+                                    uint32_t *intc_phandles,
+                                    uint32_t *aplic_phandles)
+{
+    char *aplic_name;
+    unsigned long aplic_addr;
+    MachineState *ms = MACHINE(s);
+    uint32_t aplic_m_phandle, aplic_s_phandle;
+
+    aplic_m_phandle = (*phandle)++;
+    aplic_s_phandle = (*phandle)++;
+
+    if (!kvm_enabled()) {
+        /* M-level APLIC node */
+        aplic_addr = memmap[VIRT_APLIC_M].base +
+                     (memmap[VIRT_APLIC_M].size * socket);
+        create_fdt_one_aplic(s, socket, aplic_addr, memmap[VIRT_APLIC_M].size,
+                             msi_m_phandle, intc_phandles,
+                             aplic_m_phandle, aplic_s_phandle,
+                             true);
     }
+
+    /* S-level APLIC node */
     aplic_addr = memmap[VIRT_APLIC_S].base +
                  (memmap[VIRT_APLIC_S].size * socket);
+    create_fdt_one_aplic(s, socket, aplic_addr, memmap[VIRT_APLIC_S].size,
+                         msi_s_phandle, intc_phandles,
+                         aplic_s_phandle, 0,
+                         false);
+
     aplic_name = g_strdup_printf("/soc/aplic@%lx", aplic_addr);
-    qemu_fdt_add_subnode(ms->fdt, aplic_name);
-    qemu_fdt_setprop_string(ms->fdt, aplic_name, "compatible", "riscv,aplic");
-    qemu_fdt_setprop_cell(ms->fdt, aplic_name,
-        "#interrupt-cells", FDT_APLIC_INT_CELLS);
-    qemu_fdt_setprop(ms->fdt, aplic_name, "interrupt-controller", NULL, 0);
-    if (s->aia_type == VIRT_AIA_TYPE_APLIC) {
-        qemu_fdt_setprop(ms->fdt, aplic_name, "interrupts-extended",
-            aplic_cells, s->soc[socket].num_harts * sizeof(uint32_t) * 2);
-    } else {
-        qemu_fdt_setprop_cell(ms->fdt, aplic_name, "msi-parent",
-            msi_s_phandle);
-    }
-    qemu_fdt_setprop_cells(ms->fdt, aplic_name, "reg",
-        0x0, aplic_addr, 0x0, memmap[VIRT_APLIC_S].size);
-    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "riscv,num-sources",
-        VIRT_IRQCHIP_NUM_SOURCES);
-    riscv_socket_fdt_write_id(ms, aplic_name, socket);
-    qemu_fdt_setprop_cell(ms->fdt, aplic_name, "phandle", aplic_s_phandle);
 
     if (!socket) {
         platform_bus_add_all_fdt_nodes(ms->fdt, aplic_name,
@@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aplic(RISCVVirtState *s,
 
     g_free(aplic_name);
 
-    g_free(aplic_cells);
     aplic_phandles[socket] = aplic_s_phandle;
 }
 
@@ -XXX,XX +XXX,XX @@ static DeviceState *virt_create_aia(RISCVVirtAIAType aia_type, int aia_guests,
     int i;
     hwaddr addr;
     uint32_t guest_bits;
-    DeviceState *aplic_m;
-    bool msimode = (aia_type == VIRT_AIA_TYPE_APLIC_IMSIC) ? true : false;
+    DeviceState *aplic_s = NULL;
+    DeviceState *aplic_m = NULL;
+    bool msimode = aia_type == VIRT_AIA_TYPE_APLIC_IMSIC;
 
     if (msimode) {
-        /* Per-socket M-level IMSICs */
-        addr = memmap[VIRT_IMSIC_M].base + socket * VIRT_IMSIC_GROUP_MAX_SIZE;
-        for (i = 0; i < hart_count; i++) {
-            riscv_imsic_create(addr + i * IMSIC_HART_SIZE(0),
-                               base_hartid + i, true, 1,
-                               VIRT_IRQCHIP_NUM_MSIS);
+        if (!kvm_enabled()) {
+            /* Per-socket M-level IMSICs */
+            addr = memmap[VIRT_IMSIC_M].base +
+                   socket * VIRT_IMSIC_GROUP_MAX_SIZE;
+            for (i = 0; i < hart_count; i++) {
+                riscv_imsic_create(addr + i * IMSIC_HART_SIZE(0),
+                                   base_hartid + i, true, 1,
+                                   VIRT_IRQCHIP_NUM_MSIS);
+            }
         }
 
         /* Per-socket S-level IMSICs */
@@ -XXX,XX +XXX,XX @@ static DeviceState *virt_create_aia(RISCVVirtAIAType aia_type, int aia_guests,
         }
     }
 
-    /* Per-socket M-level APLIC */
-    aplic_m = riscv_aplic_create(
-        memmap[VIRT_APLIC_M].base + socket * memmap[VIRT_APLIC_M].size,
-        memmap[VIRT_APLIC_M].size,
-        (msimode) ? 0 : base_hartid,
-        (msimode) ? 0 : hart_count,
-        VIRT_IRQCHIP_NUM_SOURCES,
-        VIRT_IRQCHIP_NUM_PRIO_BITS,
-        msimode, true, NULL);
-
-    if (aplic_m) {
-        /* Per-socket S-level APLIC */
-        riscv_aplic_create(
-            memmap[VIRT_APLIC_S].base + socket * memmap[VIRT_APLIC_S].size,
-            memmap[VIRT_APLIC_S].size,
-            (msimode) ? 0 : base_hartid,
-            (msimode) ? 0 : hart_count,
-            VIRT_IRQCHIP_NUM_SOURCES,
-            VIRT_IRQCHIP_NUM_PRIO_BITS,
-            msimode, false, aplic_m);
+    if (!kvm_enabled()) {
+        /* Per-socket M-level APLIC */
+        aplic_m = riscv_aplic_create(memmap[VIRT_APLIC_M].base +
+                                     socket * memmap[VIRT_APLIC_M].size,
+                                     memmap[VIRT_APLIC_M].size,
+                                     (msimode) ? 0 : base_hartid,
+                                     (msimode) ? 0 : hart_count,
+                                     VIRT_IRQCHIP_NUM_SOURCES,
+                                     VIRT_IRQCHIP_NUM_PRIO_BITS,
+                                     msimode, true, NULL);
     }
 
-    return aplic_m;
+    /* Per-socket S-level APLIC */
+    aplic_s = riscv_aplic_create(memmap[VIRT_APLIC_S].base +
+                                 socket * memmap[VIRT_APLIC_S].size,
+                                 memmap[VIRT_APLIC_S].size,
+                                 (msimode) ? 0 : base_hartid,
+                                 (msimode) ? 0 : hart_count,
+                                 VIRT_IRQCHIP_NUM_SOURCES,
+                                 VIRT_IRQCHIP_NUM_PRIO_BITS,
+                                 msimode, false, aplic_m);
+
+    return kvm_enabled() ? aplic_s : aplic_m;
 }
 
 static void create_platform_bus(RISCVVirtState *s, DeviceState *irqchip)
-- 
2.41.0

From: Yong-Xuan Wang <yongxuan.wang@sifive.com>

We check the in-kernel irqchip support when using KVM acceleration.

Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
Reviewed-by: Jim Shu <jim.shu@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230727102439.22554-3-yongxuan.wang@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/kvm.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -XXX,XX +XXX,XX @@ int kvm_arch_init(MachineState *ms, KVMState *s)
 
 int kvm_arch_irqchip_create(KVMState *s)
 {
-    return 0;
+    if (kvm_kernel_irqchip_split()) {
+        error_report("-machine kernel_irqchip=split is not supported on RISC-V.");
+        exit(1);
+    }
+
+    /*
+     * We can create the VAIA using the newer device control API.
+     */
+    return kvm_check_extension(s, KVM_CAP_DEVICE_CTRL);
 }
 
 int kvm_arch_process_async_events(CPUState *cs)
-- 
2.41.0

From: Yong-Xuan Wang <yongxuan.wang@sifive.com>

We create a vAIA chip by using the KVM_DEV_TYPE_RISCV_AIA and then set up
the chip with the KVM_DEV_RISCV_AIA_GRP_* APIs.
We also extend KVM accelerator to specify the KVM AIA mode. The "riscv-aia"
parameter is passed along with --accel in QEMU command-line.
1) "riscv-aia=emul": IMSIC is emulated by hypervisor
2) "riscv-aia=hwaccel": use hardware guest IMSIC
3) "riscv-aia=auto": use the hardware guest IMSICs whenever available
                     otherwise we fallback to software emulation.

Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
Reviewed-by: Jim Shu <jim.shu@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230727102439.22554-4-yongxuan.wang@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/kvm_riscv.h |   4 +
 target/riscv/kvm.c       | 186 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 190 insertions(+)

diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/kvm_riscv.h
+++ b/target/riscv/kvm_riscv.h
@@ -XXX,XX +XXX,XX @@
 void kvm_riscv_init_user_properties(Object *cpu_obj);
 void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
 void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level);
+void kvm_riscv_aia_create(MachineState *machine, uint64_t group_shift,
+                          uint64_t aia_irq_num, uint64_t aia_msi_num,
+                          uint64_t aplic_base, uint64_t imsic_base,
+                          uint64_t guest_num);
 
 #endif
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -XXX,XX +XXX,XX @@
 #include "exec/address-spaces.h"
 #include "hw/boards.h"
 #include "hw/irq.h"
+#include "hw/intc/riscv_imsic.h"
 #include "qemu/log.h"
 #include "hw/loader.h"
 #include "kvm_riscv.h"
@@ -XXX,XX +XXX,XX @@
 #include "chardev/char-fe.h"
 #include "migration/migration.h"
 #include "sysemu/runstate.h"
+#include "hw/riscv/numa.h"
 
 static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type,
                                  uint64_t idx)
@@ -XXX,XX +XXX,XX @@ bool kvm_arch_cpu_check_are_resettable(void)
     return true;
 }
 
+static int aia_mode;
+
+static const char *kvm_aia_mode_str(uint64_t mode)
+{
+    switch (mode) {
+    case KVM_DEV_RISCV_AIA_MODE_EMUL:
+        return "emul";
+    case KVM_DEV_RISCV_AIA_MODE_HWACCEL:
+        return "hwaccel";
+    case KVM_DEV_RISCV_AIA_MODE_AUTO:
+    default:
+        return "auto";
+    };
+}
+
+static char *riscv_get_kvm_aia(Object *obj, Error **errp)
+{
+    return g_strdup(kvm_aia_mode_str(aia_mode));
+}
+
+static void riscv_set_kvm_aia(Object *obj, const char *val, Error **errp)
+{
+    if (!strcmp(val, "emul")) {
+        aia_mode = KVM_DEV_RISCV_AIA_MODE_EMUL;
+    } else if (!strcmp(val, "hwaccel")) {
+        aia_mode = KVM_DEV_RISCV_AIA_MODE_HWACCEL;
+    } else if (!strcmp(val, "auto")) {
+        aia_mode = KVM_DEV_RISCV_AIA_MODE_AUTO;
+    } else {
+        error_setg(errp, "Invalid KVM AIA mode");
+        error_append_hint(errp, "Valid values are emul, hwaccel, and auto.\n");
+    }
+}
+
 void kvm_arch_accel_class_init(ObjectClass *oc)
 {
+    object_class_property_add_str(oc, "riscv-aia", riscv_get_kvm_aia,
+                                  riscv_set_kvm_aia);
+    object_class_property_set_description(oc, "riscv-aia",
+                                          "Set KVM AIA mode. Valid values are "
+                                          "emul, hwaccel, and auto. Default "
+                                          "is auto.");
+    object_property_set_default_str(object_class_property_find(oc, "riscv-aia"),
+                                    "auto");
+}
+
+void kvm_riscv_aia_create(MachineState *machine, uint64_t group_shift,
+                          uint64_t aia_irq_num, uint64_t aia_msi_num,
+                          uint64_t aplic_base, uint64_t imsic_base,
+                          uint64_t guest_num)
+{
+    int ret, i;
+    int aia_fd = -1;
+    uint64_t default_aia_mode;
+    uint64_t socket_count = riscv_socket_count(machine);
+    uint64_t max_hart_per_socket = 0;
+    uint64_t socket, base_hart, hart_count, socket_imsic_base, imsic_addr;
+    uint64_t socket_bits, hart_bits, guest_bits;
+
+    aia_fd = kvm_create_device(kvm_state, KVM_DEV_TYPE_RISCV_AIA, false);
+
+    if (aia_fd < 0) {
+        error_report("Unable to create in-kernel irqchip");
+        exit(1);
+    }
+
+    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
+                            KVM_DEV_RISCV_AIA_CONFIG_MODE,
+                            &default_aia_mode, false, NULL);
+    if (ret < 0) {
+        error_report("KVM AIA: failed to get current KVM AIA mode");
+        exit(1);
+    }
+    qemu_log("KVM AIA: default mode is %s\n",
+             kvm_aia_mode_str(default_aia_mode));
+
+    if (default_aia_mode != aia_mode) {
+        ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
+                                KVM_DEV_RISCV_AIA_CONFIG_MODE,
+                                &aia_mode, true, NULL);
+        if (ret < 0)
+            warn_report("KVM AIA: failed to set KVM AIA mode");
+        else
+            qemu_log("KVM AIA: set current mode to %s\n",
+                     kvm_aia_mode_str(aia_mode));
+    }
+
+    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
+                            KVM_DEV_RISCV_AIA_CONFIG_SRCS,
+                            &aia_irq_num, true, NULL);
+    if (ret < 0) {
+        error_report("KVM AIA: failed to set number of input irq lines");
+        exit(1);
+    }
+
+    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
+                            KVM_DEV_RISCV_AIA_CONFIG_IDS,
+                            &aia_msi_num, true, NULL);
+    if (ret < 0) {
+        error_report("KVM AIA: failed to set number of msi");
+        exit(1);
+    }
+
+    socket_bits = find_last_bit(&socket_count, BITS_PER_LONG) + 1;
+    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
+                            KVM_DEV_RISCV_AIA_CONFIG_GROUP_BITS,
+                            &socket_bits, true, NULL);
+    if (ret < 0) {
+        error_report("KVM AIA: failed to set group_bits");
+        exit(1);
+    }
+
+    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
+                            KVM_DEV_RISCV_AIA_CONFIG_GROUP_SHIFT,
+                            &group_shift, true, NULL);
+    if (ret < 0) {
+        error_report("KVM AIA: failed to set group_shift");
+        exit(1);
+    }
+
+    guest_bits = guest_num == 0 ? 0 :
+                 find_last_bit(&guest_num, BITS_PER_LONG) + 1;
+    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
+                            KVM_DEV_RISCV_AIA_CONFIG_GUEST_BITS,
+                            &guest_bits, true, NULL);
+    if (ret < 0) {
+        error_report("KVM AIA: failed to set guest_bits");
+        exit(1);
+    }
+
+    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_ADDR,
+                            KVM_DEV_RISCV_AIA_ADDR_APLIC,
+                            &aplic_base, true, NULL);
+    if (ret < 0) {
+        error_report("KVM AIA: failed to set the base address of APLIC");
+        exit(1);
+    }
+
+    for (socket = 0; socket < socket_count; socket++) {
+        socket_imsic_base = imsic_base + socket * (1U << group_shift);
+        hart_count = riscv_socket_hart_count(machine, socket);
+        base_hart = riscv_socket_first_hartid(machine, socket);
+
+        if (max_hart_per_socket < hart_count) {
+            max_hart_per_socket = hart_count;
+        }
+
+        for (i = 0; i < hart_count; i++) {
+            imsic_addr = socket_imsic_base + i * IMSIC_HART_SIZE(guest_bits);
+            ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_ADDR,
+                                    KVM_DEV_RISCV_AIA_ADDR_IMSIC(i + base_hart),
+                                    &imsic_addr, true, NULL);
+            if (ret < 0) {
+                error_report("KVM AIA: failed to set the IMSIC address for hart %d", i);
+                exit(1);
+            }
+        }
+    }
+
+    hart_bits = find_last_bit(&max_hart_per_socket, BITS_PER_LONG) + 1;
+    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CONFIG,
+                            KVM_DEV_RISCV_AIA_CONFIG_HART_BITS,
+                            &hart_bits, true, NULL);
+    if (ret < 0) {
+        error_report("KVM AIA: failed to set hart_bits");
+        exit(1);
+    }
+
+    if (kvm_has_gsi_routing()) {
+        for (uint64_t idx = 0; idx < aia_irq_num + 1; ++idx) {
+            /* KVM AIA only has one APLIC instance */
+            kvm_irqchip_add_irq_route(kvm_state, idx, 0, idx);
+        }
+        kvm_gsi_routing_allowed = true;
+        kvm_irqchip_commit_routes(kvm_state);
+    }
+
+    ret = kvm_device_access(aia_fd, KVM_DEV_RISCV_AIA_GRP_CTRL,
+                            KVM_DEV_RISCV_AIA_CTRL_INIT,
+                            NULL, true, NULL);
+    if (ret < 0) {
+        error_report("KVM AIA: initialized fail");
+        exit(1);
+    }
+
+    kvm_msi_via_irqfd_allowed = kvm_irqfds_enabled();
 }
-- 
2.41.0

From: Yong-Xuan Wang <yongxuan.wang@sifive.com>

KVM AIA can't emulate APLIC only. When "aia=aplic" parameter is passed,
APLIC devices is emulated by QEMU. For "aia=aplic-imsic", remove the
mmio operations of APLIC when using KVM AIA and send wired interrupt
signal via KVM_IRQ_LINE API.
After KVM AIA enabled, MSI messages are delivered by KVM_SIGNAL_MSI API
when the IMSICs receive mmio write requests.

Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
Reviewed-by: Jim Shu <jim.shu@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230727102439.22554-5-yongxuan.wang@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/intc/riscv_aplic.c | 56 ++++++++++++++++++++++++++++++-------------
 hw/intc/riscv_imsic.c | 25 +++++++++++++++----
 2 files changed, 61 insertions(+), 20 deletions(-)

diff --git a/hw/intc/riscv_aplic.c b/hw/intc/riscv_aplic.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/riscv_aplic.c
+++ b/hw/intc/riscv_aplic.c
@@ -XXX,XX +XXX,XX @@
 #include "hw/irq.h"
 #include "target/riscv/cpu.h"
 #include "sysemu/sysemu.h"
+#include "sysemu/kvm.h"
 #include "migration/vmstate.h"
 
 #define APLIC_MAX_IDC                  (1UL << 14)
@@ -XXX,XX +XXX,XX @@
 
 #define APLIC_IDC_CLAIMI               0x1c
 
+/*
+ * KVM AIA only supports APLIC MSI, fallback to QEMU emulation if we want to use
+ * APLIC Wired.
+ */
+static bool is_kvm_aia(bool msimode)
+{
+    return kvm_irqchip_in_kernel() && msimode;
+}
+
 static uint32_t riscv_aplic_read_input_word(RISCVAPLICState *aplic,
                                             uint32_t word)
 {
@@ -XXX,XX +XXX,XX @@ static uint32_t riscv_aplic_idc_claimi(RISCVAPLICState *aplic, uint32_t idc)
     return topi;
 }
 
+static void riscv_kvm_aplic_request(void *opaque, int irq, int level)
+{
+    kvm_set_irq(kvm_state, irq, !!level);
+}
+
 static void riscv_aplic_request(void *opaque, int irq, int level)
 {
     bool update = false;
@@ -XXX,XX +XXX,XX @@ static void riscv_aplic_realize(DeviceState *dev, Error **errp)
     uint32_t i;
     RISCVAPLICState *aplic = RISCV_APLIC(dev);
 
-    aplic->bitfield_words = (aplic->num_irqs + 31) >> 5;
-    aplic->sourcecfg = g_new0(uint32_t, aplic->num_irqs);
-    aplic->state = g_new0(uint32_t, aplic->num_irqs);
-    aplic->target = g_new0(uint32_t, aplic->num_irqs);
-    if (!aplic->msimode) {
-        for (i = 0; i < aplic->num_irqs; i++) {
-            aplic->target[i] = 1;
+    if (!is_kvm_aia(aplic->msimode)) {
+        aplic->bitfield_words = (aplic->num_irqs + 31) >> 5;
+        aplic->sourcecfg = g_new0(uint32_t, aplic->num_irqs);
+        aplic->state = g_new0(uint32_t, aplic->num_irqs);
+        aplic->target = g_new0(uint32_t, aplic->num_irqs);
+        if (!aplic->msimode) {
+            for (i = 0; i < aplic->num_irqs; i++) {
+                aplic->target[i] = 1;
+            }
         }
-    }
-    aplic->idelivery = g_new0(uint32_t, aplic->num_harts);
-    aplic->iforce = g_new0(uint32_t, aplic->num_harts);
-    aplic->ithreshold = g_new0(uint32_t, aplic->num_harts);
+        aplic->idelivery = g_new0(uint32_t, aplic->num_harts);
+        aplic->iforce = g_new0(uint32_t, aplic->num_harts);
+        aplic->ithreshold = g_new0(uint32_t, aplic->num_harts);
 
-    memory_region_init_io(&aplic->mmio, OBJECT(dev), &riscv_aplic_ops, aplic,
-                          TYPE_RISCV_APLIC, aplic->aperture_size);
-    sysbus_init_mmio(SYS_BUS_DEVICE(dev), &aplic->mmio);
+        memory_region_init_io(&aplic->mmio, OBJECT(dev), &riscv_aplic_ops,
+                              aplic, TYPE_RISCV_APLIC, aplic->aperture_size);
+        sysbus_init_mmio(SYS_BUS_DEVICE(dev), &aplic->mmio);
+    }
 
     /*
      * Only root APLICs have hardware IRQ lines. All non-root APLICs
      * have IRQ lines delegated by their parent APLIC.
      */
     if (!aplic->parent) {
-        qdev_init_gpio_in(dev, riscv_aplic_request, aplic->num_irqs);
+        if (is_kvm_aia(aplic->msimode)) {
+            qdev_init_gpio_in(dev, riscv_kvm_aplic_request, aplic->num_irqs);
+        } else {
+            qdev_init_gpio_in(dev, riscv_aplic_request, aplic->num_irqs);
+        }
     }
 
     /* Create output IRQ lines for non-MSI mode */
@@ -XXX,XX +XXX,XX @@ DeviceState *riscv_aplic_create(hwaddr addr, hwaddr size,
     qdev_prop_set_bit(dev, "mmode", mmode);
 
     sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal);
-    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, addr);
+
+    if (!is_kvm_aia(msimode)) {
+        sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, addr);
+    }
 
     if (parent) {
         riscv_aplic_add_child(parent, dev);
diff --git a/hw/intc/riscv_imsic.c b/hw/intc/riscv_imsic.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/riscv_imsic.c
+++ b/hw/intc/riscv_imsic.c
@@ -XXX,XX +XXX,XX @@
 #include "target/riscv/cpu.h"
 #include "target/riscv/cpu_bits.h"
 #include "sysemu/sysemu.h"
+#include "sysemu/kvm.h"
 #include "migration/vmstate.h"
 
 #define IMSIC_MMIO_PAGE_LE             0x00
@@ -XXX,XX +XXX,XX @@ static void riscv_imsic_write(void *opaque, hwaddr addr, uint64_t value,
         goto err;
     }
 
+#if defined(CONFIG_KVM)
+    if (kvm_irqchip_in_kernel()) {
+        struct kvm_msi msi;
+
+        msi.address_lo = extract64(imsic->mmio.addr + addr, 0, 32);
+        msi.address_hi = extract64(imsic->mmio.addr + addr, 32, 32);
+        msi.data = le32_to_cpu(value);
+
+        kvm_vm_ioctl(kvm_state, KVM_SIGNAL_MSI, &msi);
+
+        return;
+    }
+#endif
+
     /* Writes only supported for MSI little-endian registers */
     page = addr >> IMSIC_MMIO_PAGE_SHIFT;
     if ((addr & (IMSIC_MMIO_PAGE_SZ - 1)) == IMSIC_MMIO_PAGE_LE) {
@@ -XXX,XX +XXX,XX @@ static void riscv_imsic_realize(DeviceState *dev, Error **errp)
     CPUState *cpu = cpu_by_arch_id(imsic->hartid);
     CPURISCVState *env = cpu ? cpu->env_ptr : NULL;
 
-    imsic->num_eistate = imsic->num_pages * imsic->num_irqs;
-    imsic->eidelivery = g_new0(uint32_t, imsic->num_pages);
-    imsic->eithreshold = g_new0(uint32_t, imsic->num_pages);
-    imsic->eistate = g_new0(uint32_t, imsic->num_eistate);
+    if (!kvm_irqchip_in_kernel()) {
+        imsic->num_eistate = imsic->num_pages * imsic->num_irqs;
+        imsic->eidelivery = g_new0(uint32_t, imsic->num_pages);
+        imsic->eithreshold = g_new0(uint32_t, imsic->num_pages);
+        imsic->eistate = g_new0(uint32_t, imsic->num_eistate);
+    }
 
     memory_region_init_io(&imsic->mmio, OBJECT(dev), &riscv_imsic_ops,
                           imsic, TYPE_RISCV_IMSIC,
-- 
2.41.0

From: Yong-Xuan Wang <yongxuan.wang@sifive.com>

Select KVM AIA when the host kernel has in-kernel AIA chip support.
Since KVM AIA only has one APLIC instance, we map the QEMU APLIC
devices to KVM APLIC.

Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
Reviewed-by: Jim Shu <jim.shu@sifive.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230727102439.22554-6-yongxuan.wang@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/riscv/virt.c | 94 +++++++++++++++++++++++++++++++++----------------
 1 file changed, 63 insertions(+), 31 deletions(-)

diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -XXX,XX +XXX,XX @@
 #include "hw/riscv/virt.h"
 #include "hw/riscv/boot.h"
 #include "hw/riscv/numa.h"
+#include "kvm_riscv.h"
 #include "hw/intc/riscv_aclint.h"
 #include "hw/intc/riscv_aplic.h"
 #include "hw/intc/riscv_imsic.h"
@@ -XXX,XX +XXX,XX @@
 #error "Can't accommodate all IMSIC groups in address space"
 #endif
 
+/* KVM AIA only supports APLIC MSI. APLIC Wired is always emulated by QEMU. */
+static bool virt_use_kvm_aia(RISCVVirtState *s)
+{
+    return kvm_irqchip_in_kernel() && s->aia_type == VIRT_AIA_TYPE_APLIC_IMSIC;
+}
+
 static const MemMapEntry virt_memmap[] = {
     [VIRT_DEBUG] =        {        0x0,         0x100 },
     [VIRT_MROM] =         {     0x1000,        0xf000 },
@@ -XXX,XX +XXX,XX @@ static void create_fdt_one_aplic(RISCVVirtState *s, int socket,
                                  uint32_t *intc_phandles,
                                  uint32_t aplic_phandle,
                                  uint32_t aplic_child_phandle,
-                                 bool m_mode)
+                                 bool m_mode, int num_harts)
 {
     int cpu;
     char *aplic_name;
     uint32_t *aplic_cells;
     MachineState *ms = MACHINE(s);
 
-    aplic_cells = g_new0(uint32_t, s->soc[socket].num_harts * 2);
+    aplic_cells = g_new0(uint32_t, num_harts * 2);
 
-    for (cpu = 0; cpu < s->soc[socket].num_harts; cpu++) {
+    for (cpu = 0; cpu < num_harts; cpu++) {
         aplic_cells[cpu * 2 + 0] = cpu_to_be32(intc_phandles[cpu]);
         aplic_cells[cpu * 2 + 1] = cpu_to_be32(m_mode ? IRQ_M_EXT : IRQ_S_EXT);
     }
@@ -XXX,XX +XXX,XX @@ static void create_fdt_one_aplic(RISCVVirtState *s, int socket,
 
     if (s->aia_type == VIRT_AIA_TYPE_APLIC) {
         qemu_fdt_setprop(ms->fdt, aplic_name, "interrupts-extended",
-                         aplic_cells,
-                         s->soc[socket].num_harts * sizeof(uint32_t) * 2);
+                         aplic_cells, num_harts * sizeof(uint32_t) * 2);
     } else {
         qemu_fdt_setprop_cell(ms->fdt, aplic_name, "msi-parent", msi_phandle);
     }
@@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aplic(RISCVVirtState *s,
                                     uint32_t msi_s_phandle,
                                     uint32_t *phandle,
                                     uint32_t *intc_phandles,
-                                    uint32_t *aplic_phandles)
+                                    uint32_t *aplic_phandles,
+                                    int num_harts)
 {
     char *aplic_name;
     unsigned long aplic_addr;
@@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aplic(RISCVVirtState *s,
         create_fdt_one_aplic(s, socket, aplic_addr, memmap[VIRT_APLIC_M].size,
                              msi_m_phandle, intc_phandles,
                              aplic_m_phandle, aplic_s_phandle,
-                             true);
+                             true, num_harts);
     }
 
     /* S-level APLIC node */
@@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aplic(RISCVVirtState *s,
     create_fdt_one_aplic(s, socket, aplic_addr, memmap[VIRT_APLIC_S].size,
                          msi_s_phandle, intc_phandles,
                          aplic_s_phandle, 0,
-                         false);
+                         false, num_harts);
 
     aplic_name = g_strdup_printf("/soc/aplic@%lx", aplic_addr);
 
@@ -XXX,XX +XXX,XX @@ static void create_fdt_sockets(RISCVVirtState *s, const MemMapEntry *memmap,
         *msi_pcie_phandle = msi_s_phandle;
     }
 
-    phandle_pos = ms->smp.cpus;
-    for (socket = (socket_count - 1); socket >= 0; socket--) {
-        phandle_pos -= s->soc[socket].num_harts;
-
-        if (s->aia_type == VIRT_AIA_TYPE_NONE) {
-            create_fdt_socket_plic(s, memmap, socket, phandle,
-                &intc_phandles[phandle_pos], xplic_phandles);
-        } else {
-            create_fdt_socket_aplic(s, memmap, socket,
-                msi_m_phandle, msi_s_phandle, phandle,
-                &intc_phandles[phandle_pos], xplic_phandles);
+    /* KVM AIA only has one APLIC instance */
+    if (virt_use_kvm_aia(s)) {
+        create_fdt_socket_aplic(s, memmap, 0,
+                                msi_m_phandle, msi_s_phandle, phandle,
+                                &intc_phandles[0], xplic_phandles,
+                                ms->smp.cpus);
+    } else {
+        phandle_pos = ms->smp.cpus;
+        for (socket = (socket_count - 1); socket >= 0; socket--) {
+            phandle_pos -= s->soc[socket].num_harts;
+
+            if (s->aia_type == VIRT_AIA_TYPE_NONE) {
+                create_fdt_socket_plic(s, memmap, socket, phandle,
+                                       &intc_phandles[phandle_pos],
+                                       xplic_phandles);
+            } else {
+                create_fdt_socket_aplic(s, memmap, socket,
+                                        msi_m_phandle, msi_s_phandle, phandle,
+                                        &intc_phandles[phandle_pos],
+                                        xplic_phandles,
+                                        s->soc[socket].num_harts);
+            }
         }
     }
 
     g_free(intc_phandles);
 
-    for (socket = 0; socket < socket_count; socket++) {
-        if (socket == 0) {
-            *irq_mmio_phandle = xplic_phandles[socket];
-            *irq_virtio_phandle = xplic_phandles[socket];
-            *irq_pcie_phandle = xplic_phandles[socket];
-        }
-        if (socket == 1) {
-            *irq_virtio_phandle = xplic_phandles[socket];
-            *irq_pcie_phandle = xplic_phandles[socket];
-        }
-        if (socket == 2) {
-            *irq_pcie_phandle = xplic_phandles[socket];
+    if (virt_use_kvm_aia(s)) {
+        *irq_mmio_phandle = xplic_phandles[0];
+        *irq_virtio_phandle = xplic_phandles[0];
+        *irq_pcie_phandle = xplic_phandles[0];
+    } else {
+        for (socket = 0; socket < socket_count; socket++) {
+            if (socket == 0) {
+                *irq_mmio_phandle = xplic_phandles[socket];
+                *irq_virtio_phandle = xplic_phandles[socket];
+                *irq_pcie_phandle = xplic_phandles[socket];
+            }
+            if (socket == 1) {
+                *irq_virtio_phandle = xplic_phandles[socket];
+                *irq_pcie_phandle = xplic_phandles[socket];
+            }
+            if (socket == 2) {
+                *irq_pcie_phandle = xplic_phandles[socket];
+            }
         }
     }
 
@@ -XXX,XX +XXX,XX @@ static void virt_machine_init(MachineState *machine)
         }
     }
 
+    if (virt_use_kvm_aia(s)) {
+        kvm_riscv_aia_create(machine, IMSIC_MMIO_GROUP_MIN_SHIFT,
+                             VIRT_IRQCHIP_NUM_SOURCES, VIRT_IRQCHIP_NUM_MSIS,
+                             memmap[VIRT_APLIC_S].base,
+                             memmap[VIRT_IMSIC_S].base,
+                             s->aia_guests);
+    }
+
     if (riscv_is_32bit(&s->soc[0])) {
 #if HOST_LONG_BITS == 64
         /* limit RAM size in a 32-bit system */
-- 
2.41.0

From: Conor Dooley <conor.dooley@microchip.com>

On a dtb dumped from the virt machine, dt-validate complains:
soc: pmu: {'riscv,event-to-mhpmcounters': [[1, 1, 524281], [2, 2, 524284], [65561, 65561, 524280], [65563, 65563, 524280], [65569, 65569, 524280]], 'compatible': ['riscv,pmu']} should not be valid under {'type': 'object'}
        from schema $id: http://devicetree.org/schemas/simple-bus.yaml#
That's pretty cryptic, but running the dtb back through dtc produces
something a lot more reasonable:
Warning (simple_bus_reg): /soc/pmu: missing or empty reg/ranges property

Moving the riscv,pmu node out of the soc bus solves the problem.

Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Message-ID: <20230727-groom-decline-2c57ce42841c@spud>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/riscv/virt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -XXX,XX +XXX,XX @@ static void create_fdt_pmu(RISCVVirtState *s)
     MachineState *ms = MACHINE(s);
     RISCVCPU hart = s->soc[0].harts[0];
 
-    pmu_name = g_strdup_printf("/soc/pmu");
+    pmu_name = g_strdup_printf("/pmu");
     qemu_fdt_add_subnode(ms->fdt, pmu_name);
     qemu_fdt_setprop_string(ms->fdt, pmu_name, "compatible", "riscv,pmu");
     riscv_pmu_generate_fdt_node(ms->fdt, hart.cfg.pmu_num, pmu_name);
-- 
2.41.0

From: Weiwei Li <liweiwei@iscas.ac.cn>

The Svadu specification updated the name of the *envcfg bit from
HADE to ADUE.

Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn>
Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Message-ID: <20230816141916.66898-1-liweiwei@iscas.ac.cn>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu_bits.h   |  8 ++++----
 target/riscv/cpu.c        |  4 ++--
 target/riscv/cpu_helper.c |  6 +++---
 target/riscv/csr.c        | 12 ++++++------
 4 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/target/riscv/cpu_bits.h b/target/riscv/cpu_bits.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu_bits.h
+++ b/target/riscv/cpu_bits.h
@@ -XXX,XX +XXX,XX @@ typedef enum RISCVException {
 #define MENVCFG_CBIE                       (3UL << 4)
 #define MENVCFG_CBCFE                      BIT(6)
 #define MENVCFG_CBZE                       BIT(7)
-#define MENVCFG_HADE                       (1ULL << 61)
+#define MENVCFG_ADUE                       (1ULL << 61)
 #define MENVCFG_PBMTE                      (1ULL << 62)
 #define MENVCFG_STCE                       (1ULL << 63)
 
 /* For RV32 */
-#define MENVCFGH_HADE                      BIT(29)
+#define MENVCFGH_ADUE                      BIT(29)
 #define MENVCFGH_PBMTE                     BIT(30)
 #define MENVCFGH_STCE                      BIT(31)
 
@@ -XXX,XX +XXX,XX @@ typedef enum RISCVException {
 #define HENVCFG_CBIE                       MENVCFG_CBIE
 #define HENVCFG_CBCFE                      MENVCFG_CBCFE
 #define HENVCFG_CBZE                       MENVCFG_CBZE
-#define HENVCFG_HADE                       MENVCFG_HADE
+#define HENVCFG_ADUE                       MENVCFG_ADUE
 #define HENVCFG_PBMTE                      MENVCFG_PBMTE
 #define HENVCFG_STCE                       MENVCFG_STCE
 
 /* For RV32 */
-#define HENVCFGH_HADE                       MENVCFGH_HADE
+#define HENVCFGH_ADUE                       MENVCFGH_ADUE
 #define HENVCFGH_PBMTE                      MENVCFGH_PBMTE
 #define HENVCFGH_STCE                       MENVCFGH_STCE
 
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static void riscv_cpu_reset_hold(Object *obj)
     env->two_stage_lookup = false;
 
     env->menvcfg = (cpu->cfg.ext_svpbmt ? MENVCFG_PBMTE : 0) |
-                   (cpu->cfg.ext_svadu ? MENVCFG_HADE : 0);
+                   (cpu->cfg.ext_svadu ? MENVCFG_ADUE : 0);
     env->henvcfg = (cpu->cfg.ext_svpbmt ? HENVCFG_PBMTE : 0) |
-                   (cpu->cfg.ext_svadu ? HENVCFG_HADE : 0);
+                   (cpu->cfg.ext_svadu ? HENVCFG_ADUE : 0);
 
     /* Initialized default priorities of local interrupts. */
     for (i = 0; i < ARRAY_SIZE(env->miprio); i++) {
diff --git a/target/riscv/cpu_helper.c b/target/riscv/cpu_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu_helper.c
+++ b/target/riscv/cpu_helper.c
@@ -XXX,XX +XXX,XX @@ static int get_physical_address(CPURISCVState *env, hwaddr *physical,
     }
 
     bool pbmte = env->menvcfg & MENVCFG_PBMTE;
-    bool hade = env->menvcfg & MENVCFG_HADE;
+    bool adue = env->menvcfg & MENVCFG_ADUE;
 
     if (first_stage && two_stage && env->virt_enabled) {
         pbmte = pbmte && (env->henvcfg & HENVCFG_PBMTE);
-        hade = hade && (env->henvcfg & HENVCFG_HADE);
+        adue = adue && (env->henvcfg & HENVCFG_ADUE);
     }
 
     int ptshift = (levels - 1) * ptidxbits;
@@ -XXX,XX +XXX,XX @@ restart:
 
     /* Page table updates need to be atomic with MTTCG enabled */
     if (updated_pte != pte && !is_debug) {
-        if (!hade) {
+        if (!adue) {
             return TRANSLATE_FAIL;
         }
 
diff --git a/target/riscv/csr.c b/target/riscv/csr.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/csr.c
+++ b/target/riscv/csr.c
@@ -XXX,XX +XXX,XX @@ static RISCVException write_menvcfg(CPURISCVState *env, int csrno,
     if (riscv_cpu_mxl(env) == MXL_RV64) {
         mask |= (cfg->ext_svpbmt ? MENVCFG_PBMTE : 0) |
                 (cfg->ext_sstc ? MENVCFG_STCE : 0) |
-                (cfg->ext_svadu ? MENVCFG_HADE : 0);
+                (cfg->ext_svadu ? MENVCFG_ADUE : 0);
     }
     env->menvcfg = (env->menvcfg & ~mask) | (val & mask);
 
@@ -XXX,XX +XXX,XX @@ static RISCVException write_menvcfgh(CPURISCVState *env, int csrno,
     const RISCVCPUConfig *cfg = riscv_cpu_cfg(env);
     uint64_t mask = (cfg->ext_svpbmt ? MENVCFG_PBMTE : 0) |
                     (cfg->ext_sstc ? MENVCFG_STCE : 0) |
-                    (cfg->ext_svadu ? MENVCFG_HADE : 0);
+                    (cfg->ext_svadu ? MENVCFG_ADUE : 0);
     uint64_t valh = (uint64_t)val << 32;
 
     env->menvcfg = (env->menvcfg & ~mask) | (valh & mask);
@@ -XXX,XX +XXX,XX @@ static RISCVException read_henvcfg(CPURISCVState *env, int csrno,
      * henvcfg.stce is read_only 0 when menvcfg.stce = 0
      * henvcfg.hade is read_only 0 when menvcfg.hade = 0
      */
-    *val = env->henvcfg & (~(HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_HADE) |
+    *val = env->henvcfg & (~(HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_ADUE) |
                            env->menvcfg);
     return RISCV_EXCP_NONE;
 }
@@ -XXX,XX +XXX,XX @@ static RISCVException write_henvcfg(CPURISCVState *env, int csrno,
     }
 
     if (riscv_cpu_mxl(env) == MXL_RV64) {
-        mask |= env->menvcfg & (HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_HADE);
+        mask |= env->menvcfg & (HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_ADUE);
     }
 
     env->henvcfg = (env->henvcfg & ~mask) | (val & mask);
@@ -XXX,XX +XXX,XX @@ static RISCVException read_henvcfgh(CPURISCVState *env, int csrno,
         return ret;
     }
 
-    *val = (env->henvcfg & (~(HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_HADE) |
+    *val = (env->henvcfg & (~(HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_ADUE) |
                             env->menvcfg)) >> 32;
     return RISCV_EXCP_NONE;
 }
@@ -XXX,XX +XXX,XX @@ static RISCVException write_henvcfgh(CPURISCVState *env, int csrno,
                                      target_ulong val)
 {
     uint64_t mask = env->menvcfg & (HENVCFG_PBMTE | HENVCFG_STCE |
-                                    HENVCFG_HADE);
+                                    HENVCFG_ADUE);
     uint64_t valh = (uint64_t)val << 32;
     RISCVException ret;
 
-- 
2.41.0

From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>

In the same emulated RISC-V host, the 'host' KVM CPU takes 4 times
longer to boot than the 'rv64' KVM CPU.

The reason is an unintended behavior of riscv_cpu_satp_mode_finalize()
when satp_mode.supported = 0, i.e. when cpu_init() does not set
satp_mode_max_supported(). satp_mode_max_from_map(map) does:

31 - __builtin_clz(map)

This means that, if satp_mode.supported = 0, satp_mode_supported_max
wil be '31 - 32'. But this is C, so satp_mode_supported_max will gladly
set it to UINT_MAX (4294967295). After that, if the user didn't set a
satp_mode, set_satp_mode_default_map(cpu) will make

cfg.satp_mode.map = cfg.satp_mode.supported

So satp_mode.map = 0. And then satp_mode_map_max will be set to
satp_mode_max_from_map(cpu->cfg.satp_mode.map), i.e. also UINT_MAX. The
guard "satp_mode_map_max > satp_mode_supported_max" doesn't protect us
here since both are UINT_MAX.

And finally we have 2 loops:

for (int i = satp_mode_map_max - 1; i >= 0; --i) {

Which are, in fact, 2 loops from UINT_MAX -1 to -1. This is where the
extra delay when booting the 'host' CPU is coming from.

Commit 43d1de32f8 already set a precedence for satp_mode.supported = 0
in a different manner. We're doing the same here. If supported == 0,
interpret as 'the CPU wants the OS to handle satp mode alone' and skip
satp_mode_finalize().

We'll also put a guard in satp_mode_max_from_map() to assert out if map
is 0 since the function is not ready to deal with it.

Cc: Alexandre Ghiti <alexghiti@rivosinc.com>
Fixes: 6f23aaeb9b ("riscv: Allow user to set the satp mode")
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230817152903.694926-1-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu.c | 23 ++++++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)

diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static uint8_t satp_mode_from_str(const char *satp_mode_str)
 
 uint8_t satp_mode_max_from_map(uint32_t map)
 {
+    /*
+     * 'map = 0' will make us return (31 - 32), which C will
+     * happily overflow to UINT_MAX. There's no good result to
+     * return if 'map = 0' (e.g. returning 0 will be ambiguous
+     * with the result for 'map = 1').
+     *
+     * Assert out if map = 0. Callers will have to deal with
+     * it outside of this function.
+     */
+    g_assert(map > 0);
+
     /* map here has at least one bit set, so no problem with clz */
     return 31 - __builtin_clz(map);
 }
@@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
 static void riscv_cpu_satp_mode_finalize(RISCVCPU *cpu, Error **errp)
 {
     bool rv32 = riscv_cpu_mxl(&cpu->env) == MXL_RV32;
-    uint8_t satp_mode_map_max;
-    uint8_t satp_mode_supported_max =
-                        satp_mode_max_from_map(cpu->cfg.satp_mode.supported);
+    uint8_t satp_mode_map_max, satp_mode_supported_max;
+
+    /* The CPU wants the OS to decide which satp mode to use */
+    if (cpu->cfg.satp_mode.supported == 0) {
+        return;
+    }
+
+    satp_mode_supported_max =
+                    satp_mode_max_from_map(cpu->cfg.satp_mode.supported);
 
     if (cpu->cfg.satp_mode.map == 0) {
         if (cpu->cfg.satp_mode.init == 0) {
-- 
2.41.0

From: Vineet Gupta <vineetg@rivosinc.com>

zicond is now codegen supported in both llvm and gcc.

This change allows seamless enabling/testing of zicond in downstream
projects. e.g. currently riscv-gnu-toolchain parses elf attributes
to create a cmdline for qemu but fails short of enabling it because of
the "x-" prefix.

Signed-off-by: Vineet Gupta <vineetg@rivosinc.com>
Message-ID: <20230808181715.436395-1-vineetg@rivosinc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/cpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
     DEFINE_PROP_BOOL("zcf", RISCVCPU, cfg.ext_zcf, false),
     DEFINE_PROP_BOOL("zcmp", RISCVCPU, cfg.ext_zcmp, false),
     DEFINE_PROP_BOOL("zcmt", RISCVCPU, cfg.ext_zcmt, false),
+    DEFINE_PROP_BOOL("zicond", RISCVCPU, cfg.ext_zicond, false),
 
     /* Vendor-specific custom extensions */
     DEFINE_PROP_BOOL("xtheadba", RISCVCPU, cfg.ext_xtheadba, false),
@@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_extensions[] = {
     DEFINE_PROP_BOOL("xventanacondops", RISCVCPU, cfg.ext_XVentanaCondOps, false),
 
     /* These are experimental so mark with 'x-' */
-    DEFINE_PROP_BOOL("x-zicond", RISCVCPU, cfg.ext_zicond, false),
 
     /* ePMP 0.9.3 */
     DEFINE_PROP_BOOL("x-epmp", RISCVCPU, cfg.epmp, false),
-- 
2.41.0

From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>

A build with --enable-debug and without KVM will fail as follows:

/usr/bin/ld: libqemu-riscv64-softmmu.fa.p/hw_riscv_virt.c.o: in function `virt_machine_init':
./qemu/build/../hw/riscv/virt.c:1465: undefined reference to `kvm_riscv_aia_create'

This happens because the code block with "if virt_use_kvm_aia(s)" isn't
being ignored by the debug build, resulting in an undefined reference to
a KVM only function.

Add a 'kvm_enabled()' conditional together with virt_use_kvm_aia() will
make the compiler crop the kvm_riscv_aia_create() call entirely from a
non-KVM build. Note that adding the 'kvm_enabled()' conditional inside
virt_use_kvm_aia() won't fix the build because this function would need
to be inlined multiple times to make the compiler zero out the entire
block.

While we're at it, use kvm_enabled() in all instances where
virt_use_kvm_aia() is checked to allow the compiler to elide these other
kvm-only instances as well.

Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Fixes: dbdb99948e ("target/riscv: select KVM AIA in riscv virt machine")
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20230830133503.711138-2-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 hw/riscv/virt.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -XXX,XX +XXX,XX @@ static void create_fdt_sockets(RISCVVirtState *s, const MemMapEntry *memmap,
     }
 
     /* KVM AIA only has one APLIC instance */
-    if (virt_use_kvm_aia(s)) {
+    if (kvm_enabled() && virt_use_kvm_aia(s)) {
         create_fdt_socket_aplic(s, memmap, 0,
                                 msi_m_phandle, msi_s_phandle, phandle,
                                 &intc_phandles[0], xplic_phandles,
@@ -XXX,XX +XXX,XX @@ static void create_fdt_sockets(RISCVVirtState *s, const MemMapEntry *memmap,
 
     g_free(intc_phandles);
 
-    if (virt_use_kvm_aia(s)) {
+    if (kvm_enabled() && virt_use_kvm_aia(s)) {
         *irq_mmio_phandle = xplic_phandles[0];
         *irq_virtio_phandle = xplic_phandles[0];
         *irq_pcie_phandle = xplic_phandles[0];
@@ -XXX,XX +XXX,XX @@ static void virt_machine_init(MachineState *machine)
         }
     }
 
-    if (virt_use_kvm_aia(s)) {
+    if (kvm_enabled() && virt_use_kvm_aia(s)) {
         kvm_riscv_aia_create(machine, IMSIC_MMIO_GROUP_MIN_SHIFT,
                              VIRT_IRQCHIP_NUM_SOURCES, VIRT_IRQCHIP_NUM_MSIS,
                              memmap[VIRT_APLIC_S].base,
-- 
2.41.0

From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>

Commit 6df0b37e2ab breaks a --enable-debug build in a non-KVM
environment with the following error:

/usr/bin/ld: libqemu-riscv64-softmmu.fa.p/hw_intc_riscv_aplic.c.o: in function `riscv_kvm_aplic_request':
./qemu/build/../hw/intc/riscv_aplic.c:486: undefined reference to `kvm_set_irq'
collect2: error: ld returned 1 exit status

This happens because the debug build will poke into the
'if (is_kvm_aia(aplic->msimode))' block and fail to find a reference to
the KVM only function riscv_kvm_aplic_request().

There are multiple solutions to fix this. We'll go with the same
solution from the previous patch, i.e. add a kvm_enabled() conditional
to filter out the block. But there's a catch: riscv_kvm_aplic_request()
is a local function that would end up being used if the compiler crops
the block, and this won't work. Quoting Richard Henderson's explanation
in [1]:

"(...) the compiler won't eliminate entire unused functions with -O0"

We'll solve it by moving riscv_kvm_aplic_request() to kvm.c and add its
declaration in kvm_riscv.h, where all other KVM specific public
functions are already declared. Other archs handles KVM specific code in
this manner and we expect to do the same from now on.

[1] https://lore.kernel.org/qemu-riscv/d2f1ad02-eb03-138f-9d08-db676deeed05@linaro.org/

Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20230830133503.711138-3-dbarboza@ventanamicro.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/kvm_riscv.h | 1 +
 hw/intc/riscv_aplic.c    | 8 ++------
 target/riscv/kvm.c       | 5 +++++
 3 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/kvm_riscv.h
+++ b/target/riscv/kvm_riscv.h
@@ -XXX,XX +XXX,XX @@ void kvm_riscv_aia_create(MachineState *machine, uint64_t group_shift,
                           uint64_t aia_irq_num, uint64_t aia_msi_num,
                           uint64_t aplic_base, uint64_t imsic_base,
                           uint64_t guest_num);
+void riscv_kvm_aplic_request(void *opaque, int irq, int level);
 
 #endif
diff --git a/hw/intc/riscv_aplic.c b/hw/intc/riscv_aplic.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/riscv_aplic.c
+++ b/hw/intc/riscv_aplic.c
@@ -XXX,XX +XXX,XX @@
 #include "target/riscv/cpu.h"
 #include "sysemu/sysemu.h"
 #include "sysemu/kvm.h"
+#include "kvm_riscv.h"
 #include "migration/vmstate.h"
 
 #define APLIC_MAX_IDC                  (1UL << 14)
@@ -XXX,XX +XXX,XX @@ static uint32_t riscv_aplic_idc_claimi(RISCVAPLICState *aplic, uint32_t idc)
     return topi;
 }
 
-static void riscv_kvm_aplic_request(void *opaque, int irq, int level)
-{
-    kvm_set_irq(kvm_state, irq, !!level);
-}
-
 static void riscv_aplic_request(void *opaque, int irq, int level)
 {
     bool update = false;
@@ -XXX,XX +XXX,XX @@ static void riscv_aplic_realize(DeviceState *dev, Error **errp)
      * have IRQ lines delegated by their parent APLIC.
      */
     if (!aplic->parent) {
-        if (is_kvm_aia(aplic->msimode)) {
+        if (kvm_enabled() && is_kvm_aia(aplic->msimode)) {
             qdev_init_gpio_in(dev, riscv_kvm_aplic_request, aplic->num_irqs);
         } else {
             qdev_init_gpio_in(dev, riscv_aplic_request, aplic->num_irqs);
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -XXX,XX +XXX,XX @@
 #include "sysemu/runstate.h"
 #include "hw/riscv/numa.h"
 
+void riscv_kvm_aplic_request(void *opaque, int irq, int level)
+{
+    kvm_set_irq(kvm_state, irq, !!level);
+}
+
 static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type,
                                  uint64_t idx)
 {
-- 
2.41.0

From: Robbin Ehn <rehn@rivosinc.com>

This patch adds the new extensions in
linux 6.5 to the hwprobe syscall.

And fixes RVC check to OR with correct value.
The previous variable contains 0 therefore it
did work.

Signed-off-by: Robbin Ehn <rehn@rivosinc.com>
Acked-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <bc82203b72d7efb30f1b4a8f9eb3d94699799dc8.camel@rivosinc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 linux-user/syscall.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/linux-user/syscall.c b/linux-user/syscall.c
index XXXXXXX..XXXXXXX 100644
--- a/linux-user/syscall.c
+++ b/linux-user/syscall.c
@@ -XXX,XX +XXX,XX @@ static int do_getdents64(abi_long dirfd, abi_long arg2, abi_long count)
 #define RISCV_HWPROBE_KEY_IMA_EXT_0     4
 #define     RISCV_HWPROBE_IMA_FD       (1 << 0)
 #define     RISCV_HWPROBE_IMA_C        (1 << 1)
+#define     RISCV_HWPROBE_IMA_V        (1 << 2)
+#define     RISCV_HWPROBE_EXT_ZBA      (1 << 3)
+#define     RISCV_HWPROBE_EXT_ZBB      (1 << 4)
+#define     RISCV_HWPROBE_EXT_ZBS      (1 << 5)
 
 #define RISCV_HWPROBE_KEY_CPUPERF_0     5
 #define     RISCV_HWPROBE_MISALIGNED_UNKNOWN     (0 << 0)
@@ -XXX,XX +XXX,XX @@ static void risc_hwprobe_fill_pairs(CPURISCVState *env,
                     riscv_has_ext(env, RVD) ?
                     RISCV_HWPROBE_IMA_FD : 0;
             value |= riscv_has_ext(env, RVC) ?
-                     RISCV_HWPROBE_IMA_C : pair->value;
+                     RISCV_HWPROBE_IMA_C : 0;
+            value |= riscv_has_ext(env, RVV) ?
+                     RISCV_HWPROBE_IMA_V : 0;
+            value |= cfg->ext_zba ?
+                     RISCV_HWPROBE_EXT_ZBA : 0;
+            value |= cfg->ext_zbb ?
+                     RISCV_HWPROBE_EXT_ZBB : 0;
+            value |= cfg->ext_zbs ?
+                     RISCV_HWPROBE_EXT_ZBS : 0;
             __put_user(value, &pair->value);
             break;
         case RISCV_HWPROBE_KEY_CPUPERF_0:
-- 
2.41.0

From: Ard Biesheuvel <ardb@kernel.org>

Use the accelerated SubBytes/ShiftRows/AddRoundKey AES helper to
implement the first half of the key schedule derivation. This does not
actually involve shifting rows, so clone the same value into all four
columns of the AES vector to counter that operation.

Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Philippe Mathieu-Daudé <philmd@linaro.org>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20230831154118.138727-1-ardb@kernel.org>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/crypto_helper.c | 17 +++++------------
 1 file changed, 5 insertions(+), 12 deletions(-)

diff --git a/target/riscv/crypto_helper.c b/target/riscv/crypto_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/crypto_helper.c
+++ b/target/riscv/crypto_helper.c
@@ -XXX,XX +XXX,XX @@ target_ulong HELPER(aes64ks1i)(target_ulong rs1, target_ulong rnum)
 
     uint8_t enc_rnum = rnum;
     uint32_t temp = (RS1 >> 32) & 0xFFFFFFFF;
-    uint8_t rcon_ = 0;
-    target_ulong result;
+    AESState t, rc = {};
 
     if (enc_rnum != 0xA) {
         temp = ror32(temp, 8); /* Rotate right by 8 */
-        rcon_ = round_consts[enc_rnum];
+        rc.w[0] = rc.w[1] = round_consts[enc_rnum];
     }
 
-    temp = ((uint32_t)AES_sbox[(temp >> 24) & 0xFF] << 24) |
-           ((uint32_t)AES_sbox[(temp >> 16) & 0xFF] << 16) |
-           ((uint32_t)AES_sbox[(temp >> 8) & 0xFF] << 8) |
-           ((uint32_t)AES_sbox[(temp >> 0) & 0xFF] << 0);
+    t.w[0] = t.w[1] = t.w[2] = t.w[3] = temp;
+    aesenc_SB_SR_AK(&t, &t, &rc, false);
 
-    temp ^= rcon_;
-
-    result = ((uint64_t)temp << 32) | temp;
-
-    return result;
+    return t.d[0];
 }
 
 target_ulong HELPER(aes64im)(target_ulong rs1)
-- 
2.41.0

From: Akihiko Odaki <akihiko.odaki@daynix.com>

riscv_trigger_init() had been called on reset events that can happen
several times for a CPU and it allocated timers for itrigger. If old
timers were present, they were simply overwritten by the new timers,
resulting in a memory leak.

Divide riscv_trigger_init() into two functions, namely
riscv_trigger_realize() and riscv_trigger_reset() and call them in
appropriate timing. The timer allocation will happen only once for a
CPU in riscv_trigger_realize().

Fixes: 5a4ae64cac ("target/riscv: Add itrigger support when icount is enabled")
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <20230818034059.9146-1-akihiko.odaki@daynix.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/debug.h |  3 ++-
 target/riscv/cpu.c   |  8 +++++++-
 target/riscv/debug.c | 15 ++++++++++++---
 3 files changed, 21 insertions(+), 5 deletions(-)

diff --git a/target/riscv/debug.h b/target/riscv/debug.h
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/debug.h
+++ b/target/riscv/debug.h
@@ -XXX,XX +XXX,XX @@ void riscv_cpu_debug_excp_handler(CPUState *cs);
 bool riscv_cpu_debug_check_breakpoint(CPUState *cs);
 bool riscv_cpu_debug_check_watchpoint(CPUState *cs, CPUWatchpoint *wp);
 
-void riscv_trigger_init(CPURISCVState *env);
+void riscv_trigger_realize(CPURISCVState *env);
+void riscv_trigger_reset_hold(CPURISCVState *env);
 
 bool riscv_itrigger_enabled(CPURISCVState *env);
 void riscv_itrigger_update_priv(CPURISCVState *env);
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -XXX,XX +XXX,XX @@ static void riscv_cpu_reset_hold(Object *obj)
 
 #ifndef CONFIG_USER_ONLY
     if (cpu->cfg.debug) {
-        riscv_trigger_init(env);
+        riscv_trigger_reset_hold(env);
     }
 
     if (kvm_enabled()) {
@@ -XXX,XX +XXX,XX @@ static void riscv_cpu_realize(DeviceState *dev, Error **errp)
 
     riscv_cpu_register_gdb_regs_for_features(cs);
 
+#ifndef CONFIG_USER_ONLY
+    if (cpu->cfg.debug) {
+        riscv_trigger_realize(&cpu->env);
+    }
+#endif
+
     qemu_init_vcpu(cs);
     cpu_reset(cs);
 
diff --git a/target/riscv/debug.c b/target/riscv/debug.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/debug.c
+++ b/target/riscv/debug.c
@@ -XXX,XX +XXX,XX @@ bool riscv_cpu_debug_check_watchpoint(CPUState *cs, CPUWatchpoint *wp)
     return false;
 }
 
-void riscv_trigger_init(CPURISCVState *env)
+void riscv_trigger_realize(CPURISCVState *env)
+{
+    int i;
+
+    for (i = 0; i < RV_MAX_TRIGGERS; i++) {
+        env->itrigger_timer[i] = timer_new_ns(QEMU_CLOCK_VIRTUAL,
+                                              riscv_itrigger_timer_cb, env);
+    }
+}
+
+void riscv_trigger_reset_hold(CPURISCVState *env)
 {
     target_ulong tdata1 = build_tdata1(env, TRIGGER_TYPE_AD_MATCH, 0, 0);
     int i;
@@ -XXX,XX +XXX,XX @@ void riscv_trigger_init(CPURISCVState *env)
         env->tdata3[i] = 0;
         env->cpu_breakpoint[i] = NULL;
         env->cpu_watchpoint[i] = NULL;
-        env->itrigger_timer[i] = timer_new_ns(QEMU_CLOCK_VIRTUAL,
-                                              riscv_itrigger_timer_cb, env);
+        timer_del(env->itrigger_timer[i]);
     }
 }
-- 
2.41.0

From: Leon Schuermann <leons@opentitan.org>

When the rule-lock bypass (RLB) bit is set in the mseccfg CSR, the PMP
configuration lock bits must not apply. While this behavior is
implemented for the pmpcfgX CSRs, this bit is not respected for
changes to the pmpaddrX CSRs. This patch ensures that pmpaddrX CSR
writes work even on locked regions when the global rule-lock bypass is
enabled.

Signed-off-by: Leon Schuermann <leons@opentitan.org>
Reviewed-by: Mayuresh Chitale <mchitale@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <20230829215046.1430463-1-leon@is.currently.online>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/pmp.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/target/riscv/pmp.c b/target/riscv/pmp.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/pmp.c
+++ b/target/riscv/pmp.c
@@ -XXX,XX +XXX,XX @@ static inline uint8_t pmp_get_a_field(uint8_t cfg)
  */
 static inline int pmp_is_locked(CPURISCVState *env, uint32_t pmp_index)
 {
+    /* mseccfg.RLB is set */
+    if (MSECCFG_RLB_ISSET(env)) {
+        return 0;
+    }
 
     if (env->pmp_state.pmp[pmp_index].cfg_reg & PMP_LOCK) {
         return 1;
-- 
2.41.0

From: Tommy Wu <tommy.wu@sifive.com>

According to the new spec, when vsiselect has a reserved value, attempts
from M-mode or HS-mode to access vsireg, or from VS-mode to access
sireg, should preferably raise an illegal instruction exception.

Signed-off-by: Tommy Wu <tommy.wu@sifive.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Message-ID: <20230816061647.600672-1-tommy.wu@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/csr.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/target/riscv/csr.c b/target/riscv/csr.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/csr.c
+++ b/target/riscv/csr.c
@@ -XXX,XX +XXX,XX @@ static int rmw_iprio(target_ulong xlen,
 static int rmw_xireg(CPURISCVState *env, int csrno, target_ulong *val,
                      target_ulong new_val, target_ulong wr_mask)
 {
-    bool virt;
+    bool virt, isel_reserved;
     uint8_t *iprio;
     int ret = -EINVAL;
     target_ulong priv, isel, vgein;
@@ -XXX,XX +XXX,XX @@ static int rmw_xireg(CPURISCVState *env, int csrno, target_ulong *val,
 
     /* Decode register details from CSR number */
     virt = false;
+    isel_reserved = false;
     switch (csrno) {
     case CSR_MIREG:
         iprio = env->miprio;
@@ -XXX,XX +XXX,XX @@ static int rmw_xireg(CPURISCVState *env, int csrno, target_ulong *val,
                                                   riscv_cpu_mxl_bits(env)),
                                     val, new_val, wr_mask);
         }
+    } else {
+        isel_reserved = true;
     }
 
 done:
     if (ret) {
-        return (env->virt_enabled && virt) ?
+        return (env->virt_enabled && virt && !isel_reserved) ?
                RISCV_EXCP_VIRT_INSTRUCTION_FAULT : RISCV_EXCP_ILLEGAL_INST;
     }
     return RISCV_EXCP_NONE;
-- 
2.41.0

From: Nikita Shubin <n.shubin@yadro.com>

As per ISA:

"For CSRRWI, if rd=x0, then the instruction shall not read the CSR and
shall not cause any of the side effects that might occur on a CSR read."

trans_csrrwi() and trans_csrrw() call do_csrw() if rd=x0, do_csrw() calls
riscv_csrrw_do64(), via helper_csrw() passing NULL as *ret_value.

Signed-off-by: Nikita Shubin <n.shubin@yadro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <20230808090914.17634-1-nikita.shubin@maquefel.me>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/csr.c | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

diff --git a/target/riscv/csr.c b/target/riscv/csr.c
index XXXXXXX..XXXXXXX 100644
--- a/target/riscv/csr.c
+++ b/target/riscv/csr.c
@@ -XXX,XX +XXX,XX @@ static RISCVException riscv_csrrw_do64(CPURISCVState *env, int csrno,
                                        target_ulong write_mask)
 {
     RISCVException ret;
-    target_ulong old_value;
+    target_ulong old_value = 0;
 
     /* execute combined read/write operation if it exists */
     if (csr_ops[csrno].op) {
         return csr_ops[csrno].op(env, csrno, ret_value, new_value, write_mask);
     }
 
-    /* if no accessor exists then return failure */
-    if (!csr_ops[csrno].read) {
-        return RISCV_EXCP_ILLEGAL_INST;
-    }
-    /* read old value */
-    ret = csr_ops[csrno].read(env, csrno, &old_value);
-    if (ret != RISCV_EXCP_NONE) {
-        return ret;
+    /*
+     * ret_value == NULL means that rd=x0 and we're coming from helper_csrw()
+     * and we can't throw side effects caused by CSR reads.
+     */
+    if (ret_value) {
+        /* if no accessor exists then return failure */
+        if (!csr_ops[csrno].read) {
+            return RISCV_EXCP_ILLEGAL_INST;
+        }
+        /* read old value */
+        ret = csr_ops[csrno].read(env, csrno, &old_value);
+        if (ret != RISCV_EXCP_NONE) {
+            return ret;
+        }
     }
 
     /* write value if writable and write mask set, otherwise drop writes */
-- 
2.41.0