From: Alistair Francis <alistair.francis@wdc.com> The following changes since commit f22833602095b05733bceaddeb20f3edfced3c07: Merge tag 'pull-target-arm-20220428' of https://git.linaro.org/people/pmaydell/qemu-arm into staging (2022-04-28 08:34:17 -0700) are available in the Git repository at: git@github.com:alistair23/qemu.git tags/pull-riscv-to-apply-20220429 for you to fetch changes up to 325b7c4e7582c229d28c47123c3b986ed948eb84: hw/riscv: Enable TPM backends (2022-04-29 10:48:48 +1000) ---------------------------------------------------------------- Second RISC-V PR for QEMU 7.1 * Improve device tree generation * Support configuarable marchid, mvendorid, mipid CSR values * Add support for the Zbkb, Zbkc, Zbkx, Zknd/Zkne, Zknh, Zksed/Zksh and Zkr extensions * Fix incorrect PTE merge in walk_pte * Add TPM support to the virt board ---------------------------------------------------------------- Alistair Francis (6): hw/riscv: virt: Add a machine done notifier hw/core: Move the ARM sysbus-fdt to core hw/riscv: virt: Create a platform bus hw/riscv: virt: Add support for generating platform FDT entries hw/riscv: virt: Add device plug support hw/riscv: Enable TPM backends Bin Meng (2): hw/riscv: spike: Add '/chosen/stdout-path' in device tree unconditionally hw/riscv: Don't add empty bootargs to device tree Frank Chang (1): target/riscv: Support configuarable marchid, mvendorid, mipid CSR values Ralf Ramsauer (1): target/riscv: Fix incorrect PTE merge in walk_pte Weiwei Li (15): target/riscv: rvk: add cfg properties for zbk* and zk* target/riscv: rvk: add support for zbkb extension target/riscv: rvk: add support for zbkc extension target/riscv: rvk: add support for zbkx extension crypto: move sm4_sbox from target/arm target/riscv: rvk: add support for zknd/zkne extension in RV32 target/riscv: rvk: add support for zkne/zknd extension in RV64 target/riscv: rvk: add support for sha256 related instructions in zknh extension target/riscv: rvk: add support for sha512 related instructions for RV32 in zknh extension target/riscv: rvk: add support for sha512 related instructions for RV64 in zknh extension target/riscv: rvk: add support for zksed/zksh extension target/riscv: rvk: add CSR support for Zkr disas/riscv.c: rvk: add disas support for Zbk* and Zk* instructions target/riscv: rvk: expose zbk* and zk* properties target/riscv: add scalar crypto related extenstion strings to isa_string docs/system/riscv/virt.rst | 20 ++ include/crypto/sm4.h | 6 + include/hw/{arm => core}/sysbus-fdt.h | 0 include/hw/riscv/virt.h | 8 +- target/riscv/cpu.h | 17 ++ target/riscv/cpu_bits.h | 9 + target/riscv/helper.h | 22 ++ target/riscv/pmp.h | 8 +- target/riscv/insn32.decode | 97 ++++++-- crypto/sm4.c | 49 ++++ disas/riscv.c | 173 +++++++++++++- hw/arm/virt.c | 2 +- hw/arm/xlnx-versal-virt.c | 1 - hw/{arm => core}/sysbus-fdt.c | 2 +- hw/riscv/microchip_pfsoc.c | 2 +- hw/riscv/sifive_u.c | 2 +- hw/riscv/spike.c | 7 +- hw/riscv/virt.c | 319 +++++++++++++++++--------- target/arm/crypto_helper.c | 36 +-- target/riscv/bitmanip_helper.c | 80 +++++++ target/riscv/cpu.c | 58 +++++ target/riscv/crypto_helper.c | 302 ++++++++++++++++++++++++ target/riscv/csr.c | 118 +++++++++- target/riscv/monitor.c | 11 +- target/riscv/op_helper.c | 9 + target/riscv/translate.c | 8 + target/riscv/insn_trans/trans_rvb.c.inc | 116 ++++++++-- target/riscv/insn_trans/trans_rvk.c.inc | 391 ++++++++++++++++++++++++++++++++ crypto/meson.build | 1 + hw/arm/meson.build | 1 - hw/core/meson.build | 1 + hw/riscv/Kconfig | 2 + target/riscv/meson.build | 3 +- 33 files changed, 1682 insertions(+), 199 deletions(-) create mode 100644 include/crypto/sm4.h rename include/hw/{arm => core}/sysbus-fdt.h (100%) create mode 100644 crypto/sm4.c rename hw/{arm => core}/sysbus-fdt.c (99%) create mode 100644 target/riscv/crypto_helper.c create mode 100644 target/riscv/insn_trans/trans_rvk.c.inc From: Bin Meng <bin.meng@windriver.com> At present the adding '/chosen/stdout-path' property in device tree is determined by whether a kernel command line is provided, which is wrong. It should be added unconditionally. Fixes: 8d8897accb1c ("hw/riscv: spike: Allow using binary firmware as bios") Signed-off-by: Bin Meng <bin.meng@windriver.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220421055629.1177285-1-bmeng.cn@gmail.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/spike.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/hw/riscv/spike.c b/hw/riscv/spike.c index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/spike.c +++ b/hw/riscv/spike.c @@ -XXX,XX +XXX,XX @@ static void create_fdt(SpikeState *s, const MemMapEntry *memmap, riscv_socket_fdt_write_distance_matrix(mc, fdt); + qemu_fdt_add_subnode(fdt, "/chosen"); + qemu_fdt_setprop_string(fdt, "/chosen", "stdout-path", "/htif"); + if (cmdline) { - qemu_fdt_add_subnode(fdt, "/chosen"); qemu_fdt_setprop_string(fdt, "/chosen", "bootargs", cmdline); - qemu_fdt_setprop_string(fdt, "/chosen", "stdout-path", "/htif"); } } -- 2.35.1 From: Bin Meng <bin.meng@windriver.com> Commit 7c28f4da20e5 ("RISC-V: Don't add NULL bootargs to device-tree") tried to avoid adding *NULL* bootargs to device tree, but unfortunately the changes were entirely useless, due to MachineState::kernel_cmdline can't be NULL at all as the default value is given as an empty string. (see hw/core/machine.c::machine_initfn()). Note the wording of *NULL* bootargs is wrong. It can't be NULL otherwise a segfault had already been observed by dereferencing the NULL pointer. It should be worded as *empty" bootargs. Fixes: 7c28f4da20e5 ("RISC-V: Don't add NULL bootargs to device-tree") Signed-off-by: Bin Meng <bin.meng@windriver.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220421055629.1177285-2-bmeng.cn@gmail.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/microchip_pfsoc.c | 2 +- hw/riscv/sifive_u.c | 2 +- hw/riscv/spike.c | 2 +- hw/riscv/virt.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/hw/riscv/microchip_pfsoc.c b/hw/riscv/microchip_pfsoc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/microchip_pfsoc.c +++ b/hw/riscv/microchip_pfsoc.c @@ -XXX,XX +XXX,XX @@ static void microchip_icicle_kit_machine_init(MachineState *machine) "linux,initrd-end", end); } - if (machine->kernel_cmdline) { + if (machine->kernel_cmdline && *machine->kernel_cmdline) { qemu_fdt_setprop_string(machine->fdt, "/chosen", "bootargs", machine->kernel_cmdline); } diff --git a/hw/riscv/sifive_u.c b/hw/riscv/sifive_u.c index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/sifive_u.c +++ b/hw/riscv/sifive_u.c @@ -XXX,XX +XXX,XX @@ static void create_fdt(SiFiveUState *s, const MemMapEntry *memmap, g_free(nodename); update_bootargs: - if (cmdline) { + if (cmdline && *cmdline) { qemu_fdt_setprop_string(fdt, "/chosen", "bootargs", cmdline); } } diff --git a/hw/riscv/spike.c b/hw/riscv/spike.c index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/spike.c +++ b/hw/riscv/spike.c @@ -XXX,XX +XXX,XX @@ static void create_fdt(SpikeState *s, const MemMapEntry *memmap, qemu_fdt_add_subnode(fdt, "/chosen"); qemu_fdt_setprop_string(fdt, "/chosen", "stdout-path", "/htif"); - if (cmdline) { + if (cmdline && *cmdline) { qemu_fdt_setprop_string(fdt, "/chosen", "bootargs", 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 create_fdt(RISCVVirtState *s, const MemMapEntry *memmap, create_fdt_flash(s, memmap); update_bootargs: - if (cmdline) { + if (cmdline && *cmdline) { qemu_fdt_setprop_string(mc->fdt, "/chosen", "bootargs", cmdline); } } -- 2.35.1 From: Frank Chang <frank.chang@sifive.com> Allow user to set core's marchid, mvendorid, mipid CSRs through -cpu command line option. The default values of marchid and mipid are built with QEMU's version numbers. Signed-off-by: Frank Chang <frank.chang@sifive.com> Reviewed-by: Jim Shu <jim.shu@sifive.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220422040436.2233-1-frank.chang@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 4 ++++ target/riscv/cpu.c | 9 +++++++++ target/riscv/csr.c | 38 ++++++++++++++++++++++++++++++++++---- 3 files changed, 47 insertions(+), 4 deletions(-) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig { bool ext_zve32f; bool ext_zve64f; + uint32_t mvendorid; + uint64_t marchid; + uint64_t mipid; + /* Vendor-specific custom extensions */ bool ext_XVentanaCondOps; 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 @@ /* RISC-V CPU definitions */ +#define RISCV_CPU_MARCHID ((QEMU_VERSION_MAJOR << 16) | \ + (QEMU_VERSION_MINOR << 8) | \ + (QEMU_VERSION_MICRO)) +#define RISCV_CPU_MIPID RISCV_CPU_MARCHID + static const char riscv_single_letter_exts[] = "IEMAFDQCPVH"; struct isa_ext_data { @@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_properties[] = { DEFINE_PROP_UINT16("vlen", RISCVCPU, cfg.vlen, 128), DEFINE_PROP_UINT16("elen", RISCVCPU, cfg.elen, 64), + DEFINE_PROP_UINT32("mvendorid", RISCVCPU, cfg.mvendorid, 0), + DEFINE_PROP_UINT64("marchid", RISCVCPU, cfg.marchid, RISCV_CPU_MARCHID), + DEFINE_PROP_UINT64("mipid", RISCVCPU, cfg.mipid, RISCV_CPU_MIPID), + DEFINE_PROP_BOOL("svinval", RISCVCPU, cfg.ext_svinval, false), DEFINE_PROP_BOOL("svnapot", RISCVCPU, cfg.ext_svnapot, false), DEFINE_PROP_BOOL("svpbmt", RISCVCPU, cfg.ext_svpbmt, false), 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_ignore(CPURISCVState *env, int csrno, return RISCV_EXCP_NONE; } +static RISCVException read_mvendorid(CPURISCVState *env, int csrno, + target_ulong *val) +{ + CPUState *cs = env_cpu(env); + RISCVCPU *cpu = RISCV_CPU(cs); + + *val = cpu->cfg.mvendorid; + return RISCV_EXCP_NONE; +} + +static RISCVException read_marchid(CPURISCVState *env, int csrno, + target_ulong *val) +{ + CPUState *cs = env_cpu(env); + RISCVCPU *cpu = RISCV_CPU(cs); + + *val = cpu->cfg.marchid; + return RISCV_EXCP_NONE; +} + +static RISCVException read_mipid(CPURISCVState *env, int csrno, + target_ulong *val) +{ + CPUState *cs = env_cpu(env); + RISCVCPU *cpu = RISCV_CPU(cs); + + *val = cpu->cfg.mipid; + return RISCV_EXCP_NONE; +} + static RISCVException read_mhartid(CPURISCVState *env, int csrno, target_ulong *val) { @@ -XXX,XX +XXX,XX @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { [CSR_MINSTRETH] = { "minstreth", any32, read_instreth }, /* Machine Information Registers */ - [CSR_MVENDORID] = { "mvendorid", any, read_zero }, - [CSR_MARCHID] = { "marchid", any, read_zero }, - [CSR_MIMPID] = { "mimpid", any, read_zero }, - [CSR_MHARTID] = { "mhartid", any, read_mhartid }, + [CSR_MVENDORID] = { "mvendorid", any, read_mvendorid }, + [CSR_MARCHID] = { "marchid", any, read_marchid }, + [CSR_MIMPID] = { "mimpid", any, read_mipid }, + [CSR_MHARTID] = { "mhartid", any, read_mhartid }, [CSR_MCONFIGPTR] = { "mconfigptr", any, read_zero, .min_priv_ver = PRIV_VERSION_1_12_0 }, -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-2-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 13 +++++++++++++ target/riscv/cpu.c | 23 +++++++++++++++++++++++ 2 files changed, 36 insertions(+) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -XXX,XX +XXX,XX @@ struct RISCVCPUConfig { bool ext_zba; bool ext_zbb; bool ext_zbc; + bool ext_zbkb; + bool ext_zbkc; + bool ext_zbkx; bool ext_zbs; + bool ext_zk; + bool ext_zkn; + bool ext_zknd; + bool ext_zkne; + bool ext_zknh; + bool ext_zkr; + bool ext_zks; + bool ext_zksed; + bool ext_zksh; + bool ext_zkt; bool ext_counters; bool ext_ifencei; bool ext_icsr; 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(DeviceState *dev, Error **errp) cpu->cfg.ext_zfinx = true; } + if (cpu->cfg.ext_zk) { + cpu->cfg.ext_zkn = true; + cpu->cfg.ext_zkr = true; + cpu->cfg.ext_zkt = true; + } + + if (cpu->cfg.ext_zkn) { + cpu->cfg.ext_zbkb = true; + cpu->cfg.ext_zbkc = true; + cpu->cfg.ext_zbkx = true; + cpu->cfg.ext_zkne = true; + cpu->cfg.ext_zknd = true; + cpu->cfg.ext_zknh = true; + } + + if (cpu->cfg.ext_zks) { + cpu->cfg.ext_zbkb = true; + cpu->cfg.ext_zbkc = true; + cpu->cfg.ext_zbkx = true; + cpu->cfg.ext_zksed = true; + cpu->cfg.ext_zksh = true; + } + /* Set the ISA extensions, checks should have happened above */ if (cpu->cfg.ext_i) { ext |= RVI; -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - reuse partial instructions of zbb extension, update extension check for them - add brev8, pack, packh, packw, unzip, zip instructions Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Acked-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-Id: <20220423023510.30794-3-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/helper.h | 3 + target/riscv/insn32.decode | 45 +++++++----- target/riscv/bitmanip_helper.c | 53 ++++++++++++++ target/riscv/translate.c | 7 ++ target/riscv/insn_trans/trans_rvb.c.inc | 94 +++++++++++++++++++++---- 5 files changed, 174 insertions(+), 28 deletions(-) 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_FLAGS_1(fclass_d, TCG_CALL_NO_RWG_SE, tl, i64) /* Bitmanip */ DEF_HELPER_FLAGS_2(clmul, TCG_CALL_NO_RWG_SE, tl, tl, tl) DEF_HELPER_FLAGS_2(clmulr, TCG_CALL_NO_RWG_SE, tl, tl, tl) +DEF_HELPER_FLAGS_1(brev8, TCG_CALL_NO_RWG_SE, tl, tl) +DEF_HELPER_FLAGS_1(unzip, TCG_CALL_NO_RWG_SE, tl, tl) +DEF_HELPER_FLAGS_1(zip, TCG_CALL_NO_RWG_SE, tl, tl) /* Floating Point - Half Precision */ DEF_HELPER_FLAGS_3(fadd_h, TCG_CALL_NO_RWG, i64, env, i64, i64) 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 @@ sh2add_uw 0010000 .......... 100 ..... 0111011 @r sh3add_uw 0010000 .......... 110 ..... 0111011 @r slli_uw 00001 ............ 001 ..... 0011011 @sh -# *** RV32 Zbb Standard Extension *** +# *** RV32 Zbb/Zbkb Standard Extension *** andn 0100000 .......... 111 ..... 0110011 @r +rol 0110000 .......... 001 ..... 0110011 @r +ror 0110000 .......... 101 ..... 0110011 @r +rori 01100 ............ 101 ..... 0010011 @sh +# The encoding for rev8 differs between RV32 and RV64. +# rev8_32 denotes the RV32 variant. +rev8_32 011010 011000 ..... 101 ..... 0010011 @r2 +# The encoding for zext.h differs between RV32 and RV64. +# zext_h_32 denotes the RV32 variant. +{ + zext_h_32 0000100 00000 ..... 100 ..... 0110011 @r2 + pack 0000100 ..... ..... 100 ..... 0110011 @r +} +xnor 0100000 .......... 100 ..... 0110011 @r +# *** RV32 extra Zbb Standard Extension *** clz 011000 000000 ..... 001 ..... 0010011 @r2 cpop 011000 000010 ..... 001 ..... 0010011 @r2 ctz 011000 000001 ..... 001 ..... 0010011 @r2 @@ -XXX,XX +XXX,XX @@ min 0000101 .......... 100 ..... 0110011 @r minu 0000101 .......... 101 ..... 0110011 @r orc_b 001010 000111 ..... 101 ..... 0010011 @r2 orn 0100000 .......... 110 ..... 0110011 @r -# The encoding for rev8 differs between RV32 and RV64. -# rev8_32 denotes the RV32 variant. -rev8_32 011010 011000 ..... 101 ..... 0010011 @r2 -rol 0110000 .......... 001 ..... 0110011 @r -ror 0110000 .......... 101 ..... 0110011 @r -rori 01100 ............ 101 ..... 0010011 @sh sext_b 011000 000100 ..... 001 ..... 0010011 @r2 sext_h 011000 000101 ..... 001 ..... 0010011 @r2 -xnor 0100000 .......... 100 ..... 0110011 @r -# The encoding for zext.h differs between RV32 and RV64. -# zext_h_32 denotes the RV32 variant. -zext_h_32 0000100 00000 ..... 100 ..... 0110011 @r2 +# *** RV32 extra Zbkb Standard Extension *** +brev8 0110100 00111 ..... 101 ..... 0010011 @r2 #grevi +packh 0000100 .......... 111 ..... 0110011 @r +unzip 0000100 01111 ..... 101 ..... 0010011 @r2 #unshfl +zip 0000100 01111 ..... 001 ..... 0010011 @r2 #shfl -# *** RV64 Zbb Standard Extension (in addition to RV32 Zbb) *** -clzw 0110000 00000 ..... 001 ..... 0011011 @r2 -ctzw 0110000 00001 ..... 001 ..... 0011011 @r2 -cpopw 0110000 00010 ..... 001 ..... 0011011 @r2 +# *** RV64 Zbb/Zbkb Standard Extension (in addition to RV32 Zbb/Zbkb) *** # The encoding for rev8 differs between RV32 and RV64. # When executing on RV64, the encoding used in RV32 is an illegal # instruction, so we use different handler functions to differentiate. @@ -XXX,XX +XXX,XX @@ rorw 0110000 .......... 101 ..... 0111011 @r # The encoding for zext.h differs between RV32 and RV64. # When executing on RV64, the encoding used in RV32 is an illegal # instruction, so we use different handler functions to differentiate. -zext_h_64 0000100 00000 ..... 100 ..... 0111011 @r2 +{ + zext_h_64 0000100 00000 ..... 100 ..... 0111011 @r2 + packw 0000100 ..... ..... 100 ..... 0111011 @r +} +# *** RV64 extra Zbb Standard Extension (in addition to RV32 Zbb) *** +clzw 0110000 00000 ..... 001 ..... 0011011 @r2 +ctzw 0110000 00001 ..... 001 ..... 0011011 @r2 +cpopw 0110000 00010 ..... 001 ..... 0011011 @r2 # *** RV32 Zbc Standard Extension *** clmul 0000101 .......... 001 ..... 0110011 @r diff --git a/target/riscv/bitmanip_helper.c b/target/riscv/bitmanip_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/bitmanip_helper.c +++ b/target/riscv/bitmanip_helper.c @@ -XXX,XX +XXX,XX @@ target_ulong HELPER(clmulr)(target_ulong rs1, target_ulong rs2) return result; } + +static inline target_ulong do_swap(target_ulong x, uint64_t mask, int shift) +{ + return ((x & mask) << shift) | ((x & ~mask) >> shift); +} + +target_ulong HELPER(brev8)(target_ulong rs1) +{ + target_ulong x = rs1; + + x = do_swap(x, 0x5555555555555555ull, 1); + x = do_swap(x, 0x3333333333333333ull, 2); + x = do_swap(x, 0x0f0f0f0f0f0f0f0full, 4); + return x; +} + +static const uint64_t shuf_masks[] = { + dup_const(MO_8, 0x44), + dup_const(MO_8, 0x30), + dup_const(MO_16, 0x0f00), + dup_const(MO_32, 0xff0000) +}; + +static inline target_ulong do_shuf_stage(target_ulong src, uint64_t maskL, + uint64_t maskR, int shift) +{ + target_ulong x = src & ~(maskL | maskR); + + x |= ((src << shift) & maskL) | ((src >> shift) & maskR); + return x; +} + +target_ulong HELPER(unzip)(target_ulong rs1) +{ + target_ulong x = rs1; + + x = do_shuf_stage(x, shuf_masks[0], shuf_masks[0] >> 1, 1); + x = do_shuf_stage(x, shuf_masks[1], shuf_masks[1] >> 2, 2); + x = do_shuf_stage(x, shuf_masks[2], shuf_masks[2] >> 4, 4); + x = do_shuf_stage(x, shuf_masks[3], shuf_masks[3] >> 8, 8); + return x; +} + +target_ulong HELPER(zip)(target_ulong rs1) +{ + target_ulong x = rs1; + + x = do_shuf_stage(x, shuf_masks[3], shuf_masks[3] >> 8, 8); + x = do_shuf_stage(x, shuf_masks[2], shuf_masks[2] >> 4, 4); + x = do_shuf_stage(x, shuf_masks[1], shuf_masks[1] >> 2, 2); + x = do_shuf_stage(x, shuf_masks[0], shuf_masks[0] >> 1, 1); + return x; +} 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 @@ EX_SH(12) } \ } while (0) +#define REQUIRE_EITHER_EXT(ctx, A, B) do { \ + if (!ctx->cfg_ptr->ext_##A && \ + !ctx->cfg_ptr->ext_##B) { \ + return false; \ + } \ +} while (0) + static int ex_rvc_register(DisasContext *ctx, int reg) { return 8 + reg; diff --git a/target/riscv/insn_trans/trans_rvb.c.inc b/target/riscv/insn_trans/trans_rvb.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvb.c.inc +++ b/target/riscv/insn_trans/trans_rvb.c.inc @@ -XXX,XX +XXX,XX @@ /* - * RISC-V translation routines for the Zb[abcs] Standard Extension. + * RISC-V translation routines for the Zb[abcs] and Zbk[bcx] Standard Extension. * * Copyright (c) 2020 Kito Cheng, kito.cheng@sifive.com * Copyright (c) 2020 Frank Chang, frank.chang@sifive.com @@ -XXX,XX +XXX,XX @@ } \ } while (0) +#define REQUIRE_ZBKB(ctx) do { \ + if (!ctx->cfg_ptr->ext_zbkb) { \ + return false; \ + } \ +} while (0) + static void gen_clz(TCGv ret, TCGv arg1) { tcg_gen_clzi_tl(ret, arg1, TARGET_LONG_BITS); @@ -XXX,XX +XXX,XX @@ static bool trans_cpop(DisasContext *ctx, arg_cpop *a) static bool trans_andn(DisasContext *ctx, arg_andn *a) { - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); return gen_logic(ctx, a, tcg_gen_andc_tl); } static bool trans_orn(DisasContext *ctx, arg_orn *a) { - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); return gen_logic(ctx, a, tcg_gen_orc_tl); } static bool trans_xnor(DisasContext *ctx, arg_xnor *a) { - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); return gen_logic(ctx, a, tcg_gen_eqv_tl); } @@ -XXX,XX +XXX,XX @@ static void gen_rorw(TCGv ret, TCGv arg1, TCGv arg2) static bool trans_ror(DisasContext *ctx, arg_ror *a) { - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotr_tl, gen_rorw, NULL); } @@ -XXX,XX +XXX,XX @@ static void gen_roriw(TCGv ret, TCGv arg1, target_long shamt) static bool trans_rori(DisasContext *ctx, arg_rori *a) { - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); return gen_shift_imm_fn_per_ol(ctx, a, EXT_NONE, tcg_gen_rotri_tl, gen_roriw, NULL); } @@ -XXX,XX +XXX,XX @@ static void gen_rolw(TCGv ret, TCGv arg1, TCGv arg2) static bool trans_rol(DisasContext *ctx, arg_rol *a) { - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotl_tl, gen_rolw, NULL); } @@ -XXX,XX +XXX,XX @@ static void gen_rev8_32(TCGv ret, TCGv src1) static bool trans_rev8_32(DisasContext *ctx, arg_rev8_32 *a) { REQUIRE_32BIT(ctx); - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); return gen_unary(ctx, a, EXT_NONE, gen_rev8_32); } static bool trans_rev8_64(DisasContext *ctx, arg_rev8_64 *a) { REQUIRE_64BIT(ctx); - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); return gen_unary(ctx, a, EXT_NONE, tcg_gen_bswap_tl); } @@ -XXX,XX +XXX,XX @@ static bool trans_cpopw(DisasContext *ctx, arg_cpopw *a) static bool trans_rorw(DisasContext *ctx, arg_rorw *a) { REQUIRE_64BIT(ctx); - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); ctx->ol = MXL_RV32; return gen_shift(ctx, a, EXT_NONE, gen_rorw, NULL); } @@ -XXX,XX +XXX,XX @@ static bool trans_rorw(DisasContext *ctx, arg_rorw *a) static bool trans_roriw(DisasContext *ctx, arg_roriw *a) { REQUIRE_64BIT(ctx); - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); ctx->ol = MXL_RV32; return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_roriw, NULL); } @@ -XXX,XX +XXX,XX @@ static bool trans_roriw(DisasContext *ctx, arg_roriw *a) static bool trans_rolw(DisasContext *ctx, arg_rolw *a) { REQUIRE_64BIT(ctx); - REQUIRE_ZBB(ctx); + REQUIRE_EITHER_EXT(ctx, zbb, zbkb); ctx->ol = MXL_RV32; return gen_shift(ctx, a, EXT_NONE, gen_rolw, NULL); } @@ -XXX,XX +XXX,XX @@ static bool trans_clmulr(DisasContext *ctx, arg_clmulh *a) REQUIRE_ZBC(ctx); return gen_arith(ctx, a, EXT_NONE, gen_helper_clmulr, NULL); } + +static void gen_pack(TCGv ret, TCGv src1, TCGv src2) +{ + tcg_gen_deposit_tl(ret, src1, src2, + TARGET_LONG_BITS / 2, + TARGET_LONG_BITS / 2); +} + +static void gen_packh(TCGv ret, TCGv src1, TCGv src2) +{ + TCGv t = tcg_temp_new(); + + tcg_gen_ext8u_tl(t, src2); + tcg_gen_deposit_tl(ret, src1, t, 8, TARGET_LONG_BITS - 8); + tcg_temp_free(t); +} + +static void gen_packw(TCGv ret, TCGv src1, TCGv src2) +{ + TCGv t = tcg_temp_new(); + + tcg_gen_ext16s_tl(t, src2); + tcg_gen_deposit_tl(ret, src1, t, 16, TARGET_LONG_BITS - 16); + tcg_temp_free(t); +} + +static bool trans_brev8(DisasContext *ctx, arg_brev8 *a) +{ + REQUIRE_ZBKB(ctx); + return gen_unary(ctx, a, EXT_NONE, gen_helper_brev8); +} + +static bool trans_pack(DisasContext *ctx, arg_pack *a) +{ + REQUIRE_ZBKB(ctx); + return gen_arith(ctx, a, EXT_NONE, gen_pack, NULL); +} + +static bool trans_packh(DisasContext *ctx, arg_packh *a) +{ + REQUIRE_ZBKB(ctx); + return gen_arith(ctx, a, EXT_NONE, gen_packh, NULL); +} + +static bool trans_packw(DisasContext *ctx, arg_packw *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZBKB(ctx); + return gen_arith(ctx, a, EXT_NONE, gen_packw, NULL); +} + +static bool trans_unzip(DisasContext *ctx, arg_unzip *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZBKB(ctx); + return gen_unary(ctx, a, EXT_NONE, gen_helper_unzip); +} + +static bool trans_zip(DisasContext *ctx, arg_zip *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZBKB(ctx); + return gen_unary(ctx, a, EXT_NONE, gen_helper_zip); +} -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - reuse partial instructions of zbc extension, update extension check for them Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-Id: <20220423023510.30794-4-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn32.decode | 3 ++- target/riscv/insn_trans/trans_rvb.c.inc | 4 ++-- 2 files changed, 4 insertions(+), 3 deletions(-) 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 @@ clzw 0110000 00000 ..... 001 ..... 0011011 @r2 ctzw 0110000 00001 ..... 001 ..... 0011011 @r2 cpopw 0110000 00010 ..... 001 ..... 0011011 @r2 -# *** RV32 Zbc Standard Extension *** +# *** RV32 Zbc/Zbkc Standard Extension *** clmul 0000101 .......... 001 ..... 0110011 @r clmulh 0000101 .......... 011 ..... 0110011 @r +# *** RV32 extra Zbc Standard Extension *** clmulr 0000101 .......... 010 ..... 0110011 @r # *** RV32 Zbs Standard Extension *** diff --git a/target/riscv/insn_trans/trans_rvb.c.inc b/target/riscv/insn_trans/trans_rvb.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvb.c.inc +++ b/target/riscv/insn_trans/trans_rvb.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_slli_uw(DisasContext *ctx, arg_slli_uw *a) static bool trans_clmul(DisasContext *ctx, arg_clmul *a) { - REQUIRE_ZBC(ctx); + REQUIRE_EITHER_EXT(ctx, zbc, zbkc); return gen_arith(ctx, a, EXT_NONE, gen_helper_clmul, NULL); } @@ -XXX,XX +XXX,XX @@ static void gen_clmulh(TCGv dst, TCGv src1, TCGv src2) static bool trans_clmulh(DisasContext *ctx, arg_clmulr *a) { - REQUIRE_ZBC(ctx); + REQUIRE_EITHER_EXT(ctx, zbc, zbkc); return gen_arith(ctx, a, EXT_NONE, gen_clmulh, NULL); } -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - add xperm4 and xperm8 instructions Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-5-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/helper.h | 2 ++ target/riscv/insn32.decode | 4 ++++ target/riscv/bitmanip_helper.c | 27 +++++++++++++++++++++++++ target/riscv/insn_trans/trans_rvb.c.inc | 18 +++++++++++++++++ 4 files changed, 51 insertions(+) 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_FLAGS_2(clmulr, TCG_CALL_NO_RWG_SE, tl, tl, tl) DEF_HELPER_FLAGS_1(brev8, TCG_CALL_NO_RWG_SE, tl, tl) DEF_HELPER_FLAGS_1(unzip, TCG_CALL_NO_RWG_SE, tl, tl) DEF_HELPER_FLAGS_1(zip, TCG_CALL_NO_RWG_SE, tl, tl) +DEF_HELPER_FLAGS_2(xperm4, TCG_CALL_NO_RWG_SE, tl, tl, tl) +DEF_HELPER_FLAGS_2(xperm8, TCG_CALL_NO_RWG_SE, tl, tl, tl) /* Floating Point - Half Precision */ DEF_HELPER_FLAGS_3(fadd_h, TCG_CALL_NO_RWG, i64, env, i64, i64) 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 @@ clmulh 0000101 .......... 011 ..... 0110011 @r # *** RV32 extra Zbc Standard Extension *** clmulr 0000101 .......... 010 ..... 0110011 @r +# *** RV32 Zbkx Standard Extension *** +xperm4 0010100 .......... 010 ..... 0110011 @r +xperm8 0010100 .......... 100 ..... 0110011 @r + # *** RV32 Zbs Standard Extension *** bclr 0100100 .......... 001 ..... 0110011 @r bclri 01001. ........... 001 ..... 0010011 @sh diff --git a/target/riscv/bitmanip_helper.c b/target/riscv/bitmanip_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/bitmanip_helper.c +++ b/target/riscv/bitmanip_helper.c @@ -XXX,XX +XXX,XX @@ target_ulong HELPER(zip)(target_ulong rs1) x = do_shuf_stage(x, shuf_masks[0], shuf_masks[0] >> 1, 1); return x; } + +static inline target_ulong do_xperm(target_ulong rs1, target_ulong rs2, + uint32_t sz_log2) +{ + target_ulong r = 0; + target_ulong sz = 1LL << sz_log2; + target_ulong mask = (1LL << sz) - 1; + target_ulong pos; + + for (int i = 0; i < TARGET_LONG_BITS; i += sz) { + pos = ((rs2 >> i) & mask) << sz_log2; + if (pos < sizeof(target_ulong) * 8) { + r |= ((rs1 >> pos) & mask) << i; + } + } + return r; +} + +target_ulong HELPER(xperm4)(target_ulong rs1, target_ulong rs2) +{ + return do_xperm(rs1, rs2, 2); +} + +target_ulong HELPER(xperm8)(target_ulong rs1, target_ulong rs2) +{ + return do_xperm(rs1, rs2, 3); +} diff --git a/target/riscv/insn_trans/trans_rvb.c.inc b/target/riscv/insn_trans/trans_rvb.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvb.c.inc +++ b/target/riscv/insn_trans/trans_rvb.c.inc @@ -XXX,XX +XXX,XX @@ } \ } while (0) +#define REQUIRE_ZBKX(ctx) do { \ + if (!ctx->cfg_ptr->ext_zbkx) { \ + return false; \ + } \ +} while (0) + static void gen_clz(TCGv ret, TCGv arg1) { tcg_gen_clzi_tl(ret, arg1, TARGET_LONG_BITS); @@ -XXX,XX +XXX,XX @@ static bool trans_zip(DisasContext *ctx, arg_zip *a) REQUIRE_ZBKB(ctx); return gen_unary(ctx, a, EXT_NONE, gen_helper_zip); } + +static bool trans_xperm4(DisasContext *ctx, arg_xperm4 *a) +{ + REQUIRE_ZBKX(ctx); + return gen_arith(ctx, a, EXT_NONE, gen_helper_xperm4, NULL); +} + +static bool trans_xperm8(DisasContext *ctx, arg_xperm8 *a) +{ + REQUIRE_ZBKX(ctx); + return gen_arith(ctx, a, EXT_NONE, gen_helper_xperm8, NULL); +} -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - share it between target/arm and target/riscv Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-Id: <20220423023510.30794-6-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- include/crypto/sm4.h | 6 +++++ crypto/sm4.c | 49 ++++++++++++++++++++++++++++++++++++++ target/arm/crypto_helper.c | 36 +--------------------------- crypto/meson.build | 1 + 4 files changed, 57 insertions(+), 35 deletions(-) create mode 100644 include/crypto/sm4.h create mode 100644 crypto/sm4.c diff --git a/include/crypto/sm4.h b/include/crypto/sm4.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/crypto/sm4.h @@ -XXX,XX +XXX,XX @@ +#ifndef QEMU_SM4_H +#define QEMU_SM4_H + +extern const uint8_t sm4_sbox[256]; + +#endif diff --git a/crypto/sm4.c b/crypto/sm4.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/crypto/sm4.c @@ -XXX,XX +XXX,XX @@ +/* + * QEMU crypto sm4 support + * + * Copyright (C) 2013 - 2018 Linaro Ltd <ard.biesheuvel@linaro.org> + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + */ + +#include "qemu/osdep.h" +#include "crypto/sm4.h" + +uint8_t const sm4_sbox[] = { + 0xd6, 0x90, 0xe9, 0xfe, 0xcc, 0xe1, 0x3d, 0xb7, + 0x16, 0xb6, 0x14, 0xc2, 0x28, 0xfb, 0x2c, 0x05, + 0x2b, 0x67, 0x9a, 0x76, 0x2a, 0xbe, 0x04, 0xc3, + 0xaa, 0x44, 0x13, 0x26, 0x49, 0x86, 0x06, 0x99, + 0x9c, 0x42, 0x50, 0xf4, 0x91, 0xef, 0x98, 0x7a, + 0x33, 0x54, 0x0b, 0x43, 0xed, 0xcf, 0xac, 0x62, + 0xe4, 0xb3, 0x1c, 0xa9, 0xc9, 0x08, 0xe8, 0x95, + 0x80, 0xdf, 0x94, 0xfa, 0x75, 0x8f, 0x3f, 0xa6, + 0x47, 0x07, 0xa7, 0xfc, 0xf3, 0x73, 0x17, 0xba, + 0x83, 0x59, 0x3c, 0x19, 0xe6, 0x85, 0x4f, 0xa8, + 0x68, 0x6b, 0x81, 0xb2, 0x71, 0x64, 0xda, 0x8b, + 0xf8, 0xeb, 0x0f, 0x4b, 0x70, 0x56, 0x9d, 0x35, + 0x1e, 0x24, 0x0e, 0x5e, 0x63, 0x58, 0xd1, 0xa2, + 0x25, 0x22, 0x7c, 0x3b, 0x01, 0x21, 0x78, 0x87, + 0xd4, 0x00, 0x46, 0x57, 0x9f, 0xd3, 0x27, 0x52, + 0x4c, 0x36, 0x02, 0xe7, 0xa0, 0xc4, 0xc8, 0x9e, + 0xea, 0xbf, 0x8a, 0xd2, 0x40, 0xc7, 0x38, 0xb5, + 0xa3, 0xf7, 0xf2, 0xce, 0xf9, 0x61, 0x15, 0xa1, + 0xe0, 0xae, 0x5d, 0xa4, 0x9b, 0x34, 0x1a, 0x55, + 0xad, 0x93, 0x32, 0x30, 0xf5, 0x8c, 0xb1, 0xe3, + 0x1d, 0xf6, 0xe2, 0x2e, 0x82, 0x66, 0xca, 0x60, + 0xc0, 0x29, 0x23, 0xab, 0x0d, 0x53, 0x4e, 0x6f, + 0xd5, 0xdb, 0x37, 0x45, 0xde, 0xfd, 0x8e, 0x2f, + 0x03, 0xff, 0x6a, 0x72, 0x6d, 0x6c, 0x5b, 0x51, + 0x8d, 0x1b, 0xaf, 0x92, 0xbb, 0xdd, 0xbc, 0x7f, + 0x11, 0xd9, 0x5c, 0x41, 0x1f, 0x10, 0x5a, 0xd8, + 0x0a, 0xc1, 0x31, 0x88, 0xa5, 0xcd, 0x7b, 0xbd, + 0x2d, 0x74, 0xd0, 0x12, 0xb8, 0xe5, 0xb4, 0xb0, + 0x89, 0x69, 0x97, 0x4a, 0x0c, 0x96, 0x77, 0x7e, + 0x65, 0xb9, 0xf1, 0x09, 0xc5, 0x6e, 0xc6, 0x84, + 0x18, 0xf0, 0x7d, 0xec, 0x3a, 0xdc, 0x4d, 0x20, + 0x79, 0xee, 0x5f, 0x3e, 0xd7, 0xcb, 0x39, 0x48, +}; + diff --git a/target/arm/crypto_helper.c b/target/arm/crypto_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/crypto_helper.c +++ b/target/arm/crypto_helper.c @@ -XXX,XX +XXX,XX @@ #include "exec/helper-proto.h" #include "tcg/tcg-gvec-desc.h" #include "crypto/aes.h" +#include "crypto/sm4.h" #include "vec_internal.h" union CRYPTO_STATE { @@ -XXX,XX +XXX,XX @@ DO_SM3TT(crypto_sm3tt2b, 3) #undef DO_SM3TT -static uint8_t const sm4_sbox[] = { - 0xd6, 0x90, 0xe9, 0xfe, 0xcc, 0xe1, 0x3d, 0xb7, - 0x16, 0xb6, 0x14, 0xc2, 0x28, 0xfb, 0x2c, 0x05, - 0x2b, 0x67, 0x9a, 0x76, 0x2a, 0xbe, 0x04, 0xc3, - 0xaa, 0x44, 0x13, 0x26, 0x49, 0x86, 0x06, 0x99, - 0x9c, 0x42, 0x50, 0xf4, 0x91, 0xef, 0x98, 0x7a, - 0x33, 0x54, 0x0b, 0x43, 0xed, 0xcf, 0xac, 0x62, - 0xe4, 0xb3, 0x1c, 0xa9, 0xc9, 0x08, 0xe8, 0x95, - 0x80, 0xdf, 0x94, 0xfa, 0x75, 0x8f, 0x3f, 0xa6, - 0x47, 0x07, 0xa7, 0xfc, 0xf3, 0x73, 0x17, 0xba, - 0x83, 0x59, 0x3c, 0x19, 0xe6, 0x85, 0x4f, 0xa8, - 0x68, 0x6b, 0x81, 0xb2, 0x71, 0x64, 0xda, 0x8b, - 0xf8, 0xeb, 0x0f, 0x4b, 0x70, 0x56, 0x9d, 0x35, - 0x1e, 0x24, 0x0e, 0x5e, 0x63, 0x58, 0xd1, 0xa2, - 0x25, 0x22, 0x7c, 0x3b, 0x01, 0x21, 0x78, 0x87, - 0xd4, 0x00, 0x46, 0x57, 0x9f, 0xd3, 0x27, 0x52, - 0x4c, 0x36, 0x02, 0xe7, 0xa0, 0xc4, 0xc8, 0x9e, - 0xea, 0xbf, 0x8a, 0xd2, 0x40, 0xc7, 0x38, 0xb5, - 0xa3, 0xf7, 0xf2, 0xce, 0xf9, 0x61, 0x15, 0xa1, - 0xe0, 0xae, 0x5d, 0xa4, 0x9b, 0x34, 0x1a, 0x55, - 0xad, 0x93, 0x32, 0x30, 0xf5, 0x8c, 0xb1, 0xe3, - 0x1d, 0xf6, 0xe2, 0x2e, 0x82, 0x66, 0xca, 0x60, - 0xc0, 0x29, 0x23, 0xab, 0x0d, 0x53, 0x4e, 0x6f, - 0xd5, 0xdb, 0x37, 0x45, 0xde, 0xfd, 0x8e, 0x2f, - 0x03, 0xff, 0x6a, 0x72, 0x6d, 0x6c, 0x5b, 0x51, - 0x8d, 0x1b, 0xaf, 0x92, 0xbb, 0xdd, 0xbc, 0x7f, - 0x11, 0xd9, 0x5c, 0x41, 0x1f, 0x10, 0x5a, 0xd8, - 0x0a, 0xc1, 0x31, 0x88, 0xa5, 0xcd, 0x7b, 0xbd, - 0x2d, 0x74, 0xd0, 0x12, 0xb8, 0xe5, 0xb4, 0xb0, - 0x89, 0x69, 0x97, 0x4a, 0x0c, 0x96, 0x77, 0x7e, - 0x65, 0xb9, 0xf1, 0x09, 0xc5, 0x6e, 0xc6, 0x84, - 0x18, 0xf0, 0x7d, 0xec, 0x3a, 0xdc, 0x4d, 0x20, - 0x79, 0xee, 0x5f, 0x3e, 0xd7, 0xcb, 0x39, 0x48, -}; - static void do_crypto_sm4e(uint64_t *rd, uint64_t *rn, uint64_t *rm) { union CRYPTO_STATE d = { .l = { rn[0], rn[1] } }; diff --git a/crypto/meson.build b/crypto/meson.build index XXXXXXX..XXXXXXX 100644 --- a/crypto/meson.build +++ b/crypto/meson.build @@ -XXX,XX +XXX,XX @@ if have_afalg endif crypto_ss.add(when: gnutls, if_true: files('tls-cipher-suites.c')) +util_ss.add(files('sm4.c')) util_ss.add(files('aes.c')) util_ss.add(files('init.c')) if gnutls.found() -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - add aes32esmi, aes32esi, aes32dsmi and aes32dsi instructions Co-authored-by: Zewen Ye <lustrew@foxmail.com> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-7-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/helper.h | 6 ++ target/riscv/insn32.decode | 11 +++ target/riscv/crypto_helper.c | 105 ++++++++++++++++++++++++ target/riscv/translate.c | 1 + target/riscv/insn_trans/trans_rvk.c.inc | 71 ++++++++++++++++ target/riscv/meson.build | 3 +- 6 files changed, 196 insertions(+), 1 deletion(-) create mode 100644 target/riscv/crypto_helper.c create mode 100644 target/riscv/insn_trans/trans_rvk.c.inc 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(divu_i128, tl, env, tl, tl, tl, tl) DEF_HELPER_5(divs_i128, tl, env, tl, tl, tl, tl) DEF_HELPER_5(remu_i128, tl, env, tl, tl, tl, tl) DEF_HELPER_5(rems_i128, tl, env, tl, tl, tl, tl) + +/* Crypto functions */ +DEF_HELPER_FLAGS_3(aes32esmi, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) +DEF_HELPER_FLAGS_3(aes32esi, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) +DEF_HELPER_FLAGS_3(aes32dsmi, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) +DEF_HELPER_FLAGS_3(aes32dsi, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) 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_b 31:s1 7:1 25:6 8:4 !function=ex_shift_1 %imm_j 31:s1 12:8 20:1 21:10 !function=ex_shift_1 %imm_u 12:s20 !function=ex_shift_12 +%imm_bs 30:2 !function=ex_shift_3 # Argument sets: &empty @@ -XXX,XX +XXX,XX @@ &rmr vm rd rs2 &r2nfvm vm rd rs1 nf &rnfvm vm rd rs1 rs2 nf +&k_aes shamt rs2 rs1 rd # Formats 32: @r ....... ..... ..... ... ..... ....... &r %rs2 %rs1 %rd @@ -XXX,XX +XXX,XX @@ @sfence_vma ....... ..... ..... ... ..... ....... %rs2 %rs1 @sfence_vm ....... ..... ..... ... ..... ....... %rs1 +@k_aes .. ..... ..... ..... ... ..... ....... &k_aes shamt=%imm_bs %rs2 %rs1 %rd + # Formats 64: @sh5 ....... ..... ..... ... ..... ....... &shift shamt=%sh5 %rs1 %rd @@ -XXX,XX +XXX,XX @@ sfence_w_inval 0001100 00000 00000 000 00000 1110011 sfence_inval_ir 0001100 00001 00000 000 00000 1110011 hinval_vvma 0010011 ..... ..... 000 00000 1110011 @hfence_vvma hinval_gvma 0110011 ..... ..... 000 00000 1110011 @hfence_gvma + +# *** RV32 Zknd Standard Extension *** +aes32dsmi .. 10111 ..... ..... 000 ..... 0110011 @k_aes +aes32dsi .. 10101 ..... ..... 000 ..... 0110011 @k_aes +# *** RV32 Zkne Standard Extension *** +aes32esmi .. 10011 ..... ..... 000 ..... 0110011 @k_aes +aes32esi .. 10001 ..... ..... 000 ..... 0110011 @k_aes diff --git a/target/riscv/crypto_helper.c b/target/riscv/crypto_helper.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/target/riscv/crypto_helper.c @@ -XXX,XX +XXX,XX @@ +/* + * RISC-V Crypto Emulation Helpers for QEMU. + * + * Copyright (c) 2021 Ruibo Lu, luruibo2000@163.com + * Copyright (c) 2021 Zewen Ye, lustrew@foxmail.com + * + * 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 "cpu.h" +#include "exec/exec-all.h" +#include "exec/helper-proto.h" +#include "crypto/aes.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, + target_ulong rs1, target_ulong rs2, + 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); + } else { + mixed = so; + } + } else { + so = AES_isbox[si]; + if (mix) { + mixed = aes_mixcolumn_byte(so, false); + } else { + mixed = so; + } + } + mixed = rol32(mixed, shamt); + res = rs1 ^ mixed; + + return sext32_xlen(res); +} + +target_ulong HELPER(aes32esmi)(target_ulong rs1, target_ulong rs2, + target_ulong shamt) +{ + return aes32_operation(shamt, rs1, rs2, true, true); +} + +target_ulong HELPER(aes32esi)(target_ulong rs1, target_ulong rs2, + target_ulong shamt) +{ + return aes32_operation(shamt, rs1, rs2, true, false); +} + +target_ulong HELPER(aes32dsmi)(target_ulong rs1, target_ulong rs2, + target_ulong shamt) +{ + return aes32_operation(shamt, rs1, rs2, false, true); +} + +target_ulong HELPER(aes32dsi)(target_ulong rs1, target_ulong rs2, + target_ulong shamt) +{ + return aes32_operation(shamt, rs1, rs2, false, false); +} +#undef sext32_xlen 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_rvv.c.inc" #include "insn_trans/trans_rvb.c.inc" #include "insn_trans/trans_rvzfh.c.inc" +#include "insn_trans/trans_rvk.c.inc" #include "insn_trans/trans_privileged.c.inc" #include "insn_trans/trans_svinval.c.inc" #include "insn_trans/trans_xventanacondops.c.inc" diff --git a/target/riscv/insn_trans/trans_rvk.c.inc b/target/riscv/insn_trans/trans_rvk.c.inc new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/target/riscv/insn_trans/trans_rvk.c.inc @@ -XXX,XX +XXX,XX @@ +/* + * RISC-V translation routines for the Zk[nd,ne,nh,sed,sh] Standard Extension. + * + * Copyright (c) 2021 Ruibo Lu, luruibo2000@163.com + * Copyright (c) 2021 Zewen Ye, lustrew@foxmail.com + * + * 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/>. + */ + +#define REQUIRE_ZKND(ctx) do { \ + if (!ctx->cfg_ptr->ext_zknd) { \ + return false; \ + } \ +} while (0) + +#define REQUIRE_ZKNE(ctx) do { \ + if (!ctx->cfg_ptr->ext_zkne) { \ + return false; \ + } \ +} while (0) + +static bool gen_aes32_sm4(DisasContext *ctx, arg_k_aes *a, + void (*func)(TCGv, TCGv, TCGv, TCGv)) +{ + TCGv shamt = tcg_constant_tl(a->shamt); + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, EXT_NONE); + TCGv src2 = get_gpr(ctx, a->rs2, EXT_NONE); + + func(dest, src1, src2, shamt); + gen_set_gpr(ctx, a->rd, dest); + return true; +} + +static bool trans_aes32esmi(DisasContext *ctx, arg_aes32esmi *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKNE(ctx); + return gen_aes32_sm4(ctx, a, gen_helper_aes32esmi); +} + +static bool trans_aes32esi(DisasContext *ctx, arg_aes32esi *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKNE(ctx); + return gen_aes32_sm4(ctx, a, gen_helper_aes32esi); +} + +static bool trans_aes32dsmi(DisasContext *ctx, arg_aes32dsmi *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKND(ctx); + return gen_aes32_sm4(ctx, a, gen_helper_aes32dsmi); +} + +static bool trans_aes32dsi(DisasContext *ctx, arg_aes32dsi *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKND(ctx); + return gen_aes32_sm4(ctx, a, gen_helper_aes32dsi); +} 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( 'vector_helper.c', 'bitmanip_helper.c', 'translate.c', - 'm128_helper.c' + 'm128_helper.c', + 'crypto_helper.c' )) riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'), if_false: files('kvm-stub.c')) -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - add aes64dsm, aes64ds, aes64im, aes64es, aes64esm, aes64ks2, aes64ks1i instructions Co-authored-by: Ruibo Lu <luruibo2000@163.com> Co-authored-by: Zewen Ye <lustrew@foxmail.com> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-8-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/helper.h | 8 ++ target/riscv/insn32.decode | 12 ++ target/riscv/crypto_helper.c | 169 ++++++++++++++++++++++++ target/riscv/insn_trans/trans_rvk.c.inc | 54 ++++++++ 4 files changed, 243 insertions(+) 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_FLAGS_3(aes32esmi, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) DEF_HELPER_FLAGS_3(aes32esi, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) DEF_HELPER_FLAGS_3(aes32dsmi, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) DEF_HELPER_FLAGS_3(aes32dsi, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) + +DEF_HELPER_FLAGS_2(aes64esm, TCG_CALL_NO_RWG_SE, tl, tl, tl) +DEF_HELPER_FLAGS_2(aes64es, TCG_CALL_NO_RWG_SE, tl, tl, tl) +DEF_HELPER_FLAGS_2(aes64ds, TCG_CALL_NO_RWG_SE, tl, tl, tl) +DEF_HELPER_FLAGS_2(aes64dsm, TCG_CALL_NO_RWG_SE, tl, tl, tl) +DEF_HELPER_FLAGS_2(aes64ks2, TCG_CALL_NO_RWG_SE, tl, tl, tl) +DEF_HELPER_FLAGS_2(aes64ks1i, TCG_CALL_NO_RWG_SE, tl, tl, tl) +DEF_HELPER_FLAGS_1(aes64im, TCG_CALL_NO_RWG_SE, tl, tl) 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_j 31:s1 12:8 20:1 21:10 !function=ex_shift_1 %imm_u 12:s20 !function=ex_shift_12 %imm_bs 30:2 !function=ex_shift_3 +%imm_rnum 20:4 # Argument sets: &empty @@ -XXX,XX +XXX,XX @@ @sfence_vm ....... ..... ..... ... ..... ....... %rs1 @k_aes .. ..... ..... ..... ... ..... ....... &k_aes shamt=%imm_bs %rs2 %rs1 %rd +@i_aes .. ..... ..... ..... ... ..... ....... &i imm=%imm_rnum %rs1 %rd # Formats 64: @sh5 ....... ..... ..... ... ..... ....... &shift shamt=%sh5 %rs1 %rd @@ -XXX,XX +XXX,XX @@ hinval_gvma 0110011 ..... ..... 000 00000 1110011 @hfence_gvma # *** RV32 Zknd Standard Extension *** aes32dsmi .. 10111 ..... ..... 000 ..... 0110011 @k_aes aes32dsi .. 10101 ..... ..... 000 ..... 0110011 @k_aes +# *** RV64 Zknd Standard Extension *** +aes64dsm 00 11111 ..... ..... 000 ..... 0110011 @r +aes64ds 00 11101 ..... ..... 000 ..... 0110011 @r +aes64im 00 11000 00000 ..... 001 ..... 0010011 @r2 # *** RV32 Zkne Standard Extension *** aes32esmi .. 10011 ..... ..... 000 ..... 0110011 @k_aes aes32esi .. 10001 ..... ..... 000 ..... 0110011 @k_aes +# *** RV64 Zkne Standard Extension *** +aes64es 00 11001 ..... ..... 000 ..... 0110011 @r +aes64esm 00 11011 ..... ..... 000 ..... 0110011 @r +# *** RV64 Zkne/zknd Standard Extension *** +aes64ks2 01 11111 ..... ..... 000 ..... 0110011 @r +aes64ks1i 00 11000 1.... ..... 001 ..... 0010011 @i_aes 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(aes32dsi)(target_ulong rs1, target_ulong rs2, { return aes32_operation(shamt, rs1, rs2, false, false); } + +#define BY(X, I) ((X >> (8 * I)) & 0xFF) + +#define AES_SHIFROWS_LO(RS1, RS2) ( \ + (((RS1 >> 24) & 0xFF) << 56) | (((RS2 >> 48) & 0xFF) << 48) | \ + (((RS2 >> 8) & 0xFF) << 40) | (((RS1 >> 32) & 0xFF) << 32) | \ + (((RS2 >> 56) & 0xFF) << 24) | (((RS2 >> 16) & 0xFF) << 16) | \ + (((RS1 >> 40) & 0xFF) << 8) | (((RS1 >> 0) & 0xFF) << 0)) + +#define AES_INVSHIFROWS_LO(RS1, RS2) ( \ + (((RS2 >> 24) & 0xFF) << 56) | (((RS2 >> 48) & 0xFF) << 48) | \ + (((RS1 >> 8) & 0xFF) << 40) | (((RS1 >> 32) & 0xFF) << 32) | \ + (((RS1 >> 56) & 0xFF) << 24) | (((RS2 >> 16) & 0xFF) << 16) | \ + (((RS2 >> 40) & 0xFF) << 8) | (((RS1 >> 0) & 0xFF) << 0)) + +#define AES_MIXBYTE(COL, B0, B1, B2, B3) ( \ + BY(COL, B3) ^ BY(COL, B2) ^ AES_GFMUL(BY(COL, B1), 3) ^ \ + AES_GFMUL(BY(COL, B0), 2)) + +#define AES_MIXCOLUMN(COL) ( \ + AES_MIXBYTE(COL, 3, 0, 1, 2) << 24 | \ + AES_MIXBYTE(COL, 2, 3, 0, 1) << 16 | \ + AES_MIXBYTE(COL, 1, 2, 3, 0) << 8 | AES_MIXBYTE(COL, 0, 1, 2, 3) << 0) + +#define AES_INVMIXBYTE(COL, B0, B1, B2, B3) ( \ + AES_GFMUL(BY(COL, B3), 0x9) ^ AES_GFMUL(BY(COL, B2), 0xd) ^ \ + AES_GFMUL(BY(COL, B1), 0xb) ^ AES_GFMUL(BY(COL, B0), 0xe)) + +#define AES_INVMIXCOLUMN(COL) ( \ + AES_INVMIXBYTE(COL, 3, 0, 1, 2) << 24 | \ + AES_INVMIXBYTE(COL, 2, 3, 0, 1) << 16 | \ + AES_INVMIXBYTE(COL, 1, 2, 3, 0) << 8 | \ + AES_INVMIXBYTE(COL, 0, 1, 2, 3) << 0) + +static inline target_ulong aes64_operation(target_ulong rs1, target_ulong rs2, + bool enc, bool mix) +{ + uint64_t RS1 = rs1; + uint64_t RS2 = rs2; + uint64_t result; + uint64_t temp; + uint32_t col_0; + uint32_t col_1; + + if (enc) { + temp = AES_SHIFROWS_LO(RS1, RS2); + temp = (((uint64_t)AES_sbox[(temp >> 0) & 0xFF] << 0) | + ((uint64_t)AES_sbox[(temp >> 8) & 0xFF] << 8) | + ((uint64_t)AES_sbox[(temp >> 16) & 0xFF] << 16) | + ((uint64_t)AES_sbox[(temp >> 24) & 0xFF] << 24) | + ((uint64_t)AES_sbox[(temp >> 32) & 0xFF] << 32) | + ((uint64_t)AES_sbox[(temp >> 40) & 0xFF] << 40) | + ((uint64_t)AES_sbox[(temp >> 48) & 0xFF] << 48) | + ((uint64_t)AES_sbox[(temp >> 56) & 0xFF] << 56)); + if (mix) { + col_0 = temp & 0xFFFFFFFF; + col_1 = temp >> 32; + + col_0 = AES_MIXCOLUMN(col_0); + col_1 = AES_MIXCOLUMN(col_1); + + result = ((uint64_t)col_1 << 32) | col_0; + } else { + result = temp; + } + } else { + temp = AES_INVSHIFROWS_LO(RS1, RS2); + temp = (((uint64_t)AES_isbox[(temp >> 0) & 0xFF] << 0) | + ((uint64_t)AES_isbox[(temp >> 8) & 0xFF] << 8) | + ((uint64_t)AES_isbox[(temp >> 16) & 0xFF] << 16) | + ((uint64_t)AES_isbox[(temp >> 24) & 0xFF] << 24) | + ((uint64_t)AES_isbox[(temp >> 32) & 0xFF] << 32) | + ((uint64_t)AES_isbox[(temp >> 40) & 0xFF] << 40) | + ((uint64_t)AES_isbox[(temp >> 48) & 0xFF] << 48) | + ((uint64_t)AES_isbox[(temp >> 56) & 0xFF] << 56)); + if (mix) { + col_0 = temp & 0xFFFFFFFF; + col_1 = temp >> 32; + + col_0 = AES_INVMIXCOLUMN(col_0); + col_1 = AES_INVMIXCOLUMN(col_1); + + result = ((uint64_t)col_1 << 32) | col_0; + } else { + result = temp; + } + } + + return result; +} + +target_ulong HELPER(aes64esm)(target_ulong rs1, target_ulong rs2) +{ + return aes64_operation(rs1, rs2, true, true); +} + +target_ulong HELPER(aes64es)(target_ulong rs1, target_ulong rs2) +{ + return aes64_operation(rs1, rs2, true, false); +} + +target_ulong HELPER(aes64ds)(target_ulong rs1, target_ulong rs2) +{ + return aes64_operation(rs1, rs2, false, false); +} + +target_ulong HELPER(aes64dsm)(target_ulong rs1, target_ulong rs2) +{ + return aes64_operation(rs1, rs2, false, true); +} + +target_ulong HELPER(aes64ks2)(target_ulong rs1, target_ulong rs2) +{ + uint64_t RS1 = rs1; + uint64_t RS2 = rs2; + uint32_t rs1_hi = RS1 >> 32; + uint32_t rs2_lo = RS2; + uint32_t rs2_hi = RS2 >> 32; + + uint32_t r_lo = (rs1_hi ^ rs2_lo); + uint32_t r_hi = (rs1_hi ^ rs2_lo ^ rs2_hi); + target_ulong result = ((uint64_t)r_hi << 32) | r_lo; + + return result; +} + +target_ulong HELPER(aes64ks1i)(target_ulong rs1, target_ulong rnum) +{ + uint64_t RS1 = rs1; + static const uint8_t round_consts[10] = { + 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36 + }; + + uint8_t enc_rnum = rnum; + uint32_t temp = (RS1 >> 32) & 0xFFFFFFFF; + uint8_t rcon_ = 0; + target_ulong result; + + if (enc_rnum != 0xA) { + temp = ror32(temp, 8); /* Rotate right by 8 */ + rcon_ = 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); + + temp ^= rcon_; + + result = ((uint64_t)temp << 32) | temp; + + return result; +} + +target_ulong HELPER(aes64im)(target_ulong rs1) +{ + uint64_t RS1 = rs1; + uint32_t col_0 = RS1 & 0xFFFFFFFF; + uint32_t col_1 = RS1 >> 32; + target_ulong result; + + col_0 = AES_INVMIXCOLUMN(col_0); + col_1 = AES_INVMIXCOLUMN(col_1); + + result = ((uint64_t)col_1 << 32) | col_0; + + return result; +} #undef sext32_xlen diff --git a/target/riscv/insn_trans/trans_rvk.c.inc b/target/riscv/insn_trans/trans_rvk.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvk.c.inc +++ b/target/riscv/insn_trans/trans_rvk.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_aes32dsi(DisasContext *ctx, arg_aes32dsi *a) REQUIRE_ZKND(ctx); return gen_aes32_sm4(ctx, a, gen_helper_aes32dsi); } + +static bool trans_aes64es(DisasContext *ctx, arg_aes64es *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZKNE(ctx); + return gen_arith(ctx, a, EXT_NONE, gen_helper_aes64es, NULL); +} + +static bool trans_aes64esm(DisasContext *ctx, arg_aes64esm *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZKNE(ctx); + return gen_arith(ctx, a, EXT_NONE, gen_helper_aes64esm, NULL); +} + +static bool trans_aes64ds(DisasContext *ctx, arg_aes64ds *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZKND(ctx); + return gen_arith(ctx, a, EXT_NONE, gen_helper_aes64ds, NULL); +} + +static bool trans_aes64dsm(DisasContext *ctx, arg_aes64dsm *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZKND(ctx); + return gen_arith(ctx, a, EXT_NONE, gen_helper_aes64dsm, NULL); +} + +static bool trans_aes64ks2(DisasContext *ctx, arg_aes64ks2 *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_EITHER_EXT(ctx, zknd, zkne); + return gen_arith(ctx, a, EXT_NONE, gen_helper_aes64ks2, NULL); +} + +static bool trans_aes64ks1i(DisasContext *ctx, arg_aes64ks1i *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_EITHER_EXT(ctx, zknd, zkne); + + if (a->imm > 0xA) { + return false; + } + + return gen_arith_imm_tl(ctx, a, EXT_NONE, gen_helper_aes64ks1i, NULL); +} + +static bool trans_aes64im(DisasContext *ctx, arg_aes64im *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZKND(ctx); + return gen_unary(ctx, a, EXT_NONE, gen_helper_aes64im); +} -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - add sha256sig0, sha256sig1, sha256sum0 and sha256sum1 instructions Co-authored-by: Zewen Ye <lustrew@foxmail.com> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-9-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn32.decode | 5 +++ target/riscv/insn_trans/trans_rvk.c.inc | 55 +++++++++++++++++++++++++ 2 files changed, 60 insertions(+) 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 @@ aes64esm 00 11011 ..... ..... 000 ..... 0110011 @r # *** RV64 Zkne/zknd Standard Extension *** aes64ks2 01 11111 ..... ..... 000 ..... 0110011 @r aes64ks1i 00 11000 1.... ..... 001 ..... 0010011 @i_aes +# *** RV32 Zknh Standard Extension *** +sha256sig0 00 01000 00010 ..... 001 ..... 0010011 @r2 +sha256sig1 00 01000 00011 ..... 001 ..... 0010011 @r2 +sha256sum0 00 01000 00000 ..... 001 ..... 0010011 @r2 +sha256sum1 00 01000 00001 ..... 001 ..... 0010011 @r2 diff --git a/target/riscv/insn_trans/trans_rvk.c.inc b/target/riscv/insn_trans/trans_rvk.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvk.c.inc +++ b/target/riscv/insn_trans/trans_rvk.c.inc @@ -XXX,XX +XXX,XX @@ } \ } while (0) +#define REQUIRE_ZKNH(ctx) do { \ + if (!ctx->cfg_ptr->ext_zknh) { \ + return false; \ + } \ +} while (0) + static bool gen_aes32_sm4(DisasContext *ctx, arg_k_aes *a, void (*func)(TCGv, TCGv, TCGv, TCGv)) { @@ -XXX,XX +XXX,XX @@ static bool trans_aes64im(DisasContext *ctx, arg_aes64im *a) REQUIRE_ZKND(ctx); return gen_unary(ctx, a, EXT_NONE, gen_helper_aes64im); } + +static bool gen_sha256(DisasContext *ctx, arg_r2 *a, DisasExtend ext, + void (*func)(TCGv_i32, TCGv_i32, int32_t), + int32_t num1, int32_t num2, int32_t num3) +{ + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, ext); + TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_new_i32(); + TCGv_i32 t2 = tcg_temp_new_i32(); + + tcg_gen_trunc_tl_i32(t0, src1); + tcg_gen_rotri_i32(t1, t0, num1); + tcg_gen_rotri_i32(t2, t0, num2); + tcg_gen_xor_i32(t1, t1, t2); + func(t2, t0, num3); + tcg_gen_xor_i32(t1, t1, t2); + tcg_gen_ext_i32_tl(dest, t1); + + gen_set_gpr(ctx, a->rd, dest); + tcg_temp_free_i32(t0); + tcg_temp_free_i32(t1); + tcg_temp_free_i32(t2); + return true; +} + +static bool trans_sha256sig0(DisasContext *ctx, arg_sha256sig0 *a) +{ + REQUIRE_ZKNH(ctx); + return gen_sha256(ctx, a, EXT_NONE, tcg_gen_shri_i32, 7, 18, 3); +} + +static bool trans_sha256sig1(DisasContext *ctx, arg_sha256sig1 *a) +{ + REQUIRE_ZKNH(ctx); + return gen_sha256(ctx, a, EXT_NONE, tcg_gen_shri_i32, 17, 19, 10); +} + +static bool trans_sha256sum0(DisasContext *ctx, arg_sha256sum0 *a) +{ + REQUIRE_ZKNH(ctx); + return gen_sha256(ctx, a, EXT_NONE, tcg_gen_rotri_i32, 2, 13, 22); +} + +static bool trans_sha256sum1(DisasContext *ctx, arg_sha256sum1 *a) +{ + REQUIRE_ZKNH(ctx); + return gen_sha256(ctx, a, EXT_NONE, tcg_gen_rotri_i32, 6, 11, 25); +} -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - add sha512sum0r, sha512sig0l, sha512sum1r, sha512sig1l, sha512sig0h and sha512sig1h instructions Co-authored-by: Zewen Ye <lustrew@foxmail.com> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-10-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn32.decode | 6 ++ target/riscv/insn_trans/trans_rvk.c.inc | 100 ++++++++++++++++++++++++ 2 files changed, 106 insertions(+) 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 @@ sha256sig0 00 01000 00010 ..... 001 ..... 0010011 @r2 sha256sig1 00 01000 00011 ..... 001 ..... 0010011 @r2 sha256sum0 00 01000 00000 ..... 001 ..... 0010011 @r2 sha256sum1 00 01000 00001 ..... 001 ..... 0010011 @r2 +sha512sum0r 01 01000 ..... ..... 000 ..... 0110011 @r +sha512sum1r 01 01001 ..... ..... 000 ..... 0110011 @r +sha512sig0l 01 01010 ..... ..... 000 ..... 0110011 @r +sha512sig0h 01 01110 ..... ..... 000 ..... 0110011 @r +sha512sig1l 01 01011 ..... ..... 000 ..... 0110011 @r +sha512sig1h 01 01111 ..... ..... 000 ..... 0110011 @r diff --git a/target/riscv/insn_trans/trans_rvk.c.inc b/target/riscv/insn_trans/trans_rvk.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvk.c.inc +++ b/target/riscv/insn_trans/trans_rvk.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_sha256sum1(DisasContext *ctx, arg_sha256sum1 *a) REQUIRE_ZKNH(ctx); return gen_sha256(ctx, a, EXT_NONE, tcg_gen_rotri_i32, 6, 11, 25); } + +static bool gen_sha512_rv32(DisasContext *ctx, arg_r *a, DisasExtend ext, + void (*func1)(TCGv_i64, TCGv_i64, int64_t), + void (*func2)(TCGv_i64, TCGv_i64, int64_t), + int64_t num1, int64_t num2, int64_t num3) +{ + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, ext); + TCGv src2 = get_gpr(ctx, a->rs2, ext); + TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t2 = tcg_temp_new_i64(); + + tcg_gen_concat_tl_i64(t0, src1, src2); + func1(t1, t0, num1); + func2(t2, t0, num2); + tcg_gen_xor_i64(t1, t1, t2); + tcg_gen_rotri_i64(t2, t0, num3); + tcg_gen_xor_i64(t1, t1, t2); + tcg_gen_trunc_i64_tl(dest, t1); + + gen_set_gpr(ctx, a->rd, dest); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + tcg_temp_free_i64(t2); + return true; +} + +static bool trans_sha512sum0r(DisasContext *ctx, arg_sha512sum0r *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512_rv32(ctx, a, EXT_NONE, tcg_gen_rotli_i64, + tcg_gen_rotli_i64, 25, 30, 28); +} + +static bool trans_sha512sum1r(DisasContext *ctx, arg_sha512sum1r *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512_rv32(ctx, a, EXT_NONE, tcg_gen_rotli_i64, + tcg_gen_rotri_i64, 23, 14, 18); +} + +static bool trans_sha512sig0l(DisasContext *ctx, arg_sha512sig0l *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512_rv32(ctx, a, EXT_NONE, tcg_gen_rotri_i64, + tcg_gen_rotri_i64, 1, 7, 8); +} + +static bool trans_sha512sig1l(DisasContext *ctx, arg_sha512sig1l *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512_rv32(ctx, a, EXT_NONE, tcg_gen_rotli_i64, + tcg_gen_rotri_i64, 3, 6, 19); +} + +static bool gen_sha512h_rv32(DisasContext *ctx, arg_r *a, DisasExtend ext, + void (*func)(TCGv_i64, TCGv_i64, int64_t), + int64_t num1, int64_t num2, int64_t num3) +{ + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, ext); + TCGv src2 = get_gpr(ctx, a->rs2, ext); + TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t2 = tcg_temp_new_i64(); + + tcg_gen_concat_tl_i64(t0, src1, src2); + func(t1, t0, num1); + tcg_gen_ext32u_i64(t2, t0); + tcg_gen_shri_i64(t2, t2, num2); + tcg_gen_xor_i64(t1, t1, t2); + tcg_gen_rotri_i64(t2, t0, num3); + tcg_gen_xor_i64(t1, t1, t2); + tcg_gen_trunc_i64_tl(dest, t1); + + gen_set_gpr(ctx, a->rd, dest); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + tcg_temp_free_i64(t2); + return true; +} + +static bool trans_sha512sig0h(DisasContext *ctx, arg_sha512sig0h *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512h_rv32(ctx, a, EXT_NONE, tcg_gen_rotri_i64, 1, 7, 8); +} + +static bool trans_sha512sig1h(DisasContext *ctx, arg_sha512sig1h *a) +{ + REQUIRE_32BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512h_rv32(ctx, a, EXT_NONE, tcg_gen_rotli_i64, 3, 6, 19); +} -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - add sha512sum0, sha512sig0, sha512sum1 and sha512sig1 instructions Co-authored-by: Zewen Ye <lustrew@foxmail.com> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-11-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn32.decode | 5 +++ target/riscv/insn_trans/trans_rvk.c.inc | 53 +++++++++++++++++++++++++ 2 files changed, 58 insertions(+) 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 @@ sha512sig0l 01 01010 ..... ..... 000 ..... 0110011 @r sha512sig0h 01 01110 ..... ..... 000 ..... 0110011 @r sha512sig1l 01 01011 ..... ..... 000 ..... 0110011 @r sha512sig1h 01 01111 ..... ..... 000 ..... 0110011 @r +# *** RV64 Zknh Standard Extension *** +sha512sig0 00 01000 00110 ..... 001 ..... 0010011 @r2 +sha512sig1 00 01000 00111 ..... 001 ..... 0010011 @r2 +sha512sum0 00 01000 00100 ..... 001 ..... 0010011 @r2 +sha512sum1 00 01000 00101 ..... 001 ..... 0010011 @r2 diff --git a/target/riscv/insn_trans/trans_rvk.c.inc b/target/riscv/insn_trans/trans_rvk.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvk.c.inc +++ b/target/riscv/insn_trans/trans_rvk.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_sha512sig1h(DisasContext *ctx, arg_sha512sig1h *a) REQUIRE_ZKNH(ctx); return gen_sha512h_rv32(ctx, a, EXT_NONE, tcg_gen_rotli_i64, 3, 6, 19); } + +static bool gen_sha512_rv64(DisasContext *ctx, arg_r2 *a, DisasExtend ext, + void (*func)(TCGv_i64, TCGv_i64, int64_t), + int64_t num1, int64_t num2, int64_t num3) +{ + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, ext); + TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t2 = tcg_temp_new_i64(); + + tcg_gen_extu_tl_i64(t0, src1); + tcg_gen_rotri_i64(t1, t0, num1); + tcg_gen_rotri_i64(t2, t0, num2); + tcg_gen_xor_i64(t1, t1, t2); + func(t2, t0, num3); + tcg_gen_xor_i64(t1, t1, t2); + tcg_gen_trunc_i64_tl(dest, t1); + + gen_set_gpr(ctx, a->rd, dest); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + tcg_temp_free_i64(t2); + return true; +} + +static bool trans_sha512sig0(DisasContext *ctx, arg_sha512sig0 *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512_rv64(ctx, a, EXT_NONE, tcg_gen_shri_i64, 1, 8, 7); +} + +static bool trans_sha512sig1(DisasContext *ctx, arg_sha512sig1 *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512_rv64(ctx, a, EXT_NONE, tcg_gen_shri_i64, 19, 61, 6); +} + +static bool trans_sha512sum0(DisasContext *ctx, arg_sha512sum0 *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512_rv64(ctx, a, EXT_NONE, tcg_gen_rotri_i64, 28, 34, 39); +} + +static bool trans_sha512sum1(DisasContext *ctx, arg_sha512sum1 *a) +{ + REQUIRE_64BIT(ctx); + REQUIRE_ZKNH(ctx); + return gen_sha512_rv64(ctx, a, EXT_NONE, tcg_gen_rotri_i64, 14, 18, 41); +} -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - add sm3p0, sm3p1, sm4ed and sm4ks instructions Co-authored-by: Ruibo Lu <luruibo2000@163.com> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-12-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/helper.h | 3 ++ target/riscv/insn32.decode | 6 +++ target/riscv/crypto_helper.c | 28 ++++++++++++ target/riscv/insn_trans/trans_rvk.c.inc | 58 +++++++++++++++++++++++++ 4 files changed, 95 insertions(+) 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_FLAGS_2(aes64dsm, TCG_CALL_NO_RWG_SE, tl, tl, tl) DEF_HELPER_FLAGS_2(aes64ks2, TCG_CALL_NO_RWG_SE, tl, tl, tl) DEF_HELPER_FLAGS_2(aes64ks1i, TCG_CALL_NO_RWG_SE, tl, tl, tl) DEF_HELPER_FLAGS_1(aes64im, TCG_CALL_NO_RWG_SE, tl, tl) + +DEF_HELPER_FLAGS_3(sm4ed, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) +DEF_HELPER_FLAGS_3(sm4ks, TCG_CALL_NO_RWG_SE, tl, tl, tl, tl) 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 @@ sha512sig0 00 01000 00110 ..... 001 ..... 0010011 @r2 sha512sig1 00 01000 00111 ..... 001 ..... 0010011 @r2 sha512sum0 00 01000 00100 ..... 001 ..... 0010011 @r2 sha512sum1 00 01000 00101 ..... 001 ..... 0010011 @r2 +# *** RV32 Zksh Standard Extension *** +sm3p0 00 01000 01000 ..... 001 ..... 0010011 @r2 +sm3p1 00 01000 01001 ..... 001 ..... 0010011 @r2 +# *** RV32 Zksed Standard Extension *** +sm4ed .. 11000 ..... ..... 000 ..... 0110011 @k_aes +sm4ks .. 11010 ..... ..... 000 ..... 0110011 @k_aes 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(aes64im)(target_ulong rs1) return result; } + +target_ulong HELPER(sm4ed)(target_ulong rs1, target_ulong rs2, + target_ulong shamt) +{ + uint32_t sb_in = (uint8_t)(rs2 >> shamt); + uint32_t sb_out = (uint32_t)sm4_sbox[sb_in]; + + uint32_t x = sb_out ^ (sb_out << 8) ^ (sb_out << 2) ^ (sb_out << 18) ^ + ((sb_out & 0x3f) << 26) ^ ((sb_out & 0xC0) << 10); + + uint32_t rotl = rol32(x, shamt); + + return sext32_xlen(rotl ^ (uint32_t)rs1); +} + +target_ulong HELPER(sm4ks)(target_ulong rs1, target_ulong rs2, + target_ulong shamt) +{ + uint32_t sb_in = (uint8_t)(rs2 >> shamt); + uint32_t sb_out = sm4_sbox[sb_in]; + + uint32_t x = sb_out ^ ((sb_out & 0x07) << 29) ^ ((sb_out & 0xFE) << 7) ^ + ((sb_out & 0x01) << 23) ^ ((sb_out & 0xF8) << 13); + + uint32_t rotl = rol32(x, shamt); + + return sext32_xlen(rotl ^ (uint32_t)rs1); +} #undef sext32_xlen diff --git a/target/riscv/insn_trans/trans_rvk.c.inc b/target/riscv/insn_trans/trans_rvk.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvk.c.inc +++ b/target/riscv/insn_trans/trans_rvk.c.inc @@ -XXX,XX +XXX,XX @@ } \ } while (0) +#define REQUIRE_ZKSED(ctx) do { \ + if (!ctx->cfg_ptr->ext_zksed) { \ + return false; \ + } \ +} while (0) + +#define REQUIRE_ZKSH(ctx) do { \ + if (!ctx->cfg_ptr->ext_zksh) { \ + return false; \ + } \ +} while (0) + static bool gen_aes32_sm4(DisasContext *ctx, arg_k_aes *a, void (*func)(TCGv, TCGv, TCGv, TCGv)) { @@ -XXX,XX +XXX,XX @@ static bool trans_sha512sum1(DisasContext *ctx, arg_sha512sum1 *a) REQUIRE_ZKNH(ctx); return gen_sha512_rv64(ctx, a, EXT_NONE, tcg_gen_rotri_i64, 14, 18, 41); } + +/* SM3 */ +static bool gen_sm3(DisasContext *ctx, arg_r2 *a, int32_t b, int32_t c) +{ + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, EXT_NONE); + TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_new_i32(); + + tcg_gen_trunc_tl_i32(t0, src1); + tcg_gen_rotli_i32(t1, t0, b); + tcg_gen_xor_i32(t1, t0, t1); + tcg_gen_rotli_i32(t0, t0, c); + tcg_gen_xor_i32(t1, t1, t0); + tcg_gen_ext_i32_tl(dest, t1); + gen_set_gpr(ctx, a->rd, dest); + + tcg_temp_free_i32(t0); + tcg_temp_free_i32(t1); + return true; +} + +static bool trans_sm3p0(DisasContext *ctx, arg_sm3p0 *a) +{ + REQUIRE_ZKSH(ctx); + return gen_sm3(ctx, a, 9, 17); +} + +static bool trans_sm3p1(DisasContext *ctx, arg_sm3p1 *a) +{ + REQUIRE_ZKSH(ctx); + return gen_sm3(ctx, a, 15, 23); +} + +/* SM4 */ +static bool trans_sm4ed(DisasContext *ctx, arg_sm4ed *a) +{ + REQUIRE_ZKSED(ctx); + return gen_aes32_sm4(ctx, a, gen_helper_sm4ed); +} + +static bool trans_sm4ks(DisasContext *ctx, arg_sm4ks *a) +{ + REQUIRE_ZKSED(ctx); + return gen_aes32_sm4(ctx, a, gen_helper_sm4ks); +} -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - add SEED CSR which must be accessed with a read-write instruction: A read-only instruction such as CSRRS/CSRRC with rs1=x0 or CSRRSI/CSRRCI with uimm=0 will raise an illegal instruction exception. - add USEED, SSEED fields for MSECCFG CSR Co-authored-by: Ruibo Lu <luruibo2000@163.com> Co-authored-by: Zewen Ye <lustrew@foxmail.com> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-13-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu_bits.h | 9 +++++ target/riscv/pmp.h | 8 ++-- target/riscv/csr.c | 80 ++++++++++++++++++++++++++++++++++++++++ target/riscv/op_helper.c | 9 +++++ 4 files changed, 103 insertions(+), 3 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 @@ #define CSR_VSPMMASK 0x2c1 #define CSR_VSPMBASE 0x2c2 +/* Crypto Extension */ +#define CSR_SEED 0x015 + /* mstatus CSR bits */ #define MSTATUS_UIE 0x00000001 #define MSTATUS_SIE 0x00000002 @@ -XXX,XX +XXX,XX @@ typedef enum RISCVException { #define HVICTL_VALID_MASK \ (HVICTL_VTI | HVICTL_IID | HVICTL_IPRIOM | HVICTL_IPRIO) +/* seed CSR bits */ +#define SEED_OPST (0b11 << 30) +#define SEED_OPST_BIST (0b00 << 30) +#define SEED_OPST_WAIT (0b01 << 30) +#define SEED_OPST_ES16 (0b10 << 30) +#define SEED_OPST_DEAD (0b11 << 30) #endif diff --git a/target/riscv/pmp.h b/target/riscv/pmp.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/pmp.h +++ b/target/riscv/pmp.h @@ -XXX,XX +XXX,XX @@ typedef enum { } pmp_am_t; typedef enum { - MSECCFG_MML = 1 << 0, - MSECCFG_MMWP = 1 << 1, - MSECCFG_RLB = 1 << 2 + MSECCFG_MML = 1 << 0, + MSECCFG_MMWP = 1 << 1, + MSECCFG_RLB = 1 << 2, + MSECCFG_USEED = 1 << 8, + MSECCFG_SSEED = 1 << 9 } mseccfg_field_t; typedef struct { 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 @@ #include "qemu/main-loop.h" #include "exec/exec-all.h" #include "sysemu/cpu-timers.h" +#include "qemu/guest-random.h" +#include "qapi/error.h" /* CSR function table public API */ void riscv_get_csr_ops(int csrno, riscv_csr_operations *ops) @@ -XXX,XX +XXX,XX @@ static RISCVException debug(CPURISCVState *env, int csrno) } #endif +static RISCVException seed(CPURISCVState *env, int csrno) +{ + RISCVCPU *cpu = env_archcpu(env); + + if (!cpu->cfg.ext_zkr) { + return RISCV_EXCP_ILLEGAL_INST; + } + +#if !defined(CONFIG_USER_ONLY) + /* + * With a CSR read-write instruction: + * 1) The seed CSR is always available in machine mode as normal. + * 2) Attempted access to seed from virtual modes VS and VU always raises + * an exception(virtual instruction exception only if mseccfg.sseed=1). + * 3) Without the corresponding access control bit set to 1, any attempted + * access to seed from U, S or HS modes will raise an illegal instruction + * exception. + */ + if (env->priv == PRV_M) { + return RISCV_EXCP_NONE; + } else if (riscv_cpu_virt_enabled(env)) { + if (env->mseccfg & MSECCFG_SSEED) { + return RISCV_EXCP_VIRT_INSTRUCTION_FAULT; + } else { + return RISCV_EXCP_ILLEGAL_INST; + } + } else { + if (env->priv == PRV_S && (env->mseccfg & MSECCFG_SSEED)) { + return RISCV_EXCP_NONE; + } else if (env->priv == PRV_U && (env->mseccfg & MSECCFG_USEED)) { + return RISCV_EXCP_NONE; + } else { + return RISCV_EXCP_ILLEGAL_INST; + } + } +#else + return RISCV_EXCP_NONE; +#endif +} + /* User Floating-Point CSRs */ static RISCVException read_fflags(CPURISCVState *env, int csrno, target_ulong *val) @@ -XXX,XX +XXX,XX @@ static RISCVException write_upmbase(CPURISCVState *env, int csrno, #endif +/* Crypto Extension */ +static RISCVException rmw_seed(CPURISCVState *env, int csrno, + target_ulong *ret_value, + target_ulong new_value, + target_ulong write_mask) +{ + uint16_t random_v; + Error *random_e = NULL; + int random_r; + target_ulong rval; + + random_r = qemu_guest_getrandom(&random_v, 2, &random_e); + if (unlikely(random_r < 0)) { + /* + * Failed, for unknown reasons in the crypto subsystem. + * The best we can do is log the reason and return a + * failure indication to the guest. There is no reason + * we know to expect the failure to be transitory, so + * indicate DEAD to avoid having the guest spin on WAIT. + */ + qemu_log_mask(LOG_UNIMP, "%s: Crypto failure: %s", + __func__, error_get_pretty(random_e)); + error_free(random_e); + rval = SEED_OPST_DEAD; + } else { + rval = random_v | SEED_OPST_ES16; + } + + if (ret_value) { + *ret_value = rval; + } + + return RISCV_EXCP_NONE; +} + /* * riscv_csrrw - read and/or update control and status register * @@ -XXX,XX +XXX,XX @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { [CSR_TIME] = { "time", ctr, read_time }, [CSR_TIMEH] = { "timeh", ctr32, read_timeh }, + /* Crypto Extension */ + [CSR_SEED] = { "seed", seed, NULL, NULL, rmw_seed }, + #if !defined(CONFIG_USER_ONLY) /* Machine Timers and Counters */ [CSR_MCYCLE] = { "mcycle", any, read_instret }, diff --git a/target/riscv/op_helper.c b/target/riscv/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/op_helper.c +++ b/target/riscv/op_helper.c @@ -XXX,XX +XXX,XX @@ void helper_raise_exception(CPURISCVState *env, uint32_t exception) target_ulong helper_csrr(CPURISCVState *env, int csr) { + /* + * The seed CSR must be accessed with a read-write instruction. A + * read-only instruction such as CSRRS/CSRRC with rs1=x0 or CSRRSI/ + * CSRRCI with uimm=0 will raise an illegal instruction exception. + */ + if (csr == CSR_SEED) { + riscv_raise_exception(env, RISCV_EXCP_ILLEGAL_INST, GETPC()); + } + target_ulong val = 0; RISCVException ret = riscv_csrrw(env, csr, &val, 0, 0); -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> Co-authored-by: Ruibo Lu <luruibo2000@163.com> Co-authored-by: Zewen Ye <lustrew@foxmail.com> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-14-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- disas/riscv.c | 173 +++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 172 insertions(+), 1 deletion(-) 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_swsp, rv_codec_css_sdsp, rv_codec_css_sqsp, + rv_codec_k_bs, + rv_codec_k_rnum, } rv_codec; typedef enum { @@ -XXX,XX +XXX,XX @@ typedef enum { rv_op_bclr = 359, rv_op_binv = 360, rv_op_bext = 361, + rv_op_aes32esmi = 362, + rv_op_aes32esi = 363, + rv_op_aes32dsmi = 364, + rv_op_aes32dsi = 365, + rv_op_aes64ks1i = 366, + rv_op_aes64ks2 = 367, + rv_op_aes64im = 368, + rv_op_aes64esm = 369, + rv_op_aes64es = 370, + rv_op_aes64dsm = 371, + rv_op_aes64ds = 372, + rv_op_sha256sig0 = 373, + rv_op_sha256sig1 = 374, + rv_op_sha256sum0 = 375, + rv_op_sha256sum1 = 376, + rv_op_sha512sig0 = 377, + rv_op_sha512sig1 = 378, + rv_op_sha512sum0 = 379, + rv_op_sha512sum1 = 380, + rv_op_sha512sum0r = 381, + rv_op_sha512sum1r = 382, + rv_op_sha512sig0l = 383, + rv_op_sha512sig0h = 384, + rv_op_sha512sig1l = 385, + rv_op_sha512sig1h = 386, + rv_op_sm3p0 = 387, + rv_op_sm3p1 = 388, + rv_op_sm4ed = 389, + rv_op_sm4ks = 390, + rv_op_brev8 = 391, + rv_op_pack = 392, + rv_op_packh = 393, + rv_op_packw = 394, + rv_op_unzip = 395, + rv_op_zip = 396, + rv_op_xperm4 = 397, + rv_op_xperm8 = 398, } rv_op; /* structures */ @@ -XXX,XX +XXX,XX @@ typedef struct { uint8_t succ; uint8_t aq; uint8_t rl; + uint8_t bs; + uint8_t rnum; } rv_decode; typedef struct { @@ -XXX,XX +XXX,XX @@ static const char rv_freg_name_sym[32][5] = { #define rv_fmt_rd_rs2 "O\t0,2" #define rv_fmt_rs1_offset "O\t1,o" #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" /* pseudo-instruction constraints */ @@ -XXX,XX +XXX,XX @@ static const rv_comp_data rvcp_csrrw[] = { { rv_op_illegal, NULL } }; + static const rv_comp_data rvcp_csrrs[] = { { rv_op_rdcycle, rvcc_rdcycle }, { rv_op_rdtime, rvcc_rdtime }, @@ -XXX,XX +XXX,XX @@ const rv_opcode_data opcode_data[] = { { "bclr", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, { "binv", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, { "bext", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "aes32esmi", rv_codec_k_bs, rv_fmt_rs1_rs2_bs, NULL, 0, 0, 0 }, + { "aes32esi", rv_codec_k_bs, rv_fmt_rs1_rs2_bs, NULL, 0, 0, 0 }, + { "aes32dsmi", rv_codec_k_bs, rv_fmt_rs1_rs2_bs, NULL, 0, 0, 0 }, + { "aes32dsi", rv_codec_k_bs, rv_fmt_rs1_rs2_bs, NULL, 0, 0, 0 }, + { "aes64ks1i", rv_codec_k_rnum, rv_fmt_rd_rs1_rnum, NULL, 0, 0, 0 }, + { "aes64ks2", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "aes64im", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0 }, + { "aes64esm", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "aes64es", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "aes64dsm", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "aes64ds", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha256sig0", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0 }, + { "sha256sig1", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0 }, + { "sha256sum0", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0 }, + { "sha256sum1", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0 }, + { "sha512sig0", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha512sig1", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha512sum0", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha512sum1", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha512sum0r", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha512sum1r", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha512sig0l", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha512sig0h", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha512sig1l", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sha512sig1h", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "sm3p0", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0 }, + { "sm3p1", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0 }, + { "sm4ed", rv_codec_k_bs, rv_fmt_rs1_rs2_bs, NULL, 0, 0, 0 }, + { "sm4ks", rv_codec_k_bs, rv_fmt_rs1_rs2_bs, NULL, 0, 0, 0 }, + { "brev8", rv_codec_r, rv_fmt_rd_rs1, NULL, 0, 0, 0 }, + { "pack", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "packh", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "packw", rv_codec_r, rv_fmt_rd_rs1_rs2, NULL, 0, 0, 0 }, + { "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 } }; /* 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 0x0015: return "seed"; case 0x0040: return "uscratch"; case 0x0041: return "uepc"; case 0x0042: return "ucause"; @@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa) case 1: switch (((inst >> 27) & 0b11111)) { case 0b00000: op = rv_op_slli; break; + case 0b00001: + switch (((inst >> 20) & 0b1111111)) { + case 0b0001111: op = rv_op_zip; break; + } + break; + case 0b00010: + switch (((inst >> 20) & 0b1111111)) { + case 0b0000000: op = rv_op_sha256sum0; break; + case 0b0000001: op = rv_op_sha256sum1; break; + case 0b0000010: op = rv_op_sha256sig0; break; + case 0b0000011: op = rv_op_sha256sig1; break; + case 0b0000100: op = rv_op_sha512sum0; break; + case 0b0000101: op = rv_op_sha512sum1; break; + case 0b0000110: op = rv_op_sha512sig0; break; + case 0b0000111: op = rv_op_sha512sig1; break; + case 0b0001000: op = rv_op_sm3p0; break; + case 0b0001001: op = rv_op_sm3p1; break; + } + break; case 0b00101: op = rv_op_bseti; break; + case 0b00110: + switch (((inst >> 20) & 0b1111111)) { + case 0b0000000: op = rv_op_aes64im; break; + default: + if (((inst >> 24) & 0b0111) == 0b001) { + op = rv_op_aes64ks1i; + } + break; + } + break; case 0b01001: op = rv_op_bclri; break; case 0b01101: op = rv_op_binvi; break; case 0b01100: @@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa) case 5: switch (((inst >> 27) & 0b11111)) { case 0b00000: op = rv_op_srli; break; + case 0b00001: + switch (((inst >> 20) & 0b1111111)) { + case 0b0001111: op = rv_op_unzip; break; + } + break; case 0b00101: op = rv_op_orc_b; break; case 0b01000: op = rv_op_srai; break; case 0b01001: op = rv_op_bexti; break; case 0b01100: op = rv_op_rori; break; case 0b01101: switch ((inst >> 20) & 0b1111111) { + case 0b0011000: op = rv_op_rev8; break; case 0b0111000: op = rv_op_rev8; break; + case 0b0000111: op = rv_op_brev8; break; } break; } @@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa) case 36: switch ((inst >> 20) & 0b11111) { case 0: op = rv_op_zext_h; break; + default: op = rv_op_pack; break; } break; + case 39: op = rv_op_packh; break; + case 41: op = rv_op_clmul; break; case 42: op = rv_op_clmulr; break; case 43: op = rv_op_clmulh; break; @@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa) case 132: op = rv_op_sh2add; break; case 134: op = rv_op_sh3add; break; case 161: op = rv_op_bset; break; + case 162: op = rv_op_xperm4; break; + case 164: op = rv_op_xperm8; break; + case 200: op = rv_op_aes64es; break; + case 216: op = rv_op_aes64esm; break; + case 232: op = rv_op_aes64ds; break; + case 248: op = rv_op_aes64dsm; break; case 256: op = rv_op_sub; break; case 260: op = rv_op_xnor; break; case 261: op = rv_op_sra; break; @@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa) case 263: op = rv_op_andn; break; case 289: op = rv_op_bclr; break; case 293: op = rv_op_bext; break; + case 320: op = rv_op_sha512sum0r; break; + case 328: op = rv_op_sha512sum1r; break; + case 336: op = rv_op_sha512sig0l; break; + case 344: op = rv_op_sha512sig1l; break; + case 368: op = rv_op_sha512sig0h; break; + case 376: op = rv_op_sha512sig1h; break; case 385: op = rv_op_rol; break; - case 386: op = rv_op_ror; break; + case 389: op = rv_op_ror; break; case 417: op = rv_op_binv; break; + case 504: op = rv_op_aes64ks2; break; + } + switch ((inst >> 25) & 0b0011111) { + case 17: op = rv_op_aes32esi; break; + case 19: op = rv_op_aes32esmi; break; + case 21: op = rv_op_aes32dsi; break; + case 23: op = rv_op_aes32dsmi; break; + case 24: op = rv_op_sm4ed; break; + case 26: op = rv_op_sm4ks; break; } break; case 13: op = rv_op_lui; break; @@ -XXX,XX +XXX,XX @@ static void decode_inst_opcode(rv_decode *dec, rv_isa isa) case 36: switch ((inst >> 20) & 0b11111) { case 0: op = rv_op_zext_h; break; + default: op = rv_op_packw; break; } break; case 130: op = rv_op_sh1add_uw; break; @@ -XXX,XX +XXX,XX @@ static uint32_t operand_cimmq(rv_inst inst) ((inst << 57) >> 62) << 6; } +static uint32_t operand_bs(rv_inst inst) +{ + return (inst << 32) >> 62; +} + +static uint32_t operand_rnum(rv_inst inst) +{ + return (inst << 40) >> 60; +} + /* decode operands */ static void decode_inst_operands(rv_decode *dec) @@ -XXX,XX +XXX,XX @@ static void decode_inst_operands(rv_decode *dec) dec->rs2 = operand_crs2(inst); dec->imm = operand_cimmsqsp(inst); break; + case rv_codec_k_bs: + dec->rs1 = operand_rs1(inst); + dec->rs2 = operand_rs2(inst); + dec->bs = operand_bs(inst); + break; + case rv_codec_k_rnum: + dec->rd = operand_rd(inst); + dec->rs1 = operand_rs1(inst); + dec->rnum = operand_rnum(inst); + break; }; } @@ -XXX,XX +XXX,XX @@ static void format_inst(char *buf, size_t buflen, size_t tab, rv_decode *dec) case ')': append(buf, ")", buflen); break; + case 'b': + snprintf(tmp, sizeof(tmp), "%d", dec->bs); + append(buf, tmp, buflen); + break; + case 'n': + snprintf(tmp, sizeof(tmp), "%d", dec->rnum); + append(buf, tmp, buflen); + break; case '0': append(buf, rv_ireg_name_sym[dec->rd], buflen); break; -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423023510.30794-15-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.c | 13 +++++++++++++ 1 file changed, 13 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 Property riscv_cpu_properties[] = { DEFINE_PROP_BOOL("zba", RISCVCPU, cfg.ext_zba, true), DEFINE_PROP_BOOL("zbb", RISCVCPU, cfg.ext_zbb, true), DEFINE_PROP_BOOL("zbc", RISCVCPU, cfg.ext_zbc, true), + DEFINE_PROP_BOOL("zbkb", RISCVCPU, cfg.ext_zbkb, false), + DEFINE_PROP_BOOL("zbkc", RISCVCPU, cfg.ext_zbkc, false), + DEFINE_PROP_BOOL("zbkx", RISCVCPU, cfg.ext_zbkx, false), DEFINE_PROP_BOOL("zbs", RISCVCPU, cfg.ext_zbs, true), + DEFINE_PROP_BOOL("zk", RISCVCPU, cfg.ext_zk, false), + DEFINE_PROP_BOOL("zkn", RISCVCPU, cfg.ext_zkn, false), + DEFINE_PROP_BOOL("zknd", RISCVCPU, cfg.ext_zknd, false), + DEFINE_PROP_BOOL("zkne", RISCVCPU, cfg.ext_zkne, false), + DEFINE_PROP_BOOL("zknh", RISCVCPU, cfg.ext_zknh, false), + DEFINE_PROP_BOOL("zkr", RISCVCPU, cfg.ext_zkr, false), + DEFINE_PROP_BOOL("zks", RISCVCPU, cfg.ext_zks, false), + DEFINE_PROP_BOOL("zksed", RISCVCPU, cfg.ext_zksed, false), + DEFINE_PROP_BOOL("zksh", RISCVCPU, cfg.ext_zksh, false), + DEFINE_PROP_BOOL("zkt", RISCVCPU, cfg.ext_zkt, false), DEFINE_PROP_BOOL("zdinx", RISCVCPU, cfg.ext_zdinx, false), DEFINE_PROP_BOOL("zfinx", RISCVCPU, cfg.ext_zfinx, false), -- 2.35.1 From: Ralf Ramsauer <ralf.ramsauer@oth-regensburg.de> Two non-subsequent PTEs can be mapped to subsequent paddrs. In this case, walk_pte will erroneously merge them. Enforce the split up, by tracking the virtual base address. Let's say we have the mapping: 0x81200000 -> 0x89623000 (4K) 0x8120f000 -> 0x89624000 (4K) Before, walk_pte would have shown: vaddr paddr size attr ---------------- ---------------- ---------------- ------- 0000000081200000 0000000089623000 0000000000002000 rwxu-ad as it only checks for subsequent paddrs. With this patch, it becomes: vaddr paddr size attr ---------------- ---------------- ---------------- ------- 0000000081200000 0000000089623000 0000000000001000 rwxu-ad 000000008120f000 0000000089624000 0000000000001000 rwxu-ad Signed-off-by: Ralf Ramsauer <ralf.ramsauer@oth-regensburg.de> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220423215907.673663-1-ralf.ramsauer@oth-regensburg.de> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/monitor.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/target/riscv/monitor.c b/target/riscv/monitor.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/monitor.c +++ b/target/riscv/monitor.c @@ -XXX,XX +XXX,XX @@ static void walk_pte(Monitor *mon, hwaddr base, target_ulong start, { hwaddr pte_addr; hwaddr paddr; + target_ulong last_start = -1; target_ulong pgsize; target_ulong pte; int ptshift; @@ -XXX,XX +XXX,XX @@ static void walk_pte(Monitor *mon, hwaddr base, target_ulong start, * A leaf PTE has been found * * If current PTE's permission bits differ from the last one, - * or current PTE's ppn does not make a contiguous physical - * address block together with the last one, print out the last - * contiguous mapped block details. + * or the current PTE breaks up a contiguous virtual or + * physical mapping, address block together with the last one, + * print out the last contiguous mapped block details. */ if ((*last_attr != attr) || - (*last_paddr + *last_size != paddr)) { + (*last_paddr + *last_size != paddr) || + (last_start + *last_size != start)) { print_pte(mon, va_bits, *vbase, *pbase, *last_paddr + *last_size - *pbase, *last_attr); @@ -XXX,XX +XXX,XX @@ static void walk_pte(Monitor *mon, hwaddr base, target_ulong start, *last_attr = attr; } + last_start = start; *last_paddr = paddr; *last_size = pgsize; } else { -- 2.35.1 From: Weiwei Li <liweiwei@iscas.ac.cn> - add zbk* and zk* strings to isa_edata_arr Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Tested-by: Jiatai He <jiatai2021@iscas.ac.cn> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220426095204.24142-1-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.c | 13 +++++++++++++ 1 file changed, 13 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_isa_string_ext(RISCVCPU *cpu, char **isa_str, int max_str_len) ISA_EDATA_ENTRY(zba, ext_zba), ISA_EDATA_ENTRY(zbb, ext_zbb), ISA_EDATA_ENTRY(zbc, ext_zbc), + ISA_EDATA_ENTRY(zbkb, ext_zbkb), + ISA_EDATA_ENTRY(zbkc, ext_zbkc), + ISA_EDATA_ENTRY(zbkx, ext_zbkx), ISA_EDATA_ENTRY(zbs, ext_zbs), + ISA_EDATA_ENTRY(zk, ext_zk), + ISA_EDATA_ENTRY(zkn, ext_zkn), + ISA_EDATA_ENTRY(zknd, ext_zknd), + ISA_EDATA_ENTRY(zkne, ext_zkne), + ISA_EDATA_ENTRY(zknh, ext_zknh), + ISA_EDATA_ENTRY(zkr, ext_zkr), + ISA_EDATA_ENTRY(zks, ext_zks), + ISA_EDATA_ENTRY(zksed, ext_zksed), + ISA_EDATA_ENTRY(zksh, ext_zksh), + ISA_EDATA_ENTRY(zkt, ext_zkt), ISA_EDATA_ENTRY(zve32f, ext_zve32f), ISA_EDATA_ENTRY(zve64f, ext_zve64f), ISA_EDATA_ENTRY(svinval, ext_svinval), -- 2.35.1 From: Alistair Francis <alistair.francis@wdc.com> Move the binary and device tree loading code to the machine done notifier. This allows us to prepare for editing the device tree as part of the notifier. This is based on similar code in the ARM virt machine. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220427234146.1130752-2-alistair.francis@opensource.wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- include/hw/riscv/virt.h | 1 + hw/riscv/virt.c | 191 +++++++++++++++++++++------------------- 2 files changed, 102 insertions(+), 90 deletions(-) diff --git a/include/hw/riscv/virt.h b/include/hw/riscv/virt.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/riscv/virt.h +++ b/include/hw/riscv/virt.h @@ -XXX,XX +XXX,XX @@ struct RISCVVirtState { MachineState parent; /*< public >*/ + Notifier machine_done; RISCVHartArrayState soc[VIRT_SOCKETS_MAX]; DeviceState *irqchip[VIRT_SOCKETS_MAX]; PFlashCFI01 *flash[2]; 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 DeviceState *virt_create_aia(RISCVVirtAIAType aia_type, int aia_guests, return aplic_m; } +static void virt_machine_done(Notifier *notifier, void *data) +{ + RISCVVirtState *s = container_of(notifier, RISCVVirtState, + machine_done); + const MemMapEntry *memmap = virt_memmap; + MachineState *machine = MACHINE(s); + target_ulong start_addr = memmap[VIRT_DRAM].base; + target_ulong firmware_end_addr, kernel_start_addr; + uint32_t fdt_load_addr; + uint64_t kernel_entry; + + /* + * Only direct boot kernel is currently supported for KVM VM, + * so the "-bios" parameter is not supported when KVM is enabled. + */ + if (kvm_enabled()) { + if (machine->firmware) { + if (strcmp(machine->firmware, "none")) { + error_report("Machine mode firmware is not supported in " + "combination with KVM."); + exit(1); + } + } else { + machine->firmware = g_strdup("none"); + } + } + + if (riscv_is_32bit(&s->soc[0])) { + firmware_end_addr = riscv_find_and_load_firmware(machine, + RISCV32_BIOS_BIN, start_addr, NULL); + } else { + firmware_end_addr = riscv_find_and_load_firmware(machine, + RISCV64_BIOS_BIN, start_addr, NULL); + } + + if (machine->kernel_filename) { + kernel_start_addr = riscv_calc_kernel_start_addr(&s->soc[0], + firmware_end_addr); + + kernel_entry = riscv_load_kernel(machine->kernel_filename, + kernel_start_addr, NULL); + + if (machine->initrd_filename) { + hwaddr start; + hwaddr end = riscv_load_initrd(machine->initrd_filename, + machine->ram_size, kernel_entry, + &start); + qemu_fdt_setprop_cell(machine->fdt, "/chosen", + "linux,initrd-start", start); + qemu_fdt_setprop_cell(machine->fdt, "/chosen", "linux,initrd-end", + end); + } + } else { + /* + * If dynamic firmware is used, it doesn't know where is the next mode + * if kernel argument is not set. + */ + kernel_entry = 0; + } + + if (drive_get(IF_PFLASH, 0, 0)) { + /* + * Pflash was supplied, let's overwrite the address we jump to after + * reset to the base of the flash. + */ + 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); + /* load the reset vector */ + riscv_setup_rom_reset_vec(machine, &s->soc[0], start_addr, + virt_memmap[VIRT_MROM].base, + virt_memmap[VIRT_MROM].size, kernel_entry, + fdt_load_addr, machine->fdt); + + /* + * Only direct boot kernel is currently supported for KVM VM, + * So here setup kernel start address and fdt address. + * TODO:Support firmware loading and integrate to TCG start + */ + if (kvm_enabled()) { + riscv_setup_direct_kernel(kernel_entry, fdt_load_addr); + } +} + static void virt_machine_init(MachineState *machine) { const MemMapEntry *memmap = virt_memmap; @@ -XXX,XX +XXX,XX @@ static void virt_machine_init(MachineState *machine) MemoryRegion *system_memory = get_system_memory(); MemoryRegion *mask_rom = g_new(MemoryRegion, 1); char *soc_name; - target_ulong start_addr = memmap[VIRT_DRAM].base; - target_ulong firmware_end_addr, kernel_start_addr; - uint32_t fdt_load_addr; - uint64_t kernel_entry; DeviceState *mmio_irqchip, *virtio_irqchip, *pcie_irqchip; int i, base_hartid, hart_count; @@ -XXX,XX +XXX,XX @@ static void virt_machine_init(MachineState *machine) memory_region_add_subregion(system_memory, memmap[VIRT_DRAM].base, machine->ram); - /* create device tree */ - create_fdt(s, memmap, machine->ram_size, machine->kernel_cmdline, - riscv_is_32bit(&s->soc[0])); - /* boot rom */ memory_region_init_rom(mask_rom, NULL, "riscv_virt_board.mrom", memmap[VIRT_MROM].size, &error_fatal); memory_region_add_subregion(system_memory, memmap[VIRT_MROM].base, mask_rom); - /* - * Only direct boot kernel is currently supported for KVM VM, - * so the "-bios" parameter is not supported when KVM is enabled. - */ - if (kvm_enabled()) { - if (machine->firmware) { - if (strcmp(machine->firmware, "none")) { - error_report("Machine mode firmware is not supported in " - "combination with KVM."); - exit(1); - } - } else { - machine->firmware = g_strdup("none"); - } - } - - if (riscv_is_32bit(&s->soc[0])) { - firmware_end_addr = riscv_find_and_load_firmware(machine, - RISCV32_BIOS_BIN, start_addr, NULL); - } else { - firmware_end_addr = riscv_find_and_load_firmware(machine, - RISCV64_BIOS_BIN, start_addr, NULL); - } - - if (machine->kernel_filename) { - kernel_start_addr = riscv_calc_kernel_start_addr(&s->soc[0], - firmware_end_addr); - - kernel_entry = riscv_load_kernel(machine->kernel_filename, - kernel_start_addr, NULL); - - if (machine->initrd_filename) { - hwaddr start; - hwaddr end = riscv_load_initrd(machine->initrd_filename, - machine->ram_size, kernel_entry, - &start); - qemu_fdt_setprop_cell(machine->fdt, "/chosen", - "linux,initrd-start", start); - qemu_fdt_setprop_cell(machine->fdt, "/chosen", "linux,initrd-end", - end); - } - } else { - /* - * If dynamic firmware is used, it doesn't know where is the next mode - * if kernel argument is not set. - */ - kernel_entry = 0; - } - - if (drive_get(IF_PFLASH, 0, 0)) { - /* - * Pflash was supplied, let's overwrite the address we jump to after - * reset to the base of the flash. - */ - 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); - /* load the reset vector */ - riscv_setup_rom_reset_vec(machine, &s->soc[0], start_addr, - virt_memmap[VIRT_MROM].base, - virt_memmap[VIRT_MROM].size, kernel_entry, - fdt_load_addr, machine->fdt); - - /* - * Only direct boot kernel is currently supported for KVM VM, - * So here setup kernel start address and fdt address. - * TODO:Support firmware loading and integrate to TCG start - */ - if (kvm_enabled()) { - riscv_setup_direct_kernel(kernel_entry, fdt_load_addr); - } - /* SiFive Test MMIO device */ sifive_test_create(memmap[VIRT_TEST].base); @@ -XXX,XX +XXX,XX @@ static void virt_machine_init(MachineState *machine) drive_get(IF_PFLASH, 0, i)); } virt_flash_map(s, system_memory); + + /* create device tree */ + create_fdt(s, memmap, machine->ram_size, machine->kernel_cmdline, + riscv_is_32bit(&s->soc[0])); + + s->machine_done.notify = virt_machine_done; + qemu_add_machine_init_done_notifier(&s->machine_done); } static void virt_machine_instance_init(Object *obj) -- 2.35.1 From: Alistair Francis <alistair.francis@wdc.com> The ARM virt machine currently uses sysbus-fdt to create device tree entries for dynamically created MMIO devices. The RISC-V virt machine can also benefit from this, so move the code to the core directory. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220427234146.1130752-3-alistair.francis@opensource.wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- include/hw/{arm => core}/sysbus-fdt.h | 0 hw/arm/virt.c | 2 +- hw/arm/xlnx-versal-virt.c | 1 - hw/{arm => core}/sysbus-fdt.c | 2 +- hw/arm/meson.build | 1 - hw/core/meson.build | 1 + 6 files changed, 3 insertions(+), 4 deletions(-) rename include/hw/{arm => core}/sysbus-fdt.h (100%) rename hw/{arm => core}/sysbus-fdt.c (99%) diff --git a/include/hw/arm/sysbus-fdt.h b/include/hw/core/sysbus-fdt.h similarity index 100% rename from include/hw/arm/sysbus-fdt.h rename to include/hw/core/sysbus-fdt.h diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ #include "qemu/module.h" #include "hw/pci-host/gpex.h" #include "hw/virtio/virtio-pci.h" -#include "hw/arm/sysbus-fdt.h" +#include "hw/core/sysbus-fdt.h" #include "hw/platform-bus.h" #include "hw/qdev-properties.h" #include "hw/arm/fdt.h" diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xlnx-versal-virt.c +++ b/hw/arm/xlnx-versal-virt.c @@ -XXX,XX +XXX,XX @@ #include "sysemu/device_tree.h" #include "hw/boards.h" #include "hw/sysbus.h" -#include "hw/arm/sysbus-fdt.h" #include "hw/arm/fdt.h" #include "cpu.h" #include "hw/qdev-properties.h" diff --git a/hw/arm/sysbus-fdt.c b/hw/core/sysbus-fdt.c similarity index 99% rename from hw/arm/sysbus-fdt.c rename to hw/core/sysbus-fdt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/sysbus-fdt.c +++ b/hw/core/sysbus-fdt.c @@ -XXX,XX +XXX,XX @@ #ifdef CONFIG_LINUX #include <linux/vfio.h> #endif -#include "hw/arm/sysbus-fdt.h" +#include "hw/core/sysbus-fdt.h" #include "qemu/error-report.h" #include "sysemu/device_tree.h" #include "sysemu/tpm.h" diff --git a/hw/arm/meson.build b/hw/arm/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/meson.build +++ b/hw/arm/meson.build @@ -XXX,XX +XXX,XX @@ arm_ss = ss.source_set() arm_ss.add(files('boot.c'), fdt) -arm_ss.add(when: 'CONFIG_PLATFORM_BUS', if_true: files('sysbus-fdt.c')) arm_ss.add(when: 'CONFIG_ARM_VIRT', if_true: files('virt.c')) arm_ss.add(when: 'CONFIG_ACPI', if_true: files('virt-acpi-build.c')) arm_ss.add(when: 'CONFIG_DIGIC', if_true: files('digic_boards.c')) diff --git a/hw/core/meson.build b/hw/core/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/core/meson.build +++ b/hw/core/meson.build @@ -XXX,XX +XXX,XX @@ softmmu_ss.add(when: 'CONFIG_PTIMER', if_true: files('ptimer.c')) softmmu_ss.add(when: 'CONFIG_REGISTER', if_true: files('register.c')) softmmu_ss.add(when: 'CONFIG_SPLIT_IRQ', if_true: files('split-irq.c')) softmmu_ss.add(when: 'CONFIG_XILINX_AXI', if_true: files('stream.c')) +softmmu_ss.add(when: 'CONFIG_PLATFORM_BUS', if_true: files('sysbus-fdt.c')) softmmu_ss.add(files( 'cpu-sysemu.c', -- 2.35.1 From: Alistair Francis <alistair.francis@wdc.com> Create a platform bus to allow dynamic devices to be connected. This is based on the ARM implementation. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220427234146.1130752-4-alistair.francis@opensource.wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- include/hw/riscv/virt.h | 7 ++++- hw/riscv/virt.c | 68 +++++++++++++++++++++++++++++------------ hw/riscv/Kconfig | 1 + 3 files changed, 56 insertions(+), 20 deletions(-) diff --git a/include/hw/riscv/virt.h b/include/hw/riscv/virt.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/riscv/virt.h +++ b/include/hw/riscv/virt.h @@ -XXX,XX +XXX,XX @@ struct RISCVVirtState { /*< public >*/ Notifier machine_done; + DeviceState *platform_bus_dev; RISCVHartArrayState soc[VIRT_SOCKETS_MAX]; DeviceState *irqchip[VIRT_SOCKETS_MAX]; PFlashCFI01 *flash[2]; @@ -XXX,XX +XXX,XX @@ enum { VIRT_DRAM, VIRT_PCIE_MMIO, VIRT_PCIE_PIO, + VIRT_PLATFORM_BUS, VIRT_PCIE_ECAM }; @@ -XXX,XX +XXX,XX @@ enum { VIRTIO_IRQ = 1, /* 1 to 8 */ VIRTIO_COUNT = 8, PCIE_IRQ = 0x20, /* 32 to 35 */ - VIRTIO_NDEV = 0x35 /* Arbitrary maximum number of interrupts */ + VIRT_PLATFORM_BUS_IRQ = 64, /* 64 to 96 */ + VIRTIO_NDEV = 96 /* Arbitrary maximum number of interrupts */ }; +#define VIRT_PLATFORM_BUS_NUM_IRQS 32 + #define VIRT_IRQCHIP_IPI_MSI 1 #define VIRT_IRQCHIP_NUM_MSIS 255 #define VIRT_IRQCHIP_NUM_SOURCES VIRTIO_NDEV 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/intc/riscv_imsic.h" #include "hw/intc/sifive_plic.h" #include "hw/misc/sifive_test.h" +#include "hw/platform-bus.h" #include "chardev/char.h" #include "sysemu/device_tree.h" #include "sysemu/sysemu.h" @@ -XXX,XX +XXX,XX @@ #endif static const MemMapEntry virt_memmap[] = { - [VIRT_DEBUG] = { 0x0, 0x100 }, - [VIRT_MROM] = { 0x1000, 0xf000 }, - [VIRT_TEST] = { 0x100000, 0x1000 }, - [VIRT_RTC] = { 0x101000, 0x1000 }, - [VIRT_CLINT] = { 0x2000000, 0x10000 }, - [VIRT_ACLINT_SSWI] = { 0x2F00000, 0x4000 }, - [VIRT_PCIE_PIO] = { 0x3000000, 0x10000 }, - [VIRT_PLIC] = { 0xc000000, VIRT_PLIC_SIZE(VIRT_CPUS_MAX * 2) }, - [VIRT_APLIC_M] = { 0xc000000, APLIC_SIZE(VIRT_CPUS_MAX) }, - [VIRT_APLIC_S] = { 0xd000000, APLIC_SIZE(VIRT_CPUS_MAX) }, - [VIRT_UART0] = { 0x10000000, 0x100 }, - [VIRT_VIRTIO] = { 0x10001000, 0x1000 }, - [VIRT_FW_CFG] = { 0x10100000, 0x18 }, - [VIRT_FLASH] = { 0x20000000, 0x4000000 }, - [VIRT_IMSIC_M] = { 0x24000000, VIRT_IMSIC_MAX_SIZE }, - [VIRT_IMSIC_S] = { 0x28000000, VIRT_IMSIC_MAX_SIZE }, - [VIRT_PCIE_ECAM] = { 0x30000000, 0x10000000 }, - [VIRT_PCIE_MMIO] = { 0x40000000, 0x40000000 }, - [VIRT_DRAM] = { 0x80000000, 0x0 }, + [VIRT_DEBUG] = { 0x0, 0x100 }, + [VIRT_MROM] = { 0x1000, 0xf000 }, + [VIRT_TEST] = { 0x100000, 0x1000 }, + [VIRT_RTC] = { 0x101000, 0x1000 }, + [VIRT_CLINT] = { 0x2000000, 0x10000 }, + [VIRT_ACLINT_SSWI] = { 0x2F00000, 0x4000 }, + [VIRT_PCIE_PIO] = { 0x3000000, 0x10000 }, + [VIRT_PLATFORM_BUS] = { 0x4000000, 0x2000000 }, + [VIRT_PLIC] = { 0xc000000, VIRT_PLIC_SIZE(VIRT_CPUS_MAX * 2) }, + [VIRT_APLIC_M] = { 0xc000000, APLIC_SIZE(VIRT_CPUS_MAX) }, + [VIRT_APLIC_S] = { 0xd000000, APLIC_SIZE(VIRT_CPUS_MAX) }, + [VIRT_UART0] = { 0x10000000, 0x100 }, + [VIRT_VIRTIO] = { 0x10001000, 0x1000 }, + [VIRT_FW_CFG] = { 0x10100000, 0x18 }, + [VIRT_FLASH] = { 0x20000000, 0x4000000 }, + [VIRT_IMSIC_M] = { 0x24000000, VIRT_IMSIC_MAX_SIZE }, + [VIRT_IMSIC_S] = { 0x28000000, VIRT_IMSIC_MAX_SIZE }, + [VIRT_PCIE_ECAM] = { 0x30000000, 0x10000000 }, + [VIRT_PCIE_MMIO] = { 0x40000000, 0x40000000 }, + [VIRT_DRAM] = { 0x80000000, 0x0 }, }; /* PCIe high mmio is fixed for RV32 */ @@ -XXX,XX +XXX,XX @@ static DeviceState *virt_create_aia(RISCVVirtAIAType aia_type, int aia_guests, return aplic_m; } +static void create_platform_bus(RISCVVirtState *s, DeviceState *irqchip) +{ + DeviceState *dev; + SysBusDevice *sysbus; + const MemMapEntry *memmap = virt_memmap; + int i; + MemoryRegion *sysmem = get_system_memory(); + + dev = qdev_new(TYPE_PLATFORM_BUS_DEVICE); + dev->id = g_strdup(TYPE_PLATFORM_BUS_DEVICE); + qdev_prop_set_uint32(dev, "num_irqs", VIRT_PLATFORM_BUS_NUM_IRQS); + qdev_prop_set_uint32(dev, "mmio_size", memmap[VIRT_PLATFORM_BUS].size); + sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal); + s->platform_bus_dev = dev; + + sysbus = SYS_BUS_DEVICE(dev); + for (i = 0; i < VIRT_PLATFORM_BUS_NUM_IRQS; i++) { + int irq = VIRT_PLATFORM_BUS_IRQ + i; + sysbus_connect_irq(sysbus, i, qdev_get_gpio_in(irqchip, irq)); + } + + memory_region_add_subregion(sysmem, + memmap[VIRT_PLATFORM_BUS].base, + sysbus_mmio_get_region(sysbus, 0)); +} + static void virt_machine_done(Notifier *notifier, void *data) { RISCVVirtState *s = container_of(notifier, RISCVVirtState, @@ -XXX,XX +XXX,XX @@ static void virt_machine_init(MachineState *machine) memmap[VIRT_PCIE_PIO].base, DEVICE(pcie_irqchip)); + create_platform_bus(s, DEVICE(mmio_irqchip)); + serial_mm_init(system_memory, memmap[VIRT_UART0].base, 0, qdev_get_gpio_in(DEVICE(mmio_irqchip), UART0_IRQ), 399193, serial_hd(0), DEVICE_LITTLE_ENDIAN); diff --git a/hw/riscv/Kconfig b/hw/riscv/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/Kconfig +++ b/hw/riscv/Kconfig @@ -XXX,XX +XXX,XX @@ config RISCV_VIRT select SIFIVE_TEST select VIRTIO_MMIO select FW_CFG_DMA + select PLATFORM_BUS config SIFIVE_E bool -- 2.35.1 From: Alistair Francis <alistair.francis@wdc.com> Similar to the ARM virt machine add support for adding device tree entries for dynamically created devices. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com> Message-Id: <20220427234146.1130752-5-alistair.francis@opensource.wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) 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/qdev-properties.h" #include "hw/char/serial.h" #include "target/riscv/cpu.h" +#include "hw/core/sysbus-fdt.h" #include "hw/riscv/riscv_hart.h" #include "hw/riscv/virt.h" #include "hw/riscv/boot.h" @@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_plic(RISCVVirtState *s, riscv_socket_fdt_write_id(mc, mc->fdt, plic_name, socket); qemu_fdt_setprop_cell(mc->fdt, plic_name, "phandle", plic_phandles[socket]); + + platform_bus_add_all_fdt_nodes(mc->fdt, plic_name, + memmap[VIRT_PLATFORM_BUS].base, + memmap[VIRT_PLATFORM_BUS].size, + VIRT_PLATFORM_BUS_IRQ); + g_free(plic_name); g_free(plic_cells); @@ -XXX,XX +XXX,XX @@ static void create_fdt_imsic(RISCVVirtState *s, const MemMapEntry *memmap, IMSIC_MMIO_GROUP_MIN_SHIFT); } qemu_fdt_setprop_cell(mc->fdt, imsic_name, "phandle", *msi_m_phandle); + + platform_bus_add_all_fdt_nodes(mc->fdt, imsic_name, + memmap[VIRT_PLATFORM_BUS].base, + memmap[VIRT_PLATFORM_BUS].size, + VIRT_PLATFORM_BUS_IRQ); + g_free(imsic_name); /* S-level IMSIC node */ @@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aplic(RISCVVirtState *s, VIRT_IRQCHIP_NUM_SOURCES); riscv_socket_fdt_write_id(mc, mc->fdt, aplic_name, socket); qemu_fdt_setprop_cell(mc->fdt, aplic_name, "phandle", aplic_s_phandle); + + platform_bus_add_all_fdt_nodes(mc->fdt, aplic_name, + memmap[VIRT_PLATFORM_BUS].base, + memmap[VIRT_PLATFORM_BUS].size, + VIRT_PLATFORM_BUS_IRQ); + g_free(aplic_name); g_free(aplic_cells); -- 2.35.1 From: Alistair Francis <alistair.francis@wdc.com> Add support for plugging in devices, this was tested with the TPM device. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220427234146.1130752-6-alistair.francis@opensource.wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) 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_set_aclint(Object *obj, bool value, Error **errp) s->have_aclint = value; } +static HotplugHandler *virt_machine_get_hotplug_handler(MachineState *machine, + DeviceState *dev) +{ + MachineClass *mc = MACHINE_GET_CLASS(machine); + + if (device_is_dynamic_sysbus(mc, dev)) { + return HOTPLUG_HANDLER(machine); + } + return NULL; +} + +static void virt_machine_device_plug_cb(HotplugHandler *hotplug_dev, + DeviceState *dev, Error **errp) +{ + RISCVVirtState *s = RISCV_VIRT_MACHINE(hotplug_dev); + + if (s->platform_bus_dev) { + MachineClass *mc = MACHINE_GET_CLASS(s); + + if (device_is_dynamic_sysbus(mc, dev)) { + platform_bus_link_device(PLATFORM_BUS_DEVICE(s->platform_bus_dev), + SYS_BUS_DEVICE(dev)); + } + } +} + static void virt_machine_class_init(ObjectClass *oc, void *data) { char str[128]; MachineClass *mc = MACHINE_CLASS(oc); + HotplugHandlerClass *hc = HOTPLUG_HANDLER_CLASS(oc); mc->desc = "RISC-V VirtIO board"; mc->init = virt_machine_init; @@ -XXX,XX +XXX,XX @@ static void virt_machine_class_init(ObjectClass *oc, void *data) mc->get_default_cpu_node_id = riscv_numa_get_default_cpu_node_id; mc->numa_mem_supported = true; mc->default_ram_id = "riscv_virt_board.ram"; + assert(!mc->get_hotplug_handler); + mc->get_hotplug_handler = virt_machine_get_hotplug_handler; + + hc->plug = virt_machine_device_plug_cb; machine_class_allow_dynamic_sysbus_dev(mc, TYPE_RAMFB_DEVICE); @@ -XXX,XX +XXX,XX @@ static const TypeInfo virt_machine_typeinfo = { .class_init = virt_machine_class_init, .instance_init = virt_machine_instance_init, .instance_size = sizeof(RISCVVirtState), + .interfaces = (InterfaceInfo[]) { + { TYPE_HOTPLUG_HANDLER }, + { } + }, }; static void virt_machine_init_register_types(void) -- 2.35.1 From: Alistair Francis <alistair.francis@wdc.com> Imply the TPM sysbus devices. This allows users to add TPM devices to the RISC-V virt board. This was tested by first creating an emulated TPM device: swtpm socket --tpm2 -t -d --tpmstate dir=/tmp/tpm \ --ctrl type=unixio,path=swtpm-sock Then launching QEMU with: -chardev socket,id=chrtpm,path=swtpm-sock \ -tpmdev emulator,id=tpm0,chardev=chrtpm \ -device tpm-tis-device,tpmdev=tpm0 The TPM device can be seen in the memory tree and the generated device tree. Resolves: https://gitlab.com/qemu-project/qemu/-/issues/942 Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220427234146.1130752-7-alistair.francis@opensource.wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- docs/system/riscv/virt.rst | 20 ++++++++++++++++++++ hw/riscv/virt.c | 4 ++++ hw/riscv/Kconfig | 1 + 3 files changed, 25 insertions(+) diff --git a/docs/system/riscv/virt.rst b/docs/system/riscv/virt.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/riscv/virt.rst +++ b/docs/system/riscv/virt.rst @@ -XXX,XX +XXX,XX @@ The minimal QEMU commands to run U-Boot SPL are: To test 32-bit U-Boot images, switch to use qemu-riscv32_smode_defconfig and riscv32_spl_defconfig builds, and replace ``qemu-system-riscv64`` with ``qemu-system-riscv32`` in the command lines above to boot the 32-bit U-Boot. + +Enabling TPM +------------ + +A TPM device can be connected to the virt board by following the steps below. + +First launch the TPM emulator + + swtpm socket --tpm2 -t -d --tpmstate dir=/tmp/tpm \ + --ctrl type=unixio,path=swtpm-sock + +Then launch QEMU with: + + ... + -chardev socket,id=chrtpm,path=swtpm-sock \ + -tpmdev emulator,id=tpm0,chardev=chrtpm \ + -device tpm-tis-device,tpmdev=tpm0 + +The TPM device can be seen in the memory tree and the generated device +tree and should be accessible from the guest software. 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 "sysemu/device_tree.h" #include "sysemu/sysemu.h" #include "sysemu/kvm.h" +#include "sysemu/tpm.h" #include "hw/pci/pci.h" #include "hw/pci-host/gpex.h" #include "hw/display/ramfb.h" @@ -XXX,XX +XXX,XX @@ static void virt_machine_class_init(ObjectClass *oc, void *data) hc->plug = virt_machine_device_plug_cb; machine_class_allow_dynamic_sysbus_dev(mc, TYPE_RAMFB_DEVICE); +#ifdef CONFIG_TPM + machine_class_allow_dynamic_sysbus_dev(mc, TYPE_TPM_TIS_SYSBUS); +#endif object_class_property_add_bool(oc, "aclint", virt_get_aclint, virt_set_aclint); diff --git a/hw/riscv/Kconfig b/hw/riscv/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/Kconfig +++ b/hw/riscv/Kconfig @@ -XXX,XX +XXX,XX @@ config RISCV_VIRT imply PCI_DEVICES imply VIRTIO_VGA imply TEST_DEVICES + imply TPM_TIS_SYSBUS select RISCV_NUMA select GOLDFISH_RTC select MSI_NONBROKEN -- 2.35.1 From: Alistair Francis <alistair.francis@wdc.com> The following changes since commit c77283dd5d79149f4e7e9edd00f65416c648ee59: Merge tag 'pull-request-2025-07-02' of https://gitlab.com/thuth/qemu into staging (2025-07-03 06:01:41 -0400) are available in the Git repository at: https://github.com/alistair23/qemu.git tags/pull-riscv-to-apply-20250704 for you to fetch changes up to dc8bffc4eb0a93d3266cea1b17f8848dea5b915c: target: riscv: Add Svrsw60t59b extension support (2025-07-04 21:09:49 +1000) ---------------------------------------------------------------- Second RISC-V PR for 10.1 * sstc extension fixes * Fix zama16b order in isa_edata_arr * Profile handling fixes * Extend PMP region up to 64 * Remove capital 'Z' CPU properties * Add missing named features * Support atomic instruction fetch (Ziccif) * Add max_satp_mode from host cpu * Extend and configure PMP region count * Fix PPN field of Translation-reponse register * Use qemu_chr_fe_write_all() in DBCN_CONSOLE_WRITE_BYTE * Fix fcvt.s.bf16 NaN box checking * Avoid infinite delay of async xmit function * Device tree reg cleanups * Add Kunminghu CPU and platform * Fix missing exit TB flow for ldff_trans * Fix migration failure when aia is configured as aplic-imsic * Fix MEPC/SEPC bit masking for IALIGN * Add a property to set vill bit on reserved usage of vsetvli instruction * Add Svrsw60t59b extension support ---------------------------------------------------------------- Alexandre Ghiti (1): target: riscv: Add Svrsw60t59b extension support Anton Blanchard (1): target/riscv: Fix fcvt.s.bf16 NaN box checking Charalampos Mitrodimas (2): target/riscv: Fix MEPC/SEPC bit masking for IALIGN tests/tcg/riscv64: Add test for MEPC bit masking Daniel Henrique Barboza (9): target/riscv/cpu.c: fix zama16b order in isa_edata_arr[] target/riscv/tcg: restrict satp_mode changes in cpu_set_profile target/riscv/tcg: decouple profile enablement from user prop target/riscv: add profile->present flag target/riscv: remove capital 'Z' CPU properties target/riscv/cpu.c: add 'sdtrig' in riscv,isa target/riscv/cpu.c: add 'ssstrict' to riscv, isa target/riscv/cpu.c: do better with 'named features' doc target/riscv: use qemu_chr_fe_write_all() in DBCN_CONSOLE_WRITE_BYTE Florian Lugou (1): hw/char: sifive_uart: Avoid infinite delay of async xmit function Huang Borong (2): target/riscv: Add BOSC's Xiangshan Kunminghu CPU hw/riscv: Initial support for BOSC's Xiangshan Kunminghu FPGA prototype Jay Chang (2): target/riscv: Extend PMP region up to 64 target/riscv: Make PMP region count configurable Jim Shu (5): target/riscv: Add the checking into stimecmp write function. hw/intc: riscv_aclint: Fix mtime write for sstc extension target/riscv: Fix VSTIP bit in sstc extension. target/riscv: Enable/Disable S/VS-mode Timer when STCE bit is changed target/riscv: support atomic instruction fetch (Ziccif) Joel Stanley (12): hw/riscv/virt: Fix clint base address type hw/riscv/virt: Use setprop_sized_cells for clint hw/riscv/virt: Use setprop_sized_cells for memory hw/riscv/virt: Use setprop_sized_cells for aplic hw/riscv/virt: Use setprop_sized_cells for aclint hw/riscv/virt: Use setprop_sized_cells for plic hw/riscv/virt: Use setprop_sized_cells for virtio hw/riscv/virt: Use setprop_sized_cells for reset hw/riscv/virt: Use setprop_sized_cells for uart hw/riscv/virt: Use setprop_sized_cells for rtc hw/riscv/virt: Use setprop_sized_cells for iommu hw/riscv/virt: Use setprop_sized_cells for pcie Max Chou (1): target/riscv: rvv: Fix missing exit TB flow for ldff_trans Meng Zhuo (1): target/riscv/kvm: add max_satp_mode from host cpu Nutty Liu (1): hw/riscv/riscv-iommu: Fix PPN field of Translation-reponse register Vasilis Liaskovitis (1): target/riscv: Add a property to set vill bit on reserved usage of vsetvli instruction liu.xuemei1@zte.com.cn (1): migration: Fix migration failure when aia is configured as aplic-imsic MAINTAINERS | 7 + docs/system/riscv/xiangshan-kunminghu.rst | 39 +++++ docs/system/target-riscv.rst | 1 + configs/devices/riscv64-softmmu/default.mak | 1 + hw/riscv/riscv-iommu-bits.h | 1 + include/hw/riscv/xiangshan_kmh.h | 68 +++++++++ target/riscv/cpu-qom.h | 1 + target/riscv/cpu.h | 19 ++- target/riscv/cpu_bits.h | 63 +++++++- target/riscv/helper.h | 2 +- target/riscv/internals.h | 27 ++++ target/riscv/time_helper.h | 1 + target/riscv/cpu_cfg_fields.h.inc | 3 + hw/char/sifive_uart.c | 6 +- hw/intc/riscv_aclint.c | 5 + hw/intc/riscv_aplic.c | 12 +- hw/intc/riscv_imsic.c | 10 +- hw/riscv/riscv-iommu.c | 9 +- hw/riscv/virt.c | 66 ++++----- hw/riscv/xiangshan_kmh.c | 220 ++++++++++++++++++++++++++++ target/riscv/cpu.c | 144 +++++++++++++++--- target/riscv/cpu_helper.c | 3 +- target/riscv/csr.c | 192 +++++++++++++++++++++++- target/riscv/fpu_helper.c | 2 +- target/riscv/kvm/kvm-cpu.c | 18 ++- target/riscv/machine.c | 3 +- target/riscv/op_helper.c | 4 +- target/riscv/pmp.c | 28 ++-- target/riscv/riscv-qmp-cmds.c | 2 +- target/riscv/tcg/tcg-cpu.c | 186 +++++++++++------------ target/riscv/time_helper.c | 65 +++++++- target/riscv/translate.c | 46 ++++-- target/riscv/vector_helper.c | 12 +- target/riscv/insn_trans/trans_rvv.c.inc | 10 +- hw/riscv/Kconfig | 9 ++ hw/riscv/meson.build | 1 + tests/data/acpi/riscv64/virt/RHCT | Bin 400 -> 416 bytes tests/tcg/riscv64/Makefile.softmmu-target | 4 + tests/tcg/riscv64/test-mepc-masking.S | 73 +++++++++ 39 files changed, 1151 insertions(+), 212 deletions(-) create mode 100644 docs/system/riscv/xiangshan-kunminghu.rst create mode 100644 include/hw/riscv/xiangshan_kmh.h create mode 100644 hw/riscv/xiangshan_kmh.c create mode 100644 tests/tcg/riscv64/test-mepc-masking.S From: Jim Shu <jim.shu@sifive.com> Preparation commit to let aclint timer to use stimecmp write function. Aclint timer doesn't call sstc() predicate so we need to check inside the stimecmp write function. Signed-off-by: Jim Shu <jim.shu@sifive.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250519143518.11086-2-jim.shu@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/time_helper.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) diff --git a/target/riscv/time_helper.c b/target/riscv/time_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/time_helper.c +++ b/target/riscv/time_helper.c @@ -XXX,XX +XXX,XX @@ void riscv_timer_write_timecmp(CPURISCVState *env, QEMUTimer *timer, { uint64_t diff, ns_diff, next; RISCVAclintMTimerState *mtimer = env->rdtime_fn_arg; - uint32_t timebase_freq = mtimer->timebase_freq; - uint64_t rtc_r = env->rdtime_fn(env->rdtime_fn_arg) + delta; + uint32_t timebase_freq; + uint64_t rtc_r; + + if (!riscv_cpu_cfg(env)->ext_sstc || !env->rdtime_fn || + !env->rdtime_fn_arg || !get_field(env->menvcfg, MENVCFG_STCE)) { + /* S/VS Timer IRQ depends on sstc extension, rdtime_fn(), and STCE. */ + return; + } + + if (timer_irq == MIP_VSTIP && + (!riscv_has_ext(env, RVH) || !get_field(env->henvcfg, HENVCFG_STCE))) { + /* VS Timer IRQ also depends on RVH and henvcfg.STCE. */ + return; + } + + timebase_freq = mtimer->timebase_freq; + rtc_r = env->rdtime_fn(env->rdtime_fn_arg) + delta; if (timecmp <= rtc_r) { /* -- 2.50.0 From: Jim Shu <jim.shu@sifive.com> When changing the mtime value, the period of [s|vs]timecmp timers should also be updated, similar to the period of mtimecmp timer. The period of the stimecmp timer is the time until the next S-mode timer IRQ. The value is calculated as "stimecmp - time". [1] It is equal to "stimecmp - mtime" since the time CSR is a read-only shadow of the memory-mapped mtime register. Thus, changing mtime value will update the period of stimecmp timer. Similarly, the period of vstimecmp timer is calculated as "vstimecmp - (mtime + htimedelta)" [2], so changing mtime value will update the period of vstimecmp timer. [1] RISC-V Priv spec ch 9.1.1. Supervisor Timer (stimecmp) Register A supervisor timer interrupt becomes pending, as reflected in the STIP bit in the mip and sip registers whenever time contains a value greater than or equal to stimecmp. [2] RISC-V Priv spec ch19.2.1. Virtual Supervisor Timer (vstimecmp) Register A virtual supervisor timer interrupt becomes pending, as reflected in the VSTIP bit in the hip register, whenever (time + htimedelta), truncated to 64 bits, contains a value greater than or equal to vstimecmp Signed-off-by: Jim Shu <jim.shu@sifive.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250519143518.11086-3-jim.shu@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/intc/riscv_aclint.c | 5 +++++ 1 file changed, 5 insertions(+) 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 @@ #include "qemu/module.h" #include "hw/sysbus.h" #include "target/riscv/cpu.h" +#include "target/riscv/time_helper.h" #include "hw/qdev-properties.h" #include "hw/intc/riscv_aclint.h" #include "qemu/timer.h" @@ -XXX,XX +XXX,XX @@ static void riscv_aclint_mtimer_write(void *opaque, hwaddr addr, riscv_aclint_mtimer_write_timecmp(mtimer, RISCV_CPU(cpu), mtimer->hartid_base + i, mtimer->timecmp[i]); + riscv_timer_write_timecmp(env, env->stimer, env->stimecmp, 0, MIP_STIP); + riscv_timer_write_timecmp(env, env->vstimer, env->vstimecmp, + env->htimedelta, MIP_VSTIP); + } return; } -- 2.50.0 From: Jim Shu <jim.shu@sifive.com> VSTIP is only writable when both [mh]envcfg.STCE is enabled, or it will revert it's defined behavior as if sstc extension is not implemented. Signed-off-by: Jim Shu <jim.shu@sifive.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250519143518.11086-4-jim.shu@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/csr.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) 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 rmw_mip64(CPURISCVState *env, int csrno, if (riscv_cpu_cfg(env)->ext_sstc && (env->priv == PRV_M) && get_field(env->menvcfg, MENVCFG_STCE)) { /* sstc extension forbids STIP & VSTIP to be writeable in mip */ - mask = mask & ~(MIP_STIP | MIP_VSTIP); + + /* STIP is not writable when menvcfg.STCE is enabled. */ + mask = mask & ~MIP_STIP; + + /* VSTIP is not writable when both [mh]envcfg.STCE are enabled. */ + if (get_field(env->henvcfg, HENVCFG_STCE)) { + mask = mask & ~MIP_VSTIP; + } } if (mask) { -- 2.50.0 From: Jim Shu <jim.shu@sifive.com> Updating STCE will enable/disable SSTC in S-mode or/and VS-mode, so we also need to update S/VS-mode Timer and S/VSTIP bits in $mip CSR. Signed-off-by: Jim Shu <jim.shu@sifive.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250519143518.11086-5-jim.shu@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/time_helper.h | 1 + target/riscv/csr.c | 46 ++++++++++++++++++++++++++++++++++++++ target/riscv/time_helper.c | 46 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 93 insertions(+) diff --git a/target/riscv/time_helper.h b/target/riscv/time_helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/time_helper.h +++ b/target/riscv/time_helper.h @@ -XXX,XX +XXX,XX @@ void riscv_timer_write_timecmp(CPURISCVState *env, QEMUTimer *timer, uint64_t timecmp, uint64_t delta, uint32_t timer_irq); +void riscv_timer_stce_changed(CPURISCVState *env, bool is_m_mode, bool enable); void riscv_timer_init(RISCVCPU *cpu); #endif 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, const RISCVCPUConfig *cfg = riscv_cpu_cfg(env); uint64_t mask = MENVCFG_FIOM | MENVCFG_CBIE | MENVCFG_CBCFE | MENVCFG_CBZE | MENVCFG_CDE; + bool stce_changed = false; if (riscv_cpu_mxl(env) == MXL_RV64) { mask |= (cfg->ext_svpbmt ? MENVCFG_PBMTE : 0) | @@ -XXX,XX +XXX,XX @@ static RISCVException write_menvcfg(CPURISCVState *env, int csrno, if ((val & MENVCFG_DTE) == 0) { env->mstatus &= ~MSTATUS_SDT; } + + if (cfg->ext_sstc && + ((env->menvcfg & MENVCFG_STCE) != (val & MENVCFG_STCE))) { + stce_changed = true; + } } env->menvcfg = (env->menvcfg & ~mask) | (val & mask); + + if (stce_changed) { + riscv_timer_stce_changed(env, true, !!(val & MENVCFG_STCE)); + } + return write_henvcfg(env, CSR_HENVCFG, env->henvcfg, ra); } @@ -XXX,XX +XXX,XX @@ static RISCVException write_menvcfgh(CPURISCVState *env, int csrno, (cfg->ext_smcdeleg ? MENVCFG_CDE : 0) | (cfg->ext_ssdbltrp ? MENVCFG_DTE : 0); uint64_t valh = (uint64_t)val << 32; + bool stce_changed = false; + + if (cfg->ext_sstc && + ((env->menvcfg & MENVCFG_STCE) != (valh & MENVCFG_STCE))) { + stce_changed = true; + } if ((valh & MENVCFG_DTE) == 0) { env->mstatus &= ~MSTATUS_SDT; } env->menvcfg = (env->menvcfg & ~mask) | (valh & mask); + + if (stce_changed) { + riscv_timer_stce_changed(env, true, !!(valh & MENVCFG_STCE)); + } + return write_henvcfgh(env, CSR_HENVCFGH, env->henvcfg >> 32, ra); } @@ -XXX,XX +XXX,XX @@ static RISCVException read_henvcfg(CPURISCVState *env, int csrno, static RISCVException write_henvcfg(CPURISCVState *env, int csrno, target_ulong val, uintptr_t ra) { + const RISCVCPUConfig *cfg = riscv_cpu_cfg(env); uint64_t mask = HENVCFG_FIOM | HENVCFG_CBIE | HENVCFG_CBCFE | HENVCFG_CBZE; RISCVException ret; + bool stce_changed = false; ret = smstateen_acc_ok(env, 0, SMSTATEEN0_HSENVCFG); if (ret != RISCV_EXCP_NONE) { @@ -XXX,XX +XXX,XX @@ static RISCVException write_henvcfg(CPURISCVState *env, int csrno, get_field(val, HENVCFG_PMM) != PMM_FIELD_RESERVED) { mask |= HENVCFG_PMM; } + + if (cfg->ext_sstc && + ((env->henvcfg & HENVCFG_STCE) != (val & HENVCFG_STCE))) { + stce_changed = true; + } } env->henvcfg = val & mask; @@ -XXX,XX +XXX,XX @@ static RISCVException write_henvcfg(CPURISCVState *env, int csrno, env->vsstatus &= ~MSTATUS_SDT; } + if (stce_changed) { + riscv_timer_stce_changed(env, false, !!(val & HENVCFG_STCE)); + } + return RISCV_EXCP_NONE; } @@ -XXX,XX +XXX,XX @@ static RISCVException read_henvcfgh(CPURISCVState *env, int csrno, static RISCVException write_henvcfgh(CPURISCVState *env, int csrno, target_ulong val, uintptr_t ra) { + const RISCVCPUConfig *cfg = riscv_cpu_cfg(env); uint64_t mask = env->menvcfg & (HENVCFG_PBMTE | HENVCFG_STCE | HENVCFG_ADUE | HENVCFG_DTE); uint64_t valh = (uint64_t)val << 32; RISCVException ret; + bool stce_changed = false; ret = smstateen_acc_ok(env, 0, SMSTATEEN0_HSENVCFG); if (ret != RISCV_EXCP_NONE) { return ret; } + + if (cfg->ext_sstc && + ((env->henvcfg & HENVCFG_STCE) != (valh & HENVCFG_STCE))) { + stce_changed = true; + } + env->henvcfg = (env->henvcfg & 0xFFFFFFFF) | (valh & mask); if ((env->henvcfg & HENVCFG_DTE) == 0) { env->vsstatus &= ~MSTATUS_SDT; } + + if (stce_changed) { + riscv_timer_stce_changed(env, false, !!(val & HENVCFG_STCE)); + } + return RISCV_EXCP_NONE; } diff --git a/target/riscv/time_helper.c b/target/riscv/time_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/time_helper.c +++ b/target/riscv/time_helper.c @@ -XXX,XX +XXX,XX @@ void riscv_timer_write_timecmp(CPURISCVState *env, QEMUTimer *timer, timer_mod(timer, next); } +/* + * When disabling xenvcfg.STCE, the S/VS Timer may be disabled at the same time. + * It is safe to call this function regardless of whether the timer has been + * deleted or not. timer_del() will do nothing if the timer has already + * been deleted. + */ +static void riscv_timer_disable_timecmp(CPURISCVState *env, QEMUTimer *timer, + uint32_t timer_irq) +{ + /* Disable S-mode Timer IRQ and HW-based STIP */ + if ((timer_irq == MIP_STIP) && !get_field(env->menvcfg, MENVCFG_STCE)) { + riscv_cpu_update_mip(env, timer_irq, BOOL_TO_MASK(0)); + timer_del(timer); + return; + } + + /* Disable VS-mode Timer IRQ and HW-based VSTIP */ + if ((timer_irq == MIP_VSTIP) && + (!get_field(env->menvcfg, MENVCFG_STCE) || + !get_field(env->henvcfg, HENVCFG_STCE))) { + env->vstime_irq = 0; + riscv_cpu_update_mip(env, 0, BOOL_TO_MASK(0)); + timer_del(timer); + return; + } +} + +/* Enable or disable S/VS-mode Timer when xenvcfg.STCE is changed */ +void riscv_timer_stce_changed(CPURISCVState *env, bool is_m_mode, bool enable) +{ + if (enable) { + riscv_timer_write_timecmp(env, env->vstimer, env->vstimecmp, + env->htimedelta, MIP_VSTIP); + } else { + riscv_timer_disable_timecmp(env, env->vstimer, MIP_VSTIP); + } + + if (is_m_mode) { + if (enable) { + riscv_timer_write_timecmp(env, env->stimer, env->stimecmp, 0, MIP_STIP); + } else { + riscv_timer_disable_timecmp(env, env->stimer, MIP_STIP); + } + } +} + void riscv_timer_init(RISCVCPU *cpu) { CPURISCVState *env; -- 2.50.0 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Put it after zalrsc and before zawrs. Cc: qemu-trivial@nongnu.org Fixes: a60ce58fd9 ("target/riscv: Support Zama16b extension") Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250522113344.823294-1-dbarboza@ventanamicro.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 @@ const RISCVIsaExtData isa_edata_arr[] = { ISA_EXT_DATA_ENTRY(zaamo, PRIV_VERSION_1_12_0, ext_zaamo), ISA_EXT_DATA_ENTRY(zabha, PRIV_VERSION_1_13_0, ext_zabha), ISA_EXT_DATA_ENTRY(zacas, PRIV_VERSION_1_12_0, ext_zacas), - ISA_EXT_DATA_ENTRY(zama16b, PRIV_VERSION_1_13_0, ext_zama16b), ISA_EXT_DATA_ENTRY(zalrsc, PRIV_VERSION_1_12_0, ext_zalrsc), + ISA_EXT_DATA_ENTRY(zama16b, PRIV_VERSION_1_13_0, ext_zama16b), 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), -- 2.50.0 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> We're changing 'mmu' to true regardless of whether the profile is being enabled or not, and at the same time we're changing satp_mode to profile->enabled. This will promote a situation where we'll set mmu=on without a virtual memory mode, which is a mistake. Only touch 'mmu' and satp_mode if the profile is being enabled. Suggested-by: Andrew Jones <ajones@ventanamicro.com> Fixes: 55398025e7 ("target/riscv: add satp_mode profile support") Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Andrew Jones <ajones@ventanamicro.com> Reviewed-by: Björn Töpel <bjorn@rivosinc.com> Tested-by: Björn Töpel <bjorn@rivosinc.com> Message-ID: <20250528184407.1451983-2-dbarboza@ventanamicro.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/tcg/tcg-cpu.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/target/riscv/tcg/tcg-cpu.c b/target/riscv/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/tcg/tcg-cpu.c +++ b/target/riscv/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static void cpu_set_profile(Object *obj, Visitor *v, const char *name, if (profile->enabled) { cpu->env.priv_ver = profile->priv_spec; - } #ifndef CONFIG_USER_ONLY - if (profile->satp_mode != RISCV_PROFILE_ATTR_UNUSED) { - object_property_set_bool(obj, "mmu", true, NULL); - const char *satp_prop = satp_mode_str(profile->satp_mode, - riscv_cpu_is_32bit(cpu)); - object_property_set_bool(obj, satp_prop, profile->enabled, NULL); - } + if (profile->satp_mode != RISCV_PROFILE_ATTR_UNUSED) { + object_property_set_bool(obj, "mmu", true, NULL); + const char *satp_prop = satp_mode_str(profile->satp_mode, + riscv_cpu_is_32bit(cpu)); + object_property_set_bool(obj, satp_prop, true, NULL); + } #endif + } for (i = 0; misa_bits[i] != 0; i++) { uint32_t bit = misa_bits[i]; -- 2.50.0 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> We have code in riscv_cpu_add_profiles() to enable a profile right away in case a CPU chose the profile during its cpu_init(). But we're using the user callback option to do so, setting profile->user_set. Create a new helper that does all the grunt work to enable/disable a given profile. Use this new helper in the cases where we want a CPU to be compatible to a certain profile, leaving the user callback to be used exclusively by users. Fixes: fba92a92e3 ("target/riscv: add 'rva22u64' CPU") Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Andrew Jones <ajones@ventanamicro.com> Reviewed-by: Björn Töpel <bjorn@rivosinc.com> Tested-by: Björn Töpel <bjorn@rivosinc.com> Message-ID: <20250528184407.1451983-3-dbarboza@ventanamicro.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/tcg/tcg-cpu.c | 127 +++++++++++++++++++------------------ 1 file changed, 67 insertions(+), 60 deletions(-) diff --git a/target/riscv/tcg/tcg-cpu.c b/target/riscv/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/tcg/tcg-cpu.c +++ b/target/riscv/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static bool riscv_cpu_is_generic(Object *cpu_obj) return object_dynamic_cast(cpu_obj, TYPE_RISCV_DYNAMIC_CPU) != NULL; } +static void riscv_cpu_set_profile(RISCVCPU *cpu, + RISCVCPUProfile *profile, + bool enabled) +{ + int i, ext_offset; + + if (profile->u_parent != NULL) { + riscv_cpu_set_profile(cpu, profile->u_parent, enabled); + } + + if (profile->s_parent != NULL) { + riscv_cpu_set_profile(cpu, profile->s_parent, enabled); + } + + profile->enabled = enabled; + + if (profile->enabled) { + cpu->env.priv_ver = profile->priv_spec; + +#ifndef CONFIG_USER_ONLY + if (profile->satp_mode != RISCV_PROFILE_ATTR_UNUSED) { + object_property_set_bool(OBJECT(cpu), "mmu", true, NULL); + const char *satp_prop = satp_mode_str(profile->satp_mode, + riscv_cpu_is_32bit(cpu)); + object_property_set_bool(OBJECT(cpu), satp_prop, true, NULL); + } +#endif + } + + for (i = 0; misa_bits[i] != 0; i++) { + uint32_t bit = misa_bits[i]; + + if (!(profile->misa_ext & bit)) { + continue; + } + + if (bit == RVI && !profile->enabled) { + /* + * Disabling profiles will not disable the base + * ISA RV64I. + */ + continue; + } + + cpu_misa_ext_add_user_opt(bit, profile->enabled); + riscv_cpu_write_misa_bit(cpu, bit, profile->enabled); + } + + for (i = 0; profile->ext_offsets[i] != RISCV_PROFILE_EXT_LIST_END; i++) { + ext_offset = profile->ext_offsets[i]; + + if (profile->enabled) { + if (cpu_cfg_offset_is_named_feat(ext_offset)) { + riscv_cpu_enable_named_feat(cpu, ext_offset); + } + + cpu_bump_multi_ext_priv_ver(&cpu->env, ext_offset); + } + + cpu_cfg_ext_add_user_opt(ext_offset, profile->enabled); + isa_ext_update_enabled(cpu, ext_offset, profile->enabled); + } +} + /* * We'll get here via the following path: * @@ -XXX,XX +XXX,XX @@ static void cpu_set_profile(Object *obj, Visitor *v, const char *name, RISCVCPUProfile *profile = opaque; RISCVCPU *cpu = RISCV_CPU(obj); bool value; - int i, ext_offset; if (riscv_cpu_is_vendor(obj)) { error_setg(errp, "Profile %s is not available for vendor CPUs", @@ -XXX,XX +XXX,XX @@ static void cpu_set_profile(Object *obj, Visitor *v, const char *name, } profile->user_set = true; - profile->enabled = value; - - if (profile->u_parent != NULL) { - object_property_set_bool(obj, profile->u_parent->name, - profile->enabled, NULL); - } - - if (profile->s_parent != NULL) { - object_property_set_bool(obj, profile->s_parent->name, - profile->enabled, NULL); - } - - if (profile->enabled) { - cpu->env.priv_ver = profile->priv_spec; - -#ifndef CONFIG_USER_ONLY - if (profile->satp_mode != RISCV_PROFILE_ATTR_UNUSED) { - object_property_set_bool(obj, "mmu", true, NULL); - const char *satp_prop = satp_mode_str(profile->satp_mode, - riscv_cpu_is_32bit(cpu)); - object_property_set_bool(obj, satp_prop, true, NULL); - } -#endif - } - - for (i = 0; misa_bits[i] != 0; i++) { - uint32_t bit = misa_bits[i]; - - if (!(profile->misa_ext & bit)) { - continue; - } - if (bit == RVI && !profile->enabled) { - /* - * Disabling profiles will not disable the base - * ISA RV64I. - */ - continue; - } - - cpu_misa_ext_add_user_opt(bit, profile->enabled); - riscv_cpu_write_misa_bit(cpu, bit, profile->enabled); - } - - for (i = 0; profile->ext_offsets[i] != RISCV_PROFILE_EXT_LIST_END; i++) { - ext_offset = profile->ext_offsets[i]; - - if (profile->enabled) { - if (cpu_cfg_offset_is_named_feat(ext_offset)) { - riscv_cpu_enable_named_feat(cpu, ext_offset); - } - - cpu_bump_multi_ext_priv_ver(&cpu->env, ext_offset); - } - - cpu_cfg_ext_add_user_opt(ext_offset, profile->enabled); - isa_ext_update_enabled(cpu, ext_offset, profile->enabled); - } + riscv_cpu_set_profile(cpu, profile, value); } static void cpu_get_profile(Object *obj, Visitor *v, const char *name, @@ -XXX,XX +XXX,XX @@ static void cpu_get_profile(Object *obj, Visitor *v, const char *name, static void riscv_cpu_add_profiles(Object *cpu_obj) { for (int i = 0; riscv_profiles[i] != NULL; i++) { - const RISCVCPUProfile *profile = riscv_profiles[i]; + RISCVCPUProfile *profile = riscv_profiles[i]; object_property_add(cpu_obj, profile->name, "bool", cpu_get_profile, cpu_set_profile, @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_add_profiles(Object *cpu_obj) * case. */ if (profile->enabled) { - object_property_set_bool(cpu_obj, profile->name, true, NULL); + riscv_cpu_set_profile(RISCV_CPU(cpu_obj), profile, true); } } } -- 2.50.0 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Björn reported in [1] a case where a rv64 CPU is going through the profile code path to enable satp mode. In this case,the amount of extensions on top of the rv64 CPU made it compliant with the RVA22S64 profile during the validation of CPU 0. When the subsequent CPUs were initialized the static profile object has the 'enable' flag set, enabling the profile code path for those CPUs. This happens because we are initializing and realizing each CPU before going to the next, i.e. init and realize CPU0, then init and realize CPU1 and so on. If we change any persistent state during the validation of CPU N it will interfere with the init/realization of CPU N+1. We're using the 'enabled' profile flag to do two distinct things: inform cpu_init() that we want profile extensions to be enabled, and telling QMP that a profile is currently enabled in the CPU. We want to be flexible enough to recognize profile support for all CPUs that has the extension prerequisites, but we do not want to force the profile code path if a profile wasn't set too. Add a new 'present' flag for profiles that will coexist with the 'enabled' flag. Enabling a profile means "we want to switch on all its mandatory extensions". A profile is 'present' if we asserted during validation that the CPU has the needed prerequisites. This means that the case reported by Björn now results in RVA22S64.enabled=false and RVA22S64.present=true. QMP will recognize it as a RVA22 compliant CPU and we won't force the CPU into the profile path. [1] https://lore.kernel.org/qemu-riscv/87y0usiz22.fsf@all.your.base.are.belong.to.us/ Reported-by: Björn Töpel <bjorn@kernel.org> Fixes: 2af005d610 ("target/riscv/tcg: validate profiles during finalize") Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Andrew Jones <ajones@ventanamicro.com> Reviewed-by: Björn Töpel <bjorn@rivosinc.com> Tested-by: Björn Töpel <bjorn@rivosinc.com> Message-ID: <20250528184407.1451983-4-dbarboza@ventanamicro.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 15 +++++++++++++++ target/riscv/riscv-qmp-cmds.c | 2 +- target/riscv/tcg/tcg-cpu.c | 11 +++-------- 3 files changed, 19 insertions(+), 9 deletions(-) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -XXX,XX +XXX,XX @@ typedef struct riscv_cpu_profile { struct riscv_cpu_profile *s_parent; const char *name; uint32_t misa_ext; + /* + * The profile is enabled/disabled via command line or + * via cpu_init(). Enabling a profile will add all its + * mandatory extensions in the CPU during init(). + */ bool enabled; + /* + * The profile is present in the CPU, i.e. the current set of + * CPU extensions complies with it. A profile can be enabled + * and not present (e.g. the user disabled a mandatory extension) + * and the other way around (e.g. all mandatory extensions are + * present in a non-profile CPU). + * + * QMP uses this flag. + */ + bool present; bool user_set; int priv_spec; int satp_mode; diff --git a/target/riscv/riscv-qmp-cmds.c b/target/riscv/riscv-qmp-cmds.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/riscv-qmp-cmds.c +++ b/target/riscv/riscv-qmp-cmds.c @@ -XXX,XX +XXX,XX @@ static void riscv_obj_add_profiles_qdict(Object *obj, QDict *qdict_out) for (int i = 0; riscv_profiles[i] != NULL; i++) { profile = riscv_profiles[i]; - value = QOBJECT(qbool_from_bool(profile->enabled)); + value = QOBJECT(qbool_from_bool(profile->present)); qdict_put_obj(qdict_out, profile->name, value); } diff --git a/target/riscv/tcg/tcg-cpu.c b/target/riscv/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/tcg/tcg-cpu.c +++ b/target/riscv/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_check_parent_profile(RISCVCPU *cpu, RISCVCPUProfile *profile, RISCVCPUProfile *parent) { - const char *parent_name; - bool parent_enabled; - - if (!profile->enabled || !parent) { + if (!profile->present || !parent) { return; } - parent_name = parent->name; - parent_enabled = object_property_get_bool(OBJECT(cpu), parent_name, NULL); - profile->enabled = parent_enabled; + profile->present = parent->present; } static void riscv_cpu_validate_profile(RISCVCPU *cpu, @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_validate_profile(RISCVCPU *cpu, } } - profile->enabled = profile_impl; + profile->present = profile_impl; riscv_cpu_check_parent_profile(cpu, profile, profile->u_parent); riscv_cpu_check_parent_profile(cpu, profile, profile->s_parent); -- 2.50.0 From: Jay Chang <jay.chang@sifive.com> According to the RISC-V Privileged Specification (version >1.12), RV32 supports 16 CSRs (pmpcfg0–pmpcfg15) to configure 64 PMP regions (pmpaddr0–pmpaddr63). Signed-off-by: Jay Chang <jay.chang@sifive.com> Reviewed-by: Frank Chang <frank.chang@sifive.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-ID: <20250522081236.4050-2-jay.chang@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu_bits.h | 60 +++++++++++++++++++ target/riscv/csr.c | 124 +++++++++++++++++++++++++++++++++++++++- 2 files changed, 182 insertions(+), 2 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 @@ #define CSR_PMPCFG1 0x3a1 #define CSR_PMPCFG2 0x3a2 #define CSR_PMPCFG3 0x3a3 +#define CSR_PMPCFG4 0x3a4 +#define CSR_PMPCFG5 0x3a5 +#define CSR_PMPCFG6 0x3a6 +#define CSR_PMPCFG7 0x3a7 +#define CSR_PMPCFG8 0x3a8 +#define CSR_PMPCFG9 0x3a9 +#define CSR_PMPCFG10 0x3aa +#define CSR_PMPCFG11 0x3ab +#define CSR_PMPCFG12 0x3ac +#define CSR_PMPCFG13 0x3ad +#define CSR_PMPCFG14 0x3ae +#define CSR_PMPCFG15 0x3af #define CSR_PMPADDR0 0x3b0 #define CSR_PMPADDR1 0x3b1 #define CSR_PMPADDR2 0x3b2 @@ -XXX,XX +XXX,XX @@ #define CSR_PMPADDR13 0x3bd #define CSR_PMPADDR14 0x3be #define CSR_PMPADDR15 0x3bf +#define CSR_PMPADDR16 0x3c0 +#define CSR_PMPADDR17 0x3c1 +#define CSR_PMPADDR18 0x3c2 +#define CSR_PMPADDR19 0x3c3 +#define CSR_PMPADDR20 0x3c4 +#define CSR_PMPADDR21 0x3c5 +#define CSR_PMPADDR22 0x3c6 +#define CSR_PMPADDR23 0x3c7 +#define CSR_PMPADDR24 0x3c8 +#define CSR_PMPADDR25 0x3c9 +#define CSR_PMPADDR26 0x3ca +#define CSR_PMPADDR27 0x3cb +#define CSR_PMPADDR28 0x3cc +#define CSR_PMPADDR29 0x3cd +#define CSR_PMPADDR30 0x3ce +#define CSR_PMPADDR31 0x3cf +#define CSR_PMPADDR32 0x3d0 +#define CSR_PMPADDR33 0x3d1 +#define CSR_PMPADDR34 0x3d2 +#define CSR_PMPADDR35 0x3d3 +#define CSR_PMPADDR36 0x3d4 +#define CSR_PMPADDR37 0x3d5 +#define CSR_PMPADDR38 0x3d6 +#define CSR_PMPADDR39 0x3d7 +#define CSR_PMPADDR40 0x3d8 +#define CSR_PMPADDR41 0x3d9 +#define CSR_PMPADDR42 0x3da +#define CSR_PMPADDR43 0x3db +#define CSR_PMPADDR44 0x3dc +#define CSR_PMPADDR45 0x3dd +#define CSR_PMPADDR46 0x3de +#define CSR_PMPADDR47 0x3df +#define CSR_PMPADDR48 0x3e0 +#define CSR_PMPADDR49 0x3e1 +#define CSR_PMPADDR50 0x3e2 +#define CSR_PMPADDR51 0x3e3 +#define CSR_PMPADDR52 0x3e4 +#define CSR_PMPADDR53 0x3e5 +#define CSR_PMPADDR54 0x3e6 +#define CSR_PMPADDR55 0x3e7 +#define CSR_PMPADDR56 0x3e8 +#define CSR_PMPADDR57 0x3e9 +#define CSR_PMPADDR58 0x3ea +#define CSR_PMPADDR59 0x3eb +#define CSR_PMPADDR60 0x3ec +#define CSR_PMPADDR61 0x3ed +#define CSR_PMPADDR62 0x3ee +#define CSR_PMPADDR63 0x3ef /* RNMI */ #define CSR_MNSCRATCH 0x740 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 @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { [CSR_PMPCFG1] = { "pmpcfg1", pmp, read_pmpcfg, write_pmpcfg }, [CSR_PMPCFG2] = { "pmpcfg2", pmp, read_pmpcfg, write_pmpcfg }, [CSR_PMPCFG3] = { "pmpcfg3", pmp, read_pmpcfg, write_pmpcfg }, + [CSR_PMPCFG4] = { "pmpcfg4", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG5] = { "pmpcfg5", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG6] = { "pmpcfg6", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG7] = { "pmpcfg7", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG8] = { "pmpcfg8", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG9] = { "pmpcfg9", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG10] = { "pmpcfg10", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG11] = { "pmpcfg11", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG12] = { "pmpcfg12", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG13] = { "pmpcfg13", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG14] = { "pmpcfg14", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPCFG15] = { "pmpcfg15", pmp, read_pmpcfg, write_pmpcfg, + .min_priv_ver = PRIV_VERSION_1_12_0 }, [CSR_PMPADDR0] = { "pmpaddr0", pmp, read_pmpaddr, write_pmpaddr }, [CSR_PMPADDR1] = { "pmpaddr1", pmp, read_pmpaddr, write_pmpaddr }, [CSR_PMPADDR2] = { "pmpaddr2", pmp, read_pmpaddr, write_pmpaddr }, @@ -XXX,XX +XXX,XX @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { [CSR_PMPADDR11] = { "pmpaddr11", pmp, read_pmpaddr, write_pmpaddr }, [CSR_PMPADDR12] = { "pmpaddr12", pmp, read_pmpaddr, write_pmpaddr }, [CSR_PMPADDR13] = { "pmpaddr13", pmp, read_pmpaddr, write_pmpaddr }, - [CSR_PMPADDR14] = { "pmpaddr14", pmp, read_pmpaddr, write_pmpaddr }, - [CSR_PMPADDR15] = { "pmpaddr15", pmp, read_pmpaddr, write_pmpaddr }, + [CSR_PMPADDR14] = { "pmpaddr14", pmp, read_pmpaddr, write_pmpaddr }, + [CSR_PMPADDR15] = { "pmpaddr15", pmp, read_pmpaddr, write_pmpaddr }, + [CSR_PMPADDR16] = { "pmpaddr16", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR17] = { "pmpaddr17", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR18] = { "pmpaddr18", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR19] = { "pmpaddr19", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR20] = { "pmpaddr20", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR21] = { "pmpaddr21", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR22] = { "pmpaddr22", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR23] = { "pmpaddr23", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR24] = { "pmpaddr24", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR25] = { "pmpaddr25", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR26] = { "pmpaddr26", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR27] = { "pmpaddr27", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR28] = { "pmpaddr28", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR29] = { "pmpaddr29", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR30] = { "pmpaddr30", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR31] = { "pmpaddr31", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR32] = { "pmpaddr32", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR33] = { "pmpaddr33", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR34] = { "pmpaddr34", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR35] = { "pmpaddr35", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR36] = { "pmpaddr36", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR37] = { "pmpaddr37", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR38] = { "pmpaddr38", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR39] = { "pmpaddr39", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR40] = { "pmpaddr40", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR41] = { "pmpaddr41", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR42] = { "pmpaddr42", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR43] = { "pmpaddr43", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR44] = { "pmpaddr44", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR45] = { "pmpaddr45", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR46] = { "pmpaddr46", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR47] = { "pmpaddr47", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR48] = { "pmpaddr48", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR49] = { "pmpaddr49", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR50] = { "pmpaddr50", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR51] = { "pmpaddr51", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR52] = { "pmpaddr52", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR53] = { "pmpaddr53", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR54] = { "pmpaddr54", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR55] = { "pmpaddr55", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR56] = { "pmpaddr56", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR57] = { "pmpaddr57", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR58] = { "pmpaddr58", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR59] = { "pmpaddr59", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR60] = { "pmpaddr60", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR61] = { "pmpaddr61", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR62] = { "pmpaddr62", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, + [CSR_PMPADDR63] = { "pmpaddr63", pmp, read_pmpaddr, write_pmpaddr, + .min_priv_ver = PRIV_VERSION_1_12_0 }, /* Debug CSRs */ [CSR_TSELECT] = { "tselect", debug, read_tselect, write_tselect }, -- 2.50.0 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> These properties were deprecated in QEMU 8.2, commit 8043effd9b. Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250530134608.1806922-1-dbarboza@ventanamicro.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 1 - target/riscv/cpu.c | 17 ----------------- target/riscv/tcg/tcg-cpu.c | 31 +------------------------------ 3 files changed, 1 insertion(+), 48 deletions(-) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -XXX,XX +XXX,XX @@ extern const RISCVCPUMultiExtConfig riscv_cpu_extensions[]; extern const RISCVCPUMultiExtConfig riscv_cpu_vendor_exts[]; extern const RISCVCPUMultiExtConfig riscv_cpu_experimental_exts[]; extern const RISCVCPUMultiExtConfig riscv_cpu_named_features[]; -extern const RISCVCPUMultiExtConfig riscv_cpu_deprecated_exts[]; typedef struct isa_ext_data { const char *name; 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 @@ const RISCVCPUMultiExtConfig riscv_cpu_named_features[] = { { }, }; -/* Deprecated entries marked for future removal */ -const RISCVCPUMultiExtConfig riscv_cpu_deprecated_exts[] = { - MULTI_EXT_CFG_BOOL("Zifencei", ext_zifencei, true), - MULTI_EXT_CFG_BOOL("Zicsr", ext_zicsr, true), - MULTI_EXT_CFG_BOOL("Zihintntl", ext_zihintntl, true), - MULTI_EXT_CFG_BOOL("Zihintpause", ext_zihintpause, true), - MULTI_EXT_CFG_BOOL("Zawrs", ext_zawrs, true), - MULTI_EXT_CFG_BOOL("Zfa", ext_zfa, true), - MULTI_EXT_CFG_BOOL("Zfh", ext_zfh, false), - MULTI_EXT_CFG_BOOL("Zfhmin", ext_zfhmin, false), - MULTI_EXT_CFG_BOOL("Zve32f", ext_zve32f, false), - MULTI_EXT_CFG_BOOL("Zve64f", ext_zve64f, false), - MULTI_EXT_CFG_BOOL("Zve64d", ext_zve64d, false), - - { }, -}; - static void cpu_set_prop_err(RISCVCPU *cpu, const char *propname, Error **errp) { diff --git a/target/riscv/tcg/tcg-cpu.c b/target/riscv/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/tcg/tcg-cpu.c +++ b/target/riscv/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_add_profiles(Object *cpu_obj) } } -static bool cpu_ext_is_deprecated(const char *ext_name) -{ - return isupper(ext_name[0]); -} - -/* - * String will be allocated in the heap. Caller is responsible - * for freeing it. - */ -static char *cpu_ext_to_lower(const char *ext_name) -{ - char *ret = g_malloc0(strlen(ext_name) + 1); - - strcpy(ret, ext_name); - ret[0] = tolower(ret[0]); - - return ret; -} - static void cpu_set_multi_ext_cfg(Object *obj, Visitor *v, const char *name, void *opaque, Error **errp) { @@ -XXX,XX +XXX,XX @@ static void cpu_set_multi_ext_cfg(Object *obj, Visitor *v, const char *name, return; } - if (cpu_ext_is_deprecated(multi_ext_cfg->name)) { - g_autofree char *lower = cpu_ext_to_lower(multi_ext_cfg->name); - - warn_report("CPU property '%s' is deprecated. Please use '%s' instead", - multi_ext_cfg->name, lower); - } - cpu_cfg_ext_add_user_opt(multi_ext_cfg->offset, value); prev_val = isa_ext_is_enabled(cpu, multi_ext_cfg->offset); @@ -XXX,XX +XXX,XX @@ static void cpu_add_multi_ext_prop(Object *cpu_obj, const RISCVCPUMultiExtConfig *multi_cfg) { bool generic_cpu = riscv_cpu_is_generic(cpu_obj); - bool deprecated_ext = cpu_ext_is_deprecated(multi_cfg->name); object_property_add(cpu_obj, multi_cfg->name, "bool", cpu_get_multi_ext_cfg, cpu_set_multi_ext_cfg, NULL, (void *)multi_cfg); - if (!generic_cpu || deprecated_ext) { + if (!generic_cpu) { return; } @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_add_user_properties(Object *obj) riscv_cpu_add_multiext_prop_array(obj, riscv_cpu_vendor_exts); riscv_cpu_add_multiext_prop_array(obj, riscv_cpu_experimental_exts); - riscv_cpu_add_multiext_prop_array(obj, riscv_cpu_deprecated_exts); - riscv_cpu_add_profiles(obj); } -- 2.50.0 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> We have support for sdtrig for awhile but we are not advertising it. It is enabled by default via the 'debug' flag. Use the same flag to also advertise sdtrig. Add an exception in disable_priv_spec_isa_exts() to avoid spamming warnings for 'sdtrig' for vendor CPUs like sifive_u. Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250604174329.1147549-2-dbarboza@ventanamicro.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.c | 1 + target/riscv/tcg/tcg-cpu.c | 9 +++++++++ tests/data/acpi/riscv64/virt/RHCT | Bin 400 -> 406 bytes 3 files changed, 10 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 @@ const RISCVIsaExtData isa_edata_arr[] = { ISA_EXT_DATA_ENTRY(zvkt, PRIV_VERSION_1_12_0, ext_zvkt), 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(sdtrig, PRIV_VERSION_1_12_0, debug), ISA_EXT_DATA_ENTRY(shcounterenw, PRIV_VERSION_1_12_0, has_priv_1_12), ISA_EXT_DATA_ENTRY(sha, PRIV_VERSION_1_12_0, ext_sha), ISA_EXT_DATA_ENTRY(shgatpa, PRIV_VERSION_1_12_0, has_priv_1_12), diff --git a/target/riscv/tcg/tcg-cpu.c b/target/riscv/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/tcg/tcg-cpu.c +++ b/target/riscv/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_disable_priv_spec_isa_exts(RISCVCPU *cpu) continue; } + /* + * cpu.debug = true is marked as 'sdtrig', priv spec 1.12. + * Skip this warning since existing CPUs with older priv + * spec and debug = true will be impacted. + */ + if (!strcmp(edata->name, "sdtrig")) { + continue; + } + isa_ext_update_enabled(cpu, edata->ext_enable_offset, false); /* diff --git a/tests/data/acpi/riscv64/virt/RHCT b/tests/data/acpi/riscv64/virt/RHCT index XXXXXXX..XXXXXXX 100644 Binary files a/tests/data/acpi/riscv64/virt/RHCT and b/tests/data/acpi/riscv64/virt/RHCT differ -- 2.50.0 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> 'ssstrict' is a RVA23 profile-defined extension defined as follows: "No non-conforming extensions are present. Attempts to execute unimplemented opcodes or access unimplemented CSRs in the standard or reserved encoding spaces raises an illegal instruction exception that results in a contained trap to the supervisor-mode trap handler." In short, we need to throw an exception when accessing unimplemented CSRs or opcodes. We do that, so let's advertise it. Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Andrew Jones <ajones@ventanamicro.com> Message-ID: <20250529202315.1684198-3-dbarboza@ventanamicro.com> Message-ID: <20250604174329.1147549-3-dbarboza@ventanamicro.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.c | 1 + tests/data/acpi/riscv64/virt/RHCT | Bin 406 -> 416 bytes 2 files 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 @@ const RISCVIsaExtData isa_edata_arr[] = { ISA_EXT_DATA_ENTRY(ssnpm, PRIV_VERSION_1_13_0, ext_ssnpm), ISA_EXT_DATA_ENTRY(sspm, PRIV_VERSION_1_13_0, ext_sspm), ISA_EXT_DATA_ENTRY(ssstateen, PRIV_VERSION_1_12_0, ext_ssstateen), + ISA_EXT_DATA_ENTRY(ssstrict, PRIV_VERSION_1_12_0, has_priv_1_12), ISA_EXT_DATA_ENTRY(sstc, PRIV_VERSION_1_12_0, ext_sstc), ISA_EXT_DATA_ENTRY(sstvala, PRIV_VERSION_1_12_0, has_priv_1_12), ISA_EXT_DATA_ENTRY(sstvecd, PRIV_VERSION_1_12_0, has_priv_1_12), diff --git a/tests/data/acpi/riscv64/virt/RHCT b/tests/data/acpi/riscv64/virt/RHCT index XXXXXXX..XXXXXXX 100644 Binary files a/tests/data/acpi/riscv64/virt/RHCT and b/tests/data/acpi/riscv64/virt/RHCT differ -- 2.50.0 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Most of the named features are added directly in isa_edata_arr[], some of them are also added in riscv_cpu_named_features(). There is a reason for that, and the existing docs can do better explaining it. Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-ID: <20250529202315.1684198-4-dbarboza@ventanamicro.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250604174329.1147549-4-dbarboza@ventanamicro.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 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 @@ const RISCVCPUMultiExtConfig riscv_cpu_experimental_exts[] = { * 'Named features' is the name we give to extensions that we * don't want to expose to users. They are either immutable * (always enabled/disable) or they'll vary depending on - * the resulting CPU state. They have riscv,isa strings - * and priv_ver like regular extensions. + * the resulting CPU state. + * + * Some of them are always enabled depending on priv version + * of the CPU and are declared directly in isa_edata_arr[]. + * The ones listed here have special checks during finalize() + * time and require their own flags like regular extensions. + * See riscv_cpu_update_named_features() for more info. */ const RISCVCPUMultiExtConfig riscv_cpu_named_features[] = { MULTI_EXT_CFG_BOOL("zic64b", ext_zic64b, true), MULTI_EXT_CFG_BOOL("ssstateen", ext_ssstateen, true), MULTI_EXT_CFG_BOOL("sha", ext_sha, true), + + /* + * 'ziccrse' has its own flag because the KVM driver + * wants to enable/disable it on its own accord. + */ MULTI_EXT_CFG_BOOL("ziccrse", ext_ziccrse, true), { }, -- 2.50.0 From: Jim Shu <jim.shu@sifive.com> Support 4-byte atomic instruction fetch when instruction is natural aligned. Current implementation is not atomic because it loads instruction twice for first and last 2 bytes. We load 4 bytes at once to keep the atomicity. This instruction preload method only applys when instruction is 4-byte aligned. If instruction is unaligned, it could be across pages so that preload will trigger additional page fault. We encounter this issue when doing pressure test of enabling & disabling Linux kernel ftrace. Ftrace with kernel preemption requires concurrent modification and execution of instruction, so non-atomic instruction fetch will cause the race condition. We may fetch the wrong instruction which is the mixing of 2 instructions. Also, RISC-V Profile wants to provide this feature by HW. RVA20U64 Ziccif protects the atomicity of instruction fetch when it is natural aligned. This commit depends on the atomic read support of translator_ld in the commit 6a9dfe1984b0c593fb0ddb52d4e70832e6201dd6. Signed-off-by: Jim Shu <jim.shu@sifive.com> Reviewed-by: Frank Chang <frank.chang@sifive.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250508094838.19394-1-jim.shu@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/translate.c | 46 +++++++++++++++++++++++++++++----------- 1 file changed, 34 insertions(+), 12 deletions(-) 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 @@ const RISCVDecoder decoder_table[] = { const size_t decoder_table_size = ARRAY_SIZE(decoder_table); -static void decode_opc(CPURISCVState *env, DisasContext *ctx, uint16_t opcode) +static void decode_opc(CPURISCVState *env, DisasContext *ctx) { + uint32_t opcode; + bool pc_is_4byte_align = ((ctx->base.pc_next % 4) == 0); + ctx->virt_inst_excp = false; - ctx->cur_insn_len = insn_len(opcode); + if (pc_is_4byte_align) { + /* + * Load 4 bytes at once to make instruction fetch atomically. + * + * Note: When pc is 4-byte aligned, 4-byte instruction wouldn't be + * across pages. We could preload 4 bytes instruction no matter + * real one is 2 or 4 bytes. Instruction preload wouldn't trigger + * additional page fault. + */ + opcode = translator_ldl(env, &ctx->base, ctx->base.pc_next); + } else { + /* + * For unaligned pc, instruction preload may trigger additional + * page fault so we only load 2 bytes here. + */ + opcode = (uint32_t) translator_lduw(env, &ctx->base, ctx->base.pc_next); + } + ctx->ol = ctx->xl; + + ctx->cur_insn_len = insn_len((uint16_t)opcode); /* Check for compressed insn */ if (ctx->cur_insn_len == 2) { - ctx->opcode = opcode; + ctx->opcode = (uint16_t)opcode; /* * The Zca extension is added as way to refer to instructions in the C * extension that do not include the floating-point loads and stores @@ -XXX,XX +XXX,XX @@ static void decode_opc(CPURISCVState *env, DisasContext *ctx, uint16_t opcode) return; } } else { - uint32_t opcode32 = opcode; - opcode32 = deposit32(opcode32, 16, 16, - translator_lduw(env, &ctx->base, - ctx->base.pc_next + 2)); - ctx->opcode = opcode32; + if (!pc_is_4byte_align) { + /* Load last 2 bytes of instruction here */ + opcode = deposit32(opcode, 16, 16, + translator_lduw(env, &ctx->base, + ctx->base.pc_next + 2)); + } + ctx->opcode = opcode; for (guint i = 0; i < ctx->decoders->len; ++i) { riscv_cpu_decode_fn func = g_ptr_array_index(ctx->decoders, i); - if (func(ctx, opcode32)) { + if (func(ctx, opcode)) { return; } } @@ -XXX,XX +XXX,XX @@ static void riscv_tr_translate_insn(DisasContextBase *dcbase, CPUState *cpu) { DisasContext *ctx = container_of(dcbase, DisasContext, base); CPURISCVState *env = cpu_env(cpu); - uint16_t opcode16 = translator_lduw(env, &ctx->base, ctx->base.pc_next); - ctx->ol = ctx->xl; - decode_opc(env, ctx, opcode16); + decode_opc(env, ctx); ctx->base.pc_next += ctx->cur_insn_len; /* -- 2.50.0 From: Meng Zhuo <mengzhuo@iscas.ac.cn> This patch adds max_satp_mode from host kvm cpu setting. Tested on: Milkv Megrez (Eswin 7700x) Reviewed-by: Andrew Jones <ajones@ventanamicro.com> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2931 Signed-off-by: Meng Zhuo <mengzhuo@iscas.ac.cn> Message-ID: <20250606034250.181707-1-mengzhuo@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/kvm/kvm-cpu.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/target/riscv/kvm/kvm-cpu.c b/target/riscv/kvm/kvm-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/kvm/kvm-cpu.c +++ b/target/riscv/kvm/kvm-cpu.c @@ -XXX,XX +XXX,XX @@ static void kvm_riscv_destroy_scratch_vcpu(KVMScratchCPU *scratch) close(scratch->kvmfd); } +static void kvm_riscv_init_max_satp_mode(RISCVCPU *cpu, KVMScratchCPU *kvmcpu) +{ + struct kvm_one_reg reg; + int ret; + + reg.id = RISCV_CONFIG_REG(satp_mode); + reg.addr = (uint64_t)&cpu->cfg.max_satp_mode; + ret = ioctl(kvmcpu->cpufd, KVM_GET_ONE_REG, &reg); + if (ret != 0) { + error_report("Unable to retrieve satp mode from host, error %d", ret); + } +} + static void kvm_riscv_init_machine_ids(RISCVCPU *cpu, KVMScratchCPU *kvmcpu) { struct kvm_one_reg reg; @@ -XXX,XX +XXX,XX @@ static void riscv_init_kvm_registers(Object *cpu_obj) kvm_riscv_init_machine_ids(cpu, &kvmcpu); kvm_riscv_init_misa_ext_mask(cpu, &kvmcpu); kvm_riscv_init_cfg(cpu, &kvmcpu); + kvm_riscv_init_max_satp_mode(cpu, &kvmcpu); kvm_riscv_destroy_scratch_vcpu(&kvmcpu); } @@ -XXX,XX +XXX,XX @@ static bool kvm_cpu_realize(CPUState *cs, Error **errp) } } - return true; + return true; } void riscv_kvm_cpu_finalize_features(RISCVCPU *cpu, Error **errp) -- 2.50.0 From: Jay Chang <jay.chang@sifive.com> Previously, the number of PMP regions was hardcoded to 16 in QEMU. This patch replaces the fixed value with a new `pmp_regions` field, allowing platforms to configure the number of PMP regions. If no specific value is provided, the default number of PMP regions remains 16 to preserve the existing behavior. A new CPU parameter num-pmp-regions has been introduced to the QEMU command line. For example: -cpu rv64, g=true, c=true, pmp=true, num-pmp-regions=8 Signed-off-by: Jay Chang <jay.chang@sifive.com> Reviewed-by: Frank Chang <frank.chang@sifive.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250606072525.17313-3-jay.chang@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 3 +- target/riscv/cpu_cfg_fields.h.inc | 1 + target/riscv/cpu.c | 48 +++++++++++++++++++++++++++++-- target/riscv/csr.c | 5 +++- target/riscv/machine.c | 3 +- target/riscv/pmp.c | 28 ++++++++++++------ 6 files changed, 74 insertions(+), 14 deletions(-) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -XXX,XX +XXX,XX @@ extern RISCVCPUImpliedExtsRule *riscv_multi_ext_implied_rules[]; #define MMU_USER_IDX 3 -#define MAX_RISCV_PMPS (16) +#define MAX_RISCV_PMPS (64) +#define OLD_MAX_RISCV_PMPS (16) #if !defined(CONFIG_USER_ONLY) #include "pmp.h" diff --git a/target/riscv/cpu_cfg_fields.h.inc b/target/riscv/cpu_cfg_fields.h.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu_cfg_fields.h.inc +++ b/target/riscv/cpu_cfg_fields.h.inc @@ -XXX,XX +XXX,XX @@ TYPED_FIELD(uint16_t, elen, 0) TYPED_FIELD(uint16_t, cbom_blocksize, 0) TYPED_FIELD(uint16_t, cbop_blocksize, 0) TYPED_FIELD(uint16_t, cboz_blocksize, 0) +TYPED_FIELD(uint8_t, pmp_regions, 0) TYPED_FIELD(int8_t, max_satp_mode, -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 void riscv_cpu_init(Object *obj) cpu->cfg.cbom_blocksize = 64; cpu->cfg.cbop_blocksize = 64; cpu->cfg.cboz_blocksize = 64; + cpu->cfg.pmp_regions = 16; cpu->env.vext_ver = VEXT_VERSION_1_00_0; cpu->cfg.max_satp_mode = -1; @@ -XXX,XX +XXX,XX @@ static const PropertyInfo prop_pmp = { .set = prop_pmp_set, }; +static void prop_num_pmp_regions_set(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + RISCVCPU *cpu = RISCV_CPU(obj); + uint8_t value; + + visit_type_uint8(v, name, &value, errp); + + if (cpu->cfg.pmp_regions != value && riscv_cpu_is_vendor(obj)) { + cpu_set_prop_err(cpu, name, errp); + return; + } + + if (cpu->env.priv_ver < PRIV_VERSION_1_12_0 && value > OLD_MAX_RISCV_PMPS) { + error_setg(errp, "Number of PMP regions exceeds maximum available"); + return; + } else if (value > MAX_RISCV_PMPS) { + error_setg(errp, "Number of PMP regions exceeds maximum available"); + return; + } + + cpu_option_add_user_setting(name, value); + cpu->cfg.pmp_regions = value; +} + +static void prop_num_pmp_regions_get(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + uint8_t value = RISCV_CPU(obj)->cfg.pmp_regions; + + visit_type_uint8(v, name, &value, errp); +} + +static const PropertyInfo prop_num_pmp_regions = { + .type = "uint8", + .description = "num-pmp-regions", + .get = prop_num_pmp_regions_get, + .set = prop_num_pmp_regions_set, +}; + static int priv_spec_from_str(const char *priv_spec_str) { int priv_version = -1; @@ -XXX,XX +XXX,XX @@ static const Property riscv_cpu_properties[] = { {.name = "mmu", .info = &prop_mmu}, {.name = "pmp", .info = &prop_pmp}, + {.name = "num-pmp-regions", .info = &prop_num_pmp_regions}, {.name = "priv_spec", .info = &prop_priv_spec}, {.name = "vext_spec", .info = &prop_vext_spec}, @@ -XXX,XX +XXX,XX @@ static const TypeInfo riscv_cpu_type_infos[] = { .cfg.max_satp_mode = VM_1_10_MBARE, .cfg.ext_zifencei = true, .cfg.ext_zicsr = true, - .cfg.pmp = true + .cfg.pmp = true, + .cfg.pmp_regions = 8 ), DEFINE_ABSTRACT_RISCV_CPU(TYPE_RISCV_CPU_SIFIVE_U, TYPE_RISCV_VENDOR_CPU, @@ -XXX,XX +XXX,XX @@ static const TypeInfo riscv_cpu_type_infos[] = { .cfg.ext_zifencei = true, .cfg.ext_zicsr = true, .cfg.mmu = true, - .cfg.pmp = true + .cfg.pmp = true, + .cfg.pmp_regions = 8 ), #if defined(TARGET_RISCV32) || \ 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 dbltrp_hmode(CPURISCVState *env, int csrno) static RISCVException pmp(CPURISCVState *env, int csrno) { if (riscv_cpu_cfg(env)->pmp) { - if (csrno <= CSR_PMPCFG3) { + int max_pmpcfg = (env->priv_ver >= PRIV_VERSION_1_12_0) ? ++ CSR_PMPCFG15 : CSR_PMPCFG3; + + if (csrno <= max_pmpcfg) { uint32_t reg_index = csrno - CSR_PMPCFG0; /* TODO: RV128 restriction check */ diff --git a/target/riscv/machine.c b/target/riscv/machine.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/machine.c +++ b/target/riscv/machine.c @@ -XXX,XX +XXX,XX @@ static int pmp_post_load(void *opaque, int version_id) RISCVCPU *cpu = opaque; CPURISCVState *env = &cpu->env; int i; + uint8_t pmp_regions = riscv_cpu_cfg(env)->pmp_regions; - for (i = 0; i < MAX_RISCV_PMPS; i++) { + for (i = 0; i < pmp_regions; i++) { pmp_update_rule_addr(env, i); } pmp_update_rule_nums(env); 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 @@ uint32_t pmp_get_num_rules(CPURISCVState *env) */ static inline uint8_t pmp_read_cfg(CPURISCVState *env, uint32_t pmp_index) { - if (pmp_index < MAX_RISCV_PMPS) { + uint8_t pmp_regions = riscv_cpu_cfg(env)->pmp_regions; + + if (pmp_index < pmp_regions) { return env->pmp_state.pmp[pmp_index].cfg_reg; } @@ -XXX,XX +XXX,XX @@ static inline uint8_t pmp_read_cfg(CPURISCVState *env, uint32_t pmp_index) */ static bool pmp_write_cfg(CPURISCVState *env, uint32_t pmp_index, uint8_t val) { - if (pmp_index < MAX_RISCV_PMPS) { + uint8_t pmp_regions = riscv_cpu_cfg(env)->pmp_regions; + + if (pmp_index < pmp_regions) { if (env->pmp_state.pmp[pmp_index].cfg_reg == val) { /* no change */ return false; @@ -XXX,XX +XXX,XX @@ void pmp_update_rule_addr(CPURISCVState *env, uint32_t pmp_index) void pmp_update_rule_nums(CPURISCVState *env) { int i; + uint8_t pmp_regions = riscv_cpu_cfg(env)->pmp_regions; env->pmp_state.num_rules = 0; - for (i = 0; i < MAX_RISCV_PMPS; i++) { + for (i = 0; i < pmp_regions; i++) { const uint8_t a_field = pmp_get_a_field(env->pmp_state.pmp[i].cfg_reg); if (PMP_AMATCH_OFF != a_field) { @@ -XXX,XX +XXX,XX @@ bool pmp_hart_has_privs(CPURISCVState *env, hwaddr addr, int pmp_size = 0; hwaddr s = 0; hwaddr e = 0; + uint8_t pmp_regions = riscv_cpu_cfg(env)->pmp_regions; /* Short cut if no rules */ if (0 == pmp_get_num_rules(env)) { @@ -XXX,XX +XXX,XX @@ bool pmp_hart_has_privs(CPURISCVState *env, hwaddr addr, * 1.10 draft priv spec states there is an implicit order * from low to high */ - for (i = 0; i < MAX_RISCV_PMPS; i++) { + for (i = 0; i < pmp_regions; i++) { s = pmp_is_in_range(env, i, addr); e = pmp_is_in_range(env, i, addr + pmp_size - 1); @@ -XXX,XX +XXX,XX @@ void pmpaddr_csr_write(CPURISCVState *env, uint32_t addr_index, { trace_pmpaddr_csr_write(env->mhartid, addr_index, val); bool is_next_cfg_tor = false; + uint8_t pmp_regions = riscv_cpu_cfg(env)->pmp_regions; - if (addr_index < MAX_RISCV_PMPS) { + if (addr_index < pmp_regions) { if (env->pmp_state.pmp[addr_index].addr_reg == val) { /* no change */ return; @@ -XXX,XX +XXX,XX @@ void pmpaddr_csr_write(CPURISCVState *env, uint32_t addr_index, * In TOR mode, need to check the lock bit of the next pmp * (if there is a next). */ - if (addr_index + 1 < MAX_RISCV_PMPS) { + if (addr_index + 1 < pmp_regions) { uint8_t pmp_cfg = env->pmp_state.pmp[addr_index + 1].cfg_reg; is_next_cfg_tor = PMP_AMATCH_TOR == pmp_get_a_field(pmp_cfg); @@ -XXX,XX +XXX,XX @@ void pmpaddr_csr_write(CPURISCVState *env, uint32_t addr_index, target_ulong pmpaddr_csr_read(CPURISCVState *env, uint32_t addr_index) { target_ulong val = 0; + uint8_t pmp_regions = riscv_cpu_cfg(env)->pmp_regions; - if (addr_index < MAX_RISCV_PMPS) { + if (addr_index < pmp_regions) { val = env->pmp_state.pmp[addr_index].addr_reg; trace_pmpaddr_csr_read(env->mhartid, addr_index, val); } else { @@ -XXX,XX +XXX,XX @@ void mseccfg_csr_write(CPURISCVState *env, target_ulong val) { int i; uint64_t mask = MSECCFG_MMWP | MSECCFG_MML; + uint8_t pmp_regions = riscv_cpu_cfg(env)->pmp_regions; /* Update PMM field only if the value is valid according to Zjpm v1.0 */ if (riscv_cpu_cfg(env)->ext_smmpm && riscv_cpu_mxl(env) == MXL_RV64 && @@ -XXX,XX +XXX,XX @@ void mseccfg_csr_write(CPURISCVState *env, target_ulong val) /* RLB cannot be enabled if it's already 0 and if any regions are locked */ if (!MSECCFG_RLB_ISSET(env)) { - for (i = 0; i < MAX_RISCV_PMPS; i++) { + for (i = 0; i < pmp_regions; i++) { if (pmp_is_locked(env, i)) { val &= ~MSECCFG_RLB; break; @@ -XXX,XX +XXX,XX @@ target_ulong pmp_get_tlb_size(CPURISCVState *env, hwaddr addr) hwaddr tlb_sa = addr & ~(TARGET_PAGE_SIZE - 1); hwaddr tlb_ea = tlb_sa + TARGET_PAGE_SIZE - 1; int i; + uint8_t pmp_regions = riscv_cpu_cfg(env)->pmp_regions; /* * If PMP is not supported or there are no PMP rules, the TLB page will not @@ -XXX,XX +XXX,XX @@ target_ulong pmp_get_tlb_size(CPURISCVState *env, hwaddr addr) return TARGET_PAGE_SIZE; } - for (i = 0; i < MAX_RISCV_PMPS; i++) { + for (i = 0; i < pmp_regions; i++) { if (pmp_get_a_field(env->pmp_state.pmp[i].cfg_reg) == PMP_AMATCH_OFF) { continue; } -- 2.50.0 From: Nutty Liu <liujingqi@lanxincomputing.com> The original implementation incorrectly performed a bitwise AND operation between the PPN of iova and PPN Mask, leading to an incorrect PPN field in Translation-reponse register. The PPN of iova should be set entirely in the PPN field of Translation-reponse register. Also remove the code that was used to clear S field since this field is already zero. Signed-off-by: Nutty Liu <liujingqi@lanxincomputing.com> Reviewed-by: Tomasz Jeznach <tjeznach@rivosinc.com> Message-ID: <20250605124848.1248-1-liujingqi@lanxincomputing.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/riscv-iommu.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/hw/riscv/riscv-iommu.c b/hw/riscv/riscv-iommu.c index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/riscv-iommu.c +++ b/hw/riscv/riscv-iommu.c @@ -XXX,XX +XXX,XX @@ static void riscv_iommu_process_dbg(RISCVIOMMUState *s) iova = RISCV_IOMMU_TR_RESPONSE_FAULT | (((uint64_t) fault) << 10); } else { iova = iotlb.translated_addr & ~iotlb.addr_mask; - iova >>= TARGET_PAGE_BITS; - iova &= RISCV_IOMMU_TR_RESPONSE_PPN; - - /* We do not support superpages (> 4kbs) for now */ - iova &= ~RISCV_IOMMU_TR_RESPONSE_S; + iova = set_field(0, RISCV_IOMMU_TR_RESPONSE_PPN, PPN_DOWN(iova)); } riscv_iommu_reg_set64(s, RISCV_IOMMU_REG_TR_RESPONSE, iova); } -- 2.50.0 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> The SBI spec states, for console write byte: "This is a blocking SBI call and it will only return after writing the specified byte to the debug console. It will also return, with SBI_ERR_FAILED, if there are I/O errors." Being a blocker call will either succeed writing the byte or error out, it's feasible to use the blocking qemu_chr_fe_write_all() instead of qemu_chr_fe_write(). Last but not the least, we will duck possible changes in qemu_chr_fe_write() where ret = 0 will have a 'zero byte written' semantic [1] - something that we're not ready to deal in this current state. [1] https://lore.kernel.org/qemu-devel/CAFEAcA_kEndvNtw4EHySXWwQPoGs029yAzZGGBcV=zGHaj7KUQ@mail.gmail.com/ Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-ID: <20250605094456.1385105-2-dbarboza@ventanamicro.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/kvm/kvm-cpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/target/riscv/kvm/kvm-cpu.c b/target/riscv/kvm/kvm-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/kvm/kvm-cpu.c +++ b/target/riscv/kvm/kvm-cpu.c @@ -XXX,XX +XXX,XX @@ static void kvm_riscv_handle_sbi_dbcn(CPUState *cs, struct kvm_run *run) break; case SBI_EXT_DBCN_CONSOLE_WRITE_BYTE: ch = run->riscv_sbi.args[0]; - ret = qemu_chr_fe_write(serial_hd(0)->be, &ch, sizeof(ch)); + ret = qemu_chr_fe_write_all(serial_hd(0)->be, &ch, sizeof(ch)); if (ret < 0) { error_report("SBI_EXT_DBCN_CONSOLE_WRITE_BYTE: error when " -- 2.50.0 From: Anton Blanchard <antonb@tenstorrent.com> fcvt.s.bf16 uses the FP16 check_nanbox_h() which returns an FP16 quiet NaN. Add check_nanbox_bf16() which returns a BF16 quiet NaN. Signed-off-by: Anton Blanchard <antonb@tenstorrent.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250501114253.594887-1-antonb@tenstorrent.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/internals.h | 16 ++++++++++++++++ target/riscv/fpu_helper.c | 2 +- 2 files changed, 17 insertions(+), 1 deletion(-) diff --git a/target/riscv/internals.h b/target/riscv/internals.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/internals.h +++ b/target/riscv/internals.h @@ -XXX,XX +XXX,XX @@ static inline float16 check_nanbox_h(CPURISCVState *env, uint64_t f) } } +static inline float16 check_nanbox_bf16(CPURISCVState *env, uint64_t f) +{ + /* Disable nanbox check when enable zfinx */ + if (env_archcpu(env)->cfg.ext_zfinx) { + return (uint16_t)f; + } + + uint64_t mask = MAKE_64BIT_MASK(16, 48); + + if (likely((f & mask) == mask)) { + return (uint16_t)f; + } else { + return 0x7FC0u; /* default qnan */ + } +} + #ifndef CONFIG_USER_ONLY /* Our implementation of SysemuCPUOps::has_work */ bool riscv_cpu_has_work(CPUState *cs); diff --git a/target/riscv/fpu_helper.c b/target/riscv/fpu_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/fpu_helper.c +++ b/target/riscv/fpu_helper.c @@ -XXX,XX +XXX,XX @@ uint64_t helper_fcvt_bf16_s(CPURISCVState *env, uint64_t rs1) uint64_t helper_fcvt_s_bf16(CPURISCVState *env, uint64_t rs1) { - float16 frs1 = check_nanbox_h(env, rs1); + float16 frs1 = check_nanbox_bf16(env, rs1); return nanbox_s(env, bfloat16_to_float32(frs1, &env->fp_status)); } -- 2.50.0 From: Florian Lugou <florian.lugou@provenrun.com> The current handler for TXFIFO writes schedules an async callback to pop characters from the queue. When software writes to TXFIFO faster than the async callback delay (100ns), the timer may be pushed back while the previous character has not be dequeued yet. This happens in particular when using -icount with small shift values. This is especially worrysome when software repetitively issues amoor.w instructions (as suggested by SiFive specification) and the FIFO is full, leading to the callback being infinitly pushed back. This commit fixes the issue by never pushing back the timer, only updating it if it is not already active. Signed-off-by: Florian Lugou <florian.lugou@provenrun.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250605101255.797162-1-florian.lugou@provenrun.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/char/sifive_uart.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/hw/char/sifive_uart.c b/hw/char/sifive_uart.c index XXXXXXX..XXXXXXX 100644 --- a/hw/char/sifive_uart.c +++ b/hw/char/sifive_uart.c @@ -XXX,XX +XXX,XX @@ static void sifive_uart_write_tx_fifo(SiFiveUARTState *s, const uint8_t *buf, s->txfifo |= SIFIVE_UART_TXFIFO_FULL; } - timer_mod(s->fifo_trigger_handle, current_time + - TX_INTERRUPT_TRIGGER_DELAY_NS); + if (!timer_pending(s->fifo_trigger_handle)) { + timer_mod(s->fifo_trigger_handle, current_time + + TX_INTERRUPT_TRIGGER_DELAY_NS); + } } static uint64_t -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The address is a hardware address, so use hwaddr for consistency with the rest of the machine. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-2-joel@jms.id.au> 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_socket_clint(RISCVVirtState *s, int cpu; g_autofree char *clint_name = NULL; g_autofree uint32_t *clint_cells = NULL; - unsigned long clint_addr; + hwaddr clint_addr; MachineState *ms = MACHINE(s); static const char * const clint_compat[2] = { "sifive,clint0", "riscv,clint0" @@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_clint(RISCVVirtState *s, } clint_addr = s->memmap[VIRT_CLINT].base + - (s->memmap[VIRT_CLINT].size * socket); - clint_name = g_strdup_printf("/soc/clint@%lx", clint_addr); + s->memmap[VIRT_CLINT].size * socket; + clint_name = g_strdup_printf("/soc/clint@%"HWADDR_PRIx, clint_addr); qemu_fdt_add_subnode(ms->fdt, clint_name); qemu_fdt_setprop_string_array(ms->fdt, clint_name, "compatible", (char **)&clint_compat, -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-3-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 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_socket_clint(RISCVVirtState *s, qemu_fdt_setprop_string_array(ms->fdt, clint_name, "compatible", (char **)&clint_compat, ARRAY_SIZE(clint_compat)); - qemu_fdt_setprop_cells(ms->fdt, clint_name, "reg", - 0x0, clint_addr, 0x0, s->memmap[VIRT_CLINT].size); + qemu_fdt_setprop_sized_cells(ms->fdt, clint_name, "reg", + 2, clint_addr, 2, s->memmap[VIRT_CLINT].size); qemu_fdt_setprop(ms->fdt, clint_name, "interrupts-extended", clint_cells, s->soc[socket].num_harts * sizeof(uint32_t) * 4); riscv_socket_fdt_write_id(ms, clint_name, socket); -- 2.50.0 From: Joel Stanley <joel@jms.id.au> Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-4-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 3 +-- 1 file changed, 1 insertion(+), 2 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_socket_memory(RISCVVirtState *s, int socket) size = riscv_socket_mem_size(ms, socket); mem_name = g_strdup_printf("/memory@%"HWADDR_PRIx, addr); qemu_fdt_add_subnode(ms->fdt, mem_name); - qemu_fdt_setprop_cells(ms->fdt, mem_name, "reg", - addr >> 32, addr, size >> 32, size); + qemu_fdt_setprop_sized_cells(ms->fdt, mem_name, "reg", 2, addr, 2, size); qemu_fdt_setprop_string(ms->fdt, mem_name, "device_type", "memory"); riscv_socket_fdt_write_id(ms, mem_name, socket); } -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-5-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 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_one_aplic(RISCVVirtState *s, int socket, 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, aplic_size); + qemu_fdt_setprop_sized_cells(ms->fdt, aplic_name, "reg", + 2, aplic_addr, 2, aplic_size); qemu_fdt_setprop_cell(ms->fdt, aplic_name, "riscv,num-sources", VIRT_IRQCHIP_NUM_SOURCES); -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-6-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 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_socket_aclint(RISCVVirtState *s, qemu_fdt_add_subnode(ms->fdt, name); qemu_fdt_setprop_string(ms->fdt, name, "compatible", "riscv,aclint-mswi"); - qemu_fdt_setprop_cells(ms->fdt, name, "reg", - 0x0, addr, 0x0, RISCV_ACLINT_SWI_SIZE); + qemu_fdt_setprop_sized_cells(ms->fdt, name, "reg", + 2, addr, 2, RISCV_ACLINT_SWI_SIZE); qemu_fdt_setprop(ms->fdt, name, "interrupts-extended", aclint_mswi_cells, aclint_cells_size); qemu_fdt_setprop(ms->fdt, name, "interrupt-controller", NULL, 0); @@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aclint(RISCVVirtState *s, qemu_fdt_add_subnode(ms->fdt, name); qemu_fdt_setprop_string(ms->fdt, name, "compatible", "riscv,aclint-mtimer"); - qemu_fdt_setprop_cells(ms->fdt, name, "reg", - 0x0, addr + RISCV_ACLINT_DEFAULT_MTIME, - 0x0, size - RISCV_ACLINT_DEFAULT_MTIME, - 0x0, addr + RISCV_ACLINT_DEFAULT_MTIMECMP, - 0x0, RISCV_ACLINT_DEFAULT_MTIME); + qemu_fdt_setprop_sized_cells(ms->fdt, name, "reg", + 2, addr + RISCV_ACLINT_DEFAULT_MTIME, + 2, size - RISCV_ACLINT_DEFAULT_MTIME, + 2, addr + RISCV_ACLINT_DEFAULT_MTIMECMP, + 2, RISCV_ACLINT_DEFAULT_MTIME); qemu_fdt_setprop(ms->fdt, name, "interrupts-extended", aclint_mtimer_cells, aclint_cells_size); riscv_socket_fdt_write_id(ms, name, socket); @@ -XXX,XX +XXX,XX @@ static void create_fdt_socket_aclint(RISCVVirtState *s, qemu_fdt_add_subnode(ms->fdt, name); qemu_fdt_setprop_string(ms->fdt, name, "compatible", "riscv,aclint-sswi"); - qemu_fdt_setprop_cells(ms->fdt, name, "reg", - 0x0, addr, 0x0, s->memmap[VIRT_ACLINT_SSWI].size); + qemu_fdt_setprop_sized_cells(ms->fdt, name, "reg", + 2, addr, 2, s->memmap[VIRT_ACLINT_SSWI].size); qemu_fdt_setprop(ms->fdt, name, "interrupts-extended", aclint_sswi_cells, aclint_cells_size); qemu_fdt_setprop(ms->fdt, name, "interrupt-controller", NULL, 0); -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-7-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 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_socket_plic(RISCVVirtState *s, s->soc[socket].num_harts * sizeof(uint32_t) * 4); } - qemu_fdt_setprop_cells(ms->fdt, plic_name, "reg", - 0x0, plic_addr, 0x0, s->memmap[VIRT_PLIC].size); + qemu_fdt_setprop_sized_cells(ms->fdt, plic_name, "reg", + 2, plic_addr, 2, s->memmap[VIRT_PLIC].size); qemu_fdt_setprop_cell(ms->fdt, plic_name, "riscv,ndev", VIRT_IRQCHIP_NUM_SOURCES - 1); riscv_socket_fdt_write_id(ms, plic_name, socket); -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-8-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 4 +--- 1 file changed, 1 insertion(+), 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_virtio(RISCVVirtState *s, uint32_t irq_virtio_phandle) qemu_fdt_add_subnode(ms->fdt, name); qemu_fdt_setprop_string(ms->fdt, name, "compatible", "virtio,mmio"); - qemu_fdt_setprop_cells(ms->fdt, name, "reg", - 0x0, addr, - 0x0, size); + qemu_fdt_setprop_sized_cells(ms->fdt, name, "reg", 2, addr, 2, size); qemu_fdt_setprop_cell(ms->fdt, name, "interrupt-parent", irq_virtio_phandle); if (s->aia_type == VIRT_AIA_TYPE_NONE) { -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-9-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 5 +++-- 1 file changed, 3 insertions(+), 2 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_reset(RISCVVirtState *s, uint32_t *phandle) qemu_fdt_setprop_string_array(ms->fdt, name, "compatible", (char **)&compat, ARRAY_SIZE(compat)); } - qemu_fdt_setprop_cells(ms->fdt, name, "reg", - 0x0, s->memmap[VIRT_TEST].base, 0x0, s->memmap[VIRT_TEST].size); + qemu_fdt_setprop_sized_cells(ms->fdt, name, "reg", + 2, s->memmap[VIRT_TEST].base, + 2, s->memmap[VIRT_TEST].size); qemu_fdt_setprop_cell(ms->fdt, name, "phandle", test_phandle); test_phandle = qemu_fdt_get_phandle(ms->fdt, name); g_free(name); -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-10-joel@jms.id.au> 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_uart(RISCVVirtState *s, s->memmap[VIRT_UART0].base); qemu_fdt_add_subnode(ms->fdt, name); qemu_fdt_setprop_string(ms->fdt, name, "compatible", "ns16550a"); - qemu_fdt_setprop_cells(ms->fdt, name, "reg", - 0x0, s->memmap[VIRT_UART0].base, - 0x0, s->memmap[VIRT_UART0].size); + qemu_fdt_setprop_sized_cells(ms->fdt, name, "reg", + 2, s->memmap[VIRT_UART0].base, + 2, s->memmap[VIRT_UART0].size); qemu_fdt_setprop_cell(ms->fdt, name, "clock-frequency", 3686400); qemu_fdt_setprop_cell(ms->fdt, name, "interrupt-parent", irq_mmio_phandle); if (s->aia_type == VIRT_AIA_TYPE_NONE) { -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-11-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 5 +++-- 1 file changed, 3 insertions(+), 2 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_rtc(RISCVVirtState *s, qemu_fdt_add_subnode(ms->fdt, name); qemu_fdt_setprop_string(ms->fdt, name, "compatible", "google,goldfish-rtc"); - qemu_fdt_setprop_cells(ms->fdt, name, "reg", - 0x0, s->memmap[VIRT_RTC].base, 0x0, s->memmap[VIRT_RTC].size); + qemu_fdt_setprop_sized_cells(ms->fdt, name, "reg", + 2, s->memmap[VIRT_RTC].base, + 2, s->memmap[VIRT_RTC].size); qemu_fdt_setprop_cell(ms->fdt, name, "interrupt-parent", irq_mmio_phandle); if (s->aia_type == VIRT_AIA_TYPE_NONE) { -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-12-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 3 +-- 1 file changed, 1 insertion(+), 2 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_iommu_sys(RISCVVirtState *s, uint32_t irq_chip, qemu_fdt_setprop_cell(fdt, iommu_node, "#iommu-cells", 1); qemu_fdt_setprop_cell(fdt, iommu_node, "phandle", iommu_phandle); - qemu_fdt_setprop_cells(fdt, iommu_node, "reg", - addr >> 32, addr, size >> 32, size); + qemu_fdt_setprop_sized_cells(fdt, iommu_node, "reg", 2, addr, 2, size); qemu_fdt_setprop_cell(fdt, iommu_node, "interrupt-parent", irq_chip); qemu_fdt_setprop_cells(fdt, iommu_node, "interrupts", -- 2.50.0 From: Joel Stanley <joel@jms.id.au> The current device tree property uses two cells for the address (and for the size), but assumes the they are less than 32 bits by hard coding the high cell to zero. Use qemu_fdt_setprop_sized_cells to do the job of splitting the upper and lower 32 bits across cells. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Message-ID: <20250604025450.85327-13-joel@jms.id.au> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/virt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 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_pcie(RISCVVirtState *s, if (s->aia_type == VIRT_AIA_TYPE_APLIC_IMSIC) { qemu_fdt_setprop_cell(ms->fdt, name, "msi-parent", msi_pcie_phandle); } - qemu_fdt_setprop_cells(ms->fdt, name, "reg", 0, - s->memmap[VIRT_PCIE_ECAM].base, 0, s->memmap[VIRT_PCIE_ECAM].size); + qemu_fdt_setprop_sized_cells(ms->fdt, name, "reg", 2, + s->memmap[VIRT_PCIE_ECAM].base, 2, s->memmap[VIRT_PCIE_ECAM].size); qemu_fdt_setprop_sized_cells(ms->fdt, name, "ranges", 1, FDT_PCI_RANGE_IOPORT, 2, 0, 2, s->memmap[VIRT_PCIE_PIO].base, 2, s->memmap[VIRT_PCIE_PIO].size, -- 2.50.0 From: Huang Borong <3543977024@qq.com> Add a CPU entry for the Xiangshan Kunminghu CPU, an open-source, high-performance RISC-V processor. More details can be found at: https://github.com/OpenXiangShan/XiangShan Note: The ISA extensions supported by the Xiangshan Kunminghu CPU are categorized based on four RISC-V specifications: Volume I: Unprivileged Architecture, Volume II: Privileged Architecture, AIA, and RVA23. The extensions within each category are organized according to the chapter order in the specifications. Signed-off-by: Yu Hu <huyu@bosc.ac.cn> Signed-off-by: Ran Wang <wangran@bosc.ac.cn> Signed-off-by: Borong Huang <3543977024@qq.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250425122212.364-1-wangran@bosc.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu-qom.h | 1 + target/riscv/cpu.c | 58 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 59 insertions(+) diff --git a/target/riscv/cpu-qom.h b/target/riscv/cpu-qom.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu-qom.h +++ b/target/riscv/cpu-qom.h @@ -XXX,XX +XXX,XX @@ #define TYPE_RISCV_CPU_VEYRON_V1 RISCV_CPU_TYPE_NAME("veyron-v1") #define TYPE_RISCV_CPU_TT_ASCALON RISCV_CPU_TYPE_NAME("tt-ascalon") #define TYPE_RISCV_CPU_XIANGSHAN_NANHU RISCV_CPU_TYPE_NAME("xiangshan-nanhu") +#define TYPE_RISCV_CPU_XIANGSHAN_KMH RISCV_CPU_TYPE_NAME("xiangshan-kunminghu") #define TYPE_RISCV_CPU_HOST RISCV_CPU_TYPE_NAME("host") OBJECT_DECLARE_CPU_TYPE(RISCVCPU, RISCVCPUClass, RISCV_CPU) 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 TypeInfo riscv_cpu_type_infos[] = { .cfg.max_satp_mode = VM_1_10_SV39, ), + DEFINE_RISCV_CPU(TYPE_RISCV_CPU_XIANGSHAN_KMH, TYPE_RISCV_VENDOR_CPU, + .misa_mxl_max = MXL_RV64, + .misa_ext = RVG | RVC | RVB | RVS | RVU | RVH | RVV, + .priv_spec = PRIV_VERSION_1_13_0, + /* + * The RISC-V Instruction Set Manual: Volume I + * Unprivileged Architecture + */ + .cfg.ext_zicntr = true, + .cfg.ext_zihpm = true, + .cfg.ext_zihintntl = true, + .cfg.ext_zihintpause = true, + .cfg.ext_zimop = true, + .cfg.ext_zcmop = true, + .cfg.ext_zicond = true, + .cfg.ext_zawrs = true, + .cfg.ext_zacas = true, + .cfg.ext_zfh = true, + .cfg.ext_zfa = true, + .cfg.ext_zcb = true, + .cfg.ext_zbc = true, + .cfg.ext_zvfh = true, + .cfg.ext_zkn = true, + .cfg.ext_zks = true, + .cfg.ext_zkt = true, + .cfg.ext_zvbb = true, + .cfg.ext_zvkt = true, + /* + * The RISC-V Instruction Set Manual: Volume II + * Privileged Architecture + */ + .cfg.ext_smstateen = true, + .cfg.ext_smcsrind = true, + .cfg.ext_sscsrind = true, + .cfg.ext_svnapot = true, + .cfg.ext_svpbmt = true, + .cfg.ext_svinval = true, + .cfg.ext_sstc = true, + .cfg.ext_sscofpmf = true, + .cfg.ext_ssdbltrp = true, + .cfg.ext_ssnpm = true, + .cfg.ext_smnpm = true, + .cfg.ext_smmpm = true, + .cfg.ext_sspm = true, + .cfg.ext_supm = true, + /* The RISC-V Advanced Interrupt Architecture */ + .cfg.ext_smaia = true, + .cfg.ext_ssaia = true, + /* RVA23 Profiles */ + .cfg.ext_zicbom = true, + .cfg.ext_zicbop = true, + .cfg.ext_zicboz = true, + .cfg.ext_svade = true, + .cfg.mmu = true, + .cfg.pmp = true, + .cfg.max_satp_mode = VM_1_10_SV48, + ), + #if defined(CONFIG_TCG) && !defined(CONFIG_USER_ONLY) DEFINE_RISCV_CPU(TYPE_RISCV_CPU_BASE128, TYPE_RISCV_DYNAMIC_CPU, .cfg.max_satp_mode = VM_1_10_SV57, -- 2.50.0 From: Huang Borong <3543977024@qq.com> This implementation provides emulation for the Xiangshan Kunminghu FPGA prototype platform, including support for UART, CLINT, IMSIC, and APLIC devices. More details can be found at https://github.com/OpenXiangShan/XiangShan Signed-off-by: qinshaoqing <qinshaoqing@bosc.ac.cn> Signed-off-by: Yang Wang <wangyang@bosc.ac.cn> Signed-off-by: Yu Hu <819258943@qq.com> Signed-off-by: Ran Wang <wangran@bosc.ac.cn> Signed-off-by: Borong Huang <3543977024@qq.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-ID: <20250617074222.17618-1-wangran@bosc.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- MAINTAINERS | 7 + docs/system/riscv/xiangshan-kunminghu.rst | 39 ++++ docs/system/target-riscv.rst | 1 + configs/devices/riscv64-softmmu/default.mak | 1 + include/hw/riscv/xiangshan_kmh.h | 68 ++++++ hw/riscv/xiangshan_kmh.c | 220 ++++++++++++++++++++ hw/riscv/Kconfig | 9 + hw/riscv/meson.build | 1 + 8 files changed, 346 insertions(+) create mode 100644 docs/system/riscv/xiangshan-kunminghu.rst create mode 100644 include/hw/riscv/xiangshan_kmh.h create mode 100644 hw/riscv/xiangshan_kmh.c diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ S: Maintained F: hw/riscv/microblaze-v-generic.c F: docs/system/riscv/microblaze-v-generic.rst +Xiangshan Kunminghu +M: Ran Wang <wangran@bosc.ac.cn> +S: Maintained +F: docs/system/riscv/xiangshan-kunminghu.rst +F: hw/riscv/xiangshan_kmh.c +F: include/hw/riscv/xiangshan_kmh.h + RX Machines ----------- rx-gdbsim diff --git a/docs/system/riscv/xiangshan-kunminghu.rst b/docs/system/riscv/xiangshan-kunminghu.rst new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/docs/system/riscv/xiangshan-kunminghu.rst @@ -XXX,XX +XXX,XX @@ +BOSC Xiangshan Kunminghu FPGA prototype platform (``xiangshan-kunminghu``) +========================================================================== +The ``xiangshan-kunminghu`` machine is compatible with our FPGA prototype +platform. + +XiangShan is an open-source high-performance RISC-V processor project. +The third generation processor is called Kunminghu. Kunminghu is a 64-bit +RV64GCBSUHV processor core. More information can be found in our Github +repository: +https://github.com/OpenXiangShan/XiangShan + +Supported devices +----------------- +The ``xiangshan-kunminghu`` machine supports the following devices: + +* Up to 16 xiangshan-kunminghu cores +* Core Local Interruptor (CLINT) +* Incoming MSI Controller (IMSIC) +* Advanced Platform-Level Interrupt Controller (APLIC) +* 1 UART + +Boot options +------------ +The ``xiangshan-kunminghu`` machine can start using the standard ``-bios`` +functionality for loading the boot image. You need to compile and link +the firmware, kernel, and Device Tree (FDT) into a single binary file, +such as ``fw_payload.bin``. + +Running +------- +Below is an example command line for running the ``xiangshan-kunminghu`` +machine: + +.. code-block:: bash + + $ qemu-system-riscv64 -machine xiangshan-kunminghu \ + -smp 16 -m 16G \ + -bios path/to/opensbi/platform/generic/firmware/fw_payload.bin \ + -nographic diff --git a/docs/system/target-riscv.rst b/docs/system/target-riscv.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/target-riscv.rst +++ b/docs/system/target-riscv.rst @@ -XXX,XX +XXX,XX @@ undocumented; you can get a complete list by running riscv/shakti-c riscv/sifive_u riscv/virt + riscv/xiangshan-kunminghu RISC-V CPU firmware ------------------- diff --git a/configs/devices/riscv64-softmmu/default.mak b/configs/devices/riscv64-softmmu/default.mak index XXXXXXX..XXXXXXX 100644 --- a/configs/devices/riscv64-softmmu/default.mak +++ b/configs/devices/riscv64-softmmu/default.mak @@ -XXX,XX +XXX,XX @@ # CONFIG_RISCV_VIRT=n # CONFIG_MICROCHIP_PFSOC=n # CONFIG_SHAKTI_C=n +# CONFIG_XIANGSHAN_KUNMINGHU=n diff --git a/include/hw/riscv/xiangshan_kmh.h b/include/hw/riscv/xiangshan_kmh.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/riscv/xiangshan_kmh.h @@ -XXX,XX +XXX,XX @@ +/* SPDX-License-Identifier: BSD-2-Clause */ +/* + * QEMU RISC-V Board Compatible with the Xiangshan Kunminghu + * FPGA prototype platform + * + * Copyright (c) 2025 Beijing Institute of Open Source Chip (BOSC) + * + */ + +#ifndef HW_XIANGSHAN_KMH_H +#define HW_XIANGSHAN_KMH_H + +#include "hw/boards.h" +#include "hw/riscv/riscv_hart.h" + +#define XIANGSHAN_KMH_MAX_CPUS 16 + +typedef struct XiangshanKmhSoCState { + /*< private >*/ + DeviceState parent_obj; + + /*< public >*/ + RISCVHartArrayState cpus; + DeviceState *irqchip; + MemoryRegion rom; +} XiangshanKmhSoCState; + +#define TYPE_XIANGSHAN_KMH_SOC "xiangshan.kunminghu.soc" +DECLARE_INSTANCE_CHECKER(XiangshanKmhSoCState, XIANGSHAN_KMH_SOC, + TYPE_XIANGSHAN_KMH_SOC) + +typedef struct XiangshanKmhState { + /*< private >*/ + MachineState parent_obj; + + /*< public >*/ + XiangshanKmhSoCState soc; +} XiangshanKmhState; + +#define TYPE_XIANGSHAN_KMH_MACHINE MACHINE_TYPE_NAME("xiangshan-kunminghu") +DECLARE_INSTANCE_CHECKER(XiangshanKmhState, XIANGSHAN_KMH_MACHINE, + TYPE_XIANGSHAN_KMH_MACHINE) + +enum { + XIANGSHAN_KMH_ROM, + XIANGSHAN_KMH_UART0, + XIANGSHAN_KMH_CLINT, + XIANGSHAN_KMH_APLIC_M, + XIANGSHAN_KMH_APLIC_S, + XIANGSHAN_KMH_IMSIC_M, + XIANGSHAN_KMH_IMSIC_S, + XIANGSHAN_KMH_DRAM, +}; + +enum { + XIANGSHAN_KMH_UART0_IRQ = 10, +}; + +/* Indicating Timebase-freq (1MHZ) */ +#define XIANGSHAN_KMH_CLINT_TIMEBASE_FREQ 1000000 + +#define XIANGSHAN_KMH_IMSIC_NUM_IDS 255 +#define XIANGSHAN_KMH_IMSIC_NUM_GUESTS 7 +#define XIANGSHAN_KMH_IMSIC_GUEST_BITS 3 + +#define XIANGSHAN_KMH_APLIC_NUM_SOURCES 96 + +#endif diff --git a/hw/riscv/xiangshan_kmh.c b/hw/riscv/xiangshan_kmh.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/riscv/xiangshan_kmh.c @@ -XXX,XX +XXX,XX @@ +/* + * QEMU RISC-V Board Compatible with the Xiangshan Kunminghu + * FPGA prototype platform + * + * Copyright (c) 2025 Beijing Institute of Open Source Chip (BOSC) + * SPDX-License-Identifier: GPL-2.0-or-later + * + * Provides a board compatible with the Xiangshan Kunminghu + * FPGA prototype platform: + * + * 0) UART (16550A) + * 1) CLINT (Core-Local Interruptor) + * 2) IMSIC (Incoming MSI Controller) + * 3) APLIC (Advanced Platform-Level Interrupt Controller) + * + * More information can be found in our Github repository: + * https://github.com/OpenXiangShan/XiangShan + * + * 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 "qapi/error.h" +#include "system/address-spaces.h" +#include "hw/boards.h" +#include "hw/char/serial-mm.h" +#include "hw/intc/riscv_aclint.h" +#include "hw/intc/riscv_aplic.h" +#include "hw/intc/riscv_imsic.h" +#include "hw/qdev-properties.h" +#include "hw/riscv/boot.h" +#include "hw/riscv/xiangshan_kmh.h" +#include "hw/riscv/riscv_hart.h" +#include "system/system.h" + +static const MemMapEntry xiangshan_kmh_memmap[] = { + [XIANGSHAN_KMH_ROM] = { 0x1000, 0xF000 }, + [XIANGSHAN_KMH_UART0] = { 0x310B0000, 0x10000 }, + [XIANGSHAN_KMH_CLINT] = { 0x38000000, 0x10000 }, + [XIANGSHAN_KMH_APLIC_M] = { 0x31100000, 0x4000 }, + [XIANGSHAN_KMH_APLIC_S] = { 0x31120000, 0x4000 }, + [XIANGSHAN_KMH_IMSIC_M] = { 0x3A800000, 0x10000 }, + [XIANGSHAN_KMH_IMSIC_S] = { 0x3B000000, 0x80000 }, + [XIANGSHAN_KMH_DRAM] = { 0x80000000, 0x0 }, +}; + +static DeviceState *xiangshan_kmh_create_aia(uint32_t num_harts) +{ + int i; + const MemMapEntry *memmap = xiangshan_kmh_memmap; + hwaddr addr = 0; + DeviceState *aplic_m = NULL; + + /* M-level IMSICs */ + addr = memmap[XIANGSHAN_KMH_IMSIC_M].base; + for (i = 0; i < num_harts; i++) { + riscv_imsic_create(addr + i * IMSIC_HART_SIZE(0), i, true, + 1, XIANGSHAN_KMH_IMSIC_NUM_IDS); + } + + /* S-level IMSICs */ + addr = memmap[XIANGSHAN_KMH_IMSIC_S].base; + for (i = 0; i < num_harts; i++) { + riscv_imsic_create(addr + + i * IMSIC_HART_SIZE(XIANGSHAN_KMH_IMSIC_GUEST_BITS), + i, false, 1 + XIANGSHAN_KMH_IMSIC_GUEST_BITS, + XIANGSHAN_KMH_IMSIC_NUM_IDS); + } + + /* M-level APLIC */ + aplic_m = riscv_aplic_create(memmap[XIANGSHAN_KMH_APLIC_M].base, + memmap[XIANGSHAN_KMH_APLIC_M].size, + 0, 0, XIANGSHAN_KMH_APLIC_NUM_SOURCES, + 1, true, true, NULL); + + /* S-level APLIC */ + riscv_aplic_create(memmap[XIANGSHAN_KMH_APLIC_S].base, + memmap[XIANGSHAN_KMH_APLIC_S].size, + 0, 0, XIANGSHAN_KMH_APLIC_NUM_SOURCES, + 1, true, false, aplic_m); + + return aplic_m; +} + +static void xiangshan_kmh_soc_realize(DeviceState *dev, Error **errp) +{ + MachineState *ms = MACHINE(qdev_get_machine()); + XiangshanKmhSoCState *s = XIANGSHAN_KMH_SOC(dev); + const MemMapEntry *memmap = xiangshan_kmh_memmap; + MemoryRegion *system_memory = get_system_memory(); + uint32_t num_harts = ms->smp.cpus; + + qdev_prop_set_uint32(DEVICE(&s->cpus), "num-harts", num_harts); + qdev_prop_set_uint32(DEVICE(&s->cpus), "hartid-base", 0); + qdev_prop_set_string(DEVICE(&s->cpus), "cpu-type", + TYPE_RISCV_CPU_XIANGSHAN_KMH); + sysbus_realize(SYS_BUS_DEVICE(&s->cpus), &error_fatal); + + /* AIA */ + s->irqchip = xiangshan_kmh_create_aia(num_harts); + + /* UART */ + serial_mm_init(system_memory, memmap[XIANGSHAN_KMH_UART0].base, 2, + qdev_get_gpio_in(s->irqchip, XIANGSHAN_KMH_UART0_IRQ), + 115200, serial_hd(0), DEVICE_LITTLE_ENDIAN); + + /* CLINT */ + riscv_aclint_swi_create(memmap[XIANGSHAN_KMH_CLINT].base, + 0, num_harts, false); + riscv_aclint_mtimer_create(memmap[XIANGSHAN_KMH_CLINT].base + + RISCV_ACLINT_SWI_SIZE, + RISCV_ACLINT_DEFAULT_MTIMER_SIZE, + 0, num_harts, RISCV_ACLINT_DEFAULT_MTIMECMP, + RISCV_ACLINT_DEFAULT_MTIME, + XIANGSHAN_KMH_CLINT_TIMEBASE_FREQ, true); + + /* ROM */ + memory_region_init_rom(&s->rom, OBJECT(dev), "xiangshan.kunminghu.rom", + memmap[XIANGSHAN_KMH_ROM].size, &error_fatal); + memory_region_add_subregion(system_memory, + memmap[XIANGSHAN_KMH_ROM].base, &s->rom); +} + +static void xiangshan_kmh_soc_class_init(ObjectClass *klass, const void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->realize = xiangshan_kmh_soc_realize; + dc->user_creatable = false; +} + +static void xiangshan_kmh_soc_instance_init(Object *obj) +{ + XiangshanKmhSoCState *s = XIANGSHAN_KMH_SOC(obj); + + object_initialize_child(obj, "cpus", &s->cpus, TYPE_RISCV_HART_ARRAY); +} + +static const TypeInfo xiangshan_kmh_soc_info = { + .name = TYPE_XIANGSHAN_KMH_SOC, + .parent = TYPE_DEVICE, + .instance_size = sizeof(XiangshanKmhSoCState), + .instance_init = xiangshan_kmh_soc_instance_init, + .class_init = xiangshan_kmh_soc_class_init, +}; + +static void xiangshan_kmh_soc_register_types(void) +{ + type_register_static(&xiangshan_kmh_soc_info); +} +type_init(xiangshan_kmh_soc_register_types) + +static void xiangshan_kmh_machine_init(MachineState *machine) +{ + XiangshanKmhState *s = XIANGSHAN_KMH_MACHINE(machine); + const MemMapEntry *memmap = xiangshan_kmh_memmap; + MemoryRegion *system_memory = get_system_memory(); + hwaddr start_addr = memmap[XIANGSHAN_KMH_DRAM].base; + + /* Initialize SoC */ + object_initialize_child(OBJECT(machine), "soc", &s->soc, + TYPE_XIANGSHAN_KMH_SOC); + qdev_realize(DEVICE(&s->soc), NULL, &error_fatal); + + /* Register RAM */ + memory_region_add_subregion(system_memory, + memmap[XIANGSHAN_KMH_DRAM].base, + machine->ram); + + /* ROM reset vector */ + riscv_setup_rom_reset_vec(machine, &s->soc.cpus, + start_addr, + memmap[XIANGSHAN_KMH_ROM].base, + memmap[XIANGSHAN_KMH_ROM].size, 0, 0); + if (machine->firmware) { + riscv_load_firmware(machine->firmware, &start_addr, NULL); + } + + /* Note: dtb has been integrated into firmware(OpenSBI) when compiling */ +} + +static void xiangshan_kmh_machine_class_init(ObjectClass *klass, const void *data) +{ + MachineClass *mc = MACHINE_CLASS(klass); + static const char *const valid_cpu_types[] = { + TYPE_RISCV_CPU_XIANGSHAN_KMH, + NULL + }; + + mc->desc = "RISC-V Board compatible with the Xiangshan " \ + "Kunminghu FPGA prototype platform"; + mc->init = xiangshan_kmh_machine_init; + mc->max_cpus = XIANGSHAN_KMH_MAX_CPUS; + mc->default_cpu_type = TYPE_RISCV_CPU_XIANGSHAN_KMH; + mc->valid_cpu_types = valid_cpu_types; + mc->default_ram_id = "xiangshan.kunminghu.ram"; +} + +static const TypeInfo xiangshan_kmh_machine_info = { + .name = TYPE_XIANGSHAN_KMH_MACHINE, + .parent = TYPE_MACHINE, + .instance_size = sizeof(XiangshanKmhState), + .class_init = xiangshan_kmh_machine_class_init, +}; + +static void xiangshan_kmh_machine_register_types(void) +{ + type_register_static(&xiangshan_kmh_machine_info); +} +type_init(xiangshan_kmh_machine_register_types) diff --git a/hw/riscv/Kconfig b/hw/riscv/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/Kconfig +++ b/hw/riscv/Kconfig @@ -XXX,XX +XXX,XX @@ config SPIKE select HTIF select RISCV_ACLINT select SIFIVE_PLIC + +config XIANGSHAN_KUNMINGHU + bool + default y + depends on RISCV64 + select RISCV_ACLINT + select RISCV_APLIC + select RISCV_IMSIC + select SERIAL_MM diff --git a/hw/riscv/meson.build b/hw/riscv/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/meson.build +++ b/hw/riscv/meson.build @@ -XXX,XX +XXX,XX @@ riscv_ss.add(when: 'CONFIG_ACPI', if_true: files('virt-acpi-build.c')) riscv_ss.add(when: 'CONFIG_RISCV_IOMMU', if_true: files( 'riscv-iommu.c', 'riscv-iommu-pci.c', 'riscv-iommu-sys.c', 'riscv-iommu-hpm.c')) riscv_ss.add(when: 'CONFIG_MICROBLAZE_V', if_true: files('microblaze-v-generic.c')) +riscv_ss.add(when: 'CONFIG_XIANGSHAN_KUNMINGHU', if_true: files('xiangshan_kmh.c')) hw_arch += {'riscv': riscv_ss} -- 2.50.0 From: Max Chou <max.chou@sifive.com> According to the V spec, the vector fault-only-first load instructions may change the VL CSR. So the ldff_trans TCG translation function should generate the lookup_and_goto_ptr flow as the vsetvl/vsetvli translation function to make sure the vl_eq_vlmax TB flag is correct. Signed-off-by: Max Chou <max.chou@sifive.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-ID: <20250627133013.443997-1-max.chou@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn_trans/trans_rvv.c.inc | 6 ++++++ 1 file changed, 6 insertions(+) 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 ldff_trans(uint32_t vd, uint32_t rs1, uint32_t data, fn(dest, mask, base, tcg_env, desc); finalize_rvv_inst(s); + + /* vector unit-stride fault-only-first load may modify vl CSR */ + gen_update_pc(s, s->cur_insn_len); + lookup_and_goto_ptr(s); + s->base.is_jmp = DISAS_NORETURN; + return true; } -- 2.50.0 From: "liu.xuemei1@zte.com.cn" <liu.xuemei1@zte.com.cn> Address an error in migration when aia is configured as 'aplic-imsic' in riscv kvm vm by adding riscv_aplic_state_needed() and riscv_imsic_state_needed() to determine whether the corresponding sates are needed. Previously, the fields in the vmsds of 'riscv_aplic' and 'riscv_imsic' can only be initialized under certain special conditions in commit 95a97b3fd2. However, the corresponding ses of these vmsds are inserted into the savevm_state.handlers unconditionally. This led to migration failure characterized by uninitialized fields when save vm state: qemu-system-riscv64: ../migration/vmstate.c:433: vmstate_save_state_v: Assertion 'first_elem || !n_elems || !size' failed. Fixes: 95a97b3fd2 ("target/riscv: update APLIC and IMSIC to support KVM AIA") Signed-off-by: Xuemei Liu <liu.xuemei1@zte.com.cn> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20250616150034827wuHs_ffe3Qm8cqFXT7HeW@zte.com.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/intc/riscv_aplic.c | 12 ++++++++++-- hw/intc/riscv_imsic.c | 10 ++++++++-- 2 files changed, 18 insertions(+), 4 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 @@ static const Property riscv_aplic_properties[] = { DEFINE_PROP_BOOL("mmode", RISCVAPLICState, mmode, 0), }; +static bool riscv_aplic_state_needed(void *opaque) +{ + RISCVAPLICState *aplic = opaque; + + return riscv_use_emulated_aplic(aplic->msimode); +} + static const VMStateDescription vmstate_riscv_aplic = { .name = "riscv_aplic", - .version_id = 2, - .minimum_version_id = 2, + .version_id = 3, + .minimum_version_id = 3, + .needed = riscv_aplic_state_needed, .fields = (const VMStateField[]) { VMSTATE_UINT32(domaincfg, RISCVAPLICState), VMSTATE_UINT32(mmsicfgaddr, RISCVAPLICState), 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 @@ static const Property riscv_imsic_properties[] = { DEFINE_PROP_UINT32("num-irqs", RISCVIMSICState, num_irqs, 0), }; +static bool riscv_imsic_state_needed(void *opaque) +{ + return !kvm_irqchip_in_kernel(); +} + static const VMStateDescription vmstate_riscv_imsic = { .name = "riscv_imsic", - .version_id = 1, - .minimum_version_id = 1, + .version_id = 2, + .minimum_version_id = 2, + .needed = riscv_imsic_state_needed, .fields = (const VMStateField[]) { VMSTATE_VARRAY_UINT32(eidelivery, RISCVIMSICState, num_pages, 0, -- 2.50.0 From: Charalampos Mitrodimas <charmitro@posteo.net> According to the RISC-V Privileged Architecture specification, the low bit of MEPC/SEPC must always be zero. When IALIGN=32, the two low bits must be zero. This commit fixes the behavior of MEPC/SEPC CSR reads and writes, and the implicit reads by MRET/SRET instructions to properly mask the lowest bit(s) based on whether the C extension is enabled: - When C extension is enabled (IALIGN=16): mask bit 0 - When C extension is disabled (IALIGN=32): mask bits [1:0] Previously, when vectored mode bits from STVEC (which sets bit 0 for vectored mode) were written to MEPC, the bits would not be cleared correctly, causing incorrect behavior on MRET. Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2855 Signed-off-by: Charalampos Mitrodimas <charmitro@posteo.net> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-ID: <20250703182157.281320-2-charmitro@posteo.net> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/internals.h | 11 +++++++++++ target/riscv/csr.c | 8 ++++---- target/riscv/op_helper.c | 4 ++-- 3 files changed, 17 insertions(+), 6 deletions(-) diff --git a/target/riscv/internals.h b/target/riscv/internals.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/internals.h +++ b/target/riscv/internals.h @@ -XXX,XX +XXX,XX @@ static inline float16 check_nanbox_bf16(CPURISCVState *env, uint64_t f) } } +static inline target_ulong get_xepc_mask(CPURISCVState *env) +{ + /* When IALIGN=32, both low bits must be zero. + * When IALIGN=16 (has C extension), only bit 0 must be zero. */ + if (riscv_has_ext(env, RVC)) { + return ~(target_ulong)1; + } else { + return ~(target_ulong)3; + } +} + #ifndef CONFIG_USER_ONLY /* Our implementation of SysemuCPUOps::has_work */ bool riscv_cpu_has_work(CPUState *cs); 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_mscratch(CPURISCVState *env, int csrno, static RISCVException read_mepc(CPURISCVState *env, int csrno, target_ulong *val) { - *val = env->mepc; + *val = env->mepc & get_xepc_mask(env); return RISCV_EXCP_NONE; } static RISCVException write_mepc(CPURISCVState *env, int csrno, target_ulong val, uintptr_t ra) { - env->mepc = val; + env->mepc = val & get_xepc_mask(env); return RISCV_EXCP_NONE; } @@ -XXX,XX +XXX,XX @@ static RISCVException write_sscratch(CPURISCVState *env, int csrno, static RISCVException read_sepc(CPURISCVState *env, int csrno, target_ulong *val) { - *val = env->sepc; + *val = env->sepc & get_xepc_mask(env); return RISCV_EXCP_NONE; } static RISCVException write_sepc(CPURISCVState *env, int csrno, target_ulong val, uintptr_t ra) { - env->sepc = val; + env->sepc = val & get_xepc_mask(env); return RISCV_EXCP_NONE; } diff --git a/target/riscv/op_helper.c b/target/riscv/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/op_helper.c +++ b/target/riscv/op_helper.c @@ -XXX,XX +XXX,XX @@ target_ulong helper_sret(CPURISCVState *env) riscv_raise_exception(env, RISCV_EXCP_ILLEGAL_INST, GETPC()); } - target_ulong retpc = env->sepc; + target_ulong retpc = env->sepc & get_xepc_mask(env); if (!riscv_cpu_allow_16bit_insn(&env_archcpu(env)->cfg, env->priv_ver, env->misa_ext) && (retpc & 0x3)) { @@ -XXX,XX +XXX,XX @@ static target_ulong ssdbltrp_mxret(CPURISCVState *env, target_ulong mstatus, target_ulong helper_mret(CPURISCVState *env) { - target_ulong retpc = env->mepc; + target_ulong retpc = env->mepc & get_xepc_mask(env); uint64_t mstatus = env->mstatus; target_ulong prev_priv = get_field(mstatus, MSTATUS_MPP); -- 2.50.0 From: Charalampos Mitrodimas <charmitro@posteo.net> Add a regression test to verify that MEPC properly masks the lower bits when an address with mode bits is written to it, as required by the RISC-V Privileged Architecture specification. The test sets STVEC to an address with bit 0 set (vectored mode), triggers an illegal instruction exception, copies STVEC to MEPC in the trap handler, and verifies that MEPC masks bits [1:0] correctly for IALIGN=32. Without the fix, MEPC retains the mode bits (returns non-zero/FAIL). With the fix, MEPC clears bits [1:0] (returns 0/PASS). Signed-off-by: Charalampos Mitrodimas <charmitro@posteo.net> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-ID: <20250703182157.281320-3-charmitro@posteo.net> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- tests/tcg/riscv64/Makefile.softmmu-target | 4 ++ tests/tcg/riscv64/test-mepc-masking.S | 73 +++++++++++++++++++++++ 2 files changed, 77 insertions(+) create mode 100644 tests/tcg/riscv64/test-mepc-masking.S diff --git a/tests/tcg/riscv64/Makefile.softmmu-target b/tests/tcg/riscv64/Makefile.softmmu-target index XXXXXXX..XXXXXXX 100644 --- a/tests/tcg/riscv64/Makefile.softmmu-target +++ b/tests/tcg/riscv64/Makefile.softmmu-target @@ -XXX,XX +XXX,XX @@ EXTRA_RUNS += run-issue1060 run-issue1060: issue1060 $(call run-test, $<, $(QEMU) $(QEMU_OPTS)$<) +EXTRA_RUNS += run-test-mepc-masking +run-test-mepc-masking: test-mepc-masking + $(call run-test, $<, $(QEMU) $(QEMU_OPTS)$<) + # We don't currently support the multiarch system tests undefine MULTIARCH_TESTS diff --git a/tests/tcg/riscv64/test-mepc-masking.S b/tests/tcg/riscv64/test-mepc-masking.S new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/tcg/riscv64/test-mepc-masking.S @@ -XXX,XX +XXX,XX @@ +/* + * Test for MEPC masking bug fix + * + * This test verifies that MEPC properly masks the lower bits according + * to the RISC-V specification when vectored mode bits from STVEC are + * written to MEPC. + */ + + .option norvc + + .text + .global _start +_start: + /* Set up machine trap vector */ + lla t0, machine_trap_handler + csrw mtvec, t0 + + /* Set STVEC with vectored mode (mode bits = 01) */ + li t0, 0x80004001 + csrw stvec, t0 + + /* Clear medeleg to handle exceptions in M-mode */ + csrw medeleg, zero + + /* Trigger illegal instruction exception */ + .word 0xffffffff + +test_completed: + /* Exit with result in a0 */ + /* a0 = 0: success (bits [1:0] were masked) */ + /* a0 != 0: failure (some bits were not masked) */ + j _exit + +machine_trap_handler: + /* Check if illegal instruction (mcause = 2) */ + csrr t0, mcause + li t1, 2 + bne t0, t1, skip_test + + /* Test: Copy STVEC (with mode bits) to MEPC */ + csrr t0, stvec /* t0 = 0x80004001 */ + csrw mepc, t0 /* Write to MEPC */ + csrr t1, mepc /* Read back MEPC */ + + /* Check if bits [1:0] are masked (IALIGN=32 without RVC) */ + andi a0, t1, 3 /* a0 = 0 if both bits masked correctly */ + + /* Set correct return address */ + lla t0, test_completed + csrw mepc, t0 + +skip_test: + mret + +/* Exit with semihosting */ +_exit: + lla a1, semiargs + li t0, 0x20026 /* ADP_Stopped_ApplicationExit */ + sd t0, 0(a1) + sd a0, 8(a1) + li a0, 0x20 /* TARGET_SYS_EXIT_EXTENDED */ + + /* Semihosting call sequence */ + .balign 16 + slli zero, zero, 0x1f + ebreak + srai zero, zero, 0x7 + j . + + .data + .balign 8 +semiargs: + .space 16 -- 2.50.0 From: Vasilis Liaskovitis <vliaskovitis@suse.com> Usage of vsetvli instruction is reserved if VLMAX is changed when vsetvli rs1 and rd arguments are x0. In this case, if the new property is true, only the vill bit will be set. See https://github.com/riscv/riscv-isa-manual/blob/main/src/v-st-ext.adoc#avl-encoding According to the spec, the above use cases are reserved, and "Implementations may set vill in either case." Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2422 Signed-off-by: Vasilis Liaskovitis <vliaskovitis@suse.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-ID: <20250618213542.22873-1-vliaskovitis@suse.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/helper.h | 2 +- target/riscv/cpu_cfg_fields.h.inc | 1 + target/riscv/cpu.c | 1 + target/riscv/vector_helper.c | 12 +++++++++++- target/riscv/insn_trans/trans_rvv.c.inc | 4 ++-- 5 files changed, 16 insertions(+), 4 deletions(-) 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_FLAGS_3(hyp_hsv_d, TCG_CALL_NO_WG, void, env, tl, tl) #endif /* Vector functions */ -DEF_HELPER_3(vsetvl, tl, env, tl, tl) +DEF_HELPER_4(vsetvl, tl, env, tl, tl, tl) DEF_HELPER_5(vle8_v, void, ptr, ptr, tl, env, i32) DEF_HELPER_5(vle16_v, void, ptr, ptr, tl, env, i32) DEF_HELPER_5(vle32_v, void, ptr, ptr, tl, env, i32) diff --git a/target/riscv/cpu_cfg_fields.h.inc b/target/riscv/cpu_cfg_fields.h.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu_cfg_fields.h.inc +++ b/target/riscv/cpu_cfg_fields.h.inc @@ -XXX,XX +XXX,XX @@ BOOL_FIELD(ext_supm) BOOL_FIELD(rvv_ta_all_1s) BOOL_FIELD(rvv_ma_all_1s) BOOL_FIELD(rvv_vl_half_avl) +BOOL_FIELD(rvv_vsetvl_x0_vill) /* Named features */ BOOL_FIELD(ext_svade) BOOL_FIELD(ext_zic64b) 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 Property riscv_cpu_properties[] = { DEFINE_PROP_BOOL("rvv_ta_all_1s", RISCVCPU, cfg.rvv_ta_all_1s, false), DEFINE_PROP_BOOL("rvv_ma_all_1s", RISCVCPU, cfg.rvv_ma_all_1s, false), DEFINE_PROP_BOOL("rvv_vl_half_avl", RISCVCPU, cfg.rvv_vl_half_avl, false), + DEFINE_PROP_BOOL("rvv_vsetvl_x0_vill", RISCVCPU, cfg.rvv_vsetvl_x0_vill, false), /* * write_misa() is marked as experimental for now so mark 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 <math.h> target_ulong HELPER(vsetvl)(CPURISCVState *env, target_ulong s1, - target_ulong s2) + target_ulong s2, target_ulong x0) { int vlmax, vl; RISCVCPU *cpu = env_archcpu(env); @@ -XXX,XX +XXX,XX @@ target_ulong HELPER(vsetvl)(CPURISCVState *env, target_ulong s1, } else { vl = vlmax; } + + if (cpu->cfg.rvv_vsetvl_x0_vill && x0 && (env->vl != vl)) { + /* only set vill bit. */ + env->vill = 1; + env->vtype = 0; + env->vl = 0; + env->vstart = 0; + return 0; + } + env->vl = vl; env->vtype = s2; env->vstart = 0; 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 do_vsetvl(DisasContext *s, int rd, int rs1, TCGv s2) s1 = get_gpr(s, rs1, EXT_ZERO); } - gen_helper_vsetvl(dst, tcg_env, s1, s2); + gen_helper_vsetvl(dst, tcg_env, s1, s2, tcg_constant_tl((int) (rd == 0 && rs1 == 0))); gen_set_gpr(s, rd, dst); finalize_rvv_inst(s); @@ -XXX,XX +XXX,XX @@ static bool do_vsetivli(DisasContext *s, int rd, TCGv s1, TCGv s2) dst = dest_gpr(s, rd); - gen_helper_vsetvl(dst, tcg_env, s1, s2); + gen_helper_vsetvl(dst, tcg_env, s1, s2, tcg_constant_tl(0)); gen_set_gpr(s, rd, dst); finalize_rvv_inst(s); gen_update_pc(s, s->cur_insn_len); -- 2.50.0 From: Alexandre Ghiti <alexghiti@rivosinc.com> The Svrsw60t59b extension allows to free the PTE reserved bits 60 and 59 for software to use. Reviewed-by: Deepak Gupta <debug@rivosinc.com> Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Nutty Liu<liujingqi@lanxincomputing.com> Message-ID: <20250702-dev-alex-svrsw60b59b_v2-v2-1-504ddf0f8530@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/riscv/riscv-iommu-bits.h | 1 + target/riscv/cpu_bits.h | 3 ++- target/riscv/cpu_cfg_fields.h.inc | 1 + hw/riscv/riscv-iommu.c | 3 ++- target/riscv/cpu.c | 2 ++ target/riscv/cpu_helper.c | 3 ++- target/riscv/tcg/tcg-cpu.c | 8 ++++++++ 7 files changed, 18 insertions(+), 3 deletions(-) diff --git a/hw/riscv/riscv-iommu-bits.h b/hw/riscv/riscv-iommu-bits.h index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/riscv-iommu-bits.h +++ b/hw/riscv/riscv-iommu-bits.h @@ -XXX,XX +XXX,XX @@ struct riscv_iommu_pq_record { #define RISCV_IOMMU_CAP_SV39 BIT_ULL(9) #define RISCV_IOMMU_CAP_SV48 BIT_ULL(10) #define RISCV_IOMMU_CAP_SV57 BIT_ULL(11) +#define RISCV_IOMMU_CAP_SVRSW60T59B BIT_ULL(14) #define RISCV_IOMMU_CAP_SV32X4 BIT_ULL(16) #define RISCV_IOMMU_CAP_SV39X4 BIT_ULL(17) #define RISCV_IOMMU_CAP_SV48X4 BIT_ULL(18) 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 { #define PTE_SOFT 0x300 /* Reserved for Software */ #define PTE_PBMT 0x6000000000000000ULL /* Page-based memory types */ #define PTE_N 0x8000000000000000ULL /* NAPOT translation */ -#define PTE_RESERVED 0x1FC0000000000000ULL /* Reserved bits */ +#define PTE_RESERVED(svrsw60t59b) \ + (svrsw60t59b ? 0x07C0000000000000ULL : 0x1FC0000000000000ULL) /* Reserved bits */ #define PTE_ATTR (PTE_N | PTE_PBMT) /* All attributes bits */ /* Page table PPN shift amount */ diff --git a/target/riscv/cpu_cfg_fields.h.inc b/target/riscv/cpu_cfg_fields.h.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu_cfg_fields.h.inc +++ b/target/riscv/cpu_cfg_fields.h.inc @@ -XXX,XX +XXX,XX @@ BOOL_FIELD(ext_svadu) BOOL_FIELD(ext_svinval) BOOL_FIELD(ext_svnapot) BOOL_FIELD(ext_svpbmt) +BOOL_FIELD(ext_svrsw60t59b) BOOL_FIELD(ext_svvptc) BOOL_FIELD(ext_svukte) BOOL_FIELD(ext_zdinx) diff --git a/hw/riscv/riscv-iommu.c b/hw/riscv/riscv-iommu.c index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/riscv-iommu.c +++ b/hw/riscv/riscv-iommu.c @@ -XXX,XX +XXX,XX @@ static void riscv_iommu_realize(DeviceState *dev, Error **errp) } if (s->enable_g_stage) { s->cap |= RISCV_IOMMU_CAP_SV32X4 | RISCV_IOMMU_CAP_SV39X4 | - RISCV_IOMMU_CAP_SV48X4 | RISCV_IOMMU_CAP_SV57X4; + RISCV_IOMMU_CAP_SV48X4 | RISCV_IOMMU_CAP_SV57X4 | + RISCV_IOMMU_CAP_SVRSW60T59B; } if (s->hpm_cntrs > 0) { 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 @@ const RISCVIsaExtData isa_edata_arr[] = { ISA_EXT_DATA_ENTRY(svinval, PRIV_VERSION_1_12_0, ext_svinval), ISA_EXT_DATA_ENTRY(svnapot, PRIV_VERSION_1_12_0, ext_svnapot), ISA_EXT_DATA_ENTRY(svpbmt, PRIV_VERSION_1_12_0, ext_svpbmt), + ISA_EXT_DATA_ENTRY(svrsw60t59b, PRIV_VERSION_1_13_0, ext_svrsw60t59b), ISA_EXT_DATA_ENTRY(svukte, PRIV_VERSION_1_13_0, ext_svukte), ISA_EXT_DATA_ENTRY(svvptc, PRIV_VERSION_1_13_0, ext_svvptc), ISA_EXT_DATA_ENTRY(xtheadba, PRIV_VERSION_1_11_0, ext_xtheadba), @@ -XXX,XX +XXX,XX @@ const RISCVCPUMultiExtConfig riscv_cpu_extensions[] = { MULTI_EXT_CFG_BOOL("svinval", ext_svinval, false), MULTI_EXT_CFG_BOOL("svnapot", ext_svnapot, false), MULTI_EXT_CFG_BOOL("svpbmt", ext_svpbmt, false), + MULTI_EXT_CFG_BOOL("svrsw60t59b", ext_svrsw60t59b, false), MULTI_EXT_CFG_BOOL("svvptc", ext_svvptc, true), MULTI_EXT_CFG_BOOL("zicntr", ext_zicntr, true), 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 svade = riscv_cpu_cfg(env)->ext_svade; bool svadu = riscv_cpu_cfg(env)->ext_svadu; bool adue = svadu ? env->menvcfg & MENVCFG_ADUE : !svade; + bool svrsw60t59b = riscv_cpu_cfg(env)->ext_svrsw60t59b; if (first_stage && two_stage && env->virt_enabled) { pbmte = pbmte && (env->henvcfg & HENVCFG_PBMTE); @@ -XXX,XX +XXX,XX @@ static int get_physical_address(CPURISCVState *env, hwaddr *physical, if (riscv_cpu_sxl(env) == MXL_RV32) { ppn = pte >> PTE_PPN_SHIFT; } else { - if (pte & PTE_RESERVED) { + if (pte & PTE_RESERVED(svrsw60t59b)) { qemu_log_mask(LOG_GUEST_ERROR, "%s: reserved bits set in PTE: " "addr: 0x%" HWADDR_PRIx " pte: 0x" TARGET_FMT_lx "\n", __func__, pte_addr, pte); diff --git a/target/riscv/tcg/tcg-cpu.c b/target/riscv/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/tcg/tcg-cpu.c +++ b/target/riscv/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp) cpu->cfg.ext_ssctr = false; } + if (cpu->cfg.ext_svrsw60t59b && + (!cpu->cfg.mmu || mcc->def->misa_mxl_max == MXL_RV32)) { + error_setg(errp, "svrsw60t59b is not supported on RV32 and MMU-less platforms"); + return; + } + /* * Disable isa extensions based on priv spec after we * validated and set everything we need. @@ -XXX,XX +XXX,XX @@ static void riscv_init_max_cpu_extensions(Object *obj) if (env->misa_mxl != MXL_RV32) { isa_ext_update_enabled(cpu, CPU_CFG_OFFSET(ext_zcf), false); + } else { + isa_ext_update_enabled(cpu, CPU_CFG_OFFSET(ext_svrsw60t59b), false); } /* -- 2.50.0