From: Alistair Francis <alistair.francis@wdc.com> The following changes since commit d70075373af51b6aa1d637962c962120e201fc98: Merge tag 'for_upstream' of git://git.kernel.org/pub/scm/virt/kvm/mst/qemu into staging (2022-01-07 17:24:24 -0800) are available in the Git repository at: git@github.com:alistair23/qemu.git tags/pull-riscv-to-apply-20220108 for you to fetch changes up to 48eaeb56debf91817dea00a2cd9c1f6c986eb531: target/riscv: Implement the stval/mtval illegal instruction (2022-01-08 15:46:10 +1000) ---------------------------------------------------------------- Second RISC-V PR for QEMU 7.0 - Fix illegal instruction when PMP is disabled - SiFive PDMA 64-bit support - SiFive PLIC cleanups - Mark Hypervisor extension as non experimental - Enable Hypervisor extension by default - Support 32 cores on the virt machine - Corrections for the Vector extension - Experimental support for 128-bit CPUs - stval and mtval support for illegal instructions ---------------------------------------------------------------- Alistair Francis (11): hw/intc: sifive_plic: Add a reset function hw/intc: sifive_plic: Cleanup the write function hw/intc: sifive_plic: Cleanup the read function hw/intc: sifive_plic: Cleanup remaining functions target/riscv: Mark the Hypervisor extension as non experimental target/riscv: Enable the Hypervisor extension by default hw/riscv: Use error_fatal for SoC realisation hw/riscv: virt: Allow support for 32 cores target/riscv: Set the opcode in DisasContext target/riscv: Fixup setting GVA target/riscv: Implement the stval/mtval illegal instruction Bin Meng (1): roms/opensbi: Upgrade from v0.9 to v1.0 Frank Chang (3): target/riscv: rvv-1.0: Call the correct RVF/RVD check function for widening fp insns target/riscv: rvv-1.0: Call the correct RVF/RVD check function for widening fp/int type-convert insns target/riscv: rvv-1.0: Call the correct RVF/RVD check function for narrowing fp/int type-convert insns Frédéric Pétrot (18): exec/memop: Adding signedness to quad definitions exec/memop: Adding signed quad and octo defines qemu/int128: addition of div/rem 128-bit operations target/riscv: additional macros to check instruction support target/riscv: separation of bitwise logic and arithmetic helpers target/riscv: array for the 64 upper bits of 128-bit registers target/riscv: setup everything for rv64 to support rv128 execution target/riscv: moving some insns close to similar insns target/riscv: accessors to registers upper part and 128-bit load/store target/riscv: support for 128-bit bitwise instructions target/riscv: support for 128-bit U-type instructions target/riscv: support for 128-bit shift instructions target/riscv: support for 128-bit arithmetic instructions target/riscv: support for 128-bit M extension target/riscv: adding high part of some csrs target/riscv: helper functions to wrap calls to 128-bit csr insns target/riscv: modification of the trans_csrxx for 128-bit support target/riscv: actual functions to realize crs 128-bit insns Jim Shu (2): hw/dma: sifive_pdma: support high 32-bit access of 64-bit register hw/dma: sifive_pdma: permit 4/8-byte access size of PDMA registers Nikita Shubin (1): target/riscv/pmp: fix no pmp illegal intrs Philipp Tomsich (1): target/riscv: Fix position of 'experimental' comment include/disas/dis-asm.h | 1 + include/exec/memop.h | 15 +- include/hw/riscv/virt.h | 2 +- include/qemu/int128.h | 27 + include/tcg/tcg-op.h | 4 +- target/arm/translate-a32.h | 4 +- target/riscv/cpu.h | 24 + target/riscv/cpu_bits.h | 3 + target/riscv/helper.h | 9 + target/riscv/insn16.decode | 27 +- target/riscv/insn32.decode | 25 + accel/tcg/cputlb.c | 30 +- accel/tcg/user-exec.c | 8 +- disas/riscv.c | 5 + hw/dma/sifive_pdma.c | 181 ++++++- hw/intc/sifive_plic.c | 254 +++------ hw/riscv/microchip_pfsoc.c | 2 +- hw/riscv/opentitan.c | 2 +- hw/riscv/sifive_e.c | 2 +- hw/riscv/sifive_u.c | 2 +- target/alpha/translate.c | 32 +- target/arm/helper-a64.c | 8 +- target/arm/translate-a64.c | 8 +- target/arm/translate-neon.c | 6 +- target/arm/translate-sve.c | 10 +- target/arm/translate-vfp.c | 8 +- target/arm/translate.c | 2 +- target/cris/translate.c | 2 +- target/hppa/translate.c | 4 +- target/i386/tcg/mem_helper.c | 2 +- target/i386/tcg/translate.c | 36 +- target/m68k/op_helper.c | 2 +- target/mips/tcg/translate.c | 58 +- target/mips/tcg/tx79_translate.c | 8 +- target/ppc/translate.c | 32 +- target/riscv/cpu.c | 34 +- target/riscv/cpu_helper.c | 24 +- target/riscv/csr.c | 194 ++++++- target/riscv/gdbstub.c | 5 + target/riscv/m128_helper.c | 109 ++++ target/riscv/machine.c | 22 + target/riscv/op_helper.c | 47 +- target/riscv/translate.c | 257 +++++++-- target/s390x/tcg/mem_helper.c | 8 +- target/s390x/tcg/translate.c | 8 +- target/sh4/translate.c | 12 +- target/sparc/translate.c | 36 +- target/tricore/translate.c | 4 +- target/xtensa/translate.c | 4 +- tcg/tcg.c | 4 +- tcg/tci.c | 16 +- util/int128.c | 147 +++++ accel/tcg/ldst_common.c.inc | 8 +- target/mips/tcg/micromips_translate.c.inc | 10 +- target/ppc/translate/fixedpoint-impl.c.inc | 22 +- target/ppc/translate/fp-impl.c.inc | 4 +- target/ppc/translate/vsx-impl.c.inc | 42 +- target/riscv/insn_trans/trans_rva.c.inc | 22 +- target/riscv/insn_trans/trans_rvb.c.inc | 48 +- target/riscv/insn_trans/trans_rvd.c.inc | 4 +- target/riscv/insn_trans/trans_rvh.c.inc | 4 +- target/riscv/insn_trans/trans_rvi.c.inc | 716 +++++++++++++++++++++---- target/riscv/insn_trans/trans_rvm.c.inc | 192 ++++++- target/riscv/insn_trans/trans_rvv.c.inc | 78 ++- target/s390x/tcg/translate_vx.c.inc | 18 +- tcg/aarch64/tcg-target.c.inc | 2 +- tcg/arm/tcg-target.c.inc | 10 +- tcg/i386/tcg-target.c.inc | 12 +- tcg/mips/tcg-target.c.inc | 12 +- tcg/ppc/tcg-target.c.inc | 16 +- tcg/riscv/tcg-target.c.inc | 6 +- tcg/s390x/tcg-target.c.inc | 18 +- tcg/sparc/tcg-target.c.inc | 16 +- pc-bios/opensbi-riscv32-generic-fw_dynamic.bin | Bin 78680 -> 108504 bytes pc-bios/opensbi-riscv32-generic-fw_dynamic.elf | Bin 727464 -> 838904 bytes pc-bios/opensbi-riscv64-generic-fw_dynamic.bin | Bin 75096 -> 105296 bytes pc-bios/opensbi-riscv64-generic-fw_dynamic.elf | Bin 781264 -> 934696 bytes roms/opensbi | 2 +- target/riscv/meson.build | 1 + target/s390x/tcg/insn-data.def | 28 +- util/meson.build | 1 + 81 files changed, 2318 insertions(+), 750 deletions(-) create mode 100644 target/riscv/m128_helper.c create mode 100644 util/int128.c From: Nikita Shubin <n.shubin@yadro.com> As per the privilege specification, any access from S/U mode should fail if no pmp region is configured and pmp is present, othwerwise access should succeed. Fixes: d102f19a208 (target/riscv/pmp: Raise exception if no PMP entry is configured) Signed-off-by: Nikita Shubin <n.shubin@yadro.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20211214092659.15709-1-nikita.shubin@maquefel.me Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/op_helper.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) 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_mret(CPURISCVState *env, target_ulong cpu_pc_deb) uint64_t mstatus = env->mstatus; target_ulong prev_priv = get_field(mstatus, MSTATUS_MPP); - if (!pmp_get_num_rules(env) && (prev_priv != PRV_M)) { + if (riscv_feature(env, RISCV_FEATURE_PMP) && + !pmp_get_num_rules(env) && (prev_priv != PRV_M)) { riscv_raise_exception(env, RISCV_EXCP_ILLEGAL_INST, GETPC()); } -- 2.31.1 From: Jim Shu <jim.shu@sifive.com> Real PDMA supports high 32-bit read/write memory access of 64-bit register. The following result is PDMA tested in U-Boot on Unmatched board: 1. Real PDMA allows high 32-bit read/write to 64-bit register. => mw.l 0x3000000 0x0 <= Disclaim channel 0 => mw.l 0x3000000 0x1 <= Claim channel 0 => mw.l 0x3000010 0x80000000 <= Write low 32-bit NextDest (NextDest = 0x280000000) => mw.l 0x3000014 0x2 <= Write high 32-bit NextDest => md.l 0x3000010 1 <= Dump low 32-bit NextDest 03000010: 80000000 => md.l 0x3000014 1 <= Dump high 32-bit NextDest 03000014: 00000002 => mw.l 0x3000018 0x80001000 <= Write low 32-bit NextSrc (NextSrc = 0x280001000) => mw.l 0x300001c 0x2 <= Write high 32-bit NextSrc => md.l 0x3000018 1 <= Dump low 32-bit NextSrc 03000010: 80001000 => md.l 0x300001c 1 <= Dump high 32-bit NextSrc 03000014: 00000002 2. PDMA transfer from 0x280001000 to 0x280000000 is OK. => mw.q 0x3000008 0x4 <= NextBytes = 4 => mw.l 0x3000004 0x22000000 <= wsize = rsize = 2 (2^2 = 4 bytes) => mw.l 0x280000000 0x87654321 <= Fill test data to dst => mw.l 0x280001000 0x12345678 <= Fill test data to src => md.l 0x280000000 1; md.l 0x280001000 1 <= Dump src/dst memory contents 280000000: 87654321 !Ce. 280001000: 12345678 xV4. => md.l 0x3000000 8 <= Dump PDMA status 03000000: 00000001 22000000 00000004 00000000 ......."........ 03000010: 80000000 00000002 80001000 00000002 ................ => mw.l 0x3000000 0x3 <= Set channel 0 run and claim bits => md.l 0x3000000 8 <= Dump PDMA status 03000000: 40000001 22000000 00000004 00000000 ...@..."........ 03000010: 80000000 00000002 80001000 00000002 ................ => md.l 0x280000000 1; md.l 0x280001000 1 <= Dump src/dst memory contents 280000000: 12345678 xV4. 280001000: 12345678 xV4. 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> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Tested-by: Bin Meng <bmeng.cn@gmail.com> Message-id: 20220104063408.658169-2-jim.shu@sifive.com Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/dma/sifive_pdma.c | 177 +++++++++++++++++++++++++++++++++++++------ 1 file changed, 155 insertions(+), 22 deletions(-) diff --git a/hw/dma/sifive_pdma.c b/hw/dma/sifive_pdma.c index XXXXXXX..XXXXXXX 100644 --- a/hw/dma/sifive_pdma.c +++ b/hw/dma/sifive_pdma.c @@ -XXX,XX +XXX,XX @@ static inline void sifive_pdma_update_irq(SiFivePDMAState *s, int ch) s->chan[ch].state = DMA_CHAN_STATE_IDLE; } -static uint64_t sifive_pdma_read(void *opaque, hwaddr offset, unsigned size) +static uint64_t sifive_pdma_readq(SiFivePDMAState *s, int ch, hwaddr offset) { - SiFivePDMAState *s = opaque; - int ch = SIFIVE_PDMA_CHAN_NO(offset); uint64_t val = 0; - if (ch >= SIFIVE_PDMA_CHANS) { - qemu_log_mask(LOG_GUEST_ERROR, "%s: Invalid channel no %d\n", - __func__, ch); - return 0; + offset &= 0xfff; + switch (offset) { + case DMA_NEXT_BYTES: + val = s->chan[ch].next_bytes; + break; + case DMA_NEXT_DST: + val = s->chan[ch].next_dst; + break; + case DMA_NEXT_SRC: + val = s->chan[ch].next_src; + break; + case DMA_EXEC_BYTES: + val = s->chan[ch].exec_bytes; + break; + case DMA_EXEC_DST: + val = s->chan[ch].exec_dst; + break; + case DMA_EXEC_SRC: + val = s->chan[ch].exec_src; + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Unexpected 64-bit access to 0x%" HWADDR_PRIX "\n", + __func__, offset); + break; } + return val; +} + +static uint32_t sifive_pdma_readl(SiFivePDMAState *s, int ch, hwaddr offset) +{ + uint32_t val = 0; + offset &= 0xfff; switch (offset) { case DMA_CONTROL: @@ -XXX,XX +XXX,XX @@ static uint64_t sifive_pdma_read(void *opaque, hwaddr offset, unsigned size) val = s->chan[ch].next_config; break; case DMA_NEXT_BYTES: - val = s->chan[ch].next_bytes; + val = extract64(s->chan[ch].next_bytes, 0, 32); + break; + case DMA_NEXT_BYTES + 4: + val = extract64(s->chan[ch].next_bytes, 32, 32); break; case DMA_NEXT_DST: - val = s->chan[ch].next_dst; + val = extract64(s->chan[ch].next_dst, 0, 32); + break; + case DMA_NEXT_DST + 4: + val = extract64(s->chan[ch].next_dst, 32, 32); break; case DMA_NEXT_SRC: - val = s->chan[ch].next_src; + val = extract64(s->chan[ch].next_src, 0, 32); + break; + case DMA_NEXT_SRC + 4: + val = extract64(s->chan[ch].next_src, 32, 32); break; case DMA_EXEC_CONFIG: val = s->chan[ch].exec_config; break; case DMA_EXEC_BYTES: - val = s->chan[ch].exec_bytes; + val = extract64(s->chan[ch].exec_bytes, 0, 32); + break; + case DMA_EXEC_BYTES + 4: + val = extract64(s->chan[ch].exec_bytes, 32, 32); break; case DMA_EXEC_DST: - val = s->chan[ch].exec_dst; + val = extract64(s->chan[ch].exec_dst, 0, 32); + break; + case DMA_EXEC_DST + 4: + val = extract64(s->chan[ch].exec_dst, 32, 32); break; case DMA_EXEC_SRC: - val = s->chan[ch].exec_src; + val = extract64(s->chan[ch].exec_src, 0, 32); + break; + case DMA_EXEC_SRC + 4: + val = extract64(s->chan[ch].exec_src, 32, 32); break; default: - qemu_log_mask(LOG_GUEST_ERROR, "%s: Bad offset 0x%" HWADDR_PRIX "\n", + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Unexpected 32-bit access to 0x%" HWADDR_PRIX "\n", __func__, offset); break; } @@ -XXX,XX +XXX,XX @@ static uint64_t sifive_pdma_read(void *opaque, hwaddr offset, unsigned size) return val; } -static void sifive_pdma_write(void *opaque, hwaddr offset, - uint64_t value, unsigned size) +static uint64_t sifive_pdma_read(void *opaque, hwaddr offset, unsigned size) { SiFivePDMAState *s = opaque; int ch = SIFIVE_PDMA_CHAN_NO(offset); - bool claimed, run; + uint64_t val = 0; if (ch >= SIFIVE_PDMA_CHANS) { qemu_log_mask(LOG_GUEST_ERROR, "%s: Invalid channel no %d\n", __func__, ch); - return; + return 0; + } + + switch (size) { + case 8: + val = sifive_pdma_readq(s, ch, offset); + break; + case 4: + val = sifive_pdma_readl(s, ch, offset); + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, "%s: Invalid read size %u to PDMA\n", + __func__, size); + return 0; } + return val; +} + +static void sifive_pdma_writeq(SiFivePDMAState *s, int ch, + hwaddr offset, uint64_t value) +{ + offset &= 0xfff; + switch (offset) { + case DMA_NEXT_BYTES: + s->chan[ch].next_bytes = value; + break; + case DMA_NEXT_DST: + s->chan[ch].next_dst = value; + break; + case DMA_NEXT_SRC: + s->chan[ch].next_src = value; + break; + case DMA_EXEC_BYTES: + case DMA_EXEC_DST: + case DMA_EXEC_SRC: + /* these are read-only registers */ + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Unexpected 64-bit access to 0x%" HWADDR_PRIX "\n", + __func__, offset); + break; + } +} + +static void sifive_pdma_writel(SiFivePDMAState *s, int ch, + hwaddr offset, uint32_t value) +{ + bool claimed, run; + offset &= 0xfff; switch (offset) { case DMA_CONTROL: @@ -XXX,XX +XXX,XX @@ static void sifive_pdma_write(void *opaque, hwaddr offset, s->chan[ch].next_config = value; break; case DMA_NEXT_BYTES: - s->chan[ch].next_bytes = value; + s->chan[ch].next_bytes = + deposit64(s->chan[ch].next_bytes, 0, 32, value); + break; + case DMA_NEXT_BYTES + 4: + s->chan[ch].next_bytes = + deposit64(s->chan[ch].next_bytes, 32, 32, value); break; case DMA_NEXT_DST: - s->chan[ch].next_dst = value; + s->chan[ch].next_dst = deposit64(s->chan[ch].next_dst, 0, 32, value); + break; + case DMA_NEXT_DST + 4: + s->chan[ch].next_dst = deposit64(s->chan[ch].next_dst, 32, 32, value); break; case DMA_NEXT_SRC: - s->chan[ch].next_src = value; + s->chan[ch].next_src = deposit64(s->chan[ch].next_src, 0, 32, value); + break; + case DMA_NEXT_SRC + 4: + s->chan[ch].next_src = deposit64(s->chan[ch].next_src, 32, 32, value); break; case DMA_EXEC_CONFIG: case DMA_EXEC_BYTES: + case DMA_EXEC_BYTES + 4: case DMA_EXEC_DST: + case DMA_EXEC_DST + 4: case DMA_EXEC_SRC: + case DMA_EXEC_SRC + 4: /* these are read-only registers */ break; default: - qemu_log_mask(LOG_GUEST_ERROR, "%s: Bad offset 0x%" HWADDR_PRIX "\n", + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Unexpected 32-bit access to 0x%" HWADDR_PRIX "\n", __func__, offset); break; } } +static void sifive_pdma_write(void *opaque, hwaddr offset, + uint64_t value, unsigned size) +{ + SiFivePDMAState *s = opaque; + int ch = SIFIVE_PDMA_CHAN_NO(offset); + + if (ch >= SIFIVE_PDMA_CHANS) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: Invalid channel no %d\n", + __func__, ch); + return; + } + + switch (size) { + case 8: + sifive_pdma_writeq(s, ch, offset, value); + break; + case 4: + sifive_pdma_writel(s, ch, offset, (uint32_t) value); + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, "%s: Invalid write size %u to PDMA\n", + __func__, size); + break; + } +} + static const MemoryRegionOps sifive_pdma_ops = { .read = sifive_pdma_read, .write = sifive_pdma_write, -- 2.31.1 From: Jim Shu <jim.shu@sifive.com> It's obvious that PDMA supports 64-bit access of 64-bit registers, and in previous commit, we confirm that PDMA supports 32-bit access of both 32/64-bit registers. Thus, we configure 32/64-bit memory access of PDMA registers as valid in general. Signed-off-by: Jim Shu <jim.shu@sifive.com> Reviewed-by: Frank Chang <frank.chang@sifive.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Tested-by: Bin Meng <bmeng.cn@gmail.com> Message-id: 20220104063408.658169-3-jim.shu@sifive.com Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- hw/dma/sifive_pdma.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/hw/dma/sifive_pdma.c b/hw/dma/sifive_pdma.c index XXXXXXX..XXXXXXX 100644 --- a/hw/dma/sifive_pdma.c +++ b/hw/dma/sifive_pdma.c @@ -XXX,XX +XXX,XX @@ static const MemoryRegionOps sifive_pdma_ops = { .impl = { .min_access_size = 4, .max_access_size = 8, + }, + .valid = { + .min_access_size = 4, + .max_access_size = 8, } }; -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220105213937.1113508-2-alistair.francis@opensource.wdc.com> --- hw/intc/sifive_plic.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/hw/intc/sifive_plic.c b/hw/intc/sifive_plic.c index XXXXXXX..XXXXXXX 100644 --- a/hw/intc/sifive_plic.c +++ b/hw/intc/sifive_plic.c @@ -XXX,XX +XXX,XX @@ static const MemoryRegionOps sifive_plic_ops = { } }; +static void sifive_plic_reset(DeviceState *dev) +{ + SiFivePLICState *s = SIFIVE_PLIC(dev); + int i; + + memset(s->source_priority, 0, sizeof(uint32_t) * s->num_sources); + memset(s->target_priority, 0, sizeof(uint32_t) * s->num_addrs); + memset(s->pending, 0, sizeof(uint32_t) * s->bitfield_words); + memset(s->claimed, 0, sizeof(uint32_t) * s->bitfield_words); + memset(s->enable, 0, sizeof(uint32_t) * s->num_enables); + + for (i = 0; i < s->num_harts; i++) { + qemu_set_irq(s->m_external_irqs[i], 0); + qemu_set_irq(s->s_external_irqs[i], 0); + } +} + /* * parse PLIC hart/mode address offset config * @@ -XXX,XX +XXX,XX @@ static void sifive_plic_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); + dc->reset = sifive_plic_reset; device_class_set_props(dc, sifive_plic_properties); dc->realize = sifive_plic_realize; dc->vmsd = &vmstate_sifive_plic; -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220105213937.1113508-3-alistair.francis@opensource.wdc.com> --- hw/intc/sifive_plic.c | 76 +++++++++++++++---------------------------- 1 file changed, 27 insertions(+), 49 deletions(-) diff --git a/hw/intc/sifive_plic.c b/hw/intc/sifive_plic.c index XXXXXXX..XXXXXXX 100644 --- a/hw/intc/sifive_plic.c +++ b/hw/intc/sifive_plic.c @@ -XXX,XX +XXX,XX @@ #define RISCV_DEBUG_PLIC 0 +static bool addr_between(uint32_t addr, uint32_t base, uint32_t num) +{ + return addr >= base && addr - base < num; +} + static PLICMode char_to_mode(char c) { switch (c) { @@ -XXX,XX +XXX,XX @@ static void sifive_plic_write(void *opaque, hwaddr addr, uint64_t value, { SiFivePLICState *plic = opaque; - /* writes must be 4 byte words */ - if ((addr & 0x3) != 0) { - goto err; - } - - if (addr >= plic->priority_base && /* 4 bytes per source */ - addr < plic->priority_base + (plic->num_sources << 2)) - { + if (addr_between(addr, plic->priority_base, plic->num_sources << 2)) { uint32_t irq = ((addr - plic->priority_base) >> 2) + 1; + plic->source_priority[irq] = value & 7; - if (RISCV_DEBUG_PLIC) { - qemu_log("plic: write priority: irq=%d priority=%d\n", - irq, plic->source_priority[irq]); - } sifive_plic_update(plic); - return; - } else if (addr >= plic->pending_base && /* 1 bit per source */ - addr < plic->pending_base + (plic->num_sources >> 3)) - { + } else if (addr_between(addr, plic->pending_base, + plic->num_sources >> 3)) { qemu_log_mask(LOG_GUEST_ERROR, "%s: invalid pending write: 0x%" HWADDR_PRIx "", __func__, addr); - return; - } else if (addr >= plic->enable_base && /* 1 bit per source */ - addr < plic->enable_base + plic->num_addrs * plic->enable_stride) - { + } else if (addr_between(addr, plic->enable_base, + plic->num_addrs * plic->enable_stride)) { uint32_t addrid = (addr - plic->enable_base) / plic->enable_stride; uint32_t wordid = (addr & (plic->enable_stride - 1)) >> 2; + if (wordid < plic->bitfield_words) { plic->enable[addrid * plic->bitfield_words + wordid] = value; - if (RISCV_DEBUG_PLIC) { - qemu_log("plic: write enable: hart%d-%c word=%d value=%x\n", - plic->addr_config[addrid].hartid, - mode_to_char(plic->addr_config[addrid].mode), wordid, - plic->enable[addrid * plic->bitfield_words + wordid]); - } - return; + } else { + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Invalid enable write 0x%" HWADDR_PRIx "\n", + __func__, addr); } - } else if (addr >= plic->context_base && /* 4 bytes per reg */ - addr < plic->context_base + plic->num_addrs * plic->context_stride) - { + } else if (addr_between(addr, plic->context_base, + plic->num_addrs * plic->context_stride)) { uint32_t addrid = (addr - plic->context_base) / plic->context_stride; uint32_t contextid = (addr & (plic->context_stride - 1)); + if (contextid == 0) { - if (RISCV_DEBUG_PLIC) { - qemu_log("plic: write priority: hart%d-%c priority=%x\n", - plic->addr_config[addrid].hartid, - mode_to_char(plic->addr_config[addrid].mode), - plic->target_priority[addrid]); - } if (value <= plic->num_priorities) { plic->target_priority[addrid] = value; sifive_plic_update(plic); } - return; } else if (contextid == 4) { - if (RISCV_DEBUG_PLIC) { - qemu_log("plic: write claim: hart%d-%c irq=%x\n", - plic->addr_config[addrid].hartid, - mode_to_char(plic->addr_config[addrid].mode), - (uint32_t)value); - } if (value < plic->num_sources) { sifive_plic_set_claimed(plic, value, false); sifive_plic_update(plic); } - return; + } else { + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Invalid context write 0x%" HWADDR_PRIx "\n", + __func__, addr); } + } else { + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Invalid register write 0x%" HWADDR_PRIx "\n", + __func__, addr); } - -err: - qemu_log_mask(LOG_GUEST_ERROR, - "%s: Invalid register write 0x%" HWADDR_PRIx "\n", - __func__, addr); } static const MemoryRegionOps sifive_plic_ops = { -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220105213937.1113508-4-alistair.francis@opensource.wdc.com> --- hw/intc/sifive_plic.c | 55 +++++++++---------------------------------- 1 file changed, 11 insertions(+), 44 deletions(-) diff --git a/hw/intc/sifive_plic.c b/hw/intc/sifive_plic.c index XXXXXXX..XXXXXXX 100644 --- a/hw/intc/sifive_plic.c +++ b/hw/intc/sifive_plic.c @@ -XXX,XX +XXX,XX @@ static uint64_t sifive_plic_read(void *opaque, hwaddr addr, unsigned size) { SiFivePLICState *plic = opaque; - /* writes must be 4 byte words */ - if ((addr & 0x3) != 0) { - goto err; - } - - if (addr >= plic->priority_base && /* 4 bytes per source */ - addr < plic->priority_base + (plic->num_sources << 2)) - { + if (addr_between(addr, plic->priority_base, plic->num_sources << 2)) { uint32_t irq = ((addr - plic->priority_base) >> 2) + 1; - if (RISCV_DEBUG_PLIC) { - qemu_log("plic: read priority: irq=%d priority=%d\n", - irq, plic->source_priority[irq]); - } + return plic->source_priority[irq]; - } else if (addr >= plic->pending_base && /* 1 bit per source */ - addr < plic->pending_base + (plic->num_sources >> 3)) - { + } else if (addr_between(addr, plic->pending_base, plic->num_sources >> 3)) { uint32_t word = (addr - plic->pending_base) >> 2; - if (RISCV_DEBUG_PLIC) { - qemu_log("plic: read pending: word=%d value=%d\n", - word, plic->pending[word]); - } + return plic->pending[word]; - } else if (addr >= plic->enable_base && /* 1 bit per source */ - addr < plic->enable_base + plic->num_addrs * plic->enable_stride) - { + } else if (addr_between(addr, plic->enable_base, + plic->num_addrs * plic->enable_stride)) { uint32_t addrid = (addr - plic->enable_base) / plic->enable_stride; uint32_t wordid = (addr & (plic->enable_stride - 1)) >> 2; + if (wordid < plic->bitfield_words) { - if (RISCV_DEBUG_PLIC) { - qemu_log("plic: read enable: hart%d-%c word=%d value=%x\n", - plic->addr_config[addrid].hartid, - mode_to_char(plic->addr_config[addrid].mode), wordid, - plic->enable[addrid * plic->bitfield_words + wordid]); - } return plic->enable[addrid * plic->bitfield_words + wordid]; } - } else if (addr >= plic->context_base && /* 1 bit per source */ - addr < plic->context_base + plic->num_addrs * plic->context_stride) - { + } else if (addr_between(addr, plic->context_base, + plic->num_addrs * plic->context_stride)) { uint32_t addrid = (addr - plic->context_base) / plic->context_stride; uint32_t contextid = (addr & (plic->context_stride - 1)); + if (contextid == 0) { - if (RISCV_DEBUG_PLIC) { - qemu_log("plic: read priority: hart%d-%c priority=%x\n", - plic->addr_config[addrid].hartid, - mode_to_char(plic->addr_config[addrid].mode), - plic->target_priority[addrid]); - } return plic->target_priority[addrid]; } else if (contextid == 4) { uint32_t value = sifive_plic_claim(plic, addrid); - if (RISCV_DEBUG_PLIC) { - qemu_log("plic: read claim: hart%d-%c irq=%x\n", - plic->addr_config[addrid].hartid, - mode_to_char(plic->addr_config[addrid].mode), - value); - } + sifive_plic_update(plic); return value; } } -err: qemu_log_mask(LOG_GUEST_ERROR, "%s: Invalid register read 0x%" HWADDR_PRIx "\n", __func__, addr); -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> We can remove the original sifive_plic_irqs_pending() function and instead just use the sifive_plic_claim() function (renamed to sifive_plic_claimed()) to determine if any interrupts are pending. This requires move the side effects outside of sifive_plic_claimed(), but as they are only invoked once that isn't a problem. We have also removed all of the old #ifdef debugging logs, so let's cleanup the last remaining debug function while we are here. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220105213937.1113508-5-alistair.francis@opensource.wdc.com> --- hw/intc/sifive_plic.c | 109 +++++++++--------------------------------- 1 file changed, 22 insertions(+), 87 deletions(-) diff --git a/hw/intc/sifive_plic.c b/hw/intc/sifive_plic.c index XXXXXXX..XXXXXXX 100644 --- a/hw/intc/sifive_plic.c +++ b/hw/intc/sifive_plic.c @@ -XXX,XX +XXX,XX @@ #include "migration/vmstate.h" #include "hw/irq.h" -#define RISCV_DEBUG_PLIC 0 - static bool addr_between(uint32_t addr, uint32_t base, uint32_t num) { return addr >= base && addr - base < num; @@ -XXX,XX +XXX,XX @@ static PLICMode char_to_mode(char c) } } -static char mode_to_char(PLICMode m) -{ - switch (m) { - case PLICMode_U: return 'U'; - case PLICMode_S: return 'S'; - case PLICMode_H: return 'H'; - case PLICMode_M: return 'M'; - default: return '?'; - } -} - -static void sifive_plic_print_state(SiFivePLICState *plic) -{ - int i; - int addrid; - - /* pending */ - qemu_log("pending : "); - for (i = plic->bitfield_words - 1; i >= 0; i--) { - qemu_log("%08x", plic->pending[i]); - } - qemu_log("\n"); - - /* pending */ - qemu_log("claimed : "); - for (i = plic->bitfield_words - 1; i >= 0; i--) { - qemu_log("%08x", plic->claimed[i]); - } - qemu_log("\n"); - - for (addrid = 0; addrid < plic->num_addrs; addrid++) { - qemu_log("hart%d-%c enable: ", - plic->addr_config[addrid].hartid, - mode_to_char(plic->addr_config[addrid].mode)); - for (i = plic->bitfield_words - 1; i >= 0; i--) { - qemu_log("%08x", plic->enable[addrid * plic->bitfield_words + i]); - } - qemu_log("\n"); - } -} - static uint32_t atomic_set_masked(uint32_t *a, uint32_t mask, uint32_t value) { uint32_t old, new, cmp = qatomic_read(a); @@ -XXX,XX +XXX,XX @@ static void sifive_plic_set_claimed(SiFivePLICState *plic, int irq, bool level) atomic_set_masked(&plic->claimed[irq >> 5], 1 << (irq & 31), -!!level); } -static int sifive_plic_irqs_pending(SiFivePLICState *plic, uint32_t addrid) +static uint32_t sifive_plic_claimed(SiFivePLICState *plic, uint32_t addrid) { + uint32_t max_irq = 0; + uint32_t max_prio = plic->target_priority[addrid]; int i, j; + for (i = 0; i < plic->bitfield_words; i++) { uint32_t pending_enabled_not_claimed = - (plic->pending[i] & ~plic->claimed[i]) & - plic->enable[addrid * plic->bitfield_words + i]; + (plic->pending[i] & ~plic->claimed[i]) & + plic->enable[addrid * plic->bitfield_words + i]; + if (!pending_enabled_not_claimed) { continue; } + for (j = 0; j < 32; j++) { int irq = (i << 5) + j; uint32_t prio = plic->source_priority[irq]; int enabled = pending_enabled_not_claimed & (1 << j); - if (enabled && prio > plic->target_priority[addrid]) { - return 1; + + if (enabled && prio > max_prio) { + max_irq = irq; + max_prio = prio; } } } - return 0; + + return max_irq; } static void sifive_plic_update(SiFivePLICState *plic) @@ -XXX,XX +XXX,XX @@ static void sifive_plic_update(SiFivePLICState *plic) for (addrid = 0; addrid < plic->num_addrs; addrid++) { uint32_t hartid = plic->addr_config[addrid].hartid; PLICMode mode = plic->addr_config[addrid].mode; - int level = sifive_plic_irqs_pending(plic, addrid); + bool level = !!sifive_plic_claimed(plic, addrid); switch (mode) { case PLICMode_M: @@ -XXX,XX +XXX,XX @@ static void sifive_plic_update(SiFivePLICState *plic) break; } } - - if (RISCV_DEBUG_PLIC) { - sifive_plic_print_state(plic); - } -} - -static uint32_t sifive_plic_claim(SiFivePLICState *plic, uint32_t addrid) -{ - int i, j; - uint32_t max_irq = 0; - uint32_t max_prio = plic->target_priority[addrid]; - - for (i = 0; i < plic->bitfield_words; i++) { - uint32_t pending_enabled_not_claimed = - (plic->pending[i] & ~plic->claimed[i]) & - plic->enable[addrid * plic->bitfield_words + i]; - if (!pending_enabled_not_claimed) { - continue; - } - for (j = 0; j < 32; j++) { - int irq = (i << 5) + j; - uint32_t prio = plic->source_priority[irq]; - int enabled = pending_enabled_not_claimed & (1 << j); - if (enabled && prio > max_prio) { - max_irq = irq; - max_prio = prio; - } - } - } - - if (max_irq) { - sifive_plic_set_pending(plic, max_irq, false); - sifive_plic_set_claimed(plic, max_irq, true); - } - return max_irq; } static uint64_t sifive_plic_read(void *opaque, hwaddr addr, unsigned size) @@ -XXX,XX +XXX,XX @@ static uint64_t sifive_plic_read(void *opaque, hwaddr addr, unsigned size) if (contextid == 0) { return plic->target_priority[addrid]; } else if (contextid == 4) { - uint32_t value = sifive_plic_claim(plic, addrid); + uint32_t max_irq = sifive_plic_claimed(plic, addrid); + + if (max_irq) { + sifive_plic_set_pending(plic, max_irq, false); + sifive_plic_set_claimed(plic, max_irq, true); + } sifive_plic_update(plic); - return value; + return max_irq; } } -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> The Hypervisor spec is now frozen, so remove the experimental tag. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Anup Patel <anup.patel@wdc.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220105213937.1113508-6-alistair.francis@opensource.wdc.com> --- target/riscv/cpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_properties[] = { DEFINE_PROP_BOOL("s", RISCVCPU, cfg.ext_s, true), DEFINE_PROP_BOOL("u", RISCVCPU, cfg.ext_u, true), DEFINE_PROP_BOOL("v", RISCVCPU, cfg.ext_v, false), + DEFINE_PROP_BOOL("h", RISCVCPU, cfg.ext_h, false), DEFINE_PROP_BOOL("Counters", RISCVCPU, cfg.ext_counters, true), DEFINE_PROP_BOOL("Zifencei", RISCVCPU, cfg.ext_ifencei, true), DEFINE_PROP_BOOL("Zicsr", RISCVCPU, cfg.ext_icsr, true), @@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_properties[] = { DEFINE_PROP_BOOL("zbb", RISCVCPU, cfg.ext_zbb, true), DEFINE_PROP_BOOL("zbc", RISCVCPU, cfg.ext_zbc, true), DEFINE_PROP_BOOL("zbs", RISCVCPU, cfg.ext_zbs, true), - DEFINE_PROP_BOOL("x-h", RISCVCPU, cfg.ext_h, false), DEFINE_PROP_BOOL("x-j", RISCVCPU, cfg.ext_j, false), /* ePMP 0.9.3 */ DEFINE_PROP_BOOL("x-epmp", RISCVCPU, cfg.epmp, false), -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> Let's enable the Hypervisor extension by default. This doesn't affect named CPUs (such as lowrisc-ibex or sifive-u54) but does enable the Hypervisor extensions by default for the virt machine. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Anup Patel <anup.patel@wdc.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220105213937.1113508-7-alistair.francis@opensource.wdc.com> --- target/riscv/cpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_properties[] = { DEFINE_PROP_BOOL("s", RISCVCPU, cfg.ext_s, true), DEFINE_PROP_BOOL("u", RISCVCPU, cfg.ext_u, true), DEFINE_PROP_BOOL("v", RISCVCPU, cfg.ext_v, false), - DEFINE_PROP_BOOL("h", RISCVCPU, cfg.ext_h, false), + DEFINE_PROP_BOOL("h", RISCVCPU, cfg.ext_h, true), DEFINE_PROP_BOOL("Counters", RISCVCPU, cfg.ext_counters, true), DEFINE_PROP_BOOL("Zifencei", RISCVCPU, cfg.ext_ifencei, true), DEFINE_PROP_BOOL("Zicsr", RISCVCPU, cfg.ext_icsr, true), -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> When realising the SoC use error_fatal instead of error_abort as the process can fail and report useful information to the user. Currently a user can see this: $ ../qemu/bld/qemu-system-riscv64 -M sifive_u -S -monitor stdio -display none -drive if=pflash QEMU 6.1.93 monitor - type 'help' for more information (qemu) Unexpected error in sifive_u_otp_realize() at ../hw/misc/sifive_u_otp.c:229: qemu-system-riscv64: OTP drive size < 16K Aborted (core dumped) Which this patch addresses Reported-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Tested-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220105213937.1113508-8-alistair.francis@opensource.wdc.com> --- hw/riscv/microchip_pfsoc.c | 2 +- hw/riscv/opentitan.c | 2 +- hw/riscv/sifive_e.c | 2 +- hw/riscv/sifive_u.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) /* Initialize SoC */ object_initialize_child(OBJECT(machine), "soc", &s->soc, TYPE_MICROCHIP_PFSOC); - qdev_realize(DEVICE(&s->soc), NULL, &error_abort); + qdev_realize(DEVICE(&s->soc), NULL, &error_fatal); /* Split RAM into low and high regions using aliases to machine->ram */ mem_low_size = memmap[MICROCHIP_PFSOC_DRAM_LO].size; diff --git a/hw/riscv/opentitan.c b/hw/riscv/opentitan.c index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/opentitan.c +++ b/hw/riscv/opentitan.c @@ -XXX,XX +XXX,XX @@ static void opentitan_board_init(MachineState *machine) /* Initialize SoC */ object_initialize_child(OBJECT(machine), "soc", &s->soc, TYPE_RISCV_IBEX_SOC); - qdev_realize(DEVICE(&s->soc), NULL, &error_abort); + qdev_realize(DEVICE(&s->soc), NULL, &error_fatal); memory_region_add_subregion(sys_mem, memmap[IBEX_DEV_RAM].base, machine->ram); diff --git a/hw/riscv/sifive_e.c b/hw/riscv/sifive_e.c index XXXXXXX..XXXXXXX 100644 --- a/hw/riscv/sifive_e.c +++ b/hw/riscv/sifive_e.c @@ -XXX,XX +XXX,XX @@ static void sifive_e_machine_init(MachineState *machine) /* Initialize SoC */ object_initialize_child(OBJECT(machine), "soc", &s->soc, TYPE_RISCV_E_SOC); - qdev_realize(DEVICE(&s->soc), NULL, &error_abort); + qdev_realize(DEVICE(&s->soc), NULL, &error_fatal); /* Data Tightly Integrated Memory */ memory_region_add_subregion(sys_mem, 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 sifive_u_machine_init(MachineState *machine) &error_abort); object_property_set_str(OBJECT(&s->soc), "cpu-type", machine->cpu_type, &error_abort); - qdev_realize(DEVICE(&s->soc), NULL, &error_abort); + qdev_realize(DEVICE(&s->soc), NULL, &error_fatal); /* register RAM */ memory_region_add_subregion(system_memory, memmap[SIFIVE_U_DEV_DRAM].base, -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> Linux supports up to 32 cores for both 32-bit and 64-bit RISC-V, so let's set that as the maximum for the virt board. Resolves: https://gitlab.com/qemu-project/qemu/-/issues/435 Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Anup Patel <anup.patel@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-Id: <20220105213937.1113508-9-alistair.francis@opensource.wdc.com> --- include/hw/riscv/virt.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) 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 @@ #include "hw/block/flash.h" #include "qom/object.h" -#define VIRT_CPUS_MAX 8 +#define VIRT_CPUS_MAX 32 #define VIRT_SOCKETS_MAX 8 #define TYPE_RISCV_VIRT_MACHINE MACHINE_TYPE_NAME("virt") -- 2.31.1 From: Bin Meng <bmeng.cn@gmail.com> Upgrade OpenSBI from v0.9 to v1.0 and the pre-built bios images. The v1.0 release includes the following commits: ec5274b platform: implement K210 system reset 5487cf0 include: sbi: Simplify HSM state define names 8df1f9a lib: sbi: Use SBI_HSM_STATE_xyz defines instead of SBI_STATE_xyz defines 7c867fd lib: sbi: Rename sbi_hsm_hart_started_mask() function 638c948 lib: sbi: Remove redundant sbi_hsm_hart_started() function ca864a9 lib: sbi: Fix error codes returned by HSM start() and stop() functions 6290a22 include: sbi: Add HSM suspend related defines 4b05df6 lib: sbi: Add sbi_hart_reinit() function 807d71c include: sbi: Add hart_suspend() platform callback 7475689 lib: sbi: Implement SBI HSM suspend function b9cf617 include: sbi: Upgrade SBI implementation version to v0.3 50d4fde lib: Remove redundant sbi_platform_ipi_clear() calls ff5bd94 include: sbi: SBI function IDs for RFENCE extension 22d8ee9 firmware: Use lla to access all global symbols 0f20e8a firmware: Support position independent execution ddad02d lib: sbi: illegal CSR 0x306 access in hpm_allowed() bfc85c7 include: headers: Replace __ASSEMBLY__ with __ASSEMBLER__ 9190ad1 lib/utils: Support the official clint DT bindings ca3f358 lib/utils: Drop the 'compat' parameter of fdt_plic_fixup() 4edc822 lib/utils: Support fixing up the official DT bindings of PLIC 4ef2f5d firware: optimize the exception exit code 3d8a952 lib: fix csr detect support e71a7c1 firmware: Remove redundant add instruction from trap restore path d4a94ea include: types: Add __aligned(x) to define the minimum alignement d0e406f include: sbi: Allow direct initialization via SPIN_LOCK_INIT() 4d8e2f1 lib: sbi: Replace test-and-set locks by ticket locks 70ffc3e lib: sbi: fix atomic_add_return 27a16b1 docs: fix link to OpenPiton documentation b1df1ac lib: sbi: Domains can be registered only before finalizing domains 7495bce lib: sbi: Add sbi_domain_memregion_init() API 4dc0001 lib: sbi: Add sbi_domain_root_add_memregion() API 8b56980 lib: utils/sys: Add CLINT memregion in the root domain fc37c97 lib: sbi: Make the root domain instance global variable e7e4bcd lib: utils: Copy over restricted root domain memregions to FDT domains f41196a lib: sbi: Make sbi_domain_memregion_initfw() a local function c5d0645 lib: utils: Implement "64bit-mmio" property parsing 49e422c lib: utils: reset: Add T-HEAD sample platform reset driver 0d56293 lib: sbi: Fix sbi_domain_root_add_memregion() for merging memregions bf3ef53 firmware: Enable FW_PIC by default 1db8436 platform: Remove platform/thead 6d1642f docs: generic: Add T-HEAD C9xx series processors a3689db lib: sbi: Remove domains_root_regions() platform callback 068ca08 lib: sbi: Simplify console platform operations 559a8f1 lib: sbi: Simplify timer platform operations dc39c7b lib: sbi: Simplify ipi platform operations 043d088 lib: sbi: Simplify system reset platform operations a84a1dd lib: sbi: Simplify HSM platform operations e9a27ab lib: sbi: Show devices provided by platform in boot prints 632e27b docs/platform: sifive_fu540: Update U-Boot defconfig name 117fb6d lib: utils/serial: Add support for Gaisler APBUART 552f53f docs: platform: Sort platform names d4177e7 docs: platform: Describe sifive_fu540 as supported generic platform 26998f3 platform: Remove sifive/fu540 platform f90c4c2 lib: sbi: Have spinlock checks return bool e822b75 lib: utils/serial: Support Synopsys DesignWare APB UART 6139ab2 Makefile: unconditionally disable SSP c9ef2bc lib: utils: Add strncpy macro to libfdt_env.h ee7c2b2 lib: utils/fdt: Don't use sbi_string functions fe92347 lib: utils/fdt: Replace strcmp with strncmp b2dbbc0 lib: Check region base for merging in sbi_domain_root_add_memregion() 54d7def lib: utils: Try other FDT drivers when we see SBI_ENODEV d9ba653 docs: debugging OpenSBI 66c4fca lib: utils: consider ':' in stdout-path f30b189 lib: sbi_scratch: remove owner from sbi_scratch_alloc_offset a03ea2e platform: andes/ae350: Cosmetic fixes in plicsw.c b32fac4 docs/platform: andes-ae350: Fix missing spaces de446cc platform: andes/ae350: Drop plicsw_get_pending() 434198e platform: andes/ae350: Drop plicsw_ipi_sync() 1da3d80 lib: sbi_scratch: zero out scratch memory on all harts 360ab88 lib: utils: missing initialization in thead_reset_init 79f9b42 lib: sbi: Fix GET_F64_REG inline assembly eb90e0a lib: utils/libfdt: Upgrade to v1.6.1 release cdcf907 lib: sign conflict in sbi_tlb_entry_process() 9901794 lib: sign conflict in wake_coldboot_harts() 11c345f lib: simplify sbi_fifo_inplace_update() 4519e29 lib: utils/timer: Add ACLINT MTIMER library 5a049fe lib: utils/ipi: Add ACLINT MSWI library bd5d208 lib: utils: Add FDT parsing API common for both ACLINT and CLINT 56fc5f7 lib: utils/ipi: Add FDT based ACLINT MSWI IPI driver 03d6bb5 lib: utils/timer: Add FDT based ACLINT MTIMER driver a731c7e platform: Replace CLINT library usage with ACLINT library b7f2cd2 lib: utils: reset: unify naming of 'sifive_test' device 197e089 docs/platform: thead-c9xx: Remove FW_PIC=y 17e23b6 platform: generic: Terminate platform.name with null 3e8b31a docs: Add device tree bindings for SBI PMU extension fde28fa lib: sbi: Detect mcountinihibit support at runtime d3a96cc lib: sbi: Remove stray '\' character 0829f2b lib: sbi: Detect number of bits implemented in mhpmcounter 9c9b4ad lib: sbi: Disable m/scounteren & enable mcountinhibit 41ae63c include: Add a list empty check function fd9116b lib: sbi: Remove redundant boot time print statement 49966db lib: sbi: Use csr_read/write_num to read/update PMU counters e7cc7a3 lib: sbi: Add PMU specific platform hooks 13d40f2 lib: sbi: Add PMU support ae72ec0 utils: fdt: Add fdt helper functions to parse PMU DT nodes 37f9b0f lib: sbi: Implement SBI PMU extension 764a17d lib: sbi: Implement firmware counters ec1b8bb lib: sbi: Improve TLB function naming 0e12aa8 platform: generic: Add PMU support 14c7f71 firmware: Minor optimization in _scratch_init() dafaa0f docs: Correct a typo in platform_guide.md abfce9b docs: Make <xyz> visible in the rendered platform guide dcb756b firmware: Remove the sanity checks in fw_save_info() b88b366 firmware: Define a macro for version of struct fw_dynamic_info a76ac44 lib: sbi: Fix sbi_pmu_exit() for systems not having MCOUNTINHIBIT csr 7f1be8a fw_base: Don't mark fw_platform_init as both global and weak 397afe5 fw_base: Put data in .data rather than .text a3d328a firmware: Explicitly pass -pie to the linker, not just the driver 09ad811 firmware: Only default FW_PIC to y if supported 2942777 Makefile: Support building with Clang and LLVM binutils 17729d4 lib: utils: Drop dependency on libgcc by importing part of FreeBSD's libquad e931f38 lib: utils/fdt: Add fdt_parse_phandle_with_args() API 36b8eff lib: utils/gpio: Add generic GPIO configuration library c14f1fe lib: utils/gpio: Add simple FDT based GPIO framework 4c3df2a lib: utils/gpio: Add minimal SiFive GPIO driver e3d6919 lib: utils/reset: Add generic GPIO reset driver 7210e90 firmware: use __SIZEOF_LONG__ for field offsets in fw_dynamic.h f3a8f60 include: types: Use __builtin_offsetof when supported 8a1475b firmware: Remove the unhelpful alignment codes before fdt relocation a4555e5 docs: Document parameters passed to firmware and alignment requirement 2c74dc3 docs: Document FW_PIC compile time option 81eb708 README: Update toolchain information 9890391 Makefile: Manually forward RELAX_FLAG to the assembler when linking with LLD 74db0ac firmware: use _fw_start for load address 217d5e4 generic: fu740: add workaround for CIP-1200 errata ce03c88 lib: utils: remove unused variable in fdt_reset_init e928472 lib: utils: support both of gpio-poweroff, gpio-reset d244f3d lib: sbi: Fix bug in strncmp function when count is 0 47a4765 lib: utils/fdt: Change addr and size to uint64_t e0d1b9d lib: utils/timer: Allow separate base addresses for MTIME and MTIMECMP 7a3a0cc lib: utils: Extend fdt_get_node_addr_size() for multiple register sets f3a0eb8 lib: utils/fdt: Extend fdt_parse_aclint_node() function b35f782 lib: utils/timer: Allow ACLINT MTIMER supporting only 32-bit MMIO 7aa6c9a lib: utils/timer: Simplify MTIMER synchronization 33eac76 lib: sbi: Fix bug in sbi_ecall_rfence that misses checking ee27437 lib: sbi_trap: Restore redirect for access faults b1d3e91 payloads/test: Add support for SBI v0.2 ecalls bd316e2 lib: sbi: Correct typo in faults delegation CSR name c262306 lib: sbi: protect dprintf output with spinlock 1718b16 lib: sbi: Checking fifo validness in sbi_fifo_is_empty and is_full bd35521 lib: sbi: Refine the way to construct platform features 0274a96 lib: utils/reset: Sort fdt_reset driver list 395ff7e lib: utils/reset: Add a sunxi watchdog reset driver 3477f08 lib: sbi: fix ctz bug 12753d2 lib: sbi: add some macros to detect BUG at runtime 51113fe lib: sbi: Add BUG() macro for csr_read/write_num() and misa_string() 72154f4 lib: utils/fdt: Add fdt_parse_timebase_frequency() function 12e7af9 lib: sbi: Add timer frequency to struct sbi_timer_device 6355155 lib: sbi: Print timer frequency at boot time 9d0ab35 lib: sbi: Add generic timer delay loop function fa59dd3 lib: utils/reset: use sbi_timer_mdelay() in gpio reset driver 754d511 lib: utils: identify supported GPIO reset methods 516161c lib: sbi: convert reset to list 9283d50 lib: sbi: add priority for reset handler c38973e lib: sbi: Save context for all non-retentive suspend types 67cbbcb lib: sbi: system reset with invalid parameters 422eda4 Makefile: Add build time and compiler info string 78c2b19 lib: utils/irqchip: Automatically delegate T-HEAD PLIC access 309e8bd lib: utils/reset: Register separate GPIO system reset devices 723aa88 lib: sbi: Refine addr format in sbi_printf c891acc include: sbi_utils: Introduce an helper to get fdt base address 013ba4e lib: sbi: Fix GPA passed to __sbi_hfence_gvma_xyz() functions 0979ffd lib: utils/gpio: use list for drivers 2fe2f55 lib: sbi: move sbi_boot_print_general() 57f094e platform: generic: move fdt_reset_init to final_init be245ac lib: sbi: error handling in fdt_reset_init() a74daf2 riscv: Add new CSRs introduced by Sscofpmf[1] extension 7084ad9 lib: sbi: Update csr_read/write_num for PMU 867c653 lib: sbi: Detect Sscofpmf extension at run time 9134c36 lib: sbi: Delegate PMU counter overflow interrupt to S mode 730f01b lib: sbi: Support sscofpmf extension in OpenSBI 2363f95 lib: sbi: Always enable access for all counters 0c304b6 lib: sbi: Allow programmable counters to monitor cycle/instret events 1e14732 lib: sbi: Reset the mhpmevent value upon counter reset b628cfd lib: sbi: Counter info width should be zero indexed b28f070 lib: sbi: Enable PMU extension for platforms without mcountinhibit 15906a3 lib: utils: Rename the prefix in PMU DT properties b8845e4 lib: sbi: Fix initial value mask while updating the counters 31fe5a7 lib: sbi: Fix PMP address bits detection 94eba23 lib: utils/reset: add priority to gpio reset 1d462e0 lib: utils/reset: separate driver init func 2c964a2 lib: utils/i2c: Add generic I2C configuration library 6ca6bca lib: utils/i2c: Add simple FDT based I2C framework 13a1158 lib: utils/i2c: Add minimal SiFive I2C driver f374496 platform: sifive_fu740: add platform reset driver d335a17 lib: sbi: clear pmpcfg.A before setting in pmp_set() 52af6e4 lib: utils: Add LiteX UART support 22d556d lib: sbi: Fix spelling of "address" in sbi_domain.c 7a22c78 lib: sbi: Fix missing space 7e77706 lib: sbi: Resolve the uninitialized complaint in sbi_pmu 14faee6 lib: sbi: Improve fatal error handling 2428987 lib: pmu: support the event ID encoded by a bitmap. 66fbcc0 docs/platform: spike: Enhance Spike examples 460041c lib: pmu: check SSCOF before masking 69d7e53 Makefile: Fix -msave-restore compile warning with CLANG-10 (or lower) d249d65 lib: sbi: Fix compile errors using -Os option f270359 Makefile: Improve the method to disable -m(no-)save-restore option 2082153 lib: sbi: simplify pmp_set(), pmp_get() d30bde3 firmware: Move memcpy/memset mapping to fw_base.S 48f91ee include: Bump-up version to 1.0 Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- .../opensbi-riscv32-generic-fw_dynamic.bin | Bin 78680 -> 108504 bytes .../opensbi-riscv32-generic-fw_dynamic.elf | Bin 727464 -> 838904 bytes .../opensbi-riscv64-generic-fw_dynamic.bin | Bin 75096 -> 105296 bytes .../opensbi-riscv64-generic-fw_dynamic.elf | Bin 781264 -> 934696 bytes roms/opensbi | 2 +- 5 files changed, 1 insertion(+), 1 deletion(-) diff --git a/pc-bios/opensbi-riscv32-generic-fw_dynamic.bin b/pc-bios/opensbi-riscv32-generic-fw_dynamic.bin index XXXXXXX..XXXXXXX 100644 Binary files a/pc-bios/opensbi-riscv32-generic-fw_dynamic.bin and b/pc-bios/opensbi-riscv32-generic-fw_dynamic.bin differ diff --git a/pc-bios/opensbi-riscv32-generic-fw_dynamic.elf b/pc-bios/opensbi-riscv32-generic-fw_dynamic.elf index XXXXXXX..XXXXXXX 100644 Binary files a/pc-bios/opensbi-riscv32-generic-fw_dynamic.elf and b/pc-bios/opensbi-riscv32-generic-fw_dynamic.elf differ diff --git a/pc-bios/opensbi-riscv64-generic-fw_dynamic.bin b/pc-bios/opensbi-riscv64-generic-fw_dynamic.bin index XXXXXXX..XXXXXXX 100644 Binary files a/pc-bios/opensbi-riscv64-generic-fw_dynamic.bin and b/pc-bios/opensbi-riscv64-generic-fw_dynamic.bin differ diff --git a/pc-bios/opensbi-riscv64-generic-fw_dynamic.elf b/pc-bios/opensbi-riscv64-generic-fw_dynamic.elf index XXXXXXX..XXXXXXX 100644 Binary files a/pc-bios/opensbi-riscv64-generic-fw_dynamic.elf and b/pc-bios/opensbi-riscv64-generic-fw_dynamic.elf differ diff --git a/roms/opensbi b/roms/opensbi index XXXXXXX..XXXXXXX 160000 --- a/roms/opensbi +++ b/roms/opensbi @@ -1 +1 @@ -Subproject commit 234ed8e427f4d92903123199f6590d144e0d9351 +Subproject commit 48f91ee9c960f048c4a7d1da4447d31e04931e38 -- 2.31.1 From: Frank Chang <frank.chang@sifive.com> Vector widening floating-point instructions should use require_scale_rvf() instead of require_rvf() to check whether RVF/RVD is enabled. Signed-off-by: Frank Chang <frank.chang@sifive.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220105022247.21131-2-frank.chang@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn_trans/trans_rvv.c.inc | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/target/riscv/insn_trans/trans_rvv.c.inc b/target/riscv/insn_trans/trans_rvv.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvv.c.inc +++ b/target/riscv/insn_trans/trans_rvv.c.inc @@ -XXX,XX +XXX,XX @@ GEN_OPFVF_TRANS(vfrsub_vf, opfvf_check) static bool opfvv_widen_check(DisasContext *s, arg_rmrr *a) { return require_rvv(s) && - require_rvf(s) && + require_scale_rvf(s) && + (s->sew != MO_8) && vext_check_isa_ill(s) && vext_check_dss(s, a->rd, a->rs1, a->rs2, a->vm); } @@ -XXX,XX +XXX,XX @@ GEN_OPFVV_WIDEN_TRANS(vfwsub_vv, opfvv_widen_check) static bool opfvf_widen_check(DisasContext *s, arg_rmrr *a) { return require_rvv(s) && - require_rvf(s) && + require_scale_rvf(s) && + (s->sew != MO_8) && vext_check_isa_ill(s) && vext_check_ds(s, a->rd, a->rs2, a->vm); } @@ -XXX,XX +XXX,XX @@ GEN_OPFVF_WIDEN_TRANS(vfwsub_vf) static bool opfwv_widen_check(DisasContext *s, arg_rmrr *a) { return require_rvv(s) && - require_rvf(s) && + require_scale_rvf(s) && + (s->sew != MO_8) && vext_check_isa_ill(s) && vext_check_dds(s, a->rd, a->rs1, a->rs2, a->vm); } @@ -XXX,XX +XXX,XX @@ GEN_OPFWV_WIDEN_TRANS(vfwsub_wv) static bool opfwf_widen_check(DisasContext *s, arg_rmrr *a) { return require_rvv(s) && - require_rvf(s) && + require_scale_rvf(s) && + (s->sew != MO_8) && vext_check_isa_ill(s) && vext_check_dd(s, a->rd, a->rs2, a->vm); } -- 2.31.1 From: Frank Chang <frank.chang@sifive.com> vfwcvt.xu.f.v, vfwcvt.x.f.v, vfwcvt.rtz.xu.f.v and vfwcvt.rtz.x.f.v convert single-width floating-point to double-width integer. Therefore, should use require_rvf() to check whether RVF/RVD is enabled. vfwcvt.f.xu.v, vfwcvt.f.x.v convert single-width integer to double-width floating-point, and vfwcvt.f.f.v convert double-width floating-point to single-width floating-point. Therefore, should use require_scale_rvf() to check whether RVF/RVD is enabled. Signed-off-by: Frank Chang <frank.chang@sifive.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220105022247.21131-3-frank.chang@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn_trans/trans_rvv.c.inc | 34 ++++++++++++++++++------- 1 file changed, 25 insertions(+), 9 deletions(-) diff --git a/target/riscv/insn_trans/trans_rvv.c.inc b/target/riscv/insn_trans/trans_rvv.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvv.c.inc +++ b/target/riscv/insn_trans/trans_rvv.c.inc @@ -XXX,XX +XXX,XX @@ GEN_OPFV_CVT_TRANS(vfcvt_rtz_x_f_v, vfcvt_x_f_v, RISCV_FRM_RTZ) static bool opfv_widen_check(DisasContext *s, arg_rmr *a) { return require_rvv(s) && - require_scale_rvf(s) && - (s->sew != MO_8) && vext_check_isa_ill(s) && vext_check_ds(s, a->rd, a->rs2, a->vm); } -#define GEN_OPFV_WIDEN_TRANS(NAME, HELPER, FRM) \ +static bool opxfv_widen_check(DisasContext *s, arg_rmr *a) +{ + return opfv_widen_check(s, a) && + require_rvf(s); +} + +static bool opffv_widen_check(DisasContext *s, arg_rmr *a) +{ + return opfv_widen_check(s, a) && + require_scale_rvf(s) && + (s->sew != MO_8); +} + +#define GEN_OPFV_WIDEN_TRANS(NAME, CHECK, HELPER, FRM) \ static bool trans_##NAME(DisasContext *s, arg_rmr *a) \ { \ - if (opfv_widen_check(s, a)) { \ + if (CHECK(s, a)) { \ if (FRM != RISCV_FRM_DYN) { \ gen_set_rm(s, RISCV_FRM_DYN); \ } \ @@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a) \ return false; \ } -GEN_OPFV_WIDEN_TRANS(vfwcvt_xu_f_v, vfwcvt_xu_f_v, RISCV_FRM_DYN) -GEN_OPFV_WIDEN_TRANS(vfwcvt_x_f_v, vfwcvt_x_f_v, RISCV_FRM_DYN) -GEN_OPFV_WIDEN_TRANS(vfwcvt_f_f_v, vfwcvt_f_f_v, RISCV_FRM_DYN) +GEN_OPFV_WIDEN_TRANS(vfwcvt_xu_f_v, opxfv_widen_check, vfwcvt_xu_f_v, + RISCV_FRM_DYN) +GEN_OPFV_WIDEN_TRANS(vfwcvt_x_f_v, opxfv_widen_check, vfwcvt_x_f_v, + RISCV_FRM_DYN) +GEN_OPFV_WIDEN_TRANS(vfwcvt_f_f_v, opffv_widen_check, vfwcvt_f_f_v, + RISCV_FRM_DYN) /* Reuse the helper functions from vfwcvt.xu.f.v and vfwcvt.x.f.v */ -GEN_OPFV_WIDEN_TRANS(vfwcvt_rtz_xu_f_v, vfwcvt_xu_f_v, RISCV_FRM_RTZ) -GEN_OPFV_WIDEN_TRANS(vfwcvt_rtz_x_f_v, vfwcvt_x_f_v, RISCV_FRM_RTZ) +GEN_OPFV_WIDEN_TRANS(vfwcvt_rtz_xu_f_v, opxfv_widen_check, vfwcvt_xu_f_v, + RISCV_FRM_RTZ) +GEN_OPFV_WIDEN_TRANS(vfwcvt_rtz_x_f_v, opxfv_widen_check, vfwcvt_x_f_v, + RISCV_FRM_RTZ) static bool opfxv_widen_check(DisasContext *s, arg_rmr *a) { -- 2.31.1 From: Frank Chang <frank.chang@sifive.com> vfncvt.f.xu.w, vfncvt.f.x.w convert double-width integer to single-width floating-point. Therefore, should use require_rvf() to check whether RVF/RVD is enabled. vfncvt.f.f.w, vfncvt.rod.f.f.w convert double-width floating-point to single-width integer. Therefore, should use require_scale_rvf() to check whether RVF/RVD is enabled. Signed-off-by: Frank Chang <frank.chang@sifive.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20220105022247.21131-4-frank.chang@sifive.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn_trans/trans_rvv.c.inc | 32 ++++++++++++++++++------- 1 file changed, 24 insertions(+), 8 deletions(-) diff --git a/target/riscv/insn_trans/trans_rvv.c.inc b/target/riscv/insn_trans/trans_rvv.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvv.c.inc +++ b/target/riscv/insn_trans/trans_rvv.c.inc @@ -XXX,XX +XXX,XX @@ GEN_OPFXV_WIDEN_TRANS(vfwcvt_f_x_v) static bool opfv_narrow_check(DisasContext *s, arg_rmr *a) { return require_rvv(s) && - require_rvf(s) && - (s->sew != MO_64) && vext_check_isa_ill(s) && /* OPFV narrowing instructions ignore vs1 check */ vext_check_sd(s, a->rd, a->rs2, a->vm); } -#define GEN_OPFV_NARROW_TRANS(NAME, HELPER, FRM) \ +static bool opfxv_narrow_check(DisasContext *s, arg_rmr *a) +{ + return opfv_narrow_check(s, a) && + require_rvf(s) && + (s->sew != MO_64); +} + +static bool opffv_narrow_check(DisasContext *s, arg_rmr *a) +{ + return opfv_narrow_check(s, a) && + require_scale_rvf(s) && + (s->sew != MO_8); +} + +#define GEN_OPFV_NARROW_TRANS(NAME, CHECK, HELPER, FRM) \ static bool trans_##NAME(DisasContext *s, arg_rmr *a) \ { \ - if (opfv_narrow_check(s, a)) { \ + if (CHECK(s, a)) { \ if (FRM != RISCV_FRM_DYN) { \ gen_set_rm(s, RISCV_FRM_DYN); \ } \ @@ -XXX,XX +XXX,XX @@ static bool trans_##NAME(DisasContext *s, arg_rmr *a) \ return false; \ } -GEN_OPFV_NARROW_TRANS(vfncvt_f_xu_w, vfncvt_f_xu_w, RISCV_FRM_DYN) -GEN_OPFV_NARROW_TRANS(vfncvt_f_x_w, vfncvt_f_x_w, RISCV_FRM_DYN) -GEN_OPFV_NARROW_TRANS(vfncvt_f_f_w, vfncvt_f_f_w, RISCV_FRM_DYN) +GEN_OPFV_NARROW_TRANS(vfncvt_f_xu_w, opfxv_narrow_check, vfncvt_f_xu_w, + RISCV_FRM_DYN) +GEN_OPFV_NARROW_TRANS(vfncvt_f_x_w, opfxv_narrow_check, vfncvt_f_x_w, + RISCV_FRM_DYN) +GEN_OPFV_NARROW_TRANS(vfncvt_f_f_w, opffv_narrow_check, vfncvt_f_f_w, + RISCV_FRM_DYN) /* Reuse the helper function from vfncvt.f.f.w */ -GEN_OPFV_NARROW_TRANS(vfncvt_rod_f_f_w, vfncvt_f_f_w, RISCV_FRM_ROD) +GEN_OPFV_NARROW_TRANS(vfncvt_rod_f_f_w, opffv_narrow_check, vfncvt_f_f_w, + RISCV_FRM_ROD) static bool opxfv_narrow_check(DisasContext *s, arg_rmr *a) { -- 2.31.1 From: Philipp Tomsich <philipp.tomsich@vrull.eu> When commit 0643c12e4b dropped the 'x-' prefix for Zb[abcs] and set them to be enabled by default, the comment about experimental extensions was kept in place above them. This moves it down a few lines to only cover experimental extensions. References: 0643c12e4b ("target/riscv: Enable bitmanip Zb[abcs] instructions") Signed-off-by: Philipp Tomsich <philipp.tomsich@vrull.eu> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106134020.1628889-1-philipp.tomsich@vrull.eu Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -XXX,XX +XXX,XX @@ static Property riscv_cpu_properties[] = { DEFINE_PROP_UINT16("vlen", RISCVCPU, cfg.vlen, 128), DEFINE_PROP_UINT16("elen", RISCVCPU, cfg.elen, 64), - /* These are experimental so mark with 'x-' */ 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("zbs", RISCVCPU, cfg.ext_zbs, true), + + /* These are experimental so mark with 'x-' */ DEFINE_PROP_BOOL("x-j", RISCVCPU, cfg.ext_j, false), /* ePMP 0.9.3 */ DEFINE_PROP_BOOL("x-epmp", RISCVCPU, cfg.epmp, false), -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Renaming defines for quad in their various forms so that their signedness is now explicit. Done using git grep as suggested by Philippe, with a bit of hand edition to keep assignments aligned. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-2-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- include/exec/memop.h | 8 +-- include/tcg/tcg-op.h | 4 +- target/arm/translate-a32.h | 4 +- accel/tcg/cputlb.c | 30 +++++------ accel/tcg/user-exec.c | 8 +-- target/alpha/translate.c | 32 ++++++------ target/arm/helper-a64.c | 8 +-- target/arm/translate-a64.c | 8 +-- target/arm/translate-neon.c | 6 +-- target/arm/translate-sve.c | 10 ++-- target/arm/translate-vfp.c | 8 +-- target/arm/translate.c | 2 +- target/cris/translate.c | 2 +- target/hppa/translate.c | 4 +- target/i386/tcg/mem_helper.c | 2 +- target/i386/tcg/translate.c | 36 +++++++------- target/m68k/op_helper.c | 2 +- target/mips/tcg/translate.c | 58 +++++++++++----------- target/mips/tcg/tx79_translate.c | 8 +-- target/ppc/translate.c | 32 ++++++------ target/s390x/tcg/mem_helper.c | 8 +-- target/s390x/tcg/translate.c | 8 +-- target/sh4/translate.c | 12 ++--- target/sparc/translate.c | 36 +++++++------- target/tricore/translate.c | 4 +- target/xtensa/translate.c | 4 +- tcg/tcg.c | 4 +- tcg/tci.c | 16 +++--- accel/tcg/ldst_common.c.inc | 8 +-- target/mips/tcg/micromips_translate.c.inc | 10 ++-- target/ppc/translate/fixedpoint-impl.c.inc | 22 ++++---- target/ppc/translate/fp-impl.c.inc | 4 +- target/ppc/translate/vsx-impl.c.inc | 42 ++++++++-------- target/riscv/insn_trans/trans_rva.c.inc | 22 ++++---- target/riscv/insn_trans/trans_rvd.c.inc | 4 +- target/riscv/insn_trans/trans_rvh.c.inc | 4 +- target/riscv/insn_trans/trans_rvi.c.inc | 4 +- target/s390x/tcg/translate_vx.c.inc | 18 +++---- tcg/aarch64/tcg-target.c.inc | 2 +- tcg/arm/tcg-target.c.inc | 10 ++-- tcg/i386/tcg-target.c.inc | 12 ++--- tcg/mips/tcg-target.c.inc | 12 ++--- tcg/ppc/tcg-target.c.inc | 16 +++--- tcg/riscv/tcg-target.c.inc | 6 +-- tcg/s390x/tcg-target.c.inc | 18 +++---- tcg/sparc/tcg-target.c.inc | 16 +++--- target/s390x/tcg/insn-data.def | 28 +++++------ 47 files changed, 311 insertions(+), 311 deletions(-) diff --git a/include/exec/memop.h b/include/exec/memop.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/memop.h +++ b/include/exec/memop.h @@ -XXX,XX +XXX,XX @@ typedef enum MemOp { MO_UB = MO_8, MO_UW = MO_16, MO_UL = MO_32, + MO_UQ = MO_64, MO_SB = MO_SIGN | MO_8, MO_SW = MO_SIGN | MO_16, MO_SL = MO_SIGN | MO_32, - MO_Q = MO_64, MO_LEUW = MO_LE | MO_UW, MO_LEUL = MO_LE | MO_UL, + MO_LEUQ = MO_LE | MO_UQ, MO_LESW = MO_LE | MO_SW, MO_LESL = MO_LE | MO_SL, - MO_LEQ = MO_LE | MO_Q, MO_BEUW = MO_BE | MO_UW, MO_BEUL = MO_BE | MO_UL, + MO_BEUQ = MO_BE | MO_UQ, MO_BESW = MO_BE | MO_SW, MO_BESL = MO_BE | MO_SL, - MO_BEQ = MO_BE | MO_Q, #ifdef NEED_CPU_H MO_TEUW = MO_TE | MO_UW, MO_TEUL = MO_TE | MO_UL, + MO_TEUQ = MO_TE | MO_UQ, MO_TESW = MO_TE | MO_SW, MO_TESL = MO_TE | MO_SL, - MO_TEQ = MO_TE | MO_Q, #endif MO_SSIZE = MO_SIZE | MO_SIGN, diff --git a/include/tcg/tcg-op.h b/include/tcg/tcg-op.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg-op.h +++ b/include/tcg/tcg-op.h @@ -XXX,XX +XXX,XX @@ static inline void tcg_gen_qemu_ld32s(TCGv ret, TCGv addr, int mem_index) static inline void tcg_gen_qemu_ld64(TCGv_i64 ret, TCGv addr, int mem_index) { - tcg_gen_qemu_ld_i64(ret, addr, mem_index, MO_TEQ); + tcg_gen_qemu_ld_i64(ret, addr, mem_index, MO_TEUQ); } static inline void tcg_gen_qemu_st8(TCGv arg, TCGv addr, int mem_index) @@ -XXX,XX +XXX,XX @@ static inline void tcg_gen_qemu_st32(TCGv arg, TCGv addr, int mem_index) static inline void tcg_gen_qemu_st64(TCGv_i64 arg, TCGv addr, int mem_index) { - tcg_gen_qemu_st_i64(arg, addr, mem_index, MO_TEQ); + tcg_gen_qemu_st_i64(arg, addr, mem_index, MO_TEUQ); } void tcg_gen_atomic_cmpxchg_i32(TCGv_i32, TCGv, TCGv_i32, TCGv_i32, diff --git a/target/arm/translate-a32.h b/target/arm/translate-a32.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a32.h +++ b/target/arm/translate-a32.h @@ -XXX,XX +XXX,XX @@ void gen_aa32_st_i64(DisasContext *s, TCGv_i64 val, TCGv_i32 a32, static inline void gen_aa32_ld64(DisasContext *s, TCGv_i64 val, TCGv_i32 a32, int index) { - gen_aa32_ld_i64(s, val, a32, index, MO_Q); + gen_aa32_ld_i64(s, val, a32, index, MO_UQ); } static inline void gen_aa32_st64(DisasContext *s, TCGv_i64 val, TCGv_i32 a32, int index) { - gen_aa32_st_i64(s, val, a32, index, MO_Q); + gen_aa32_st_i64(s, val, a32, index, MO_UQ); } DO_GEN_LD(8u, MO_UB) diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ load_memop(const void *haddr, MemOp op) return (uint32_t)ldl_be_p(haddr); case MO_LEUL: return (uint32_t)ldl_le_p(haddr); - case MO_BEQ: + case MO_BEUQ: return ldq_be_p(haddr); - case MO_LEQ: + case MO_LEUQ: return ldq_le_p(haddr); default: qemu_build_not_reached(); @@ -XXX,XX +XXX,XX @@ tcg_target_ulong helper_be_ldul_mmu(CPUArchState *env, target_ulong addr, uint64_t helper_le_ldq_mmu(CPUArchState *env, target_ulong addr, MemOpIdx oi, uintptr_t retaddr) { - validate_memop(oi, MO_LEQ); - return load_helper(env, addr, oi, retaddr, MO_LEQ, false, + validate_memop(oi, MO_LEUQ); + return load_helper(env, addr, oi, retaddr, MO_LEUQ, false, helper_le_ldq_mmu); } uint64_t helper_be_ldq_mmu(CPUArchState *env, target_ulong addr, MemOpIdx oi, uintptr_t retaddr) { - validate_memop(oi, MO_BEQ); - return load_helper(env, addr, oi, retaddr, MO_BEQ, false, + validate_memop(oi, MO_BEUQ); + return load_helper(env, addr, oi, retaddr, MO_BEUQ, false, helper_be_ldq_mmu); } @@ -XXX,XX +XXX,XX @@ uint32_t cpu_ldl_be_mmu(CPUArchState *env, abi_ptr addr, uint64_t cpu_ldq_be_mmu(CPUArchState *env, abi_ptr addr, MemOpIdx oi, uintptr_t ra) { - return cpu_load_helper(env, addr, oi, MO_BEQ, helper_be_ldq_mmu); + return cpu_load_helper(env, addr, oi, MO_BEUQ, helper_be_ldq_mmu); } uint16_t cpu_ldw_le_mmu(CPUArchState *env, abi_ptr addr, @@ -XXX,XX +XXX,XX @@ store_memop(void *haddr, uint64_t val, MemOp op) case MO_LEUL: stl_le_p(haddr, val); break; - case MO_BEQ: + case MO_BEUQ: stq_be_p(haddr, val); break; - case MO_LEQ: + case MO_LEUQ: stq_le_p(haddr, val); break; default: @@ -XXX,XX +XXX,XX @@ void helper_be_stl_mmu(CPUArchState *env, target_ulong addr, uint32_t val, void helper_le_stq_mmu(CPUArchState *env, target_ulong addr, uint64_t val, MemOpIdx oi, uintptr_t retaddr) { - validate_memop(oi, MO_LEQ); - store_helper(env, addr, val, oi, retaddr, MO_LEQ); + validate_memop(oi, MO_LEUQ); + store_helper(env, addr, val, oi, retaddr, MO_LEUQ); } void helper_be_stq_mmu(CPUArchState *env, target_ulong addr, uint64_t val, MemOpIdx oi, uintptr_t retaddr) { - validate_memop(oi, MO_BEQ); - store_helper(env, addr, val, oi, retaddr, MO_BEQ); + validate_memop(oi, MO_BEUQ); + store_helper(env, addr, val, oi, retaddr, MO_BEUQ); } /* @@ -XXX,XX +XXX,XX @@ uint32_t cpu_ldl_code(CPUArchState *env, abi_ptr addr) static uint64_t full_ldq_code(CPUArchState *env, target_ulong addr, MemOpIdx oi, uintptr_t retaddr) { - return load_helper(env, addr, oi, retaddr, MO_TEQ, true, full_ldq_code); + return load_helper(env, addr, oi, retaddr, MO_TEUQ, true, full_ldq_code); } uint64_t cpu_ldq_code(CPUArchState *env, abi_ptr addr) { - MemOpIdx oi = make_memop_idx(MO_TEQ, cpu_mmu_index(env, true)); + MemOpIdx oi = make_memop_idx(MO_TEUQ, cpu_mmu_index(env, true)); return full_ldq_code(env, addr, oi, 0); } diff --git a/accel/tcg/user-exec.c b/accel/tcg/user-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/user-exec.c +++ b/accel/tcg/user-exec.c @@ -XXX,XX +XXX,XX @@ uint64_t cpu_ldq_be_mmu(CPUArchState *env, abi_ptr addr, void *haddr; uint64_t ret; - validate_memop(oi, MO_BEQ); + validate_memop(oi, MO_BEUQ); trace_guest_ld_before_exec(env_cpu(env), addr, oi); haddr = cpu_mmu_lookup(env, addr, oi, ra, MMU_DATA_LOAD); ret = ldq_be_p(haddr); @@ -XXX,XX +XXX,XX @@ uint64_t cpu_ldq_le_mmu(CPUArchState *env, abi_ptr addr, void *haddr; uint64_t ret; - validate_memop(oi, MO_LEQ); + validate_memop(oi, MO_LEUQ); trace_guest_ld_before_exec(env_cpu(env), addr, oi); haddr = cpu_mmu_lookup(env, addr, oi, ra, MMU_DATA_LOAD); ret = ldq_le_p(haddr); @@ -XXX,XX +XXX,XX @@ void cpu_stq_be_mmu(CPUArchState *env, abi_ptr addr, uint64_t val, { void *haddr; - validate_memop(oi, MO_BEQ); + validate_memop(oi, MO_BEUQ); trace_guest_st_before_exec(env_cpu(env), addr, oi); haddr = cpu_mmu_lookup(env, addr, oi, ra, MMU_DATA_STORE); stq_be_p(haddr, val); @@ -XXX,XX +XXX,XX @@ void cpu_stq_le_mmu(CPUArchState *env, abi_ptr addr, uint64_t val, { void *haddr; - validate_memop(oi, MO_LEQ); + validate_memop(oi, MO_LEUQ); trace_guest_st_before_exec(env_cpu(env), addr, oi); haddr = cpu_mmu_lookup(env, addr, oi, ra, MMU_DATA_STORE); stq_le_p(haddr, val); diff --git a/target/alpha/translate.c b/target/alpha/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/alpha/translate.c +++ b/target/alpha/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_ldf(DisasContext *ctx, TCGv dest, TCGv addr) static void gen_ldg(DisasContext *ctx, TCGv dest, TCGv addr) { TCGv tmp = tcg_temp_new(); - tcg_gen_qemu_ld_i64(tmp, addr, ctx->mem_idx, MO_LEQ | UNALIGN(ctx)); + tcg_gen_qemu_ld_i64(tmp, addr, ctx->mem_idx, MO_LEUQ | UNALIGN(ctx)); gen_helper_memory_to_g(dest, tmp); tcg_temp_free(tmp); } @@ -XXX,XX +XXX,XX @@ static void gen_lds(DisasContext *ctx, TCGv dest, TCGv addr) static void gen_ldt(DisasContext *ctx, TCGv dest, TCGv addr) { - tcg_gen_qemu_ld_i64(dest, addr, ctx->mem_idx, MO_LEQ | UNALIGN(ctx)); + tcg_gen_qemu_ld_i64(dest, addr, ctx->mem_idx, MO_LEUQ | UNALIGN(ctx)); } static void gen_load_fp(DisasContext *ctx, int ra, int rb, int32_t disp16, @@ -XXX,XX +XXX,XX @@ static void gen_stg(DisasContext *ctx, TCGv src, TCGv addr) { TCGv tmp = tcg_temp_new(); gen_helper_g_to_memory(tmp, src); - tcg_gen_qemu_st_i64(tmp, addr, ctx->mem_idx, MO_LEQ | UNALIGN(ctx)); + tcg_gen_qemu_st_i64(tmp, addr, ctx->mem_idx, MO_LEUQ | UNALIGN(ctx)); tcg_temp_free(tmp); } @@ -XXX,XX +XXX,XX @@ static void gen_sts(DisasContext *ctx, TCGv src, TCGv addr) static void gen_stt(DisasContext *ctx, TCGv src, TCGv addr) { - tcg_gen_qemu_st_i64(src, addr, ctx->mem_idx, MO_LEQ | UNALIGN(ctx)); + tcg_gen_qemu_st_i64(src, addr, ctx->mem_idx, MO_LEUQ | UNALIGN(ctx)); } static void gen_store_fp(DisasContext *ctx, int ra, int rb, int32_t disp16, @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) break; case 0x0B: /* LDQ_U */ - gen_load_int(ctx, ra, rb, disp16, MO_LEQ, 1, 0); + gen_load_int(ctx, ra, rb, disp16, MO_LEUQ, 1, 0); break; case 0x0C: /* LDWU */ @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) break; case 0x0F: /* STQ_U */ - gen_store_int(ctx, ra, rb, disp16, MO_LEQ, 1); + gen_store_int(ctx, ra, rb, disp16, MO_LEUQ, 1); break; case 0x10: @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) break; case 0x1: /* Quadword physical access (hw_ldq/p) */ - tcg_gen_qemu_ld_i64(va, addr, MMU_PHYS_IDX, MO_LEQ); + tcg_gen_qemu_ld_i64(va, addr, MMU_PHYS_IDX, MO_LEUQ); break; case 0x2: /* Longword physical access with lock (hw_ldl_l/p) */ @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) break; case 0x3: /* Quadword physical access with lock (hw_ldq_l/p) */ - tcg_gen_qemu_ld_i64(va, addr, MMU_PHYS_IDX, MO_LEQ); + tcg_gen_qemu_ld_i64(va, addr, MMU_PHYS_IDX, MO_LEUQ); tcg_gen_mov_i64(cpu_lock_addr, addr); tcg_gen_mov_i64(cpu_lock_value, va); break; @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) break; case 0xB: /* Quadword virtual access with protection check (hw_ldq/w) */ - tcg_gen_qemu_ld_i64(va, addr, MMU_KERNEL_IDX, MO_LEQ); + tcg_gen_qemu_ld_i64(va, addr, MMU_KERNEL_IDX, MO_LEUQ); break; case 0xC: /* Longword virtual access with alt access mode (hw_ldl/a)*/ @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) case 0xF: /* Quadword virtual access with alternate access mode and protection checks (hw_ldq/wa) */ - tcg_gen_qemu_ld_i64(va, addr, MMU_USER_IDX, MO_LEQ); + tcg_gen_qemu_ld_i64(va, addr, MMU_USER_IDX, MO_LEUQ); break; } tcg_temp_free(addr); @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) vb = load_gpr(ctx, rb); tmp = tcg_temp_new(); tcg_gen_addi_i64(tmp, vb, disp12); - tcg_gen_qemu_st_i64(va, tmp, MMU_PHYS_IDX, MO_LEQ); + tcg_gen_qemu_st_i64(va, tmp, MMU_PHYS_IDX, MO_LEUQ); tcg_temp_free(tmp); break; case 0x2: @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) case 0x3: /* Quadword physical access with lock */ ret = gen_store_conditional(ctx, ra, rb, disp12, - MMU_PHYS_IDX, MO_LEQ); + MMU_PHYS_IDX, MO_LEUQ); break; case 0x4: /* Longword virtual access */ @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) break; case 0x29: /* LDQ */ - gen_load_int(ctx, ra, rb, disp16, MO_LEQ, 0, 0); + gen_load_int(ctx, ra, rb, disp16, MO_LEUQ, 0, 0); break; case 0x2A: /* LDL_L */ @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) break; case 0x2B: /* LDQ_L */ - gen_load_int(ctx, ra, rb, disp16, MO_LEQ, 0, 1); + gen_load_int(ctx, ra, rb, disp16, MO_LEUQ, 0, 1); break; case 0x2C: /* STL */ @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) break; case 0x2D: /* STQ */ - gen_store_int(ctx, ra, rb, disp16, MO_LEQ, 0); + gen_store_int(ctx, ra, rb, disp16, MO_LEUQ, 0); break; case 0x2E: /* STL_C */ @@ -XXX,XX +XXX,XX @@ static DisasJumpType translate_one(DisasContext *ctx, uint32_t insn) case 0x2F: /* STQ_C */ ret = gen_store_conditional(ctx, ra, rb, disp16, - ctx->mem_idx, MO_LEQ); + ctx->mem_idx, MO_LEUQ); break; case 0x30: /* BR */ diff --git a/target/arm/helper-a64.c b/target/arm/helper-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper-a64.c +++ b/target/arm/helper-a64.c @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(paired_cmpxchg64_le)(CPUARMState *env, uint64_t addr, uint64_t o0, o1; bool success; int mem_idx = cpu_mmu_index(env, false); - MemOpIdx oi0 = make_memop_idx(MO_LEQ | MO_ALIGN_16, mem_idx); - MemOpIdx oi1 = make_memop_idx(MO_LEQ, mem_idx); + MemOpIdx oi0 = make_memop_idx(MO_LEUQ | MO_ALIGN_16, mem_idx); + MemOpIdx oi1 = make_memop_idx(MO_LEUQ, mem_idx); o0 = cpu_ldq_le_mmu(env, addr + 0, oi0, ra); o1 = cpu_ldq_le_mmu(env, addr + 8, oi1, ra); @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(paired_cmpxchg64_be)(CPUARMState *env, uint64_t addr, uint64_t o0, o1; bool success; int mem_idx = cpu_mmu_index(env, false); - MemOpIdx oi0 = make_memop_idx(MO_BEQ | MO_ALIGN_16, mem_idx); - MemOpIdx oi1 = make_memop_idx(MO_BEQ, mem_idx); + MemOpIdx oi0 = make_memop_idx(MO_BEUQ | MO_ALIGN_16, mem_idx); + MemOpIdx oi1 = make_memop_idx(MO_BEUQ, mem_idx); o1 = cpu_ldq_be_mmu(env, addr + 0, oi0, ra); o0 = cpu_ldq_be_mmu(env, addr + 8, oi1, ra); diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void do_fp_st(DisasContext *s, int srcidx, TCGv_i64 tcg_addr, int size) tcg_gen_ld_i64(tmphi, cpu_env, fp_reg_hi_offset(s, srcidx)); - mop = s->be_data | MO_Q; + mop = s->be_data | MO_UQ; tcg_gen_qemu_st_i64(be ? tmphi : tmplo, tcg_addr, get_mem_index(s), mop | (s->align_mem ? MO_ALIGN_16 : 0)); tcg_gen_addi_i64(tcg_hiaddr, tcg_addr, 8); @@ -XXX,XX +XXX,XX @@ static void do_fp_ld(DisasContext *s, int destidx, TCGv_i64 tcg_addr, int size) tmphi = tcg_temp_new_i64(); tcg_hiaddr = tcg_temp_new_i64(); - mop = s->be_data | MO_Q; + mop = s->be_data | MO_UQ; tcg_gen_qemu_ld_i64(be ? tmphi : tmplo, tcg_addr, get_mem_index(s), mop | (s->align_mem ? MO_ALIGN_16 : 0)); tcg_gen_addi_i64(tcg_hiaddr, tcg_addr, 8); @@ -XXX,XX +XXX,XX @@ static void disas_ldst_tag(DisasContext *s, uint32_t insn) int i, n = (1 + is_pair) << LOG2_TAG_GRANULE; tcg_gen_qemu_st_i64(tcg_zero, clean_addr, mem_index, - MO_Q | MO_ALIGN_16); + MO_UQ | MO_ALIGN_16); for (i = 8; i < n; i += 8) { tcg_gen_addi_i64(clean_addr, clean_addr, 8); - tcg_gen_qemu_st_i64(tcg_zero, clean_addr, mem_index, MO_Q); + tcg_gen_qemu_st_i64(tcg_zero, clean_addr, mem_index, MO_UQ); } tcg_temp_free_i64(tcg_zero); } diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.c +++ b/target/arm/translate-neon.c @@ -XXX,XX +XXX,XX @@ static void neon_load_element64(TCGv_i64 var, int reg, int ele, MemOp mop) case MO_UL: tcg_gen_ld32u_i64(var, cpu_env, offset); break; - case MO_Q: + case MO_UQ: tcg_gen_ld_i64(var, cpu_env, offset); break; default: @@ -XXX,XX +XXX,XX @@ static bool do_prewiden_3d(DisasContext *s, arg_3diff *a, return false; } - if ((a->vd & 1) || (src1_mop == MO_Q && (a->vn & 1))) { + if ((a->vd & 1) || (src1_mop == MO_UQ && (a->vn & 1))) { return false; } @@ -XXX,XX +XXX,XX @@ static bool do_prewiden_3d(DisasContext *s, arg_3diff *a, }; \ int narrow_mop = a->size == MO_32 ? MO_32 | SIGN : -1; \ return do_prewiden_3d(s, a, widenfn[a->size], addfn[a->size], \ - SRC1WIDE ? MO_Q : narrow_mop, \ + SRC1WIDE ? MO_UQ : narrow_mop, \ narrow_mop); \ } diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static void do_ldr(DisasContext *s, uint32_t vofs, int len, int rn, int imm) t0 = tcg_temp_new_i64(); for (i = 0; i < len_align; i += 8) { - tcg_gen_qemu_ld_i64(t0, clean_addr, midx, MO_LEQ); + tcg_gen_qemu_ld_i64(t0, clean_addr, midx, MO_LEUQ); tcg_gen_st_i64(t0, cpu_env, vofs + i); tcg_gen_addi_i64(clean_addr, clean_addr, 8); } @@ -XXX,XX +XXX,XX @@ static void do_ldr(DisasContext *s, uint32_t vofs, int len, int rn, int imm) gen_set_label(loop); t0 = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(t0, clean_addr, midx, MO_LEQ); + tcg_gen_qemu_ld_i64(t0, clean_addr, midx, MO_LEUQ); tcg_gen_addi_i64(clean_addr, clean_addr, 8); tp = tcg_temp_new_ptr(); @@ -XXX,XX +XXX,XX @@ static void do_str(DisasContext *s, uint32_t vofs, int len, int rn, int imm) t0 = tcg_temp_new_i64(); for (i = 0; i < len_align; i += 8) { tcg_gen_ld_i64(t0, cpu_env, vofs + i); - tcg_gen_qemu_st_i64(t0, clean_addr, midx, MO_LEQ); + tcg_gen_qemu_st_i64(t0, clean_addr, midx, MO_LEUQ); tcg_gen_addi_i64(clean_addr, clean_addr, 8); } tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ static void do_str(DisasContext *s, uint32_t vofs, int len, int rn, int imm) tcg_gen_addi_ptr(i, i, 8); tcg_temp_free_ptr(tp); - tcg_gen_qemu_st_i64(t0, clean_addr, midx, MO_LEQ); + tcg_gen_qemu_st_i64(t0, clean_addr, midx, MO_LEUQ); tcg_gen_addi_i64(clean_addr, clean_addr, 8); tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ static const MemOp dtype_mop[16] = { MO_UB, MO_UB, MO_UB, MO_UB, MO_SL, MO_UW, MO_UW, MO_UW, MO_SW, MO_SW, MO_UL, MO_UL, - MO_SB, MO_SB, MO_SB, MO_Q + MO_SB, MO_SB, MO_SB, MO_UQ }; #define dtype_msz(x) (dtype_mop[x] & MO_SIZE) diff --git a/target/arm/translate-vfp.c b/target/arm/translate-vfp.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-vfp.c +++ b/target/arm/translate-vfp.c @@ -XXX,XX +XXX,XX @@ static bool trans_VLDR_VSTR_dp(DisasContext *s, arg_VLDR_VSTR_dp *a) addr = add_reg_for_lit(s, a->rn, offset); tmp = tcg_temp_new_i64(); if (a->l) { - gen_aa32_ld_i64(s, tmp, addr, get_mem_index(s), MO_Q | MO_ALIGN_4); + gen_aa32_ld_i64(s, tmp, addr, get_mem_index(s), MO_UQ | MO_ALIGN_4); vfp_store_reg64(tmp, a->vd); } else { vfp_load_reg64(tmp, a->vd); - gen_aa32_st_i64(s, tmp, addr, get_mem_index(s), MO_Q | MO_ALIGN_4); + gen_aa32_st_i64(s, tmp, addr, get_mem_index(s), MO_UQ | MO_ALIGN_4); } tcg_temp_free_i64(tmp); tcg_temp_free_i32(addr); @@ -XXX,XX +XXX,XX @@ static bool trans_VLDM_VSTM_dp(DisasContext *s, arg_VLDM_VSTM_dp *a) for (i = 0; i < n; i++) { if (a->l) { /* load */ - gen_aa32_ld_i64(s, tmp, addr, get_mem_index(s), MO_Q | MO_ALIGN_4); + gen_aa32_ld_i64(s, tmp, addr, get_mem_index(s), MO_UQ | MO_ALIGN_4); vfp_store_reg64(tmp, a->vd + i); } else { /* store */ vfp_load_reg64(tmp, a->vd + i); - gen_aa32_st_i64(s, tmp, addr, get_mem_index(s), MO_Q | MO_ALIGN_4); + gen_aa32_st_i64(s, tmp, addr, get_mem_index(s), MO_UQ | MO_ALIGN_4); } tcg_gen_addi_i32(addr, addr, offset); } diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ void read_neon_element64(TCGv_i64 dest, int reg, int ele, MemOp memop) case MO_UL: tcg_gen_ld32u_i64(dest, cpu_env, off); break; - case MO_Q: + case MO_UQ: tcg_gen_ld_i64(dest, cpu_env, off); break; default: diff --git a/target/cris/translate.c b/target/cris/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/cris/translate.c +++ b/target/cris/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_load64(DisasContext *dc, TCGv_i64 dst, TCGv addr) cris_store_direct_jmp(dc); } - tcg_gen_qemu_ld_i64(dst, addr, mem_index, MO_TEQ); + tcg_gen_qemu_ld_i64(dst, addr, mem_index, MO_TEUQ); } static void gen_load(DisasContext *dc, TCGv dst, TCGv addr, diff --git a/target/hppa/translate.c b/target/hppa/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/hppa/translate.c +++ b/target/hppa/translate.c @@ -XXX,XX +XXX,XX @@ static bool do_floadd(DisasContext *ctx, unsigned rt, unsigned rb, nullify_over(ctx); tmp = tcg_temp_new_i64(); - do_load_64(ctx, tmp, rb, rx, scale, disp, sp, modify, MO_TEQ); + do_load_64(ctx, tmp, rb, rx, scale, disp, sp, modify, MO_TEUQ); save_frd(rt, tmp); tcg_temp_free_i64(tmp); @@ -XXX,XX +XXX,XX @@ static bool do_fstored(DisasContext *ctx, unsigned rt, unsigned rb, nullify_over(ctx); tmp = load_frd(rt); - do_store_64(ctx, tmp, rb, rx, scale, disp, sp, modify, MO_TEQ); + do_store_64(ctx, tmp, rb, rx, scale, disp, sp, modify, MO_TEUQ); tcg_temp_free_i64(tmp); return nullify_end(ctx); diff --git a/target/i386/tcg/mem_helper.c b/target/i386/tcg/mem_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/mem_helper.c +++ b/target/i386/tcg/mem_helper.c @@ -XXX,XX +XXX,XX @@ void helper_cmpxchg8b(CPUX86State *env, target_ulong a0) { uintptr_t ra = GETPC(); int mem_idx = cpu_mmu_index(env, false); - MemOpIdx oi = make_memop_idx(MO_TEQ, mem_idx); + MemOpIdx oi = make_memop_idx(MO_TEUQ, mem_idx); oldv = cpu_atomic_cmpxchgq_le_mmu(env, a0, cmpv, newv, oi, ra); } diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/translate.c +++ b/target/i386/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_jmp(DisasContext *s, target_ulong eip) static inline void gen_ldq_env_A0(DisasContext *s, int offset) { - tcg_gen_qemu_ld_i64(s->tmp1_i64, s->A0, s->mem_index, MO_LEQ); + tcg_gen_qemu_ld_i64(s->tmp1_i64, s->A0, s->mem_index, MO_LEUQ); tcg_gen_st_i64(s->tmp1_i64, cpu_env, offset); } static inline void gen_stq_env_A0(DisasContext *s, int offset) { tcg_gen_ld_i64(s->tmp1_i64, cpu_env, offset); - tcg_gen_qemu_st_i64(s->tmp1_i64, s->A0, s->mem_index, MO_LEQ); + tcg_gen_qemu_st_i64(s->tmp1_i64, s->A0, s->mem_index, MO_LEUQ); } static inline void gen_ldo_env_A0(DisasContext *s, int offset) { int mem_index = s->mem_index; - tcg_gen_qemu_ld_i64(s->tmp1_i64, s->A0, mem_index, MO_LEQ); + tcg_gen_qemu_ld_i64(s->tmp1_i64, s->A0, mem_index, MO_LEUQ); tcg_gen_st_i64(s->tmp1_i64, cpu_env, offset + offsetof(ZMMReg, ZMM_Q(0))); tcg_gen_addi_tl(s->tmp0, s->A0, 8); - tcg_gen_qemu_ld_i64(s->tmp1_i64, s->tmp0, mem_index, MO_LEQ); + tcg_gen_qemu_ld_i64(s->tmp1_i64, s->tmp0, mem_index, MO_LEUQ); tcg_gen_st_i64(s->tmp1_i64, cpu_env, offset + offsetof(ZMMReg, ZMM_Q(1))); } @@ -XXX,XX +XXX,XX @@ static inline void gen_sto_env_A0(DisasContext *s, int offset) { int mem_index = s->mem_index; tcg_gen_ld_i64(s->tmp1_i64, cpu_env, offset + offsetof(ZMMReg, ZMM_Q(0))); - tcg_gen_qemu_st_i64(s->tmp1_i64, s->A0, mem_index, MO_LEQ); + tcg_gen_qemu_st_i64(s->tmp1_i64, s->A0, mem_index, MO_LEUQ); tcg_gen_addi_tl(s->tmp0, s->A0, 8); tcg_gen_ld_i64(s->tmp1_i64, cpu_env, offset + offsetof(ZMMReg, ZMM_Q(1))); - tcg_gen_qemu_st_i64(s->tmp1_i64, s->tmp0, mem_index, MO_LEQ); + tcg_gen_qemu_st_i64(s->tmp1_i64, s->tmp0, mem_index, MO_LEUQ); } static inline void gen_op_movo(DisasContext *s, int d_offset, int s_offset) @@ -XXX,XX +XXX,XX @@ static void gen_sse(CPUX86State *env, DisasContext *s, int b, tcg_gen_mov_i64(cpu_regs[rm], s->tmp1_i64); } else { tcg_gen_qemu_st_i64(s->tmp1_i64, s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); } #else goto illegal_op; @@ -XXX,XX +XXX,XX @@ static void gen_sse(CPUX86State *env, DisasContext *s, int b, gen_op_mov_v_reg(s, ot, s->tmp1_i64, rm); } else { tcg_gen_qemu_ld_i64(s->tmp1_i64, s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); } tcg_gen_st_i64(s->tmp1_i64, cpu_env, offsetof(CPUX86State, @@ -XXX,XX +XXX,XX @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu) break; case 2: tcg_gen_qemu_ld_i64(s->tmp1_i64, s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); gen_helper_fldl_FT0(cpu_env, s->tmp1_i64); break; case 3: @@ -XXX,XX +XXX,XX @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu) break; case 2: tcg_gen_qemu_ld_i64(s->tmp1_i64, s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); gen_helper_fldl_ST0(cpu_env, s->tmp1_i64); break; case 3: @@ -XXX,XX +XXX,XX @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu) case 2: gen_helper_fisttll_ST0(s->tmp1_i64, cpu_env); tcg_gen_qemu_st_i64(s->tmp1_i64, s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); break; case 3: default: @@ -XXX,XX +XXX,XX @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu) case 2: gen_helper_fstl_ST0(s->tmp1_i64, cpu_env); tcg_gen_qemu_st_i64(s->tmp1_i64, s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); break; case 3: default: @@ -XXX,XX +XXX,XX @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu) break; case 0x3d: /* fildll */ tcg_gen_qemu_ld_i64(s->tmp1_i64, s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); gen_helper_fildll_ST0(cpu_env, s->tmp1_i64); break; case 0x3f: /* fistpll */ gen_helper_fistll_ST0(s->tmp1_i64, cpu_env); tcg_gen_qemu_st_i64(s->tmp1_i64, s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); gen_helper_fpop(cpu_env); break; default: @@ -XXX,XX +XXX,XX @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu) gen_lea_modrm(env, s, modrm); if (CODE64(s)) { tcg_gen_qemu_ld_i64(cpu_bndl[reg], s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); tcg_gen_addi_tl(s->A0, s->A0, 8); tcg_gen_qemu_ld_i64(cpu_bndu[reg], s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); } else { tcg_gen_qemu_ld_i64(cpu_bndl[reg], s->A0, s->mem_index, MO_LEUL); @@ -XXX,XX +XXX,XX @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu) gen_lea_modrm(env, s, modrm); if (CODE64(s)) { tcg_gen_qemu_st_i64(cpu_bndl[reg], s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); tcg_gen_addi_tl(s->A0, s->A0, 8); tcg_gen_qemu_st_i64(cpu_bndu[reg], s->A0, - s->mem_index, MO_LEQ); + s->mem_index, MO_LEUQ); } else { tcg_gen_qemu_st_i64(cpu_bndl[reg], s->A0, s->mem_index, MO_LEUL); diff --git a/target/m68k/op_helper.c b/target/m68k/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/m68k/op_helper.c +++ b/target/m68k/op_helper.c @@ -XXX,XX +XXX,XX @@ static void do_cas2l(CPUM68KState *env, uint32_t regs, uint32_t a1, uint32_t a2, uintptr_t ra = GETPC(); #if defined(CONFIG_ATOMIC64) int mmu_idx = cpu_mmu_index(env, 0); - MemOpIdx oi = make_memop_idx(MO_BEQ, mmu_idx); + MemOpIdx oi = make_memop_idx(MO_BEUQ, mmu_idx); #endif if (parallel) { diff --git a/target/mips/tcg/translate.c b/target/mips/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/translate.c +++ b/target/mips/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_ld(DisasContext *ctx, uint32_t opc, gen_store_gpr(t0, rt); break; case OPC_LD: - tcg_gen_qemu_ld_tl(t0, t0, mem_idx, MO_TEQ | + tcg_gen_qemu_ld_tl(t0, t0, mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_store_gpr(t0, rt); break; @@ -XXX,XX +XXX,XX @@ static void gen_ld(DisasContext *ctx, uint32_t opc, } tcg_gen_shli_tl(t1, t1, 3); tcg_gen_andi_tl(t0, t0, ~7); - tcg_gen_qemu_ld_tl(t0, t0, mem_idx, MO_TEQ); + tcg_gen_qemu_ld_tl(t0, t0, mem_idx, MO_TEUQ); tcg_gen_shl_tl(t0, t0, t1); t2 = tcg_const_tl(-1); tcg_gen_shl_tl(t2, t2, t1); @@ -XXX,XX +XXX,XX @@ static void gen_ld(DisasContext *ctx, uint32_t opc, } tcg_gen_shli_tl(t1, t1, 3); tcg_gen_andi_tl(t0, t0, ~7); - tcg_gen_qemu_ld_tl(t0, t0, mem_idx, MO_TEQ); + tcg_gen_qemu_ld_tl(t0, t0, mem_idx, MO_TEUQ); tcg_gen_shr_tl(t0, t0, t1); tcg_gen_xori_tl(t1, t1, 63); t2 = tcg_const_tl(0xfffffffffffffffeull); @@ -XXX,XX +XXX,XX @@ static void gen_ld(DisasContext *ctx, uint32_t opc, t1 = tcg_const_tl(pc_relative_pc(ctx)); gen_op_addr_add(ctx, t0, t0, t1); tcg_temp_free(t1); - tcg_gen_qemu_ld_tl(t0, t0, mem_idx, MO_TEQ); + tcg_gen_qemu_ld_tl(t0, t0, mem_idx, MO_TEUQ); gen_store_gpr(t0, rt); break; #endif @@ -XXX,XX +XXX,XX @@ static void gen_st(DisasContext *ctx, uint32_t opc, int rt, switch (opc) { #if defined(TARGET_MIPS64) case OPC_SD: - tcg_gen_qemu_st_tl(t1, t0, mem_idx, MO_TEQ | + tcg_gen_qemu_st_tl(t1, t0, mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); break; case OPC_SDL: @@ -XXX,XX +XXX,XX @@ static void gen_flt_ldst(DisasContext *ctx, uint32_t opc, int ft, case OPC_LDC1: { TCGv_i64 fp0 = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(fp0, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_ld_i64(fp0, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_store_fpr64(ctx, fp0, ft); tcg_temp_free_i64(fp0); @@ -XXX,XX +XXX,XX @@ static void gen_flt_ldst(DisasContext *ctx, uint32_t opc, int ft, { TCGv_i64 fp0 = tcg_temp_new_i64(); gen_load_fpr64(ctx, fp0, ft); - tcg_gen_qemu_st_i64(fp0, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_st_i64(fp0, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); tcg_temp_free_i64(fp0); } @@ -XXX,XX +XXX,XX @@ static inline void gen_pcrel(DisasContext *ctx, int opc, target_ulong pc, check_mips_64(ctx); offset = sextract32(ctx->opcode << 3, 0, 21); addr = addr_add(ctx, (pc & ~0x7), offset); - gen_r6_ld(addr, rs, ctx->mem_idx, MO_TEQ); + gen_r6_ld(addr, rs, ctx->mem_idx, MO_TEUQ); break; #endif default: @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lswc2(DisasContext *ctx, int rt, case OPC_GSLQ: t1 = tcg_temp_new(); gen_base_offset_addr(ctx, t0, rs, lsq_offset); - tcg_gen_qemu_ld_tl(t1, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_ld_tl(t1, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_base_offset_addr(ctx, t0, rs, lsq_offset + 8); - tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_store_gpr(t1, rt); gen_store_gpr(t0, lsq_rt1); @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lswc2(DisasContext *ctx, int rt, check_cp1_enabled(ctx); t1 = tcg_temp_new(); gen_base_offset_addr(ctx, t0, rs, lsq_offset); - tcg_gen_qemu_ld_tl(t1, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_ld_tl(t1, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_base_offset_addr(ctx, t0, rs, lsq_offset + 8); - tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_store_fpr64(ctx, t1, rt); gen_store_fpr64(ctx, t0, lsq_rt1); @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lswc2(DisasContext *ctx, int rt, t1 = tcg_temp_new(); gen_base_offset_addr(ctx, t0, rs, lsq_offset); gen_load_gpr(t1, rt); - tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_base_offset_addr(ctx, t0, rs, lsq_offset + 8); gen_load_gpr(t1, lsq_rt1); - tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); tcg_temp_free(t1); break; @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lswc2(DisasContext *ctx, int rt, t1 = tcg_temp_new(); gen_base_offset_addr(ctx, t0, rs, lsq_offset); gen_load_fpr64(ctx, t1, rt); - tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_base_offset_addr(ctx, t0, rs, lsq_offset + 8); gen_load_fpr64(ctx, t1, lsq_rt1); - tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); tcg_temp_free(t1); break; @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lswc2(DisasContext *ctx, int rt, } tcg_gen_shli_tl(t1, t1, 3); tcg_gen_andi_tl(t0, t0, ~7); - tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEUQ); tcg_gen_shl_tl(t0, t0, t1); t2 = tcg_const_tl(-1); tcg_gen_shl_tl(t2, t2, t1); @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lswc2(DisasContext *ctx, int rt, } tcg_gen_shli_tl(t1, t1, 3); tcg_gen_andi_tl(t0, t0, ~7); - tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEUQ); tcg_gen_shr_tl(t0, t0, t1); tcg_gen_xori_tl(t1, t1, 63); t2 = tcg_const_tl(0xfffffffffffffffeull); @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lsdc2(DisasContext *ctx, int rt, if (rd) { gen_op_addr_add(ctx, t0, cpu_gpr[rd], t0); } - tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_store_gpr(t0, rt); break; @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lsdc2(DisasContext *ctx, int rt, if (rd) { gen_op_addr_add(ctx, t0, cpu_gpr[rd], t0); } - tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); gen_store_fpr64(ctx, t0, rt); break; @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lsdc2(DisasContext *ctx, int rt, case OPC_GSSDX: t1 = tcg_temp_new(); gen_load_gpr(t1, rt); - tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); tcg_temp_free(t1); break; @@ -XXX,XX +XXX,XX @@ static void gen_loongson_lsdc2(DisasContext *ctx, int rt, case OPC_GSSDXC1: t1 = tcg_temp_new(); gen_load_fpr64(ctx, t1, rt); - tcg_gen_qemu_st_i64(t1, t0, ctx->mem_idx, MO_TEQ | + tcg_gen_qemu_st_i64(t1, t0, ctx->mem_idx, MO_TEUQ | ctx->default_tcg_memop_mask); tcg_temp_free(t1); break; @@ -XXX,XX +XXX,XX @@ static void gen_flt3_ldst(DisasContext *ctx, uint32_t opc, check_cp1_registers(ctx, fd); { TCGv_i64 fp0 = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(fp0, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_i64(fp0, t0, ctx->mem_idx, MO_TEUQ); gen_store_fpr64(ctx, fp0, fd); tcg_temp_free_i64(fp0); } @@ -XXX,XX +XXX,XX @@ static void gen_flt3_ldst(DisasContext *ctx, uint32_t opc, { TCGv_i64 fp0 = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(fp0, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_i64(fp0, t0, ctx->mem_idx, MO_TEUQ); gen_store_fpr64(ctx, fp0, fd); tcg_temp_free_i64(fp0); } @@ -XXX,XX +XXX,XX @@ static void gen_flt3_ldst(DisasContext *ctx, uint32_t opc, { TCGv_i64 fp0 = tcg_temp_new_i64(); gen_load_fpr64(ctx, fp0, fs); - tcg_gen_qemu_st_i64(fp0, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_st_i64(fp0, t0, ctx->mem_idx, MO_TEUQ); tcg_temp_free_i64(fp0); } break; @@ -XXX,XX +XXX,XX @@ static void gen_flt3_ldst(DisasContext *ctx, uint32_t opc, { TCGv_i64 fp0 = tcg_temp_new_i64(); gen_load_fpr64(ctx, fp0, fs); - tcg_gen_qemu_st_i64(fp0, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_st_i64(fp0, t0, ctx->mem_idx, MO_TEUQ); tcg_temp_free_i64(fp0); } break; @@ -XXX,XX +XXX,XX @@ static void gen_mipsdsp_ld(DisasContext *ctx, uint32_t opc, break; #if defined(TARGET_MIPS64) case OPC_LDX: - tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TEUQ); gen_store_gpr(t0, rd); break; #endif @@ -XXX,XX +XXX,XX @@ static void decode_opc_special3_r6(CPUMIPSState *env, DisasContext *ctx) #endif #if defined(TARGET_MIPS64) case R6_OPC_SCD: - gen_st_cond(ctx, rt, rs, imm, MO_TEQ, false); + gen_st_cond(ctx, rt, rs, imm, MO_TEUQ, false); break; case R6_OPC_LLD: gen_ld(ctx, op1, rt, rs, imm); @@ -XXX,XX +XXX,XX @@ static bool decode_opc_legacy(CPUMIPSState *env, DisasContext *ctx) check_insn_opc_user_only(ctx, INSN_R5900); } check_mips_64(ctx); - gen_st_cond(ctx, rt, rs, imm, MO_TEQ, false); + gen_st_cond(ctx, rt, rs, imm, MO_TEUQ, false); break; case OPC_BNVC: /* OPC_BNEZALC, OPC_BNEC, OPC_DADDI */ if (ctx->insn_flags & ISA_MIPS_R6) { diff --git a/target/mips/tcg/tx79_translate.c b/target/mips/tcg/tx79_translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/tx79_translate.c +++ b/target/mips/tcg/tx79_translate.c @@ -XXX,XX +XXX,XX @@ static bool trans_LQ(DisasContext *ctx, arg_i *a) tcg_gen_andi_tl(addr, addr, ~0xf); /* Lower half */ - tcg_gen_qemu_ld_i64(t0, addr, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_i64(t0, addr, ctx->mem_idx, MO_TEUQ); gen_store_gpr(t0, a->rt); /* Upper half */ tcg_gen_addi_i64(addr, addr, 8); - tcg_gen_qemu_ld_i64(t0, addr, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_i64(t0, addr, ctx->mem_idx, MO_TEUQ); gen_store_gpr_hi(t0, a->rt); tcg_temp_free(t0); @@ -XXX,XX +XXX,XX @@ static bool trans_SQ(DisasContext *ctx, arg_i *a) /* Lower half */ gen_load_gpr(t0, a->rt); - tcg_gen_qemu_st_i64(t0, addr, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_st_i64(t0, addr, ctx->mem_idx, MO_TEUQ); /* Upper half */ tcg_gen_addi_i64(addr, addr, 8); gen_load_gpr_hi(t0, a->rt); - tcg_gen_qemu_st_i64(t0, addr, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_st_i64(t0, addr, ctx->mem_idx, MO_TEUQ); tcg_temp_free(addr); tcg_temp_free(t0); diff --git a/target/ppc/translate.c b/target/ppc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/translate.c +++ b/target/ppc/translate.c @@ -XXX,XX +XXX,XX @@ GEN_QEMU_LOAD_64(ld8u, DEF_MEMOP(MO_UB)) GEN_QEMU_LOAD_64(ld16u, DEF_MEMOP(MO_UW)) GEN_QEMU_LOAD_64(ld32u, DEF_MEMOP(MO_UL)) GEN_QEMU_LOAD_64(ld32s, DEF_MEMOP(MO_SL)) -GEN_QEMU_LOAD_64(ld64, DEF_MEMOP(MO_Q)) +GEN_QEMU_LOAD_64(ld64, DEF_MEMOP(MO_UQ)) #if defined(TARGET_PPC64) -GEN_QEMU_LOAD_64(ld64ur, BSWAP_MEMOP(MO_Q)) +GEN_QEMU_LOAD_64(ld64ur, BSWAP_MEMOP(MO_UQ)) #endif #define GEN_QEMU_STORE_TL(stop, op) \ @@ -XXX,XX +XXX,XX @@ static void glue(gen_qemu_, glue(stop, _i64))(DisasContext *ctx, \ GEN_QEMU_STORE_64(st8, DEF_MEMOP(MO_UB)) GEN_QEMU_STORE_64(st16, DEF_MEMOP(MO_UW)) GEN_QEMU_STORE_64(st32, DEF_MEMOP(MO_UL)) -GEN_QEMU_STORE_64(st64, DEF_MEMOP(MO_Q)) +GEN_QEMU_STORE_64(st64, DEF_MEMOP(MO_UQ)) #if defined(TARGET_PPC64) -GEN_QEMU_STORE_64(st64r, BSWAP_MEMOP(MO_Q)) +GEN_QEMU_STORE_64(st64r, BSWAP_MEMOP(MO_UQ)) #endif #define GEN_LDX_E(name, ldop, opc2, opc3, type, type2, chk) \ @@ -XXX,XX +XXX,XX @@ GEN_LDEPX(lb, DEF_MEMOP(MO_UB), 0x1F, 0x02) GEN_LDEPX(lh, DEF_MEMOP(MO_UW), 0x1F, 0x08) GEN_LDEPX(lw, DEF_MEMOP(MO_UL), 0x1F, 0x00) #if defined(TARGET_PPC64) -GEN_LDEPX(ld, DEF_MEMOP(MO_Q), 0x1D, 0x00) +GEN_LDEPX(ld, DEF_MEMOP(MO_UQ), 0x1D, 0x00) #endif #if defined(TARGET_PPC64) @@ -XXX,XX +XXX,XX @@ GEN_STEPX(stb, DEF_MEMOP(MO_UB), 0x1F, 0x06) GEN_STEPX(sth, DEF_MEMOP(MO_UW), 0x1F, 0x0C) GEN_STEPX(stw, DEF_MEMOP(MO_UL), 0x1F, 0x04) #if defined(TARGET_PPC64) -GEN_STEPX(std, DEF_MEMOP(MO_Q), 0x1d, 0x04) +GEN_STEPX(std, DEF_MEMOP(MO_UQ), 0x1d, 0x04) #endif #if defined(TARGET_PPC64) @@ -XXX,XX +XXX,XX @@ static void gen_lwat(DisasContext *ctx) #ifdef TARGET_PPC64 static void gen_ldat(DisasContext *ctx) { - gen_ld_atomic(ctx, DEF_MEMOP(MO_Q)); + gen_ld_atomic(ctx, DEF_MEMOP(MO_UQ)); } #endif @@ -XXX,XX +XXX,XX @@ static void gen_stwat(DisasContext *ctx) #ifdef TARGET_PPC64 static void gen_stdat(DisasContext *ctx) { - gen_st_atomic(ctx, DEF_MEMOP(MO_Q)); + gen_st_atomic(ctx, DEF_MEMOP(MO_UQ)); } #endif @@ -XXX,XX +XXX,XX @@ STCX(stwcx_, DEF_MEMOP(MO_UL)) #if defined(TARGET_PPC64) /* ldarx */ -LARX(ldarx, DEF_MEMOP(MO_Q)) +LARX(ldarx, DEF_MEMOP(MO_UQ)) /* stdcx. */ -STCX(stdcx_, DEF_MEMOP(MO_Q)) +STCX(stdcx_, DEF_MEMOP(MO_UQ)) /* lqarx */ static void gen_lqarx(DisasContext *ctx) @@ -XXX,XX +XXX,XX @@ static void gen_lqarx(DisasContext *ctx) return; } } else if (ctx->le_mode) { - tcg_gen_qemu_ld_i64(lo, EA, ctx->mem_idx, MO_LEQ | MO_ALIGN_16); + tcg_gen_qemu_ld_i64(lo, EA, ctx->mem_idx, MO_LEUQ | MO_ALIGN_16); tcg_gen_mov_tl(cpu_reserve, EA); gen_addr_add(ctx, EA, EA, 8); - tcg_gen_qemu_ld_i64(hi, EA, ctx->mem_idx, MO_LEQ); + tcg_gen_qemu_ld_i64(hi, EA, ctx->mem_idx, MO_LEUQ); } else { - tcg_gen_qemu_ld_i64(hi, EA, ctx->mem_idx, MO_BEQ | MO_ALIGN_16); + tcg_gen_qemu_ld_i64(hi, EA, ctx->mem_idx, MO_BEUQ | MO_ALIGN_16); tcg_gen_mov_tl(cpu_reserve, EA); gen_addr_add(ctx, EA, EA, 8); - tcg_gen_qemu_ld_i64(lo, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_ld_i64(lo, EA, ctx->mem_idx, MO_BEUQ); } tcg_temp_free(EA); @@ -XXX,XX +XXX,XX @@ GEN_LDEPX(lb, DEF_MEMOP(MO_UB), 0x1F, 0x02) GEN_LDEPX(lh, DEF_MEMOP(MO_UW), 0x1F, 0x08) GEN_LDEPX(lw, DEF_MEMOP(MO_UL), 0x1F, 0x00) #if defined(TARGET_PPC64) -GEN_LDEPX(ld, DEF_MEMOP(MO_Q), 0x1D, 0x00) +GEN_LDEPX(ld, DEF_MEMOP(MO_UQ), 0x1D, 0x00) #endif #undef GEN_STX_E @@ -XXX,XX +XXX,XX @@ GEN_STEPX(stb, DEF_MEMOP(MO_UB), 0x1F, 0x06) GEN_STEPX(sth, DEF_MEMOP(MO_UW), 0x1F, 0x0C) GEN_STEPX(stw, DEF_MEMOP(MO_UL), 0x1F, 0x04) #if defined(TARGET_PPC64) -GEN_STEPX(std, DEF_MEMOP(MO_Q), 0x1D, 0x04) +GEN_STEPX(std, DEF_MEMOP(MO_UQ), 0x1D, 0x04) #endif #undef GEN_CRLOGIC diff --git a/target/s390x/tcg/mem_helper.c b/target/s390x/tcg/mem_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/tcg/mem_helper.c +++ b/target/s390x/tcg/mem_helper.c @@ -XXX,XX +XXX,XX @@ static uint32_t do_csst(CPUS390XState *env, uint32_t r3, uint64_t a1, if (parallel) { #ifdef CONFIG_ATOMIC64 - MemOpIdx oi = make_memop_idx(MO_TEQ | MO_ALIGN, mem_idx); + MemOpIdx oi = make_memop_idx(MO_TEUQ | MO_ALIGN, mem_idx); ov = cpu_atomic_cmpxchgq_be_mmu(env, a1, cv, nv, oi, ra); #else /* Note that we asserted !parallel above. */ @@ -XXX,XX +XXX,XX @@ static uint32_t do_csst(CPUS390XState *env, uint32_t r3, uint64_t a1, cpu_stq_data_ra(env, a2 + 0, svh, ra); cpu_stq_data_ra(env, a2 + 8, svl, ra); } else if (HAVE_ATOMIC128) { - MemOpIdx oi = make_memop_idx(MO_TEQ | MO_ALIGN_16, mem_idx); + MemOpIdx oi = make_memop_idx(MO_TEUQ | MO_ALIGN_16, mem_idx); Int128 sv = int128_make128(svl, svh); cpu_atomic_sto_be_mmu(env, a2, sv, oi, ra); } else { @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(lpq_parallel)(CPUS390XState *env, uint64_t addr) assert(HAVE_ATOMIC128); mem_idx = cpu_mmu_index(env, false); - oi = make_memop_idx(MO_TEQ | MO_ALIGN_16, mem_idx); + oi = make_memop_idx(MO_TEUQ | MO_ALIGN_16, mem_idx); v = cpu_atomic_ldo_be_mmu(env, addr, oi, ra); hi = int128_gethi(v); lo = int128_getlo(v); @@ -XXX,XX +XXX,XX @@ void HELPER(stpq_parallel)(CPUS390XState *env, uint64_t addr, assert(HAVE_ATOMIC128); mem_idx = cpu_mmu_index(env, false); - oi = make_memop_idx(MO_TEQ | MO_ALIGN_16, mem_idx); + oi = make_memop_idx(MO_TEUQ | MO_ALIGN_16, mem_idx); v = int128_make128(low, high); cpu_atomic_sto_be_mmu(env, addr, v, oi, ra); } diff --git a/target/s390x/tcg/translate.c b/target/s390x/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/tcg/translate.c +++ b/target/s390x/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static DisasJumpType op_lpswe(DisasContext *s, DisasOps *o) t1 = tcg_temp_new_i64(); t2 = tcg_temp_new_i64(); tcg_gen_qemu_ld_i64(t1, o->in2, get_mem_index(s), - MO_TEQ | MO_ALIGN_8); + MO_TEUQ | MO_ALIGN_8); tcg_gen_addi_i64(o->in2, o->in2, 8); tcg_gen_qemu_ld64(t2, o->in2, get_mem_index(s)); gen_helper_load_psw(cpu_env, t1, t2); @@ -XXX,XX +XXX,XX @@ static DisasJumpType op_stcke(DisasContext *s, DisasOps *o) #ifndef CONFIG_USER_ONLY static DisasJumpType op_sck(DisasContext *s, DisasOps *o) { - tcg_gen_qemu_ld_i64(o->in1, o->addr1, get_mem_index(s), MO_TEQ | MO_ALIGN); + tcg_gen_qemu_ld_i64(o->in1, o->addr1, get_mem_index(s), MO_TEUQ | MO_ALIGN); gen_helper_sck(cc_op, cpu_env, o->in1); set_cc_static(s); return DISAS_NEXT; @@ -XXX,XX +XXX,XX @@ static void wout_m1_64(DisasContext *s, DisasOps *o) #ifndef CONFIG_USER_ONLY static void wout_m1_64a(DisasContext *s, DisasOps *o) { - tcg_gen_qemu_st_i64(o->out, o->addr1, get_mem_index(s), MO_TEQ | MO_ALIGN); + tcg_gen_qemu_st_i64(o->out, o->addr1, get_mem_index(s), MO_TEUQ | MO_ALIGN); } #define SPEC_wout_m1_64a 0 #endif @@ -XXX,XX +XXX,XX @@ static void in2_m2_64w(DisasContext *s, DisasOps *o) static void in2_m2_64a(DisasContext *s, DisasOps *o) { in2_a2(s, o); - tcg_gen_qemu_ld_i64(o->in2, o->in2, get_mem_index(s), MO_TEQ | MO_ALIGN); + tcg_gen_qemu_ld_i64(o->in2, o->in2, get_mem_index(s), MO_TEUQ | MO_ALIGN); } #define SPEC_in2_m2_64a 0 #endif diff --git a/target/sh4/translate.c b/target/sh4/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/translate.c +++ b/target/sh4/translate.c @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) if (ctx->tbflags & FPSCR_SZ) { TCGv_i64 fp = tcg_temp_new_i64(); gen_load_fpr64(ctx, fp, XHACK(B7_4)); - tcg_gen_qemu_st_i64(fp, REG(B11_8), ctx->memidx, MO_TEQ); + tcg_gen_qemu_st_i64(fp, REG(B11_8), ctx->memidx, MO_TEUQ); tcg_temp_free_i64(fp); } else { tcg_gen_qemu_st_i32(FREG(B7_4), REG(B11_8), ctx->memidx, MO_TEUL); @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) CHECK_FPU_ENABLED if (ctx->tbflags & FPSCR_SZ) { TCGv_i64 fp = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(fp, REG(B7_4), ctx->memidx, MO_TEQ); + tcg_gen_qemu_ld_i64(fp, REG(B7_4), ctx->memidx, MO_TEUQ); gen_store_fpr64(ctx, fp, XHACK(B11_8)); tcg_temp_free_i64(fp); } else { @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) CHECK_FPU_ENABLED if (ctx->tbflags & FPSCR_SZ) { TCGv_i64 fp = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(fp, REG(B7_4), ctx->memidx, MO_TEQ); + tcg_gen_qemu_ld_i64(fp, REG(B7_4), ctx->memidx, MO_TEUQ); gen_store_fpr64(ctx, fp, XHACK(B11_8)); tcg_temp_free_i64(fp); tcg_gen_addi_i32(REG(B7_4), REG(B7_4), 8); @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) TCGv_i64 fp = tcg_temp_new_i64(); gen_load_fpr64(ctx, fp, XHACK(B7_4)); tcg_gen_subi_i32(addr, REG(B11_8), 8); - tcg_gen_qemu_st_i64(fp, addr, ctx->memidx, MO_TEQ); + tcg_gen_qemu_st_i64(fp, addr, ctx->memidx, MO_TEUQ); tcg_temp_free_i64(fp); } else { tcg_gen_subi_i32(addr, REG(B11_8), 4); @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) tcg_gen_add_i32(addr, REG(B7_4), REG(0)); if (ctx->tbflags & FPSCR_SZ) { TCGv_i64 fp = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(fp, addr, ctx->memidx, MO_TEQ); + tcg_gen_qemu_ld_i64(fp, addr, ctx->memidx, MO_TEUQ); gen_store_fpr64(ctx, fp, XHACK(B11_8)); tcg_temp_free_i64(fp); } else { @@ -XXX,XX +XXX,XX @@ static void _decode_opc(DisasContext * ctx) if (ctx->tbflags & FPSCR_SZ) { TCGv_i64 fp = tcg_temp_new_i64(); gen_load_fpr64(ctx, fp, XHACK(B7_4)); - tcg_gen_qemu_st_i64(fp, addr, ctx->memidx, MO_TEQ); + tcg_gen_qemu_st_i64(fp, addr, ctx->memidx, MO_TEUQ); tcg_temp_free_i64(fp); } else { tcg_gen_qemu_st_i32(FREG(B7_4), addr, ctx->memidx, MO_TEUL); diff --git a/target/sparc/translate.c b/target/sparc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/translate.c +++ b/target/sparc/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_ldstub_asi(DisasContext *dc, TCGv dst, TCGv addr, int insn) static void gen_ldf_asi(DisasContext *dc, TCGv addr, int insn, int size, int rd) { - DisasASI da = get_asi(dc, insn, (size == 4 ? MO_TEUL : MO_TEQ)); + DisasASI da = get_asi(dc, insn, (size == 4 ? MO_TEUL : MO_TEUQ)); TCGv_i32 d32; TCGv_i64 d64; @@ -XXX,XX +XXX,XX @@ static void gen_ldf_asi(DisasContext *dc, TCGv addr, static void gen_stf_asi(DisasContext *dc, TCGv addr, int insn, int size, int rd) { - DisasASI da = get_asi(dc, insn, (size == 4 ? MO_TEUL : MO_TEQ)); + DisasASI da = get_asi(dc, insn, (size == 4 ? MO_TEUL : MO_TEUQ)); TCGv_i32 d32; switch (da.type) { @@ -XXX,XX +XXX,XX @@ static void gen_stf_asi(DisasContext *dc, TCGv addr, static void gen_ldda_asi(DisasContext *dc, TCGv addr, int insn, int rd) { - DisasASI da = get_asi(dc, insn, MO_TEQ); + DisasASI da = get_asi(dc, insn, MO_TEUQ); TCGv_i64 hi = gen_dest_gpr(dc, rd); TCGv_i64 lo = gen_dest_gpr(dc, rd + 1); @@ -XXX,XX +XXX,XX @@ static void gen_ldda_asi(DisasContext *dc, TCGv addr, int insn, int rd) static void gen_stda_asi(DisasContext *dc, TCGv hi, TCGv addr, int insn, int rd) { - DisasASI da = get_asi(dc, insn, MO_TEQ); + DisasASI da = get_asi(dc, insn, MO_TEUQ); TCGv lo = gen_load_gpr(dc, rd + 1); switch (da.type) { @@ -XXX,XX +XXX,XX @@ static void gen_stda_asi(DisasContext *dc, TCGv hi, TCGv addr, static void gen_casx_asi(DisasContext *dc, TCGv addr, TCGv cmpv, int insn, int rd) { - DisasASI da = get_asi(dc, insn, MO_TEQ); + DisasASI da = get_asi(dc, insn, MO_TEUQ); TCGv oldv; switch (da.type) { @@ -XXX,XX +XXX,XX @@ static void gen_ldda_asi(DisasContext *dc, TCGv addr, int insn, int rd) TCGv lo = gen_dest_gpr(dc, rd | 1); TCGv hi = gen_dest_gpr(dc, rd); TCGv_i64 t64 = tcg_temp_new_i64(); - DisasASI da = get_asi(dc, insn, MO_TEQ); + DisasASI da = get_asi(dc, insn, MO_TEUQ); switch (da.type) { case GET_ASI_EXCP: @@ -XXX,XX +XXX,XX @@ static void gen_ldda_asi(DisasContext *dc, TCGv addr, int insn, int rd) default: { TCGv_i32 r_asi = tcg_const_i32(da.asi); - TCGv_i32 r_mop = tcg_const_i32(MO_Q); + TCGv_i32 r_mop = tcg_const_i32(MO_UQ); save_state(dc); gen_helper_ld_asi(t64, cpu_env, addr, r_asi, r_mop); @@ -XXX,XX +XXX,XX @@ static void gen_ldda_asi(DisasContext *dc, TCGv addr, int insn, int rd) static void gen_stda_asi(DisasContext *dc, TCGv hi, TCGv addr, int insn, int rd) { - DisasASI da = get_asi(dc, insn, MO_TEQ); + DisasASI da = get_asi(dc, insn, MO_TEUQ); TCGv lo = gen_load_gpr(dc, rd + 1); TCGv_i64 t64 = tcg_temp_new_i64(); @@ -XXX,XX +XXX,XX @@ static void gen_stda_asi(DisasContext *dc, TCGv hi, TCGv addr, default: { TCGv_i32 r_asi = tcg_const_i32(da.asi); - TCGv_i32 r_mop = tcg_const_i32(MO_Q); + TCGv_i32 r_mop = tcg_const_i32(MO_UQ); save_state(dc); gen_helper_st_asi(cpu_env, addr, t64, r_asi, r_mop); @@ -XXX,XX +XXX,XX @@ static void disas_sparc_insn(DisasContext * dc, unsigned int insn) gen_ld_asi(dc, cpu_val, cpu_addr, insn, MO_TESL); break; case 0x1b: /* V9 ldxa */ - gen_ld_asi(dc, cpu_val, cpu_addr, insn, MO_TEQ); + gen_ld_asi(dc, cpu_val, cpu_addr, insn, MO_TEUQ); break; case 0x2d: /* V9 prefetch, no effect */ goto skip_move; @@ -XXX,XX +XXX,XX @@ static void disas_sparc_insn(DisasContext * dc, unsigned int insn) if (rd == 1) { TCGv_i64 t64 = tcg_temp_new_i64(); tcg_gen_qemu_ld_i64(t64, cpu_addr, - dc->mem_idx, MO_TEQ); + dc->mem_idx, MO_TEUQ); gen_helper_ldxfsr(cpu_fsr, cpu_env, cpu_fsr, t64); tcg_temp_free_i64(t64); break; @@ -XXX,XX +XXX,XX @@ static void disas_sparc_insn(DisasContext * dc, unsigned int insn) gen_address_mask(dc, cpu_addr); cpu_src1_64 = tcg_temp_new_i64(); tcg_gen_qemu_ld_i64(cpu_src1_64, cpu_addr, dc->mem_idx, - MO_TEQ | MO_ALIGN_4); + MO_TEUQ | MO_ALIGN_4); tcg_gen_addi_tl(cpu_addr, cpu_addr, 8); cpu_src2_64 = tcg_temp_new_i64(); tcg_gen_qemu_ld_i64(cpu_src2_64, cpu_addr, dc->mem_idx, - MO_TEQ | MO_ALIGN_4); + MO_TEUQ | MO_ALIGN_4); gen_store_fpr_Q(dc, rd, cpu_src1_64, cpu_src2_64); tcg_temp_free_i64(cpu_src1_64); tcg_temp_free_i64(cpu_src2_64); @@ -XXX,XX +XXX,XX @@ static void disas_sparc_insn(DisasContext * dc, unsigned int insn) gen_address_mask(dc, cpu_addr); cpu_dst_64 = gen_dest_fpr_D(dc, rd); tcg_gen_qemu_ld_i64(cpu_dst_64, cpu_addr, dc->mem_idx, - MO_TEQ | MO_ALIGN_4); + MO_TEUQ | MO_ALIGN_4); gen_store_fpr_D(dc, rd, cpu_dst_64); break; default: @@ -XXX,XX +XXX,XX @@ static void disas_sparc_insn(DisasContext * dc, unsigned int insn) tcg_gen_qemu_st64(cpu_val, cpu_addr, dc->mem_idx); break; case 0x1e: /* V9 stxa */ - gen_st_asi(dc, cpu_val, cpu_addr, insn, MO_TEQ); + gen_st_asi(dc, cpu_val, cpu_addr, insn, MO_TEUQ); break; #endif default: @@ -XXX,XX +XXX,XX @@ static void disas_sparc_insn(DisasContext * dc, unsigned int insn) before performing the first write. */ cpu_src1_64 = gen_load_fpr_Q0(dc, rd); tcg_gen_qemu_st_i64(cpu_src1_64, cpu_addr, - dc->mem_idx, MO_TEQ | MO_ALIGN_16); + dc->mem_idx, MO_TEUQ | MO_ALIGN_16); tcg_gen_addi_tl(cpu_addr, cpu_addr, 8); cpu_src2_64 = gen_load_fpr_Q1(dc, rd); tcg_gen_qemu_st_i64(cpu_src1_64, cpu_addr, - dc->mem_idx, MO_TEQ); + dc->mem_idx, MO_TEUQ); break; #else /* !TARGET_SPARC64 */ /* stdfq, store floating point queue */ @@ -XXX,XX +XXX,XX @@ static void disas_sparc_insn(DisasContext * dc, unsigned int insn) gen_address_mask(dc, cpu_addr); cpu_src1_64 = gen_load_fpr_D(dc, rd); tcg_gen_qemu_st_i64(cpu_src1_64, cpu_addr, dc->mem_idx, - MO_TEQ | MO_ALIGN_4); + MO_TEUQ | MO_ALIGN_4); break; default: goto illegal_insn; diff --git a/target/tricore/translate.c b/target/tricore/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/tricore/translate.c +++ b/target/tricore/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_st_2regs_64(TCGv rh, TCGv rl, TCGv address, DisasContext *ctx) TCGv_i64 temp = tcg_temp_new_i64(); tcg_gen_concat_i32_i64(temp, rl, rh); - tcg_gen_qemu_st_i64(temp, address, ctx->mem_idx, MO_LEQ); + tcg_gen_qemu_st_i64(temp, address, ctx->mem_idx, MO_LEUQ); tcg_temp_free_i64(temp); } @@ -XXX,XX +XXX,XX @@ static void gen_ld_2regs_64(TCGv rh, TCGv rl, TCGv address, DisasContext *ctx) { TCGv_i64 temp = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(temp, address, ctx->mem_idx, MO_LEQ); + tcg_gen_qemu_ld_i64(temp, address, ctx->mem_idx, MO_LEUQ); /* write back to two 32 bit regs */ tcg_gen_extr_i64_i32(rl, rh, temp); diff --git a/target/xtensa/translate.c b/target/xtensa/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/xtensa/translate.c +++ b/target/xtensa/translate.c @@ -XXX,XX +XXX,XX @@ static void translate_ldsti_d(DisasContext *dc, const OpcodeArg arg[], } else { addr = arg[1].in; } - mop = gen_load_store_alignment(dc, MO_TEQ, addr); + mop = gen_load_store_alignment(dc, MO_TEUQ, addr); if (par[0]) { tcg_gen_qemu_st_i64(arg[0].in, addr, dc->cring, mop); } else { @@ -XXX,XX +XXX,XX @@ static void translate_ldstx_d(DisasContext *dc, const OpcodeArg arg[], } else { addr = arg[1].in; } - mop = gen_load_store_alignment(dc, MO_TEQ, addr); + mop = gen_load_store_alignment(dc, MO_TEUQ, addr); if (par[0]) { tcg_gen_qemu_st_i64(arg[0].in, addr, dc->cring, mop); } else { diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ static const char * const ldst_name[] = [MO_LESW] = "lesw", [MO_LEUL] = "leul", [MO_LESL] = "lesl", - [MO_LEQ] = "leq", + [MO_LEUQ] = "leq", [MO_BEUW] = "beuw", [MO_BESW] = "besw", [MO_BEUL] = "beul", [MO_BESL] = "besl", - [MO_BEQ] = "beq", + [MO_BEUQ] = "beq", }; static const char * const alignment_name[(MO_AMASK >> MO_ASHIFT) + 1] = { diff --git a/tcg/tci.c b/tcg/tci.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -XXX,XX +XXX,XX @@ static uint64_t tci_qemu_ld(CPUArchState *env, target_ulong taddr, return helper_le_ldul_mmu(env, taddr, oi, ra); case MO_LESL: return helper_le_ldsl_mmu(env, taddr, oi, ra); - case MO_LEQ: + case MO_LEUQ: return helper_le_ldq_mmu(env, taddr, oi, ra); case MO_BEUW: return helper_be_lduw_mmu(env, taddr, oi, ra); @@ -XXX,XX +XXX,XX @@ static uint64_t tci_qemu_ld(CPUArchState *env, target_ulong taddr, return helper_be_ldul_mmu(env, taddr, oi, ra); case MO_BESL: return helper_be_ldsl_mmu(env, taddr, oi, ra); - case MO_BEQ: + case MO_BEUQ: return helper_be_ldq_mmu(env, taddr, oi, ra); default: g_assert_not_reached(); @@ -XXX,XX +XXX,XX @@ static uint64_t tci_qemu_ld(CPUArchState *env, target_ulong taddr, case MO_LESL: ret = (int32_t)ldl_le_p(haddr); break; - case MO_LEQ: + case MO_LEUQ: ret = ldq_le_p(haddr); break; case MO_BEUW: @@ -XXX,XX +XXX,XX @@ static uint64_t tci_qemu_ld(CPUArchState *env, target_ulong taddr, case MO_BESL: ret = (int32_t)ldl_be_p(haddr); break; - case MO_BEQ: + case MO_BEUQ: ret = ldq_be_p(haddr); break; default: @@ -XXX,XX +XXX,XX @@ static void tci_qemu_st(CPUArchState *env, target_ulong taddr, uint64_t val, case MO_LEUL: helper_le_stl_mmu(env, taddr, val, oi, ra); break; - case MO_LEQ: + case MO_LEUQ: helper_le_stq_mmu(env, taddr, val, oi, ra); break; case MO_BEUW: @@ -XXX,XX +XXX,XX @@ static void tci_qemu_st(CPUArchState *env, target_ulong taddr, uint64_t val, case MO_BEUL: helper_be_stl_mmu(env, taddr, val, oi, ra); break; - case MO_BEQ: + case MO_BEUQ: helper_be_stq_mmu(env, taddr, val, oi, ra); break; default: @@ -XXX,XX +XXX,XX @@ static void tci_qemu_st(CPUArchState *env, target_ulong taddr, uint64_t val, case MO_LEUL: stl_le_p(haddr, val); break; - case MO_LEQ: + case MO_LEUQ: stq_le_p(haddr, val); break; case MO_BEUW: @@ -XXX,XX +XXX,XX @@ static void tci_qemu_st(CPUArchState *env, target_ulong taddr, uint64_t val, case MO_BEUL: stl_be_p(haddr, val); break; - case MO_BEQ: + case MO_BEUQ: stq_be_p(haddr, val); break; default: diff --git a/accel/tcg/ldst_common.c.inc b/accel/tcg/ldst_common.c.inc index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/ldst_common.c.inc +++ b/accel/tcg/ldst_common.c.inc @@ -XXX,XX +XXX,XX @@ uint32_t cpu_ldl_be_mmuidx_ra(CPUArchState *env, abi_ptr addr, uint64_t cpu_ldq_be_mmuidx_ra(CPUArchState *env, abi_ptr addr, int mmu_idx, uintptr_t ra) { - MemOpIdx oi = make_memop_idx(MO_BEQ | MO_UNALN, mmu_idx); + MemOpIdx oi = make_memop_idx(MO_BEUQ | MO_UNALN, mmu_idx); return cpu_ldq_be_mmu(env, addr, oi, ra); } @@ -XXX,XX +XXX,XX @@ uint32_t cpu_ldl_le_mmuidx_ra(CPUArchState *env, abi_ptr addr, uint64_t cpu_ldq_le_mmuidx_ra(CPUArchState *env, abi_ptr addr, int mmu_idx, uintptr_t ra) { - MemOpIdx oi = make_memop_idx(MO_LEQ | MO_UNALN, mmu_idx); + MemOpIdx oi = make_memop_idx(MO_LEUQ | MO_UNALN, mmu_idx); return cpu_ldq_le_mmu(env, addr, oi, ra); } @@ -XXX,XX +XXX,XX @@ void cpu_stl_be_mmuidx_ra(CPUArchState *env, abi_ptr addr, uint32_t val, void cpu_stq_be_mmuidx_ra(CPUArchState *env, abi_ptr addr, uint64_t val, int mmu_idx, uintptr_t ra) { - MemOpIdx oi = make_memop_idx(MO_BEQ | MO_UNALN, mmu_idx); + MemOpIdx oi = make_memop_idx(MO_BEUQ | MO_UNALN, mmu_idx); cpu_stq_be_mmu(env, addr, val, oi, ra); } @@ -XXX,XX +XXX,XX @@ void cpu_stl_le_mmuidx_ra(CPUArchState *env, abi_ptr addr, uint32_t val, void cpu_stq_le_mmuidx_ra(CPUArchState *env, abi_ptr addr, uint64_t val, int mmu_idx, uintptr_t ra) { - MemOpIdx oi = make_memop_idx(MO_LEQ | MO_UNALN, mmu_idx); + MemOpIdx oi = make_memop_idx(MO_LEUQ | MO_UNALN, mmu_idx); cpu_stq_le_mmu(env, addr, val, oi, ra); } diff --git a/target/mips/tcg/micromips_translate.c.inc b/target/mips/tcg/micromips_translate.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/micromips_translate.c.inc +++ b/target/mips/tcg/micromips_translate.c.inc @@ -XXX,XX +XXX,XX @@ static void gen_ldst_pair(DisasContext *ctx, uint32_t opc, int rd, gen_reserved_instruction(ctx); return; } - tcg_gen_qemu_ld_tl(t1, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_tl(t1, t0, ctx->mem_idx, MO_TEUQ); gen_store_gpr(t1, rd); tcg_gen_movi_tl(t1, 8); gen_op_addr_add(ctx, t0, t0, t1); - tcg_gen_qemu_ld_tl(t1, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_tl(t1, t0, ctx->mem_idx, MO_TEUQ); gen_store_gpr(t1, rd + 1); break; case SDP: gen_load_gpr(t1, rd); - tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEUQ); tcg_gen_movi_tl(t1, 8); gen_op_addr_add(ctx, t0, t0, t1); gen_load_gpr(t1, rd + 1); - tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_st_tl(t1, t0, ctx->mem_idx, MO_TEUQ); break; #endif } @@ -XXX,XX +XXX,XX @@ static void decode_micromips32_opc(CPUMIPSState *env, DisasContext *ctx) case SCD: check_insn(ctx, ISA_MIPS3); check_mips_64(ctx); - gen_st_cond(ctx, rt, rs, offset, MO_TEQ, false); + gen_st_cond(ctx, rt, rs, offset, MO_TEUQ, false); break; #endif case LD_EVA: diff --git a/target/ppc/translate/fixedpoint-impl.c.inc b/target/ppc/translate/fixedpoint-impl.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/translate/fixedpoint-impl.c.inc +++ b/target/ppc/translate/fixedpoint-impl.c.inc @@ -XXX,XX +XXX,XX @@ static bool do_ldst_quad(DisasContext *ctx, arg_D *a, bool store, bool prefixed) ctx->base.is_jmp = DISAS_NORETURN; } } else { - mop = DEF_MEMOP(MO_Q); + mop = DEF_MEMOP(MO_UQ); if (store) { tcg_gen_qemu_st_i64(low_addr_gpr, ea, ctx->mem_idx, mop); } else { @@ -XXX,XX +XXX,XX @@ TRANS64(LWAUX, do_ldst_X, true, false, MO_SL) TRANS64(PLWA, do_ldst_PLS_D, false, false, MO_SL) /* Load Doubleword */ -TRANS64(LD, do_ldst_D, false, false, MO_Q) -TRANS64(LDX, do_ldst_X, false, false, MO_Q) -TRANS64(LDU, do_ldst_D, true, false, MO_Q) -TRANS64(LDUX, do_ldst_X, true, false, MO_Q) -TRANS64(PLD, do_ldst_PLS_D, false, false, MO_Q) +TRANS64(LD, do_ldst_D, false, false, MO_UQ) +TRANS64(LDX, do_ldst_X, false, false, MO_UQ) +TRANS64(LDU, do_ldst_D, true, false, MO_UQ) +TRANS64(LDUX, do_ldst_X, true, false, MO_UQ) +TRANS64(PLD, do_ldst_PLS_D, false, false, MO_UQ) /* Load Quadword */ TRANS64(LQ, do_ldst_quad, false, false); @@ -XXX,XX +XXX,XX @@ TRANS(STWUX, do_ldst_X, true, true, MO_UL) TRANS(PSTW, do_ldst_PLS_D, false, true, MO_UL) /* Store Doubleword */ -TRANS64(STD, do_ldst_D, false, true, MO_Q) -TRANS64(STDX, do_ldst_X, false, true, MO_Q) -TRANS64(STDU, do_ldst_D, true, true, MO_Q) -TRANS64(STDUX, do_ldst_X, true, true, MO_Q) -TRANS64(PSTD, do_ldst_PLS_D, false, true, MO_Q) +TRANS64(STD, do_ldst_D, false, true, MO_UQ) +TRANS64(STDX, do_ldst_X, false, true, MO_UQ) +TRANS64(STDU, do_ldst_D, true, true, MO_UQ) +TRANS64(STDUX, do_ldst_X, true, true, MO_UQ) +TRANS64(PSTD, do_ldst_PLS_D, false, true, MO_UQ) /* Store Quadword */ TRANS64(STQ, do_ldst_quad, true, false); diff --git a/target/ppc/translate/fp-impl.c.inc b/target/ppc/translate/fp-impl.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/translate/fp-impl.c.inc +++ b/target/ppc/translate/fp-impl.c.inc @@ -XXX,XX +XXX,XX @@ static void gen_lfdepx(DisasContext *ctx) EA = tcg_temp_new(); t0 = tcg_temp_new_i64(); gen_addr_reg_index(ctx, EA); - tcg_gen_qemu_ld_i64(t0, EA, PPC_TLB_EPID_LOAD, DEF_MEMOP(MO_Q)); + tcg_gen_qemu_ld_i64(t0, EA, PPC_TLB_EPID_LOAD, DEF_MEMOP(MO_UQ)); set_fpr(rD(ctx->opcode), t0); tcg_temp_free(EA); tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ static void gen_stfdepx(DisasContext *ctx) t0 = tcg_temp_new_i64(); gen_addr_reg_index(ctx, EA); get_fpr(t0, rD(ctx->opcode)); - tcg_gen_qemu_st_i64(t0, EA, PPC_TLB_EPID_STORE, DEF_MEMOP(MO_Q)); + tcg_gen_qemu_st_i64(t0, EA, PPC_TLB_EPID_STORE, DEF_MEMOP(MO_UQ)); tcg_temp_free(EA); tcg_temp_free_i64(t0); } diff --git a/target/ppc/translate/vsx-impl.c.inc b/target/ppc/translate/vsx-impl.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/translate/vsx-impl.c.inc +++ b/target/ppc/translate/vsx-impl.c.inc @@ -XXX,XX +XXX,XX @@ static void gen_lxvw4x(DisasContext *ctx) TCGv_i64 t0 = tcg_temp_new_i64(); TCGv_i64 t1 = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(t0, EA, ctx->mem_idx, MO_LEQ); + tcg_gen_qemu_ld_i64(t0, EA, ctx->mem_idx, MO_LEUQ); tcg_gen_shri_i64(t1, t0, 32); tcg_gen_deposit_i64(xth, t1, t0, 32, 32); tcg_gen_addi_tl(EA, EA, 8); - tcg_gen_qemu_ld_i64(t0, EA, ctx->mem_idx, MO_LEQ); + tcg_gen_qemu_ld_i64(t0, EA, ctx->mem_idx, MO_LEUQ); tcg_gen_shri_i64(t1, t0, 32); tcg_gen_deposit_i64(xtl, t1, t0, 32, 32); tcg_temp_free_i64(t0); tcg_temp_free_i64(t1); } else { - tcg_gen_qemu_ld_i64(xth, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_ld_i64(xth, EA, ctx->mem_idx, MO_BEUQ); tcg_gen_addi_tl(EA, EA, 8); - tcg_gen_qemu_ld_i64(xtl, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_ld_i64(xtl, EA, ctx->mem_idx, MO_BEUQ); } set_cpu_vsr(xT(ctx->opcode), xth, true); set_cpu_vsr(xT(ctx->opcode), xtl, false); @@ -XXX,XX +XXX,XX @@ static void gen_lxvdsx(DisasContext *ctx) gen_addr_reg_index(ctx, EA); data = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(data, EA, ctx->mem_idx, DEF_MEMOP(MO_Q)); - tcg_gen_gvec_dup_i64(MO_Q, vsr_full_offset(xT(ctx->opcode)), 16, 16, data); + tcg_gen_qemu_ld_i64(data, EA, ctx->mem_idx, DEF_MEMOP(MO_UQ)); + tcg_gen_gvec_dup_i64(MO_UQ, vsr_full_offset(xT(ctx->opcode)), 16, 16, data); tcg_temp_free(EA); tcg_temp_free_i64(data); @@ -XXX,XX +XXX,XX @@ static void gen_lxvh8x(DisasContext *ctx) EA = tcg_temp_new(); gen_addr_reg_index(ctx, EA); - tcg_gen_qemu_ld_i64(xth, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_ld_i64(xth, EA, ctx->mem_idx, MO_BEUQ); tcg_gen_addi_tl(EA, EA, 8); - tcg_gen_qemu_ld_i64(xtl, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_ld_i64(xtl, EA, ctx->mem_idx, MO_BEUQ); if (ctx->le_mode) { gen_bswap16x8(xth, xtl, xth, xtl); } @@ -XXX,XX +XXX,XX @@ static void gen_lxvb16x(DisasContext *ctx) gen_set_access_type(ctx, ACCESS_INT); EA = tcg_temp_new(); gen_addr_reg_index(ctx, EA); - tcg_gen_qemu_ld_i64(xth, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_ld_i64(xth, EA, ctx->mem_idx, MO_BEUQ); tcg_gen_addi_tl(EA, EA, 8); - tcg_gen_qemu_ld_i64(xtl, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_ld_i64(xtl, EA, ctx->mem_idx, MO_BEUQ); set_cpu_vsr(xT(ctx->opcode), xth, true); set_cpu_vsr(xT(ctx->opcode), xtl, false); tcg_temp_free(EA); @@ -XXX,XX +XXX,XX @@ static void gen_stxvw4x(DisasContext *ctx) tcg_gen_shri_i64(t0, xsh, 32); tcg_gen_deposit_i64(t1, t0, xsh, 32, 32); - tcg_gen_qemu_st_i64(t1, EA, ctx->mem_idx, MO_LEQ); + tcg_gen_qemu_st_i64(t1, EA, ctx->mem_idx, MO_LEUQ); tcg_gen_addi_tl(EA, EA, 8); tcg_gen_shri_i64(t0, xsl, 32); tcg_gen_deposit_i64(t1, t0, xsl, 32, 32); - tcg_gen_qemu_st_i64(t1, EA, ctx->mem_idx, MO_LEQ); + tcg_gen_qemu_st_i64(t1, EA, ctx->mem_idx, MO_LEUQ); tcg_temp_free_i64(t0); tcg_temp_free_i64(t1); } else { - tcg_gen_qemu_st_i64(xsh, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_st_i64(xsh, EA, ctx->mem_idx, MO_BEUQ); tcg_gen_addi_tl(EA, EA, 8); - tcg_gen_qemu_st_i64(xsl, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_st_i64(xsl, EA, ctx->mem_idx, MO_BEUQ); } tcg_temp_free(EA); tcg_temp_free_i64(xsh); @@ -XXX,XX +XXX,XX @@ static void gen_stxvh8x(DisasContext *ctx) TCGv_i64 outl = tcg_temp_new_i64(); gen_bswap16x8(outh, outl, xsh, xsl); - tcg_gen_qemu_st_i64(outh, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_st_i64(outh, EA, ctx->mem_idx, MO_BEUQ); tcg_gen_addi_tl(EA, EA, 8); - tcg_gen_qemu_st_i64(outl, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_st_i64(outl, EA, ctx->mem_idx, MO_BEUQ); tcg_temp_free_i64(outh); tcg_temp_free_i64(outl); } else { - tcg_gen_qemu_st_i64(xsh, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_st_i64(xsh, EA, ctx->mem_idx, MO_BEUQ); tcg_gen_addi_tl(EA, EA, 8); - tcg_gen_qemu_st_i64(xsl, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_st_i64(xsl, EA, ctx->mem_idx, MO_BEUQ); } tcg_temp_free(EA); tcg_temp_free_i64(xsh); @@ -XXX,XX +XXX,XX @@ static void gen_stxvb16x(DisasContext *ctx) gen_set_access_type(ctx, ACCESS_INT); EA = tcg_temp_new(); gen_addr_reg_index(ctx, EA); - tcg_gen_qemu_st_i64(xsh, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_st_i64(xsh, EA, ctx->mem_idx, MO_BEUQ); tcg_gen_addi_tl(EA, EA, 8); - tcg_gen_qemu_st_i64(xsl, EA, ctx->mem_idx, MO_BEQ); + tcg_gen_qemu_st_i64(xsl, EA, ctx->mem_idx, MO_BEUQ); tcg_temp_free(EA); tcg_temp_free_i64(xsh); tcg_temp_free_i64(xsl); @@ -XXX,XX +XXX,XX @@ static bool do_lstxv(DisasContext *ctx, int ra, TCGv displ, xt = tcg_temp_new_i64(); - mop = DEF_MEMOP(MO_Q); + mop = DEF_MEMOP(MO_UQ); gen_set_access_type(ctx, ACCESS_INT); ea = do_ea_calc(ctx, ra, displ); diff --git a/target/riscv/insn_trans/trans_rva.c.inc b/target/riscv/insn_trans/trans_rva.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rva.c.inc +++ b/target/riscv/insn_trans/trans_rva.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_amomaxu_w(DisasContext *ctx, arg_amomaxu_w *a) static bool trans_lr_d(DisasContext *ctx, arg_lr_d *a) { REQUIRE_64BIT(ctx); - return gen_lr(ctx, a, MO_ALIGN | MO_TEQ); + return gen_lr(ctx, a, MO_ALIGN | MO_TEUQ); } static bool trans_sc_d(DisasContext *ctx, arg_sc_d *a) { REQUIRE_64BIT(ctx); - return gen_sc(ctx, a, (MO_ALIGN | MO_TEQ)); + return gen_sc(ctx, a, (MO_ALIGN | MO_TEUQ)); } static bool trans_amoswap_d(DisasContext *ctx, arg_amoswap_d *a) { REQUIRE_64BIT(ctx); - return gen_amo(ctx, a, &tcg_gen_atomic_xchg_tl, (MO_ALIGN | MO_TEQ)); + return gen_amo(ctx, a, &tcg_gen_atomic_xchg_tl, (MO_ALIGN | MO_TEUQ)); } static bool trans_amoadd_d(DisasContext *ctx, arg_amoadd_d *a) { REQUIRE_64BIT(ctx); - return gen_amo(ctx, a, &tcg_gen_atomic_fetch_add_tl, (MO_ALIGN | MO_TEQ)); + return gen_amo(ctx, a, &tcg_gen_atomic_fetch_add_tl, (MO_ALIGN | MO_TEUQ)); } static bool trans_amoxor_d(DisasContext *ctx, arg_amoxor_d *a) { REQUIRE_64BIT(ctx); - return gen_amo(ctx, a, &tcg_gen_atomic_fetch_xor_tl, (MO_ALIGN | MO_TEQ)); + return gen_amo(ctx, a, &tcg_gen_atomic_fetch_xor_tl, (MO_ALIGN | MO_TEUQ)); } static bool trans_amoand_d(DisasContext *ctx, arg_amoand_d *a) { REQUIRE_64BIT(ctx); - return gen_amo(ctx, a, &tcg_gen_atomic_fetch_and_tl, (MO_ALIGN | MO_TEQ)); + return gen_amo(ctx, a, &tcg_gen_atomic_fetch_and_tl, (MO_ALIGN | MO_TEUQ)); } static bool trans_amoor_d(DisasContext *ctx, arg_amoor_d *a) { REQUIRE_64BIT(ctx); - return gen_amo(ctx, a, &tcg_gen_atomic_fetch_or_tl, (MO_ALIGN | MO_TEQ)); + return gen_amo(ctx, a, &tcg_gen_atomic_fetch_or_tl, (MO_ALIGN | MO_TEUQ)); } static bool trans_amomin_d(DisasContext *ctx, arg_amomin_d *a) { REQUIRE_64BIT(ctx); - return gen_amo(ctx, a, &tcg_gen_atomic_fetch_smin_tl, (MO_ALIGN | MO_TEQ)); + return gen_amo(ctx, a, &tcg_gen_atomic_fetch_smin_tl, (MO_ALIGN | MO_TEUQ)); } static bool trans_amomax_d(DisasContext *ctx, arg_amomax_d *a) { REQUIRE_64BIT(ctx); - return gen_amo(ctx, a, &tcg_gen_atomic_fetch_smax_tl, (MO_ALIGN | MO_TEQ)); + return gen_amo(ctx, a, &tcg_gen_atomic_fetch_smax_tl, (MO_ALIGN | MO_TEUQ)); } static bool trans_amominu_d(DisasContext *ctx, arg_amominu_d *a) { REQUIRE_64BIT(ctx); - return gen_amo(ctx, a, &tcg_gen_atomic_fetch_umin_tl, (MO_ALIGN | MO_TEQ)); + return gen_amo(ctx, a, &tcg_gen_atomic_fetch_umin_tl, (MO_ALIGN | MO_TEUQ)); } static bool trans_amomaxu_d(DisasContext *ctx, arg_amomaxu_d *a) { REQUIRE_64BIT(ctx); - return gen_amo(ctx, a, &tcg_gen_atomic_fetch_umax_tl, (MO_ALIGN | MO_TEQ)); + return gen_amo(ctx, a, &tcg_gen_atomic_fetch_umax_tl, (MO_ALIGN | MO_TEUQ)); } diff --git a/target/riscv/insn_trans/trans_rvd.c.inc b/target/riscv/insn_trans/trans_rvd.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvd.c.inc +++ b/target/riscv/insn_trans/trans_rvd.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_fld(DisasContext *ctx, arg_fld *a) } addr = gen_pm_adjust_address(ctx, addr); - tcg_gen_qemu_ld_i64(cpu_fpr[a->rd], addr, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_ld_i64(cpu_fpr[a->rd], addr, ctx->mem_idx, MO_TEUQ); mark_fs_dirty(ctx); return true; @@ -XXX,XX +XXX,XX @@ static bool trans_fsd(DisasContext *ctx, arg_fsd *a) } addr = gen_pm_adjust_address(ctx, addr); - tcg_gen_qemu_st_i64(cpu_fpr[a->rs2], addr, ctx->mem_idx, MO_TEQ); + tcg_gen_qemu_st_i64(cpu_fpr[a->rs2], addr, ctx->mem_idx, MO_TEUQ); return true; } diff --git a/target/riscv/insn_trans/trans_rvh.c.inc b/target/riscv/insn_trans/trans_rvh.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvh.c.inc +++ b/target/riscv/insn_trans/trans_rvh.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_hlv_d(DisasContext *ctx, arg_hlv_d *a) { REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVH); - return do_hlv(ctx, a, MO_TEQ); + return do_hlv(ctx, a, MO_TEUQ); } static bool trans_hsv_d(DisasContext *ctx, arg_hsv_d *a) { REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVH); - return do_hsv(ctx, a, MO_TEQ); + return do_hsv(ctx, a, MO_TEUQ); } #ifndef CONFIG_USER_ONLY diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_lwu(DisasContext *ctx, arg_lwu *a) static bool trans_ld(DisasContext *ctx, arg_ld *a) { REQUIRE_64BIT(ctx); - return gen_load(ctx, a, MO_TEQ); + return gen_load(ctx, a, MO_TEUQ); } static bool trans_sd(DisasContext *ctx, arg_sd *a) { REQUIRE_64BIT(ctx); - return gen_store(ctx, a, MO_TEQ); + return gen_store(ctx, a, MO_TEUQ); } static bool trans_addi(DisasContext *ctx, arg_addi *a) diff --git a/target/s390x/tcg/translate_vx.c.inc b/target/s390x/tcg/translate_vx.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/tcg/translate_vx.c.inc +++ b/target/s390x/tcg/translate_vx.c.inc @@ -XXX,XX +XXX,XX @@ static DisasJumpType op_vl(DisasContext *s, DisasOps *o) TCGv_i64 t0 = tcg_temp_new_i64(); TCGv_i64 t1 = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(t0, o->addr1, get_mem_index(s), MO_TEQ); + tcg_gen_qemu_ld_i64(t0, o->addr1, get_mem_index(s), MO_TEUQ); gen_addi_and_wrap_i64(s, o->addr1, o->addr1, 8); - tcg_gen_qemu_ld_i64(t1, o->addr1, get_mem_index(s), MO_TEQ); + tcg_gen_qemu_ld_i64(t1, o->addr1, get_mem_index(s), MO_TEUQ); write_vec_element_i64(t0, get_field(s, v1), 0, ES_64); write_vec_element_i64(t1, get_field(s, v1), 1, ES_64); tcg_temp_free(t0); @@ -XXX,XX +XXX,XX @@ static DisasJumpType op_vlm(DisasContext *s, DisasOps *o) t0 = tcg_temp_new_i64(); t1 = tcg_temp_new_i64(); gen_addi_and_wrap_i64(s, t0, o->addr1, (v3 - v1) * 16 + 8); - tcg_gen_qemu_ld_i64(t0, t0, get_mem_index(s), MO_TEQ); + tcg_gen_qemu_ld_i64(t0, t0, get_mem_index(s), MO_TEUQ); for (;; v1++) { - tcg_gen_qemu_ld_i64(t1, o->addr1, get_mem_index(s), MO_TEQ); + tcg_gen_qemu_ld_i64(t1, o->addr1, get_mem_index(s), MO_TEUQ); write_vec_element_i64(t1, v1, 0, ES_64); if (v1 == v3) { break; } gen_addi_and_wrap_i64(s, o->addr1, o->addr1, 8); - tcg_gen_qemu_ld_i64(t1, o->addr1, get_mem_index(s), MO_TEQ); + tcg_gen_qemu_ld_i64(t1, o->addr1, get_mem_index(s), MO_TEUQ); write_vec_element_i64(t1, v1, 1, ES_64); gen_addi_and_wrap_i64(s, o->addr1, o->addr1, 8); } @@ -XXX,XX +XXX,XX @@ static DisasJumpType op_vst(DisasContext *s, DisasOps *o) gen_helper_probe_write_access(cpu_env, o->addr1, tmp); read_vec_element_i64(tmp, get_field(s, v1), 0, ES_64); - tcg_gen_qemu_st_i64(tmp, o->addr1, get_mem_index(s), MO_TEQ); + tcg_gen_qemu_st_i64(tmp, o->addr1, get_mem_index(s), MO_TEUQ); gen_addi_and_wrap_i64(s, o->addr1, o->addr1, 8); read_vec_element_i64(tmp, get_field(s, v1), 1, ES_64); - tcg_gen_qemu_st_i64(tmp, o->addr1, get_mem_index(s), MO_TEQ); + tcg_gen_qemu_st_i64(tmp, o->addr1, get_mem_index(s), MO_TEUQ); tcg_temp_free_i64(tmp); return DISAS_NEXT; } @@ -XXX,XX +XXX,XX @@ static DisasJumpType op_vstm(DisasContext *s, DisasOps *o) for (;; v1++) { read_vec_element_i64(tmp, v1, 0, ES_64); - tcg_gen_qemu_st_i64(tmp, o->addr1, get_mem_index(s), MO_TEQ); + tcg_gen_qemu_st_i64(tmp, o->addr1, get_mem_index(s), MO_TEUQ); gen_addi_and_wrap_i64(s, o->addr1, o->addr1, 8); read_vec_element_i64(tmp, v1, 1, ES_64); - tcg_gen_qemu_st_i64(tmp, o->addr1, get_mem_index(s), MO_TEQ); + tcg_gen_qemu_st_i64(tmp, o->addr1, get_mem_index(s), MO_TEUQ); if (v1 == v3) { break; } diff --git a/tcg/aarch64/tcg-target.c.inc b/tcg/aarch64/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/aarch64/tcg-target.c.inc +++ b/tcg/aarch64/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void tcg_out_qemu_ld_direct(TCGContext *s, MemOp memop, TCGType ext, case MO_SL: tcg_out_ldst_r(s, I3312_LDRSWX, data_r, addr_r, otype, off_r); break; - case MO_Q: + case MO_UQ: tcg_out_ldst_r(s, I3312_LDRX, data_r, addr_r, otype, off_r); break; default: diff --git a/tcg/arm/tcg-target.c.inc b/tcg/arm/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/arm/tcg-target.c.inc +++ b/tcg/arm/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void * const qemu_ld_helpers[MO_SSIZE + 1] = { #ifdef HOST_WORDS_BIGENDIAN [MO_UW] = helper_be_lduw_mmu, [MO_UL] = helper_be_ldul_mmu, - [MO_Q] = helper_be_ldq_mmu, + [MO_UQ] = helper_be_ldq_mmu, [MO_SW] = helper_be_ldsw_mmu, [MO_SL] = helper_be_ldul_mmu, #else [MO_UW] = helper_le_lduw_mmu, [MO_UL] = helper_le_ldul_mmu, - [MO_Q] = helper_le_ldq_mmu, + [MO_UQ] = helper_le_ldq_mmu, [MO_SW] = helper_le_ldsw_mmu, [MO_SL] = helper_le_ldul_mmu, #endif @@ -XXX,XX +XXX,XX @@ static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *lb) default: tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_R0); break; - case MO_Q: + case MO_UQ: if (datalo != TCG_REG_R1) { tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_R0); tcg_out_mov_reg(s, COND_AL, datahi, TCG_REG_R1); @@ -XXX,XX +XXX,XX @@ static void tcg_out_qemu_ld_index(TCGContext *s, MemOp opc, case MO_UL: tcg_out_ld32_r(s, COND_AL, datalo, addrlo, addend); break; - case MO_Q: + case MO_UQ: /* Avoid ldrd for user-only emulation, to handle unaligned. */ if (USING_SOFTMMU && use_armv6_instructions && (datalo & 1) == 0 && datahi == datalo + 1) { @@ -XXX,XX +XXX,XX @@ static void tcg_out_qemu_ld_direct(TCGContext *s, MemOp opc, TCGReg datalo, case MO_UL: tcg_out_ld32_12(s, COND_AL, datalo, addrlo, 0); break; - case MO_Q: + case MO_UQ: /* Avoid ldrd for user-only emulation, to handle unaligned. */ if (USING_SOFTMMU && use_armv6_instructions && (datalo & 1) == 0 && datahi == datalo + 1) { diff --git a/tcg/i386/tcg-target.c.inc b/tcg/i386/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/i386/tcg-target.c.inc +++ b/tcg/i386/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void * const qemu_ld_helpers[(MO_SIZE | MO_BSWAP) + 1] = { [MO_UB] = helper_ret_ldub_mmu, [MO_LEUW] = helper_le_lduw_mmu, [MO_LEUL] = helper_le_ldul_mmu, - [MO_LEQ] = helper_le_ldq_mmu, + [MO_LEUQ] = helper_le_ldq_mmu, [MO_BEUW] = helper_be_lduw_mmu, [MO_BEUL] = helper_be_ldul_mmu, - [MO_BEQ] = helper_be_ldq_mmu, + [MO_BEUQ] = helper_be_ldq_mmu, }; /* helper signature: helper_ret_st_mmu(CPUState *env, target_ulong addr, @@ -XXX,XX +XXX,XX @@ static void * const qemu_st_helpers[(MO_SIZE | MO_BSWAP) + 1] = { [MO_UB] = helper_ret_stb_mmu, [MO_LEUW] = helper_le_stw_mmu, [MO_LEUL] = helper_le_stl_mmu, - [MO_LEQ] = helper_le_stq_mmu, + [MO_LEUQ] = helper_le_stq_mmu, [MO_BEUW] = helper_be_stw_mmu, [MO_BEUL] = helper_be_stl_mmu, - [MO_BEQ] = helper_be_stq_mmu, + [MO_BEUQ] = helper_be_stq_mmu, }; /* Perform the TLB load and compare. @@ -XXX,XX +XXX,XX @@ static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *l) case MO_UL: tcg_out_mov(s, TCG_TYPE_I32, data_reg, TCG_REG_EAX); break; - case MO_Q: + case MO_UQ: if (TCG_TARGET_REG_BITS == 64) { tcg_out_mov(s, TCG_TYPE_I64, data_reg, TCG_REG_RAX); } else if (data_reg == TCG_REG_EDX) { @@ -XXX,XX +XXX,XX @@ static void tcg_out_qemu_ld_direct(TCGContext *s, TCGReg datalo, TCGReg datahi, } break; #endif - case MO_Q: + case MO_UQ: if (TCG_TARGET_REG_BITS == 64) { tcg_out_modrm_sib_offset(s, movop + P_REXW + seg, datalo, base, index, 0, ofs); diff --git a/tcg/mips/tcg-target.c.inc b/tcg/mips/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/mips/tcg-target.c.inc +++ b/tcg/mips/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void * const qemu_ld_helpers[(MO_SSIZE | MO_BSWAP) + 1] = { [MO_LEUW] = helper_le_lduw_mmu, [MO_LESW] = helper_le_ldsw_mmu, [MO_LEUL] = helper_le_ldul_mmu, - [MO_LEQ] = helper_le_ldq_mmu, + [MO_LEUQ] = helper_le_ldq_mmu, [MO_BEUW] = helper_be_lduw_mmu, [MO_BESW] = helper_be_ldsw_mmu, [MO_BEUL] = helper_be_ldul_mmu, - [MO_BEQ] = helper_be_ldq_mmu, + [MO_BEUQ] = helper_be_ldq_mmu, #if TCG_TARGET_REG_BITS == 64 [MO_LESL] = helper_le_ldsl_mmu, [MO_BESL] = helper_be_ldsl_mmu, @@ -XXX,XX +XXX,XX @@ static void * const qemu_st_helpers[(MO_SIZE | MO_BSWAP) + 1] = { [MO_UB] = helper_ret_stb_mmu, [MO_LEUW] = helper_le_stw_mmu, [MO_LEUL] = helper_le_stl_mmu, - [MO_LEQ] = helper_le_stq_mmu, + [MO_LEUQ] = helper_le_stq_mmu, [MO_BEUW] = helper_be_stw_mmu, [MO_BEUL] = helper_be_stl_mmu, - [MO_BEQ] = helper_be_stq_mmu, + [MO_BEUQ] = helper_be_stq_mmu, }; /* Helper routines for marshalling helper function arguments into @@ -XXX,XX +XXX,XX @@ static void tcg_out_qemu_ld_direct(TCGContext *s, TCGReg lo, TCGReg hi, case MO_SL: tcg_out_opc_imm(s, OPC_LW, lo, base, 0); break; - case MO_Q | MO_BSWAP: + case MO_UQ | MO_BSWAP: if (TCG_TARGET_REG_BITS == 64) { if (use_mips32r2_instructions) { tcg_out_opc_imm(s, OPC_LD, lo, base, 0); @@ -XXX,XX +XXX,XX @@ static void tcg_out_qemu_ld_direct(TCGContext *s, TCGReg lo, TCGReg hi, tcg_out_mov(s, TCG_TYPE_I32, MIPS_BE ? hi : lo, TCG_TMP3); } break; - case MO_Q: + case MO_UQ: /* Prefer to load from offset 0 first, but allow for overlap. */ if (TCG_TARGET_REG_BITS == 64) { tcg_out_opc_imm(s, OPC_LD, lo, base, 0); diff --git a/tcg/ppc/tcg-target.c.inc b/tcg/ppc/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/ppc/tcg-target.c.inc +++ b/tcg/ppc/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static const uint32_t qemu_ldx_opc[(MO_SSIZE + MO_BSWAP) + 1] = { [MO_UB] = LBZX, [MO_UW] = LHZX, [MO_UL] = LWZX, - [MO_Q] = LDX, + [MO_UQ] = LDX, [MO_SW] = LHAX, [MO_SL] = LWAX, [MO_BSWAP | MO_UB] = LBZX, [MO_BSWAP | MO_UW] = LHBRX, [MO_BSWAP | MO_UL] = LWBRX, - [MO_BSWAP | MO_Q] = LDBRX, + [MO_BSWAP | MO_UQ] = LDBRX, }; static const uint32_t qemu_stx_opc[(MO_SIZE + MO_BSWAP) + 1] = { [MO_UB] = STBX, [MO_UW] = STHX, [MO_UL] = STWX, - [MO_Q] = STDX, + [MO_UQ] = STDX, [MO_BSWAP | MO_UB] = STBX, [MO_BSWAP | MO_UW] = STHBRX, [MO_BSWAP | MO_UL] = STWBRX, - [MO_BSWAP | MO_Q] = STDBRX, + [MO_BSWAP | MO_UQ] = STDBRX, }; static const uint32_t qemu_exts_opc[4] = { @@ -XXX,XX +XXX,XX @@ static void * const qemu_ld_helpers[(MO_SIZE | MO_BSWAP) + 1] = { [MO_UB] = helper_ret_ldub_mmu, [MO_LEUW] = helper_le_lduw_mmu, [MO_LEUL] = helper_le_ldul_mmu, - [MO_LEQ] = helper_le_ldq_mmu, + [MO_LEUQ] = helper_le_ldq_mmu, [MO_BEUW] = helper_be_lduw_mmu, [MO_BEUL] = helper_be_ldul_mmu, - [MO_BEQ] = helper_be_ldq_mmu, + [MO_BEUQ] = helper_be_ldq_mmu, }; /* helper signature: helper_st_mmu(CPUState *env, target_ulong addr, @@ -XXX,XX +XXX,XX @@ static void * const qemu_st_helpers[(MO_SIZE | MO_BSWAP) + 1] = { [MO_UB] = helper_ret_stb_mmu, [MO_LEUW] = helper_le_stw_mmu, [MO_LEUL] = helper_le_stl_mmu, - [MO_LEQ] = helper_le_stq_mmu, + [MO_LEUQ] = helper_le_stq_mmu, [MO_BEUW] = helper_be_stw_mmu, [MO_BEUL] = helper_be_stl_mmu, - [MO_BEQ] = helper_be_stq_mmu, + [MO_BEUQ] = helper_be_stq_mmu, }; /* We expect to use a 16-bit negative offset from ENV. */ diff --git a/tcg/riscv/tcg-target.c.inc b/tcg/riscv/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/riscv/tcg-target.c.inc +++ b/tcg/riscv/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void * const qemu_ld_helpers[MO_SSIZE + 1] = { #if TCG_TARGET_REG_BITS == 64 [MO_SL] = helper_be_ldsl_mmu, #endif - [MO_Q] = helper_be_ldq_mmu, + [MO_UQ] = helper_be_ldq_mmu, #else [MO_UW] = helper_le_lduw_mmu, [MO_SW] = helper_le_ldsw_mmu, @@ -XXX,XX +XXX,XX @@ static void * const qemu_ld_helpers[MO_SSIZE + 1] = { #if TCG_TARGET_REG_BITS == 64 [MO_SL] = helper_le_ldsl_mmu, #endif - [MO_Q] = helper_le_ldq_mmu, + [MO_UQ] = helper_le_ldq_mmu, #endif }; @@ -XXX,XX +XXX,XX @@ static void tcg_out_qemu_ld_direct(TCGContext *s, TCGReg lo, TCGReg hi, case MO_SL: tcg_out_opc_imm(s, OPC_LW, lo, base, 0); break; - case MO_Q: + case MO_UQ: /* Prefer to load from offset 0 first, but allow for overlap. */ if (TCG_TARGET_REG_BITS == 64) { tcg_out_opc_imm(s, OPC_LD, lo, base, 0); diff --git a/tcg/s390x/tcg-target.c.inc b/tcg/s390x/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/s390x/tcg-target.c.inc +++ b/tcg/s390x/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void * const qemu_ld_helpers[(MO_SSIZE | MO_BSWAP) + 1] = { [MO_LESW] = helper_le_ldsw_mmu, [MO_LEUL] = helper_le_ldul_mmu, [MO_LESL] = helper_le_ldsl_mmu, - [MO_LEQ] = helper_le_ldq_mmu, + [MO_LEUQ] = helper_le_ldq_mmu, [MO_BEUW] = helper_be_lduw_mmu, [MO_BESW] = helper_be_ldsw_mmu, [MO_BEUL] = helper_be_ldul_mmu, [MO_BESL] = helper_be_ldsl_mmu, - [MO_BEQ] = helper_be_ldq_mmu, + [MO_BEUQ] = helper_be_ldq_mmu, }; static void * const qemu_st_helpers[(MO_SIZE | MO_BSWAP) + 1] = { [MO_UB] = helper_ret_stb_mmu, [MO_LEUW] = helper_le_stw_mmu, [MO_LEUL] = helper_le_stl_mmu, - [MO_LEQ] = helper_le_stq_mmu, + [MO_LEUQ] = helper_le_stq_mmu, [MO_BEUW] = helper_be_stw_mmu, [MO_BEUL] = helper_be_stl_mmu, - [MO_BEQ] = helper_be_stq_mmu, + [MO_BEUQ] = helper_be_stq_mmu, }; #endif @@ -XXX,XX +XXX,XX @@ static void tcg_out_qemu_ld_direct(TCGContext *s, MemOp opc, TCGReg data, tcg_out_insn(s, RXY, LGF, data, base, index, disp); break; - case MO_Q | MO_BSWAP: + case MO_UQ | MO_BSWAP: tcg_out_insn(s, RXY, LRVG, data, base, index, disp); break; - case MO_Q: + case MO_UQ: tcg_out_insn(s, RXY, LG, data, base, index, disp); break; @@ -XXX,XX +XXX,XX @@ static void tcg_out_qemu_st_direct(TCGContext *s, MemOp opc, TCGReg data, } break; - case MO_Q | MO_BSWAP: + case MO_UQ | MO_BSWAP: tcg_out_insn(s, RXY, STRVG, data, base, index, disp); break; - case MO_Q: + case MO_UQ: tcg_out_insn(s, RXY, STG, data, base, index, disp); break; @@ -XXX,XX +XXX,XX @@ static bool tcg_out_qemu_st_slow_path(TCGContext *s, TCGLabelQemuLdst *lb) case MO_UL: tgen_ext32u(s, TCG_REG_R4, data_reg); break; - case MO_Q: + case MO_UQ: tcg_out_mov(s, TCG_TYPE_I64, TCG_REG_R4, data_reg); break; default: diff --git a/tcg/sparc/tcg-target.c.inc b/tcg/sparc/tcg-target.c.inc index XXXXXXX..XXXXXXX 100644 --- a/tcg/sparc/tcg-target.c.inc +++ b/tcg/sparc/tcg-target.c.inc @@ -XXX,XX +XXX,XX @@ static void build_trampolines(TCGContext *s) [MO_LEUW] = helper_le_lduw_mmu, [MO_LESW] = helper_le_ldsw_mmu, [MO_LEUL] = helper_le_ldul_mmu, - [MO_LEQ] = helper_le_ldq_mmu, + [MO_LEUQ] = helper_le_ldq_mmu, [MO_BEUW] = helper_be_lduw_mmu, [MO_BESW] = helper_be_ldsw_mmu, [MO_BEUL] = helper_be_ldul_mmu, - [MO_BEQ] = helper_be_ldq_mmu, + [MO_BEUQ] = helper_be_ldq_mmu, }; static void * const qemu_st_helpers[] = { [MO_UB] = helper_ret_stb_mmu, [MO_LEUW] = helper_le_stw_mmu, [MO_LEUL] = helper_le_stl_mmu, - [MO_LEQ] = helper_le_stq_mmu, + [MO_LEUQ] = helper_le_stq_mmu, [MO_BEUW] = helper_be_stw_mmu, [MO_BEUL] = helper_be_stl_mmu, - [MO_BEQ] = helper_be_stq_mmu, + [MO_BEUQ] = helper_be_stq_mmu, }; int i; @@ -XXX,XX +XXX,XX @@ static const int qemu_ld_opc[(MO_SSIZE | MO_BSWAP) + 1] = { [MO_BESW] = LDSH, [MO_BEUL] = LDUW, [MO_BESL] = LDSW, - [MO_BEQ] = LDX, + [MO_BEUQ] = LDX, [MO_LEUW] = LDUH_LE, [MO_LESW] = LDSH_LE, [MO_LEUL] = LDUW_LE, [MO_LESL] = LDSW_LE, - [MO_LEQ] = LDX_LE, + [MO_LEUQ] = LDX_LE, }; static const int qemu_st_opc[(MO_SIZE | MO_BSWAP) + 1] = { @@ -XXX,XX +XXX,XX @@ static const int qemu_st_opc[(MO_SIZE | MO_BSWAP) + 1] = { [MO_BEUW] = STH, [MO_BEUL] = STW, - [MO_BEQ] = STX, + [MO_BEUQ] = STX, [MO_LEUW] = STH_LE, [MO_LEUL] = STW_LE, - [MO_LEQ] = STX_LE, + [MO_LEUQ] = STX_LE, }; static void tcg_out_qemu_ld(TCGContext *s, TCGReg data, TCGReg addr, diff --git a/target/s390x/tcg/insn-data.def b/target/s390x/tcg/insn-data.def index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/tcg/insn-data.def +++ b/target/s390x/tcg/insn-data.def @@ -XXX,XX +XXX,XX @@ D(0xeb6a, ASI, SIY, GIE, la1, i2, new, 0, asi, adds32, MO_TESL) C(0xecd8, AHIK, RIE_d, DO, r3, i2, new, r1_32, add, adds32) C(0xc208, AGFI, RIL_a, EI, r1, i2, r1, 0, add, adds64) - D(0xeb7a, AGSI, SIY, GIE, la1, i2, new, 0, asi, adds64, MO_TEQ) + D(0xeb7a, AGSI, SIY, GIE, la1, i2, new, 0, asi, adds64, MO_TEUQ) C(0xecd9, AGHIK, RIE_d, DO, r3, i2, r1, 0, add, adds64) /* ADD IMMEDIATE HIGH */ C(0xcc08, AIH, RIL_a, HW, r1_sr32, i2, new, r1_32h, add, adds32) @@ -XXX,XX +XXX,XX @@ /* ADD LOGICAL WITH SIGNED IMMEDIATE */ D(0xeb6e, ALSI, SIY, GIE, la1, i2_32u, new, 0, asi, addu32, MO_TEUL) C(0xecda, ALHSIK, RIE_d, DO, r3_32u, i2_32u, new, r1_32, add, addu32) - D(0xeb7e, ALGSI, SIY, GIE, la1, i2, new, 0, asiu64, addu64, MO_TEQ) + D(0xeb7e, ALGSI, SIY, GIE, la1, i2, new, 0, asiu64, addu64, MO_TEUQ) C(0xecdb, ALGHSIK, RIE_d, DO, r3, i2, r1, 0, addu64, addu64) /* ADD LOGICAL WITH SIGNED IMMEDIATE HIGH */ C(0xcc0a, ALSIH, RIL_a, HW, r1_sr32, i2_32u, new, r1_32h, add, addu32) @@ -XXX,XX +XXX,XX @@ /* COMPARE AND SWAP */ D(0xba00, CS, RS_a, Z, r3_32u, r1_32u, new, r1_32, cs, 0, MO_TEUL) D(0xeb14, CSY, RSY_a, LD, r3_32u, r1_32u, new, r1_32, cs, 0, MO_TEUL) - D(0xeb30, CSG, RSY_a, Z, r3_o, r1_o, new, r1, cs, 0, MO_TEQ) + D(0xeb30, CSG, RSY_a, Z, r3_o, r1_o, new, r1, cs, 0, MO_TEUQ) /* COMPARE DOUBLE AND SWAP */ - D(0xbb00, CDS, RS_a, Z, r3_D32, r1_D32, new, r1_D32, cs, 0, MO_TEQ) - D(0xeb31, CDSY, RSY_a, LD, r3_D32, r1_D32, new, r1_D32, cs, 0, MO_TEQ) + D(0xbb00, CDS, RS_a, Z, r3_D32, r1_D32, new, r1_D32, cs, 0, MO_TEUQ) + D(0xeb31, CDSY, RSY_a, LD, r3_D32, r1_D32, new, r1_D32, cs, 0, MO_TEUQ) C(0xeb3e, CDSG, RSY_a, Z, 0, 0, 0, 0, cdsg, 0) /* COMPARE AND SWAP AND STORE */ C(0xc802, CSST, SSF, CASS, la1, a2, 0, 0, csst, 0) @@ -XXX,XX +XXX,XX @@ C(0xc000, LARL, RIL_b, Z, 0, ri2, 0, r1, mov2, 0) /* LOAD AND ADD */ D(0xebf8, LAA, RSY_a, ILA, r3_32s, a2, new, in2_r1_32, laa, adds32, MO_TESL) - D(0xebe8, LAAG, RSY_a, ILA, r3, a2, new, in2_r1, laa, adds64, MO_TEQ) + D(0xebe8, LAAG, RSY_a, ILA, r3, a2, new, in2_r1, laa, adds64, MO_TEUQ) /* LOAD AND ADD LOGICAL */ D(0xebfa, LAAL, RSY_a, ILA, r3_32u, a2, new, in2_r1_32, laa, addu32, MO_TEUL) - D(0xebea, LAALG, RSY_a, ILA, r3, a2, new, in2_r1, laa, addu64, MO_TEQ) + D(0xebea, LAALG, RSY_a, ILA, r3, a2, new, in2_r1, laa, addu64, MO_TEUQ) /* LOAD AND AND */ D(0xebf4, LAN, RSY_a, ILA, r3_32s, a2, new, in2_r1_32, lan, nz32, MO_TESL) - D(0xebe4, LANG, RSY_a, ILA, r3, a2, new, in2_r1, lan, nz64, MO_TEQ) + D(0xebe4, LANG, RSY_a, ILA, r3, a2, new, in2_r1, lan, nz64, MO_TEUQ) /* LOAD AND EXCLUSIVE OR */ D(0xebf7, LAX, RSY_a, ILA, r3_32s, a2, new, in2_r1_32, lax, nz32, MO_TESL) - D(0xebe7, LAXG, RSY_a, ILA, r3, a2, new, in2_r1, lax, nz64, MO_TEQ) + D(0xebe7, LAXG, RSY_a, ILA, r3, a2, new, in2_r1, lax, nz64, MO_TEUQ) /* LOAD AND OR */ D(0xebf6, LAO, RSY_a, ILA, r3_32s, a2, new, in2_r1_32, lao, nz32, MO_TESL) - D(0xebe6, LAOG, RSY_a, ILA, r3, a2, new, in2_r1, lao, nz64, MO_TEQ) + D(0xebe6, LAOG, RSY_a, ILA, r3, a2, new, in2_r1, lao, nz64, MO_TEUQ) /* LOAD AND TEST */ C(0x1200, LTR, RR_a, Z, 0, r2_o, 0, cond_r1r2_32, mov2, s32) C(0xb902, LTGR, RRE, Z, 0, r2_o, 0, r1, mov2, s64) @@ -XXX,XX +XXX,XX @@ C(0xebe0, LOCFH, RSY_b, LOC2, r1_sr32, m2_32u, new, r1_32h, loc, 0) /* LOAD PAIR DISJOINT */ D(0xc804, LPD, SSF, ILA, 0, 0, new_P, r3_P32, lpd, 0, MO_TEUL) - D(0xc805, LPDG, SSF, ILA, 0, 0, new_P, r3_P64, lpd, 0, MO_TEQ) + D(0xc805, LPDG, SSF, ILA, 0, 0, new_P, r3_P64, lpd, 0, MO_TEUQ) /* LOAD PAIR FROM QUADWORD */ C(0xe38f, LPQ, RXY_a, Z, 0, a2, r1_P, 0, lpq, 0) /* LOAD POSITIVE */ @@ -XXX,XX +XXX,XX @@ #ifndef CONFIG_USER_ONLY /* COMPARE AND SWAP AND PURGE */ E(0xb250, CSP, RRE, Z, r1_32u, ra2, r1_P, 0, csp, 0, MO_TEUL, IF_PRIV) - E(0xb98a, CSPG, RRE, DAT_ENH, r1_o, ra2, r1_P, 0, csp, 0, MO_TEQ, IF_PRIV) + E(0xb98a, CSPG, RRE, DAT_ENH, r1_o, ra2, r1_P, 0, csp, 0, MO_TEUQ, IF_PRIV) /* DIAGNOSE (KVM hypercall) */ F(0x8300, DIAG, RSI, Z, 0, 0, 0, 0, diag, 0, IF_PRIV | IF_IO) /* INSERT STORAGE KEY EXTENDED */ @@ -XXX,XX +XXX,XX @@ F(0xe303, LRAG, RXY_a, Z, 0, a2, r1, 0, lra, 0, IF_PRIV) /* LOAD USING REAL ADDRESS */ E(0xb24b, LURA, RRE, Z, 0, ra2, new, r1_32, lura, 0, MO_TEUL, IF_PRIV) - E(0xb905, LURAG, RRE, Z, 0, ra2, r1, 0, lura, 0, MO_TEQ, IF_PRIV) + E(0xb905, LURAG, RRE, Z, 0, ra2, r1, 0, lura, 0, MO_TEUQ, IF_PRIV) /* MOVE TO PRIMARY */ F(0xda00, MVCP, SS_d, Z, la1, a2, 0, 0, mvcp, 0, IF_PRIV) /* MOVE TO SECONDARY */ @@ -XXX,XX +XXX,XX @@ F(0xad00, STOSM, SI, Z, la1, 0, 0, 0, stnosm, 0, IF_PRIV) /* STORE USING REAL ADDRESS */ E(0xb246, STURA, RRE, Z, r1_o, ra2, 0, 0, stura, 0, MO_TEUL, IF_PRIV) - E(0xb925, STURG, RRE, Z, r1_o, ra2, 0, 0, stura, 0, MO_TEQ, IF_PRIV) + E(0xb925, STURG, RRE, Z, r1_o, ra2, 0, 0, stura, 0, MO_TEUQ, IF_PRIV) /* TEST BLOCK */ F(0xb22c, TB, RRE, Z, 0, r2_o, 0, 0, testblock, 0, IF_PRIV) /* TEST PROTECTION */ -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Adding defines to handle signed 64-bit and unsigned 128-bit quantities in memory accesses. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-3-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- include/exec/memop.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/include/exec/memop.h b/include/exec/memop.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/memop.h +++ b/include/exec/memop.h @@ -XXX,XX +XXX,XX @@ typedef enum MemOp { MO_UW = MO_16, MO_UL = MO_32, MO_UQ = MO_64, + MO_UO = MO_128, MO_SB = MO_SIGN | MO_8, MO_SW = MO_SIGN | MO_16, MO_SL = MO_SIGN | MO_32, + MO_SQ = MO_SIGN | MO_64, + MO_SO = MO_SIGN | MO_128, MO_LEUW = MO_LE | MO_UW, MO_LEUL = MO_LE | MO_UL, MO_LEUQ = MO_LE | MO_UQ, MO_LESW = MO_LE | MO_SW, MO_LESL = MO_LE | MO_SL, + MO_LESQ = MO_LE | MO_SQ, MO_BEUW = MO_BE | MO_UW, MO_BEUL = MO_BE | MO_UL, MO_BEUQ = MO_BE | MO_UQ, MO_BESW = MO_BE | MO_SW, MO_BESL = MO_BE | MO_SL, + MO_BESQ = MO_BE | MO_SQ, #ifdef NEED_CPU_H MO_TEUW = MO_TE | MO_UW, MO_TEUL = MO_TE | MO_UL, MO_TEUQ = MO_TE | MO_UQ, + MO_TEUO = MO_TE | MO_UO, MO_TESW = MO_TE | MO_SW, MO_TESL = MO_TE | MO_SL, + MO_TESQ = MO_TE | MO_SQ, #endif MO_SSIZE = MO_SIZE | MO_SIGN, -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Addition of div and rem on 128-bit integers, using the 128/64->128 divu and 64x64->128 mulu in host-utils. These operations will be used within div/rem helpers in the 128-bit riscv target. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-4-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- include/qemu/int128.h | 27 ++++++++ util/int128.c | 147 ++++++++++++++++++++++++++++++++++++++++++ util/meson.build | 1 + 3 files changed, 175 insertions(+) create mode 100644 util/int128.c diff --git a/include/qemu/int128.h b/include/qemu/int128.h index XXXXXXX..XXXXXXX 100644 --- a/include/qemu/int128.h +++ b/include/qemu/int128.h @@ -XXX,XX +XXX,XX @@ static inline Int128 bswap128(Int128 a) #endif } +static inline Int128 int128_divu(Int128 a, Int128 b) +{ + return (__uint128_t)a / (__uint128_t)b; +} + +static inline Int128 int128_remu(Int128 a, Int128 b) +{ + return (__uint128_t)a % (__uint128_t)b; +} + +static inline Int128 int128_divs(Int128 a, Int128 b) +{ + return a / b; +} + +static inline Int128 int128_rems(Int128 a, Int128 b) +{ + return a % b; +} + #else /* !CONFIG_INT128 */ typedef struct Int128 Int128; @@ -XXX,XX +XXX,XX @@ static inline Int128 bswap128(Int128 a) return int128_make128(bswap64(a.hi), bswap64(a.lo)); } +Int128 int128_divu(Int128, Int128); +Int128 int128_remu(Int128, Int128); +Int128 int128_divs(Int128, Int128); +Int128 int128_rems(Int128, Int128); + #endif /* CONFIG_INT128 */ static inline void bswap128s(Int128 *s) @@ -XXX,XX +XXX,XX @@ static inline void bswap128s(Int128 *s) *s = bswap128(*s); } +#define UINT128_MAX int128_make128(~0LL, ~0LL) + #endif /* INT128_H */ diff --git a/util/int128.c b/util/int128.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/util/int128.c @@ -XXX,XX +XXX,XX @@ +/* + * 128-bit division and remainder for compilers not supporting __int128 + * + * Copyright (c) 2021 Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ + +#include "qemu/osdep.h" +#include "qemu/host-utils.h" +#include "qemu/int128.h" + +#ifndef CONFIG_INT128 + +/* + * Division and remainder algorithms for 128-bit due to Stefan Kanthak, + * https://skanthak.homepage.t-online.de/integer.html#udivmodti4 + * Preconditions: + * - function should never be called with v equals to 0, it has to + * be dealt with beforehand + * - quotien pointer must be valid + */ +static Int128 divrem128(Int128 u, Int128 v, Int128 *q) +{ + Int128 qq; + uint64_t hi, lo, tmp; + int s = clz64(v.hi); + + if (s == 64) { + /* we have uu÷0v => let's use divu128 */ + hi = u.hi; + lo = u.lo; + tmp = divu128(&lo, &hi, v.lo); + *q = int128_make128(lo, hi); + return int128_make128(tmp, 0); + } else { + hi = int128_gethi(int128_lshift(v, s)); + + if (hi > u.hi) { + lo = u.lo; + tmp = u.hi; + divu128(&lo, &tmp, hi); + lo = int128_gethi(int128_lshift(int128_make128(lo, 0), s)); + } else { /* prevent overflow */ + lo = u.lo; + tmp = u.hi - hi; + divu128(&lo, &tmp, hi); + lo = int128_gethi(int128_lshift(int128_make128(lo, 1), s)); + } + + qq = int128_make64(lo); + + tmp = lo * v.hi; + mulu64(&lo, &hi, lo, v.lo); + hi += tmp; + + if (hi < tmp /* quotient * divisor >= 2**128 > dividend */ + || hi > u.hi /* quotient * divisor > dividend */ + || (hi == u.hi && lo > u.lo)) { + qq.lo -= 1; + mulu64(&lo, &hi, qq.lo, v.lo); + hi += qq.lo * v.hi; + } + + *q = qq; + u.hi -= hi + (u.lo < lo); + u.lo -= lo; + return u; + } +} + +Int128 int128_divu(Int128 a, Int128 b) +{ + Int128 q; + divrem128(a, b, &q); + return q; +} + +Int128 int128_remu(Int128 a, Int128 b) +{ + Int128 q; + return divrem128(a, b, &q); +} + +Int128 int128_divs(Int128 a, Int128 b) +{ + Int128 q; + bool sgna = !int128_nonneg(a); + bool sgnb = !int128_nonneg(b); + + if (sgna) { + a = int128_neg(a); + } + + if (sgnb) { + b = int128_neg(b); + } + + divrem128(a, b, &q); + + if (sgna != sgnb) { + q = int128_neg(q); + } + + return q; +} + +Int128 int128_rems(Int128 a, Int128 b) +{ + Int128 q, r; + bool sgna = !int128_nonneg(a); + bool sgnb = !int128_nonneg(b); + + if (sgna) { + a = int128_neg(a); + } + + if (sgnb) { + b = int128_neg(b); + } + + r = divrem128(a, b, &q); + + if (sgna) { + r = int128_neg(r); + } + + return r; +} + +#endif diff --git a/util/meson.build b/util/meson.build index XXXXXXX..XXXXXXX 100644 --- a/util/meson.build +++ b/util/meson.build @@ -XXX,XX +XXX,XX @@ util_ss.add(files('transactions.c')) util_ss.add(when: 'CONFIG_POSIX', if_true: files('drm.c')) util_ss.add(files('guest-random.c')) util_ss.add(files('yank.c')) +util_ss.add(files('int128.c')) if have_user util_ss.add(files('selfmap.c')) -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Given that the 128-bit version of the riscv spec adds new instructions, and that some instructions that were previously only available in 64-bit mode are now available for both 64-bit and 128-bit, we added new macros to check for the processor mode during translation. Although RV128 is a superset of RV64, we keep for now the RV64 only tests for extensions other than RVI and RVM. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-5-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/translate.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 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 @@ EX_SH(12) } \ } while (0) -#define REQUIRE_64BIT(ctx) do { \ - if (get_xl(ctx) < MXL_RV64) { \ - return false; \ - } \ +#define REQUIRE_64BIT(ctx) do { \ + if (get_xl(ctx) != MXL_RV64) { \ + return false; \ + } \ +} while (0) + +#define REQUIRE_128BIT(ctx) do { \ + if (get_xl(ctx) != MXL_RV128) { \ + return false; \ + } \ +} while (0) + +#define REQUIRE_64_OR_128BIT(ctx) do { \ + if (get_xl(ctx) == MXL_RV32) { \ + return false; \ + } \ } while (0) static int ex_rvc_register(DisasContext *ctx, int reg) -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Introduction of a gen_logic function for bitwise logic to implement instructions in which no propagation of information occurs between bits and use of this function on the bitwise instructions. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-6-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/translate.c | 27 +++++++++++++++++++++++++ target/riscv/insn_trans/trans_rvb.c.inc | 6 +++--- target/riscv/insn_trans/trans_rvi.c.inc | 12 +++++------ 3 files changed, 36 insertions(+), 9 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 @@ static int ex_rvc_shifti(DisasContext *ctx, int imm) /* Include the auto-generated decoder for 32 bit insn */ #include "decode-insn32.c.inc" +static bool gen_logic_imm_fn(DisasContext *ctx, arg_i *a, + void (*func)(TCGv, TCGv, target_long)) +{ + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, EXT_NONE); + + func(dest, src1, a->imm); + + gen_set_gpr(ctx, a->rd, dest); + + return true; +} + +static bool gen_logic(DisasContext *ctx, arg_r *a, + void (*func)(TCGv, TCGv, TCGv)) +{ + 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); + + gen_set_gpr(ctx, a->rd, dest); + + return true; +} + static bool gen_arith_imm_fn(DisasContext *ctx, arg_i *a, DisasExtend ext, void (*func)(TCGv, TCGv, target_long)) { 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_cpop(DisasContext *ctx, arg_cpop *a) static bool trans_andn(DisasContext *ctx, arg_andn *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_andc_tl); + return gen_logic(ctx, a, tcg_gen_andc_tl); } static bool trans_orn(DisasContext *ctx, arg_orn *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_orc_tl); + return gen_logic(ctx, a, tcg_gen_orc_tl); } static bool trans_xnor(DisasContext *ctx, arg_xnor *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_eqv_tl); + return gen_logic(ctx, a, tcg_gen_eqv_tl); } static bool trans_min(DisasContext *ctx, arg_min *a) diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_sltiu(DisasContext *ctx, arg_sltiu *a) static bool trans_xori(DisasContext *ctx, arg_xori *a) { - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_xori_tl); + return gen_logic_imm_fn(ctx, a, tcg_gen_xori_tl); } static bool trans_ori(DisasContext *ctx, arg_ori *a) { - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_ori_tl); + return gen_logic_imm_fn(ctx, a, tcg_gen_ori_tl); } static bool trans_andi(DisasContext *ctx, arg_andi *a) { - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_andi_tl); + return gen_logic_imm_fn(ctx, a, tcg_gen_andi_tl); } static bool trans_slli(DisasContext *ctx, arg_slli *a) @@ -XXX,XX +XXX,XX @@ static bool trans_sltu(DisasContext *ctx, arg_sltu *a) static bool trans_xor(DisasContext *ctx, arg_xor *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_xor_tl); + return gen_logic(ctx, a, tcg_gen_xor_tl); } static bool trans_srl(DisasContext *ctx, arg_srl *a) @@ -XXX,XX +XXX,XX @@ static bool trans_sra(DisasContext *ctx, arg_sra *a) static bool trans_or(DisasContext *ctx, arg_or *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_or_tl); + return gen_logic(ctx, a, tcg_gen_or_tl); } static bool trans_and(DisasContext *ctx, arg_and *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_and_tl); + return gen_logic(ctx, a, tcg_gen_and_tl); } static bool trans_addiw(DisasContext *ctx, arg_addiw *a) -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> The upper 64-bit of the 128-bit registers have now a place inside the cpu state structure, and are created as globals for future use. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-7-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 2 ++ target/riscv/cpu.c | 9 +++++++++ target/riscv/machine.c | 20 ++++++++++++++++++++ target/riscv/translate.c | 5 ++++- 4 files changed, 35 insertions(+), 1 deletion(-) 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 @@ FIELD(VTYPE, VILL, sizeof(target_ulong) * 8 - 1, 1) struct CPURISCVState { target_ulong gpr[32]; + target_ulong gprh[32]; /* 64 top bits of the 128-bit registers */ uint64_t fpr[32]; /* assume both F and D extensions */ /* vector coprocessor state. */ @@ -XXX,XX +XXX,XX @@ static inline bool riscv_feature(CPURISCVState *env, int feature) #include "cpu_user.h" extern const char * const riscv_int_regnames[]; +extern const char * const riscv_int_regnamesh[]; extern const char * const riscv_fpr_regnames[]; const char *riscv_cpu_get_trap_name(target_ulong cause, bool async); 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 char * const riscv_int_regnames[] = { "x28/t3", "x29/t4", "x30/t5", "x31/t6" }; +const char * const riscv_int_regnamesh[] = { + "x0h/zeroh", "x1h/rah", "x2h/sph", "x3h/gph", "x4h/tph", "x5h/t0h", + "x6h/t1h", "x7h/t2h", "x8h/s0h", "x9h/s1h", "x10h/a0h", "x11h/a1h", + "x12h/a2h", "x13h/a3h", "x14h/a4h", "x15h/a5h", "x16h/a6h", "x17h/a7h", + "x18h/s2h", "x19h/s3h", "x20h/s4h", "x21h/s5h", "x22h/s6h", "x23h/s7h", + "x24h/s8h", "x25h/s9h", "x26h/s10h", "x27h/s11h", "x28h/t3h", "x29h/t4h", + "x30h/t5h", "x31h/t6h" +}; + const char * const riscv_fpr_regnames[] = { "f0/ft0", "f1/ft1", "f2/ft2", "f3/ft3", "f4/ft4", "f5/ft5", "f6/ft6", "f7/ft7", "f8/fs0", "f9/fs1", "f10/fa0", "f11/fa1", 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 const VMStateDescription vmstate_pointermasking = { } }; +static bool rv128_needed(void *opaque) +{ + RISCVCPU *cpu = opaque; + CPURISCVState *env = &cpu->env; + + return env->misa_mxl_max == MXL_RV128; +} + +static const VMStateDescription vmstate_rv128 = { + .name = "cpu/rv128", + .version_id = 1, + .minimum_version_id = 1, + .needed = rv128_needed, + .fields = (VMStateField[]) { + VMSTATE_UINTTL_ARRAY(env.gprh, RISCVCPU, 32), + VMSTATE_END_OF_LIST() + } +}; + const VMStateDescription vmstate_riscv_cpu = { .name = "cpu", .version_id = 3, @@ -XXX,XX +XXX,XX @@ const VMStateDescription vmstate_riscv_cpu = { &vmstate_hyper, &vmstate_vector, &vmstate_pointermasking, + &vmstate_rv128, NULL } }; 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 @@ #include "internals.h" /* global register indices */ -static TCGv cpu_gpr[32], cpu_pc, cpu_vl, cpu_vstart; +static TCGv cpu_gpr[32], cpu_gprh[32], cpu_pc, cpu_vl, cpu_vstart; static TCGv_i64 cpu_fpr[32]; /* assume F and D extensions */ static TCGv load_res; static TCGv load_val; @@ -XXX,XX +XXX,XX @@ void riscv_translate_init(void) * unless you specifically block reads/writes to reg 0. */ cpu_gpr[0] = NULL; + cpu_gprh[0] = NULL; for (i = 1; i < 32; i++) { cpu_gpr[i] = tcg_global_mem_new(cpu_env, offsetof(CPURISCVState, gpr[i]), riscv_int_regnames[i]); + cpu_gprh[i] = tcg_global_mem_new(cpu_env, + offsetof(CPURISCVState, gprh[i]), riscv_int_regnamesh[i]); } for (i = 0; i < 32; i++) { -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> This patch adds the support of the '-cpu rv128' option to qemu-system-riscv64 so that we can indicate that we want to run rv128 executables. Still, there is no support for 128-bit insns at that stage so qemu fails miserably (as expected) if launched with this option. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-8-frederic.petrot@univ-grenoble-alpes.fr [ Changed by AF - Rename CPU to "x-rv128" ] Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- include/disas/dis-asm.h | 1 + target/riscv/cpu.h | 1 + disas/riscv.c | 5 +++++ target/riscv/cpu.c | 20 ++++++++++++++++++++ target/riscv/gdbstub.c | 5 +++++ 5 files changed, 32 insertions(+) diff --git a/include/disas/dis-asm.h b/include/disas/dis-asm.h index XXXXXXX..XXXXXXX 100644 --- a/include/disas/dis-asm.h +++ b/include/disas/dis-asm.h @@ -XXX,XX +XXX,XX @@ int print_insn_nios2(bfd_vma, disassemble_info*); int print_insn_xtensa (bfd_vma, disassemble_info*); int print_insn_riscv32 (bfd_vma, disassemble_info*); int print_insn_riscv64 (bfd_vma, disassemble_info*); +int print_insn_riscv128 (bfd_vma, disassemble_info*); int print_insn_rx(bfd_vma, disassemble_info *); int print_insn_hexagon(bfd_vma, disassemble_info *); 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 @@ #define TYPE_RISCV_CPU_ANY RISCV_CPU_TYPE_NAME("any") #define TYPE_RISCV_CPU_BASE32 RISCV_CPU_TYPE_NAME("rv32") #define TYPE_RISCV_CPU_BASE64 RISCV_CPU_TYPE_NAME("rv64") +#define TYPE_RISCV_CPU_BASE128 RISCV_CPU_TYPE_NAME("x-rv128") #define TYPE_RISCV_CPU_IBEX RISCV_CPU_TYPE_NAME("lowrisc-ibex") #define TYPE_RISCV_CPU_SHAKTI_C RISCV_CPU_TYPE_NAME("shakti-c") #define TYPE_RISCV_CPU_SIFIVE_E31 RISCV_CPU_TYPE_NAME("sifive-e31") 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 @@ int print_insn_riscv64(bfd_vma memaddr, struct disassemble_info *info) { return print_insn_riscv(memaddr, info, rv64); } + +int print_insn_riscv128(bfd_vma memaddr, struct disassemble_info *info) +{ + return print_insn_riscv(memaddr, info, rv128); +} 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 rv64_sifive_e_cpu_init(Object *obj) set_priv_version(env, PRIV_VERSION_1_10_0); qdev_prop_set_bit(DEVICE(obj), "mmu", false); } + +static void rv128_base_cpu_init(Object *obj) +{ + if (qemu_tcg_mttcg_enabled()) { + /* Missing 128-bit aligned atomics */ + error_report("128-bit RISC-V currently does not work with Multi " + "Threaded TCG. Please use: -accel tcg,thread=single"); + exit(EXIT_FAILURE); + } + CPURISCVState *env = &RISCV_CPU(obj)->env; + /* We set this in the realise function */ + set_misa(env, MXL_RV128, 0); +} #else static void rv32_base_cpu_init(Object *obj) { @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_disas_set_info(CPUState *s, disassemble_info *info) case MXL_RV64: info->print_insn = print_insn_riscv64; break; + case MXL_RV128: + info->print_insn = print_insn_riscv128; + break; default: g_assert_not_reached(); } @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_realize(DeviceState *dev, Error **errp) #ifdef TARGET_RISCV64 case MXL_RV64: break; + case MXL_RV128: + break; #endif case MXL_RV32: break; @@ -XXX,XX +XXX,XX @@ static gchar *riscv_gdb_arch_name(CPUState *cs) case MXL_RV32: return g_strdup("riscv:rv32"); case MXL_RV64: + case MXL_RV128: return g_strdup("riscv:rv64"); default: g_assert_not_reached(); @@ -XXX,XX +XXX,XX @@ static const TypeInfo riscv_cpu_type_infos[] = { DEFINE_CPU(TYPE_RISCV_CPU_SIFIVE_E51, rv64_sifive_e_cpu_init), DEFINE_CPU(TYPE_RISCV_CPU_SIFIVE_U54, rv64_sifive_u_cpu_init), DEFINE_CPU(TYPE_RISCV_CPU_SHAKTI_C, rv64_sifive_u_cpu_init), + DEFINE_CPU(TYPE_RISCV_CPU_BASE128, rv128_base_cpu_init), #endif }; diff --git a/target/riscv/gdbstub.c b/target/riscv/gdbstub.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/gdbstub.c +++ b/target/riscv/gdbstub.c @@ -XXX,XX +XXX,XX @@ static int riscv_gen_dynamic_csr_xml(CPUState *cs, int base_reg) int bitsize = 16 << env->misa_mxl_max; int i; + /* Until gdb knows about 128-bit registers */ + if (bitsize > 64) { + bitsize = 64; + } + g_string_printf(s, "<?xml version=\"1.0\"?>"); g_string_append_printf(s, "<!DOCTYPE feature SYSTEM \"gdb-target.dtd\">"); g_string_append_printf(s, "<feature name=\"org.gnu.gdb.riscv.csr\">"); -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> lwu and ld are functionally close to the other loads, but were after the stores in the source file. Similarly, xor was away from or and and by two arithmetic functions, while the immediate versions were nicely put together. This patch moves the aforementioned loads after lhu, and xor above or, where they more logically belong. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-9-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn_trans/trans_rvi.c.inc | 34 ++++++++++++------------- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_lhu(DisasContext *ctx, arg_lhu *a) return gen_load(ctx, a, MO_TEUW); } +static bool trans_lwu(DisasContext *ctx, arg_lwu *a) +{ + REQUIRE_64BIT(ctx); + return gen_load(ctx, a, MO_TEUL); +} + +static bool trans_ld(DisasContext *ctx, arg_ld *a) +{ + REQUIRE_64BIT(ctx); + return gen_load(ctx, a, MO_TEUQ); +} + static bool gen_store(DisasContext *ctx, arg_sb *a, MemOp memop) { TCGv addr = get_gpr(ctx, a->rs1, EXT_NONE); @@ -XXX,XX +XXX,XX @@ static bool trans_sw(DisasContext *ctx, arg_sw *a) return gen_store(ctx, a, MO_TESL); } -static bool trans_lwu(DisasContext *ctx, arg_lwu *a) -{ - REQUIRE_64BIT(ctx); - return gen_load(ctx, a, MO_TEUL); -} - -static bool trans_ld(DisasContext *ctx, arg_ld *a) -{ - REQUIRE_64BIT(ctx); - return gen_load(ctx, a, MO_TEUQ); -} - static bool trans_sd(DisasContext *ctx, arg_sd *a) { REQUIRE_64BIT(ctx); @@ -XXX,XX +XXX,XX @@ static bool trans_sltu(DisasContext *ctx, arg_sltu *a) return gen_arith(ctx, a, EXT_SIGN, gen_sltu); } -static bool trans_xor(DisasContext *ctx, arg_xor *a) -{ - return gen_logic(ctx, a, tcg_gen_xor_tl); -} - static bool trans_srl(DisasContext *ctx, arg_srl *a) { return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl); @@ -XXX,XX +XXX,XX @@ static bool trans_sra(DisasContext *ctx, arg_sra *a) return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl); } +static bool trans_xor(DisasContext *ctx, arg_xor *a) +{ + return gen_logic(ctx, a, tcg_gen_xor_tl); +} + static bool trans_or(DisasContext *ctx, arg_or *a) { return gen_logic(ctx, a, tcg_gen_or_tl); -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Get function to retrieve the 64 top bits of a register, stored in the gprh field of the cpu state. Set function that writes the 128-bit value at once. The access to the gprh field can not be protected at compile time to make sure it is accessed only in the 128-bit version of the processor because we have no way to indicate that the misa_mxl_max field is const. The 128-bit ISA adds ldu, lq and sq. We provide support for these instructions. Note that (a) we compute only 64-bit addresses to actually access memory, cowardly utilizing the existing address translation mechanism of QEMU, and (b) we assume for now little-endian memory accesses. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-10-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn16.decode | 27 ++++++- target/riscv/insn32.decode | 5 ++ target/riscv/translate.c | 41 ++++++++++ target/riscv/insn_trans/trans_rvi.c.inc | 100 ++++++++++++++++++++++-- 4 files changed, 163 insertions(+), 10 deletions(-) diff --git a/target/riscv/insn16.decode b/target/riscv/insn16.decode index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn16.decode +++ b/target/riscv/insn16.decode @@ -XXX,XX +XXX,XX @@ # Immediates: %imm_ci 12:s1 2:5 %nzuimm_ciw 7:4 11:2 5:1 6:1 !function=ex_shift_2 +%uimm_cl_q 10:1 5:2 11:2 !function=ex_shift_4 %uimm_cl_d 5:2 10:3 !function=ex_shift_3 %uimm_cl_w 5:1 10:3 6:1 !function=ex_shift_2 %imm_cb 12:s1 5:2 2:1 10:2 3:2 !function=ex_shift_1 %imm_cj 12:s1 8:1 9:2 6:1 7:1 2:1 11:1 3:3 !function=ex_shift_1 %shimm_6bit 12:1 2:5 !function=ex_rvc_shifti +%uimm_6bit_lq 2:4 12:1 6:1 !function=ex_shift_4 %uimm_6bit_ld 2:3 12:1 5:2 !function=ex_shift_3 %uimm_6bit_lw 2:2 12:1 4:3 !function=ex_shift_2 +%uimm_6bit_sq 7:4 11:2 !function=ex_shift_4 %uimm_6bit_sd 7:3 10:3 !function=ex_shift_3 %uimm_6bit_sw 7:2 9:4 !function=ex_shift_2 @@ -XXX,XX +XXX,XX @@ # Formats 16: @cr .... ..... ..... .. &r rs2=%rs2_5 rs1=%rd %rd @ci ... . ..... ..... .. &i imm=%imm_ci rs1=%rd %rd +@cl_q ... . ..... ..... .. &i imm=%uimm_cl_q rs1=%rs1_3 rd=%rs2_3 @cl_d ... ... ... .. ... .. &i imm=%uimm_cl_d rs1=%rs1_3 rd=%rs2_3 @cl_w ... ... ... .. ... .. &i imm=%uimm_cl_w rs1=%rs1_3 rd=%rs2_3 @cs_2 ... ... ... .. ... .. &r rs2=%rs2_3 rs1=%rs1_3 rd=%rs1_3 +@cs_q ... ... ... .. ... .. &s imm=%uimm_cl_q rs1=%rs1_3 rs2=%rs2_3 @cs_d ... ... ... .. ... .. &s imm=%uimm_cl_d rs1=%rs1_3 rs2=%rs2_3 @cs_w ... ... ... .. ... .. &s imm=%uimm_cl_w rs1=%rs1_3 rs2=%rs2_3 @cj ... ........... .. &j imm=%imm_cj @cb_z ... ... ... .. ... .. &b imm=%imm_cb rs1=%rs1_3 rs2=0 +@c_lqsp ... . ..... ..... .. &i imm=%uimm_6bit_lq rs1=2 %rd @c_ldsp ... . ..... ..... .. &i imm=%uimm_6bit_ld rs1=2 %rd @c_lwsp ... . ..... ..... .. &i imm=%uimm_6bit_lw rs1=2 %rd +@c_sqsp ... . ..... ..... .. &s imm=%uimm_6bit_sq rs1=2 rs2=%rs2_5 @c_sdsp ... . ..... ..... .. &s imm=%uimm_6bit_sd rs1=2 rs2=%rs2_5 @c_swsp ... . ..... ..... .. &s imm=%uimm_6bit_sw rs1=2 rs2=%rs2_5 @c_li ... . ..... ..... .. &i imm=%imm_ci rs1=0 %rd @@ -XXX,XX +XXX,XX @@ illegal 000 000 000 00 --- 00 addi 000 ... ... .. ... 00 @c_addi4spn } -fld 001 ... ... .. ... 00 @cl_d +{ + lq 001 ... ... .. ... 00 @cl_q + fld 001 ... ... .. ... 00 @cl_d +} lw 010 ... ... .. ... 00 @cl_w -fsd 101 ... ... .. ... 00 @cs_d +{ + sq 101 ... ... .. ... 00 @cs_q + fsd 101 ... ... .. ... 00 @cs_d +} sw 110 ... ... .. ... 00 @cs_w # *** RV32C and RV64C specific Standard Extension (Quadrant 0) *** @@ -XXX,XX +XXX,XX @@ addw 100 1 11 ... 01 ... 01 @cs_2 # *** RV32/64C Standard Extension (Quadrant 2) *** slli 000 . ..... ..... 10 @c_shift2 -fld 001 . ..... ..... 10 @c_ldsp +{ + lq 001 ... ... .. ... 10 @c_lqsp + fld 001 . ..... ..... 10 @c_ldsp +} { illegal 010 - 00000 ----- 10 # c.lwsp, RES rd=0 lw 010 . ..... ..... 10 @c_lwsp @@ -XXX,XX +XXX,XX @@ fld 001 . ..... ..... 10 @c_ldsp jalr 100 1 ..... 00000 10 @c_jalr rd=1 # C.JALR add 100 1 ..... ..... 10 @cr } -fsd 101 ...... ..... 10 @c_sdsp +{ + sq 101 ... ... .. ... 10 @c_sqsp + fsd 101 ...... ..... 10 @c_sdsp +} sw 110 . ..... ..... 10 @c_swsp # *** RV32C and RV64C specific Standard Extension (Quadrant 2) *** 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 @@ sllw 0000000 ..... ..... 001 ..... 0111011 @r srlw 0000000 ..... ..... 101 ..... 0111011 @r sraw 0100000 ..... ..... 101 ..... 0111011 @r +# *** RV128I Base Instruction Set (in addition to RV64I) *** +ldu ............ ..... 111 ..... 0000011 @i +lq ............ ..... 010 ..... 0001111 @i +sq ............ ..... 100 ..... 0100011 @s + # *** RV32M Standard Extension *** mul 0000001 ..... ..... 000 ..... 0110011 @r mulh 0000001 ..... ..... 001 ..... 0110011 @r 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 @@ typedef struct DisasContext { /* pc_succ_insn points to the instruction following base.pc_next */ target_ulong pc_succ_insn; target_ulong priv_ver; + RISCVMXL misa_mxl_max; RISCVMXL xl; uint32_t misa_ext; uint32_t opcode; @@ -XXX,XX +XXX,XX @@ static inline int get_olen(DisasContext *ctx) return 16 << get_ol(ctx); } +/* The maximum register length */ +#ifdef TARGET_RISCV32 +#define get_xl_max(ctx) MXL_RV32 +#else +#define get_xl_max(ctx) ((ctx)->misa_mxl_max) +#endif + /* * RISC-V requires NaN-boxing of narrower width floating point values. * This applies when a 32-bit value is assigned to a 64-bit FP register. @@ -XXX,XX +XXX,XX @@ static TCGv get_gpr(DisasContext *ctx, int reg_num, DisasExtend ext) } break; case MXL_RV64: + case MXL_RV128: break; default: g_assert_not_reached(); @@ -XXX,XX +XXX,XX @@ static TCGv get_gpr(DisasContext *ctx, int reg_num, DisasExtend ext) return cpu_gpr[reg_num]; } +static TCGv get_gprh(DisasContext *ctx, int reg_num) +{ + assert(get_xl(ctx) == MXL_RV128); + if (reg_num == 0) { + return ctx->zero; + } + return cpu_gprh[reg_num]; +} + static TCGv dest_gpr(DisasContext *ctx, int reg_num) { if (reg_num == 0 || get_olen(ctx) < TARGET_LONG_BITS) { @@ -XXX,XX +XXX,XX @@ static TCGv dest_gpr(DisasContext *ctx, int reg_num) return cpu_gpr[reg_num]; } +static TCGv dest_gprh(DisasContext *ctx, int reg_num) +{ + if (reg_num == 0) { + return temp_new(ctx); + } + return cpu_gprh[reg_num]; +} + static void gen_set_gpr(DisasContext *ctx, int reg_num, TCGv t) { if (reg_num != 0) { @@ -XXX,XX +XXX,XX @@ static void gen_set_gpr(DisasContext *ctx, int reg_num, TCGv t) tcg_gen_ext32s_tl(cpu_gpr[reg_num], t); break; case MXL_RV64: + case MXL_RV128: tcg_gen_mov_tl(cpu_gpr[reg_num], t); break; default: g_assert_not_reached(); } + + if (get_xl_max(ctx) == MXL_RV128) { + tcg_gen_sari_tl(cpu_gprh[reg_num], cpu_gpr[reg_num], 63); + } + } +} + +static void gen_set_gpr128(DisasContext *ctx, int reg_num, TCGv rl, TCGv rh) +{ + assert(get_ol(ctx) == MXL_RV128); + if (reg_num != 0) { + tcg_gen_mov_tl(cpu_gpr[reg_num], rl); + tcg_gen_mov_tl(cpu_gprh[reg_num], rh); } } @@ -XXX,XX +XXX,XX @@ static void riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs) ctx->lmul = sextract32(FIELD_EX32(tb_flags, TB_FLAGS, LMUL), 0, 3); ctx->vstart = env->vstart; ctx->vl_eq_vlmax = FIELD_EX32(tb_flags, TB_FLAGS, VL_EQ_VLMAX); + ctx->misa_mxl_max = env->misa_mxl_max; ctx->xl = FIELD_EX32(tb_flags, TB_FLAGS, XL); ctx->cs = cs; ctx->ntemp = 0; diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_bgeu(DisasContext *ctx, arg_bgeu *a) return gen_branch(ctx, a, TCG_COND_GEU); } -static bool gen_load(DisasContext *ctx, arg_lb *a, MemOp memop) +static bool gen_load_tl(DisasContext *ctx, arg_lb *a, MemOp memop) { TCGv dest = dest_gpr(ctx, a->rd); TCGv addr = get_gpr(ctx, a->rs1, EXT_NONE); @@ -XXX,XX +XXX,XX @@ static bool gen_load(DisasContext *ctx, arg_lb *a, MemOp memop) return true; } +/* Compute only 64-bit addresses to use the address translation mechanism */ +static bool gen_load_i128(DisasContext *ctx, arg_lb *a, MemOp memop) +{ + TCGv src1l = get_gpr(ctx, a->rs1, EXT_NONE); + TCGv destl = dest_gpr(ctx, a->rd); + TCGv desth = dest_gprh(ctx, a->rd); + TCGv addrl = tcg_temp_new(); + + tcg_gen_addi_tl(addrl, src1l, a->imm); + + if ((memop & MO_SIZE) <= MO_64) { + tcg_gen_qemu_ld_tl(destl, addrl, ctx->mem_idx, memop); + if (memop & MO_SIGN) { + tcg_gen_sari_tl(desth, destl, 63); + } else { + tcg_gen_movi_tl(desth, 0); + } + } else { + /* assume little-endian memory access for now */ + tcg_gen_qemu_ld_tl(destl, addrl, ctx->mem_idx, MO_TEUQ); + tcg_gen_addi_tl(addrl, addrl, 8); + tcg_gen_qemu_ld_tl(desth, addrl, ctx->mem_idx, MO_TEUQ); + } + + gen_set_gpr128(ctx, a->rd, destl, desth); + + tcg_temp_free(addrl); + return true; +} + +static bool gen_load(DisasContext *ctx, arg_lb *a, MemOp memop) +{ + if (get_xl(ctx) == MXL_RV128) { + return gen_load_i128(ctx, a, memop); + } else { + return gen_load_tl(ctx, a, memop); + } +} + static bool trans_lb(DisasContext *ctx, arg_lb *a) { return gen_load(ctx, a, MO_SB); @@ -XXX,XX +XXX,XX @@ static bool trans_lw(DisasContext *ctx, arg_lw *a) return gen_load(ctx, a, MO_TESL); } +static bool trans_ld(DisasContext *ctx, arg_ld *a) +{ + REQUIRE_64_OR_128BIT(ctx); + return gen_load(ctx, a, MO_TESQ); +} + +static bool trans_lq(DisasContext *ctx, arg_lq *a) +{ + REQUIRE_128BIT(ctx); + return gen_load(ctx, a, MO_TEUO); +} + static bool trans_lbu(DisasContext *ctx, arg_lbu *a) { return gen_load(ctx, a, MO_UB); @@ -XXX,XX +XXX,XX @@ static bool trans_lhu(DisasContext *ctx, arg_lhu *a) static bool trans_lwu(DisasContext *ctx, arg_lwu *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); return gen_load(ctx, a, MO_TEUL); } -static bool trans_ld(DisasContext *ctx, arg_ld *a) +static bool trans_ldu(DisasContext *ctx, arg_ldu *a) { - REQUIRE_64BIT(ctx); + REQUIRE_128BIT(ctx); return gen_load(ctx, a, MO_TEUQ); } -static bool gen_store(DisasContext *ctx, arg_sb *a, MemOp memop) +static bool gen_store_tl(DisasContext *ctx, arg_sb *a, MemOp memop) { TCGv addr = get_gpr(ctx, a->rs1, EXT_NONE); TCGv data = get_gpr(ctx, a->rs2, EXT_NONE); @@ -XXX,XX +XXX,XX @@ static bool gen_store(DisasContext *ctx, arg_sb *a, MemOp memop) return true; } +static bool gen_store_i128(DisasContext *ctx, arg_sb *a, MemOp memop) +{ + TCGv src1l = get_gpr(ctx, a->rs1, EXT_NONE); + TCGv src2l = get_gpr(ctx, a->rs2, EXT_NONE); + TCGv src2h = get_gprh(ctx, a->rs2); + TCGv addrl = tcg_temp_new(); + + tcg_gen_addi_tl(addrl, src1l, a->imm); + + if ((memop & MO_SIZE) <= MO_64) { + tcg_gen_qemu_st_tl(src2l, addrl, ctx->mem_idx, memop); + } else { + /* little-endian memory access assumed for now */ + tcg_gen_qemu_st_tl(src2l, addrl, ctx->mem_idx, MO_TEUQ); + tcg_gen_addi_tl(addrl, addrl, 8); + tcg_gen_qemu_st_tl(src2h, addrl, ctx->mem_idx, MO_TEUQ); + } + + tcg_temp_free(addrl); + return true; +} + +static bool gen_store(DisasContext *ctx, arg_sb *a, MemOp memop) +{ + if (get_xl(ctx) == MXL_RV128) { + return gen_store_i128(ctx, a, memop); + } else { + return gen_store_tl(ctx, a, memop); + } +} + static bool trans_sb(DisasContext *ctx, arg_sb *a) { return gen_store(ctx, a, MO_SB); @@ -XXX,XX +XXX,XX @@ static bool trans_sw(DisasContext *ctx, arg_sw *a) static bool trans_sd(DisasContext *ctx, arg_sd *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); return gen_store(ctx, a, MO_TEUQ); } +static bool trans_sq(DisasContext *ctx, arg_sq *a) +{ + REQUIRE_128BIT(ctx); + return gen_store(ctx, a, MO_TEUO); +} + static bool trans_addi(DisasContext *ctx, arg_addi *a) { return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl); -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> The 128-bit bitwise instructions do not need any function prototype change as the functions can be applied independently on the lower and upper part of the registers. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-11-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/translate.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 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 @@ static bool gen_logic_imm_fn(DisasContext *ctx, arg_i *a, func(dest, src1, a->imm); - gen_set_gpr(ctx, a->rd, dest); + if (get_xl(ctx) == MXL_RV128) { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv desth = dest_gprh(ctx, a->rd); + + func(desth, src1h, -(a->imm < 0)); + gen_set_gpr128(ctx, a->rd, dest, desth); + } else { + gen_set_gpr(ctx, a->rd, dest); + } return true; } @@ -XXX,XX +XXX,XX @@ static bool gen_logic(DisasContext *ctx, arg_r *a, func(dest, src1, src2); - gen_set_gpr(ctx, a->rd, dest); + if (get_xl(ctx) == MXL_RV128) { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv src2h = get_gprh(ctx, a->rs2); + TCGv desth = dest_gprh(ctx, a->rd); + + func(desth, src1h, src2h); + gen_set_gpr128(ctx, a->rd, dest, desth); + } else { + gen_set_gpr(ctx, a->rd, dest); + } return true; } -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Adding the 128-bit version of lui and auipc, and introducing to that end a "set register with immediat" function to handle extension on 128 bits. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-12-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/translate.c | 21 +++++++++++++++++++++ target/riscv/insn_trans/trans_rvi.c.inc | 8 ++++---- 2 files changed, 25 insertions(+), 4 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 @@ static void gen_set_gpr(DisasContext *ctx, int reg_num, TCGv t) } } +static void gen_set_gpri(DisasContext *ctx, int reg_num, target_long imm) +{ + if (reg_num != 0) { + switch (get_ol(ctx)) { + case MXL_RV32: + tcg_gen_movi_tl(cpu_gpr[reg_num], (int32_t)imm); + break; + case MXL_RV64: + case MXL_RV128: + tcg_gen_movi_tl(cpu_gpr[reg_num], imm); + break; + default: + g_assert_not_reached(); + } + + if (get_xl_max(ctx) == MXL_RV128) { + tcg_gen_movi_tl(cpu_gprh[reg_num], -(imm < 0)); + } + } +} + static void gen_set_gpr128(DisasContext *ctx, int reg_num, TCGv rl, TCGv rh) { assert(get_ol(ctx) == MXL_RV128); diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_illegal(DisasContext *ctx, arg_empty *a) static bool trans_c64_illegal(DisasContext *ctx, arg_empty *a) { - REQUIRE_64BIT(ctx); - return trans_illegal(ctx, a); + REQUIRE_64_OR_128BIT(ctx); + return trans_illegal(ctx, a); } static bool trans_lui(DisasContext *ctx, arg_lui *a) { if (a->rd != 0) { - tcg_gen_movi_tl(cpu_gpr[a->rd], a->imm); + gen_set_gpri(ctx, a->rd, a->imm); } return true; } @@ -XXX,XX +XXX,XX @@ static bool trans_lui(DisasContext *ctx, arg_lui *a) static bool trans_auipc(DisasContext *ctx, arg_auipc *a) { if (a->rd != 0) { - tcg_gen_movi_tl(cpu_gpr[a->rd], a->imm + ctx->base.pc_next); + gen_set_gpri(ctx, a->rd, a->imm + ctx->base.pc_next); } return true; } -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Handling shifts for 32, 64 and 128 operation length for RV128, following the general framework for handling various olens proposed by Richard. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-13-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn32.decode | 10 ++ target/riscv/translate.c | 58 ++++-- target/riscv/insn_trans/trans_rvb.c.inc | 22 +-- target/riscv/insn_trans/trans_rvi.c.inc | 224 ++++++++++++++++++++++-- 4 files changed, 270 insertions(+), 44 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 @@ %rs1 15:5 %rd 7:5 %sh5 20:5 +%sh6 20:6 %sh7 20:7 %csr 20:12 @@ -XXX,XX +XXX,XX @@ # Formats 64: @sh5 ....... ..... ..... ... ..... ....... &shift shamt=%sh5 %rs1 %rd +# Formats 128: +@sh6 ...... ...... ..... ... ..... ....... &shift shamt=%sh6 %rs1 %rd + # *** Privileged Instructions *** ecall 000000000000 00000 000 00000 1110011 ebreak 000000000001 00000 000 00000 1110011 @@ -XXX,XX +XXX,XX @@ sraw 0100000 ..... ..... 101 ..... 0111011 @r ldu ............ ..... 111 ..... 0000011 @i lq ............ ..... 010 ..... 0001111 @i sq ............ ..... 100 ..... 0100011 @s +sllid 000000 ...... ..... 001 ..... 1011011 @sh6 +srlid 000000 ...... ..... 101 ..... 1011011 @sh6 +sraid 010000 ...... ..... 101 ..... 1011011 @sh6 +slld 0000000 ..... ..... 001 ..... 1111011 @r +srld 0000000 ..... ..... 101 ..... 1111011 @r +srad 0100000 ..... ..... 101 ..... 1111011 @r # *** RV32M Standard Extension *** mul 0000001 ..... ..... 000 ..... 0110011 @r 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 bool gen_arith_per_ol(DisasContext *ctx, arg_r *a, DisasExtend ext, } static bool gen_shift_imm_fn(DisasContext *ctx, arg_shift *a, DisasExtend ext, - void (*func)(TCGv, TCGv, target_long)) + void (*func)(TCGv, TCGv, target_long), + void (*f128)(TCGv, TCGv, TCGv, TCGv, target_long)) { TCGv dest, src1; int max_len = get_olen(ctx); @@ -XXX,XX +XXX,XX @@ static bool gen_shift_imm_fn(DisasContext *ctx, arg_shift *a, DisasExtend ext, dest = dest_gpr(ctx, a->rd); src1 = get_gpr(ctx, a->rs1, ext); - func(dest, src1, a->shamt); + if (max_len < 128) { + func(dest, src1, a->shamt); + gen_set_gpr(ctx, a->rd, dest); + } else { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv desth = dest_gprh(ctx, a->rd); - gen_set_gpr(ctx, a->rd, dest); + if (f128 == NULL) { + return false; + } + f128(dest, desth, src1, src1h, a->shamt); + gen_set_gpr128(ctx, a->rd, dest, desth); + } return true; } static bool gen_shift_imm_fn_per_ol(DisasContext *ctx, arg_shift *a, DisasExtend ext, void (*f_tl)(TCGv, TCGv, target_long), - void (*f_32)(TCGv, TCGv, target_long)) + void (*f_32)(TCGv, TCGv, target_long), + void (*f_128)(TCGv, TCGv, TCGv, TCGv, + target_long)) { int olen = get_olen(ctx); if (olen != TARGET_LONG_BITS) { if (olen == 32) { f_tl = f_32; - } else { + } else if (olen != 128) { g_assert_not_reached(); } } - return gen_shift_imm_fn(ctx, a, ext, f_tl); + return gen_shift_imm_fn(ctx, a, ext, f_tl, f_128); } static bool gen_shift_imm_tl(DisasContext *ctx, arg_shift *a, DisasExtend ext, @@ -XXX,XX +XXX,XX @@ static bool gen_shift_imm_tl(DisasContext *ctx, arg_shift *a, DisasExtend ext, } static bool gen_shift(DisasContext *ctx, arg_r *a, DisasExtend ext, - void (*func)(TCGv, TCGv, TCGv)) + void (*func)(TCGv, TCGv, TCGv), + void (*f128)(TCGv, TCGv, TCGv, TCGv, TCGv)) { - TCGv dest = dest_gpr(ctx, a->rd); - TCGv src1 = get_gpr(ctx, a->rs1, ext); TCGv src2 = get_gpr(ctx, a->rs2, EXT_NONE); TCGv ext2 = tcg_temp_new(); + int max_len = get_olen(ctx); - tcg_gen_andi_tl(ext2, src2, get_olen(ctx) - 1); - func(dest, src1, ext2); + tcg_gen_andi_tl(ext2, src2, max_len - 1); - gen_set_gpr(ctx, a->rd, dest); + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, ext); + + if (max_len < 128) { + func(dest, src1, ext2); + gen_set_gpr(ctx, a->rd, dest); + } else { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv desth = dest_gprh(ctx, a->rd); + + if (f128 == NULL) { + return false; + } + f128(dest, desth, src1, src1h, ext2); + gen_set_gpr128(ctx, a->rd, dest, desth); + } tcg_temp_free(ext2); return true; } static bool gen_shift_per_ol(DisasContext *ctx, arg_r *a, DisasExtend ext, void (*f_tl)(TCGv, TCGv, TCGv), - void (*f_32)(TCGv, TCGv, TCGv)) + void (*f_32)(TCGv, TCGv, TCGv), + void (*f_128)(TCGv, TCGv, TCGv, TCGv, TCGv)) { int olen = get_olen(ctx); if (olen != TARGET_LONG_BITS) { if (olen == 32) { f_tl = f_32; - } else { + } else if (olen != 128) { g_assert_not_reached(); } } - return gen_shift(ctx, a, ext, f_tl); + return gen_shift(ctx, a, ext, f_tl, f_128); } static bool gen_unary(DisasContext *ctx, arg_r2 *a, DisasExtend ext, 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 void gen_bset(TCGv ret, TCGv arg1, TCGv shamt) static bool trans_bset(DisasContext *ctx, arg_bset *a) { REQUIRE_ZBS(ctx); - return gen_shift(ctx, a, EXT_NONE, gen_bset); + return gen_shift(ctx, a, EXT_NONE, gen_bset, NULL); } static bool trans_bseti(DisasContext *ctx, arg_bseti *a) @@ -XXX,XX +XXX,XX @@ static void gen_bclr(TCGv ret, TCGv arg1, TCGv shamt) static bool trans_bclr(DisasContext *ctx, arg_bclr *a) { REQUIRE_ZBS(ctx); - return gen_shift(ctx, a, EXT_NONE, gen_bclr); + return gen_shift(ctx, a, EXT_NONE, gen_bclr, NULL); } static bool trans_bclri(DisasContext *ctx, arg_bclri *a) @@ -XXX,XX +XXX,XX @@ static void gen_binv(TCGv ret, TCGv arg1, TCGv shamt) static bool trans_binv(DisasContext *ctx, arg_binv *a) { REQUIRE_ZBS(ctx); - return gen_shift(ctx, a, EXT_NONE, gen_binv); + return gen_shift(ctx, a, EXT_NONE, gen_binv, NULL); } static bool trans_binvi(DisasContext *ctx, arg_binvi *a) @@ -XXX,XX +XXX,XX @@ static void gen_bext(TCGv ret, TCGv arg1, TCGv shamt) static bool trans_bext(DisasContext *ctx, arg_bext *a) { REQUIRE_ZBS(ctx); - return gen_shift(ctx, a, EXT_NONE, gen_bext); + return gen_shift(ctx, a, EXT_NONE, gen_bext, NULL); } static bool trans_bexti(DisasContext *ctx, arg_bexti *a) @@ -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); - return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotr_tl, gen_rorw); + return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotr_tl, gen_rorw, NULL); } static void gen_roriw(TCGv ret, TCGv arg1, target_long shamt) @@ -XXX,XX +XXX,XX @@ static bool trans_rori(DisasContext *ctx, arg_rori *a) { REQUIRE_ZBB(ctx); return gen_shift_imm_fn_per_ol(ctx, a, EXT_NONE, - tcg_gen_rotri_tl, gen_roriw); + tcg_gen_rotri_tl, gen_roriw, NULL); } static void gen_rolw(TCGv ret, TCGv arg1, TCGv arg2) @@ -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); - return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotl_tl, gen_rolw); + return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotl_tl, gen_rolw, NULL); } static void gen_rev8_32(TCGv ret, TCGv src1) @@ -XXX,XX +XXX,XX @@ static bool trans_rorw(DisasContext *ctx, arg_rorw *a) REQUIRE_64BIT(ctx); REQUIRE_ZBB(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_NONE, gen_rorw); + return gen_shift(ctx, a, EXT_NONE, gen_rorw, NULL); } static bool trans_roriw(DisasContext *ctx, arg_roriw *a) @@ -XXX,XX +XXX,XX @@ static bool trans_roriw(DisasContext *ctx, arg_roriw *a) REQUIRE_64BIT(ctx); REQUIRE_ZBB(ctx); ctx->ol = MXL_RV32; - return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_roriw); + return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_roriw, NULL); } static bool trans_rolw(DisasContext *ctx, arg_rolw *a) @@ -XXX,XX +XXX,XX @@ static bool trans_rolw(DisasContext *ctx, arg_rolw *a) REQUIRE_64BIT(ctx); REQUIRE_ZBB(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_NONE, gen_rolw); + return gen_shift(ctx, a, EXT_NONE, gen_rolw, NULL); } #define GEN_SHADD_UW(SHAMT) \ @@ -XXX,XX +XXX,XX @@ static bool trans_slli_uw(DisasContext *ctx, arg_slli_uw *a) { REQUIRE_64BIT(ctx); REQUIRE_ZBA(ctx); - return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_slli_uw); + return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_slli_uw, NULL); } static bool trans_clmul(DisasContext *ctx, arg_clmul *a) diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_andi(DisasContext *ctx, arg_andi *a) return gen_logic_imm_fn(ctx, a, tcg_gen_andi_tl); } +static void gen_slli_i128(TCGv retl, TCGv reth, + TCGv src1l, TCGv src1h, + target_long shamt) +{ + if (shamt >= 64) { + tcg_gen_shli_tl(reth, src1l, shamt - 64); + tcg_gen_movi_tl(retl, 0); + } else { + tcg_gen_extract2_tl(reth, src1l, src1h, 64 - shamt); + tcg_gen_shli_tl(retl, src1l, shamt); + } +} + static bool trans_slli(DisasContext *ctx, arg_slli *a) { - return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl); + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl, gen_slli_i128); } static void gen_srliw(TCGv dst, TCGv src, target_long shamt) @@ -XXX,XX +XXX,XX @@ static void gen_srliw(TCGv dst, TCGv src, target_long shamt) tcg_gen_extract_tl(dst, src, shamt, 32 - shamt); } +static void gen_srli_i128(TCGv retl, TCGv reth, + TCGv src1l, TCGv src1h, + target_long shamt) +{ + if (shamt >= 64) { + tcg_gen_shri_tl(retl, src1h, shamt - 64); + tcg_gen_movi_tl(reth, 0); + } else { + tcg_gen_extract2_tl(retl, src1l, src1h, shamt); + tcg_gen_shri_tl(reth, src1h, shamt); + } +} + static bool trans_srli(DisasContext *ctx, arg_srli *a) { return gen_shift_imm_fn_per_ol(ctx, a, EXT_NONE, - tcg_gen_shri_tl, gen_srliw); + tcg_gen_shri_tl, gen_srliw, gen_srli_i128); } static void gen_sraiw(TCGv dst, TCGv src, target_long shamt) @@ -XXX,XX +XXX,XX @@ static void gen_sraiw(TCGv dst, TCGv src, target_long shamt) tcg_gen_sextract_tl(dst, src, shamt, 32 - shamt); } +static void gen_srai_i128(TCGv retl, TCGv reth, + TCGv src1l, TCGv src1h, + target_long shamt) +{ + if (shamt >= 64) { + tcg_gen_sari_tl(retl, src1h, shamt - 64); + tcg_gen_sari_tl(reth, src1h, 63); + } else { + tcg_gen_extract2_tl(retl, src1l, src1h, shamt); + tcg_gen_sari_tl(reth, src1h, shamt); + } +} + static bool trans_srai(DisasContext *ctx, arg_srai *a) { return gen_shift_imm_fn_per_ol(ctx, a, EXT_NONE, - tcg_gen_sari_tl, gen_sraiw); + tcg_gen_sari_tl, gen_sraiw, gen_srai_i128); } static bool trans_add(DisasContext *ctx, arg_add *a) @@ -XXX,XX +XXX,XX @@ static bool trans_sub(DisasContext *ctx, arg_sub *a) return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl); } +static void gen_sll_i128(TCGv destl, TCGv desth, + TCGv src1l, TCGv src1h, TCGv shamt) +{ + TCGv ls = tcg_temp_new(); + TCGv rs = tcg_temp_new(); + TCGv hs = tcg_temp_new(); + TCGv ll = tcg_temp_new(); + TCGv lr = tcg_temp_new(); + TCGv h0 = tcg_temp_new(); + TCGv h1 = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_andi_tl(hs, shamt, 64); + tcg_gen_andi_tl(ls, shamt, 63); + tcg_gen_neg_tl(shamt, shamt); + tcg_gen_andi_tl(rs, shamt, 63); + + tcg_gen_shl_tl(ll, src1l, ls); + tcg_gen_shl_tl(h0, src1h, ls); + tcg_gen_shr_tl(lr, src1l, rs); + tcg_gen_movcond_tl(TCG_COND_NE, lr, shamt, zero, lr, zero); + tcg_gen_or_tl(h1, h0, lr); + + tcg_gen_movcond_tl(TCG_COND_NE, destl, hs, zero, zero, ll); + tcg_gen_movcond_tl(TCG_COND_NE, desth, hs, zero, ll, h1); + + tcg_temp_free(ls); + tcg_temp_free(rs); + tcg_temp_free(hs); + tcg_temp_free(ll); + tcg_temp_free(lr); + tcg_temp_free(h0); + tcg_temp_free(h1); +} + static bool trans_sll(DisasContext *ctx, arg_sll *a) { - return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl); + return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl, gen_sll_i128); } static bool trans_slt(DisasContext *ctx, arg_slt *a) @@ -XXX,XX +XXX,XX @@ static bool trans_sltu(DisasContext *ctx, arg_sltu *a) return gen_arith(ctx, a, EXT_SIGN, gen_sltu); } +static void gen_srl_i128(TCGv destl, TCGv desth, + TCGv src1l, TCGv src1h, TCGv shamt) +{ + TCGv ls = tcg_temp_new(); + TCGv rs = tcg_temp_new(); + TCGv hs = tcg_temp_new(); + TCGv ll = tcg_temp_new(); + TCGv lr = tcg_temp_new(); + TCGv h0 = tcg_temp_new(); + TCGv h1 = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_andi_tl(hs, shamt, 64); + tcg_gen_andi_tl(rs, shamt, 63); + tcg_gen_neg_tl(shamt, shamt); + tcg_gen_andi_tl(ls, shamt, 63); + + tcg_gen_shr_tl(lr, src1l, rs); + tcg_gen_shr_tl(h1, src1h, rs); + tcg_gen_shl_tl(ll, src1h, ls); + tcg_gen_movcond_tl(TCG_COND_NE, ll, shamt, zero, ll, zero); + tcg_gen_or_tl(h0, ll, lr); + + tcg_gen_movcond_tl(TCG_COND_NE, destl, hs, zero, h1, h0); + tcg_gen_movcond_tl(TCG_COND_NE, desth, hs, zero, zero, h1); + + tcg_temp_free(ls); + tcg_temp_free(rs); + tcg_temp_free(hs); + tcg_temp_free(ll); + tcg_temp_free(lr); + tcg_temp_free(h0); + tcg_temp_free(h1); +} + static bool trans_srl(DisasContext *ctx, arg_srl *a) { - return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl); + return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl, gen_srl_i128); +} + +static void gen_sra_i128(TCGv destl, TCGv desth, + TCGv src1l, TCGv src1h, TCGv shamt) +{ + TCGv ls = tcg_temp_new(); + TCGv rs = tcg_temp_new(); + TCGv hs = tcg_temp_new(); + TCGv ll = tcg_temp_new(); + TCGv lr = tcg_temp_new(); + TCGv h0 = tcg_temp_new(); + TCGv h1 = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_andi_tl(hs, shamt, 64); + tcg_gen_andi_tl(rs, shamt, 63); + tcg_gen_neg_tl(shamt, shamt); + tcg_gen_andi_tl(ls, shamt, 63); + + tcg_gen_shr_tl(lr, src1l, rs); + tcg_gen_sar_tl(h1, src1h, rs); + tcg_gen_shl_tl(ll, src1h, ls); + tcg_gen_movcond_tl(TCG_COND_NE, ll, shamt, zero, ll, zero); + tcg_gen_or_tl(h0, ll, lr); + tcg_gen_sari_tl(lr, src1h, 63); + + tcg_gen_movcond_tl(TCG_COND_NE, destl, hs, zero, h1, h0); + tcg_gen_movcond_tl(TCG_COND_NE, desth, hs, zero, lr, h1); + + tcg_temp_free(ls); + tcg_temp_free(rs); + tcg_temp_free(hs); + tcg_temp_free(ll); + tcg_temp_free(lr); + tcg_temp_free(h0); + tcg_temp_free(h1); } static bool trans_sra(DisasContext *ctx, arg_sra *a) { - return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl); + return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl, gen_sra_i128); } static bool trans_xor(DisasContext *ctx, arg_xor *a) @@ -XXX,XX +XXX,XX @@ static bool trans_addiw(DisasContext *ctx, arg_addiw *a) static bool trans_slliw(DisasContext *ctx, arg_slliw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl); + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl, NULL); } static bool trans_srliw(DisasContext *ctx, arg_srliw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_srliw); + return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_srliw, NULL); } static bool trans_sraiw(DisasContext *ctx, arg_sraiw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_sraiw); + return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_sraiw, NULL); +} + +static bool trans_sllid(DisasContext *ctx, arg_sllid *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl, NULL); +} + +static bool trans_srlid(DisasContext *ctx, arg_srlid *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shri_tl, NULL); +} + +static bool trans_sraid(DisasContext *ctx, arg_sraid *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_sari_tl, NULL); } static bool trans_addw(DisasContext *ctx, arg_addw *a) @@ -XXX,XX +XXX,XX @@ static bool trans_subw(DisasContext *ctx, arg_subw *a) static bool trans_sllw(DisasContext *ctx, arg_sllw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl); + return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl, NULL); } static bool trans_srlw(DisasContext *ctx, arg_srlw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl); + return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl, NULL); } static bool trans_sraw(DisasContext *ctx, arg_sraw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl); + return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl, NULL); +} + +static bool trans_slld(DisasContext *ctx, arg_slld *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl, NULL); } +static bool trans_srld(DisasContext *ctx, arg_srld *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl, NULL); +} + +static bool trans_srad(DisasContext *ctx, arg_srad *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl, NULL); +} + + static bool trans_fence(DisasContext *ctx, arg_fence *a) { /* FENCE is a full memory barrier. */ -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Addition of 128-bit adds and subs in their various sizes, "set if less than"s and branches. Refactored the code to have a comparison function used for both stls and branches. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-14-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn32.decode | 3 + target/riscv/translate.c | 63 ++++++++-- target/riscv/insn_trans/trans_rvb.c.inc | 20 +-- target/riscv/insn_trans/trans_rvi.c.inc | 159 +++++++++++++++++++++--- target/riscv/insn_trans/trans_rvm.c.inc | 26 ++-- 5 files changed, 222 insertions(+), 49 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 @@ sraw 0100000 ..... ..... 101 ..... 0111011 @r ldu ............ ..... 111 ..... 0000011 @i lq ............ ..... 010 ..... 0001111 @i sq ............ ..... 100 ..... 0100011 @s +addid ............ ..... 000 ..... 1011011 @i sllid 000000 ...... ..... 001 ..... 1011011 @sh6 srlid 000000 ...... ..... 101 ..... 1011011 @sh6 sraid 010000 ...... ..... 101 ..... 1011011 @sh6 +addd 0000000 ..... ..... 000 ..... 1111011 @r +subd 0100000 ..... ..... 000 ..... 1111011 @r slld 0000000 ..... ..... 001 ..... 1111011 @r srld 0000000 ..... ..... 101 ..... 1111011 @r srad 0100000 ..... ..... 101 ..... 1111011 @r 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 bool gen_logic(DisasContext *ctx, arg_r *a, } static bool gen_arith_imm_fn(DisasContext *ctx, arg_i *a, DisasExtend ext, - void (*func)(TCGv, TCGv, target_long)) + void (*func)(TCGv, TCGv, target_long), + void (*f128)(TCGv, TCGv, TCGv, TCGv, target_long)) { TCGv dest = dest_gpr(ctx, a->rd); TCGv src1 = get_gpr(ctx, a->rs1, ext); - func(dest, src1, a->imm); + if (get_ol(ctx) < MXL_RV128) { + func(dest, src1, a->imm); + gen_set_gpr(ctx, a->rd, dest); + } else { + if (f128 == NULL) { + return false; + } - gen_set_gpr(ctx, a->rd, dest); + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv desth = dest_gprh(ctx, a->rd); + + f128(dest, desth, src1, src1h, a->imm); + gen_set_gpr128(ctx, a->rd, dest, desth); + } return true; } static bool gen_arith_imm_tl(DisasContext *ctx, arg_i *a, DisasExtend ext, - void (*func)(TCGv, TCGv, TCGv)) + void (*func)(TCGv, TCGv, TCGv), + void (*f128)(TCGv, TCGv, TCGv, TCGv, TCGv, TCGv)) { TCGv dest = dest_gpr(ctx, a->rd); TCGv src1 = get_gpr(ctx, a->rs1, ext); TCGv src2 = tcg_constant_tl(a->imm); - func(dest, src1, src2); + if (get_ol(ctx) < MXL_RV128) { + func(dest, src1, src2); + gen_set_gpr(ctx, a->rd, dest); + } else { + if (f128 == NULL) { + return false; + } - gen_set_gpr(ctx, a->rd, dest); + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv src2h = tcg_constant_tl(-(a->imm < 0)); + TCGv desth = dest_gprh(ctx, a->rd); + + f128(dest, desth, src1, src1h, src2, src2h); + gen_set_gpr128(ctx, a->rd, dest, desth); + } return true; } static bool gen_arith(DisasContext *ctx, arg_r *a, DisasExtend ext, - void (*func)(TCGv, TCGv, TCGv)) + void (*func)(TCGv, TCGv, TCGv), + void (*f128)(TCGv, TCGv, TCGv, TCGv, TCGv, TCGv)) { TCGv dest = dest_gpr(ctx, a->rd); TCGv src1 = get_gpr(ctx, a->rs1, ext); TCGv src2 = get_gpr(ctx, a->rs2, ext); - func(dest, src1, src2); + if (get_ol(ctx) < MXL_RV128) { + func(dest, src1, src2); + gen_set_gpr(ctx, a->rd, dest); + } else { + if (f128 == NULL) { + return false; + } - gen_set_gpr(ctx, a->rd, dest); + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv src2h = get_gprh(ctx, a->rs2); + TCGv desth = dest_gprh(ctx, a->rd); + + f128(dest, desth, src1, src1h, src2, src2h); + gen_set_gpr128(ctx, a->rd, dest, desth); + } return true; } static bool gen_arith_per_ol(DisasContext *ctx, arg_r *a, DisasExtend ext, void (*f_tl)(TCGv, TCGv, TCGv), - void (*f_32)(TCGv, TCGv, TCGv)) + void (*f_32)(TCGv, TCGv, TCGv), + void (*f_128)(TCGv, TCGv, TCGv, TCGv, TCGv, TCGv)) { int olen = get_olen(ctx); if (olen != TARGET_LONG_BITS) { if (olen == 32) { f_tl = f_32; - } else { + } else if (olen != 128) { g_assert_not_reached(); } } - return gen_arith(ctx, a, ext, f_tl); + return gen_arith(ctx, a, ext, f_tl, f_128); } static bool gen_shift_imm_fn(DisasContext *ctx, arg_shift *a, DisasExtend ext, 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_xnor(DisasContext *ctx, arg_xnor *a) static bool trans_min(DisasContext *ctx, arg_min *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_SIGN, tcg_gen_smin_tl); + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_smin_tl, NULL); } static bool trans_max(DisasContext *ctx, arg_max *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_SIGN, tcg_gen_smax_tl); + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_smax_tl, NULL); } static bool trans_minu(DisasContext *ctx, arg_minu *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_SIGN, tcg_gen_umin_tl); + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_umin_tl, NULL); } static bool trans_maxu(DisasContext *ctx, arg_maxu *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_SIGN, tcg_gen_umax_tl); + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_umax_tl, NULL); } static bool trans_sext_b(DisasContext *ctx, arg_sext_b *a) @@ -XXX,XX +XXX,XX @@ GEN_SHADD(3) static bool trans_sh##SHAMT##add(DisasContext *ctx, arg_sh##SHAMT##add *a) \ { \ REQUIRE_ZBA(ctx); \ - return gen_arith(ctx, a, EXT_NONE, gen_sh##SHAMT##add); \ + return gen_arith(ctx, a, EXT_NONE, gen_sh##SHAMT##add, NULL); \ } GEN_TRANS_SHADD(1) @@ -XXX,XX +XXX,XX @@ static bool trans_sh##SHAMT##add_uw(DisasContext *ctx, \ { \ REQUIRE_64BIT(ctx); \ REQUIRE_ZBA(ctx); \ - return gen_arith(ctx, a, EXT_NONE, gen_sh##SHAMT##add_uw); \ + return gen_arith(ctx, a, EXT_NONE, gen_sh##SHAMT##add_uw, NULL); \ } GEN_TRANS_SHADD_UW(1) @@ -XXX,XX +XXX,XX @@ static bool trans_add_uw(DisasContext *ctx, arg_add_uw *a) { REQUIRE_64BIT(ctx); REQUIRE_ZBA(ctx); - return gen_arith(ctx, a, EXT_NONE, gen_add_uw); + return gen_arith(ctx, a, EXT_NONE, gen_add_uw, NULL); } static void gen_slli_uw(TCGv dest, TCGv src, target_long shamt) @@ -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); - return gen_arith(ctx, a, EXT_NONE, gen_helper_clmul); + return gen_arith(ctx, a, EXT_NONE, gen_helper_clmul, NULL); } static void gen_clmulh(TCGv dst, TCGv src1, TCGv src2) @@ -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); - return gen_arith(ctx, a, EXT_NONE, gen_clmulh); + return gen_arith(ctx, a, EXT_NONE, gen_clmulh, NULL); } static bool trans_clmulr(DisasContext *ctx, arg_clmulh *a) { REQUIRE_ZBC(ctx); - return gen_arith(ctx, a, EXT_NONE, gen_helper_clmulr); + return gen_arith(ctx, a, EXT_NONE, gen_helper_clmulr, NULL); } diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_jalr(DisasContext *ctx, arg_jalr *a) return true; } +static TCGCond gen_compare_i128(bool bz, TCGv rl, + TCGv al, TCGv ah, TCGv bl, TCGv bh, + TCGCond cond) +{ + TCGv rh = tcg_temp_new(); + bool invert = false; + + switch (cond) { + case TCG_COND_EQ: + case TCG_COND_NE: + if (bz) { + tcg_gen_or_tl(rl, al, ah); + } else { + tcg_gen_xor_tl(rl, al, bl); + tcg_gen_xor_tl(rh, ah, bh); + tcg_gen_or_tl(rl, rl, rh); + } + break; + + case TCG_COND_GE: + case TCG_COND_LT: + if (bz) { + tcg_gen_mov_tl(rl, ah); + } else { + TCGv tmp = tcg_temp_new(); + + tcg_gen_sub2_tl(rl, rh, al, ah, bl, bh); + tcg_gen_xor_tl(rl, rh, ah); + tcg_gen_xor_tl(tmp, ah, bh); + tcg_gen_and_tl(rl, rl, tmp); + tcg_gen_xor_tl(rl, rh, rl); + + tcg_temp_free(tmp); + } + break; + + case TCG_COND_LTU: + invert = true; + /* fallthrough */ + case TCG_COND_GEU: + { + TCGv tmp = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + TCGv one = tcg_constant_tl(1); + + cond = TCG_COND_NE; + /* borrow in to second word */ + tcg_gen_setcond_tl(TCG_COND_LTU, tmp, al, bl); + /* seed third word with 1, which will be result */ + tcg_gen_sub2_tl(tmp, rh, ah, one, tmp, zero); + tcg_gen_sub2_tl(tmp, rl, tmp, rh, bh, zero); + + tcg_temp_free(tmp); + } + break; + + default: + g_assert_not_reached(); + } + + if (invert) { + cond = tcg_invert_cond(cond); + } + + tcg_temp_free(rh); + return cond; +} + +static void gen_setcond_i128(TCGv rl, TCGv rh, + TCGv src1l, TCGv src1h, + TCGv src2l, TCGv src2h, + TCGCond cond) +{ + cond = gen_compare_i128(false, rl, src1l, src1h, src2l, src2h, cond); + tcg_gen_setcondi_tl(cond, rl, rl, 0); + tcg_gen_movi_tl(rh, 0); +} + static bool gen_branch(DisasContext *ctx, arg_b *a, TCGCond cond) { TCGLabel *l = gen_new_label(); TCGv src1 = get_gpr(ctx, a->rs1, EXT_SIGN); TCGv src2 = get_gpr(ctx, a->rs2, EXT_SIGN); - tcg_gen_brcond_tl(cond, src1, src2, l); + if (get_xl(ctx) == MXL_RV128) { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv src2h = get_gprh(ctx, a->rs2); + TCGv tmp = tcg_temp_new(); + + cond = gen_compare_i128(a->rs2 == 0, + tmp, src1, src1h, src2, src2h, cond); + tcg_gen_brcondi_tl(cond, tmp, 0, l); + + tcg_temp_free(tmp); + } else { + tcg_gen_brcond_tl(cond, src1, src2, l); + } gen_goto_tb(ctx, 1, ctx->pc_succ_insn); gen_set_label(l); /* branch taken */ @@ -XXX,XX +XXX,XX @@ static bool trans_sq(DisasContext *ctx, arg_sq *a) return gen_store(ctx, a, MO_TEUO); } +static bool trans_addd(DisasContext *ctx, arg_addd *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl, NULL); +} + +static bool trans_addid(DisasContext *ctx, arg_addid *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl, NULL); +} + +static bool trans_subd(DisasContext *ctx, arg_subd *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl, NULL); +} + +static void gen_addi2_i128(TCGv retl, TCGv reth, + TCGv srcl, TCGv srch, target_long imm) +{ + TCGv imml = tcg_constant_tl(imm); + TCGv immh = tcg_constant_tl(-(imm < 0)); + tcg_gen_add2_tl(retl, reth, srcl, srch, imml, immh); +} + static bool trans_addi(DisasContext *ctx, arg_addi *a) { - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl); + return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl, gen_addi2_i128); } static void gen_slt(TCGv ret, TCGv s1, TCGv s2) @@ -XXX,XX +XXX,XX @@ static void gen_slt(TCGv ret, TCGv s1, TCGv s2) tcg_gen_setcond_tl(TCG_COND_LT, ret, s1, s2); } +static void gen_slt_i128(TCGv retl, TCGv reth, + TCGv s1l, TCGv s1h, TCGv s2l, TCGv s2h) +{ + gen_setcond_i128(retl, reth, s1l, s1h, s2l, s2h, TCG_COND_LT); +} + static void gen_sltu(TCGv ret, TCGv s1, TCGv s2) { tcg_gen_setcond_tl(TCG_COND_LTU, ret, s1, s2); } +static void gen_sltu_i128(TCGv retl, TCGv reth, + TCGv s1l, TCGv s1h, TCGv s2l, TCGv s2h) +{ + gen_setcond_i128(retl, reth, s1l, s1h, s2l, s2h, TCG_COND_LTU); +} + static bool trans_slti(DisasContext *ctx, arg_slti *a) { - return gen_arith_imm_tl(ctx, a, EXT_SIGN, gen_slt); + return gen_arith_imm_tl(ctx, a, EXT_SIGN, gen_slt, gen_slt_i128); } static bool trans_sltiu(DisasContext *ctx, arg_sltiu *a) { - return gen_arith_imm_tl(ctx, a, EXT_SIGN, gen_sltu); + return gen_arith_imm_tl(ctx, a, EXT_SIGN, gen_sltu, gen_sltu_i128); } static bool trans_xori(DisasContext *ctx, arg_xori *a) @@ -XXX,XX +XXX,XX @@ static bool trans_srai(DisasContext *ctx, arg_srai *a) static bool trans_add(DisasContext *ctx, arg_add *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl, tcg_gen_add2_tl); } static bool trans_sub(DisasContext *ctx, arg_sub *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl, tcg_gen_sub2_tl); } static void gen_sll_i128(TCGv destl, TCGv desth, @@ -XXX,XX +XXX,XX @@ static bool trans_sll(DisasContext *ctx, arg_sll *a) static bool trans_slt(DisasContext *ctx, arg_slt *a) { - return gen_arith(ctx, a, EXT_SIGN, gen_slt); + return gen_arith(ctx, a, EXT_SIGN, gen_slt, gen_slt_i128); } static bool trans_sltu(DisasContext *ctx, arg_sltu *a) { - return gen_arith(ctx, a, EXT_SIGN, gen_sltu); + return gen_arith(ctx, a, EXT_SIGN, gen_sltu, gen_sltu_i128); } static void gen_srl_i128(TCGv destl, TCGv desth, @@ -XXX,XX +XXX,XX @@ static bool trans_and(DisasContext *ctx, arg_and *a) static bool trans_addiw(DisasContext *ctx, arg_addiw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl); + return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl, NULL); } static bool trans_slliw(DisasContext *ctx, arg_slliw *a) @@ -XXX,XX +XXX,XX @@ static bool trans_sraid(DisasContext *ctx, arg_sraid *a) static bool trans_addw(DisasContext *ctx, arg_addw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl, NULL); } static bool trans_subw(DisasContext *ctx, arg_subw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl, NULL); } static bool trans_sllw(DisasContext *ctx, arg_sllw *a) diff --git a/target/riscv/insn_trans/trans_rvm.c.inc b/target/riscv/insn_trans/trans_rvm.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvm.c.inc +++ b/target/riscv/insn_trans/trans_rvm.c.inc @@ -XXX,XX +XXX,XX @@ static bool trans_mul(DisasContext *ctx, arg_mul *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, NULL); } static void gen_mulh(TCGv ret, TCGv s1, TCGv s2) @@ -XXX,XX +XXX,XX @@ static void gen_mulh_w(TCGv ret, TCGv s1, TCGv s2) static bool trans_mulh(DisasContext *ctx, arg_mulh *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith_per_ol(ctx, a, EXT_SIGN, gen_mulh, gen_mulh_w); + return gen_arith_per_ol(ctx, a, EXT_SIGN, gen_mulh, gen_mulh_w, NULL); } static void gen_mulhsu(TCGv ret, TCGv arg1, TCGv arg2) @@ -XXX,XX +XXX,XX @@ static void gen_mulhsu_w(TCGv ret, TCGv arg1, TCGv arg2) static bool trans_mulhsu(DisasContext *ctx, arg_mulhsu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith_per_ol(ctx, a, EXT_NONE, gen_mulhsu, gen_mulhsu_w); + return gen_arith_per_ol(ctx, a, EXT_NONE, gen_mulhsu, gen_mulhsu_w, NULL); } static void gen_mulhu(TCGv ret, TCGv s1, TCGv s2) @@ -XXX,XX +XXX,XX @@ static bool trans_mulhu(DisasContext *ctx, arg_mulhu *a) { REQUIRE_EXT(ctx, RVM); /* gen_mulh_w works for either sign as input. */ - return gen_arith_per_ol(ctx, a, EXT_ZERO, gen_mulhu, gen_mulh_w); + return gen_arith_per_ol(ctx, a, EXT_ZERO, gen_mulhu, gen_mulh_w, NULL); } static void gen_div(TCGv ret, TCGv source1, TCGv source2) @@ -XXX,XX +XXX,XX @@ static void gen_div(TCGv ret, TCGv source1, TCGv source2) static bool trans_div(DisasContext *ctx, arg_div *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_SIGN, gen_div); + return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); } static void gen_divu(TCGv ret, TCGv source1, TCGv source2) @@ -XXX,XX +XXX,XX @@ static void gen_divu(TCGv ret, TCGv source1, TCGv source2) static bool trans_divu(DisasContext *ctx, arg_divu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_ZERO, gen_divu); + return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); } static void gen_rem(TCGv ret, TCGv source1, TCGv source2) @@ -XXX,XX +XXX,XX @@ static void gen_rem(TCGv ret, TCGv source1, TCGv source2) static bool trans_rem(DisasContext *ctx, arg_rem *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_SIGN, gen_rem); + return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); } static void gen_remu(TCGv ret, TCGv source1, TCGv source2) @@ -XXX,XX +XXX,XX @@ static void gen_remu(TCGv ret, TCGv source1, TCGv source2) static bool trans_remu(DisasContext *ctx, arg_remu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_ZERO, gen_remu); + return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); } static bool trans_mulw(DisasContext *ctx, arg_mulw *a) @@ -XXX,XX +XXX,XX @@ static bool trans_mulw(DisasContext *ctx, arg_mulw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, NULL); } static bool trans_divw(DisasContext *ctx, arg_divw *a) @@ -XXX,XX +XXX,XX @@ static bool trans_divw(DisasContext *ctx, arg_divw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_SIGN, gen_div); + return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); } static bool trans_divuw(DisasContext *ctx, arg_divuw *a) @@ -XXX,XX +XXX,XX @@ static bool trans_divuw(DisasContext *ctx, arg_divuw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_ZERO, gen_divu); + return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); } static bool trans_remw(DisasContext *ctx, arg_remw *a) @@ -XXX,XX +XXX,XX @@ static bool trans_remw(DisasContext *ctx, arg_remw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_SIGN, gen_rem); + return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); } static bool trans_remuw(DisasContext *ctx, arg_remuw *a) @@ -XXX,XX +XXX,XX @@ static bool trans_remuw(DisasContext *ctx, arg_remuw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_ZERO, gen_remu); + return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); } -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Mult are generated inline (using a cool trick pointed out by Richard), but for div and rem, given the complexity of the implementation of these instructions, we call helpers to produce their behavior. From an implementation standpoint, the helpers return the low part of the results, while the high part is temporarily stored in a dedicated field of cpu_env that is used to update the architectural register in the generation wrapper. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-15-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 3 + target/riscv/helper.h | 6 + target/riscv/insn32.decode | 7 + target/riscv/m128_helper.c | 109 ++++++++++++++ target/riscv/insn_trans/trans_rvm.c.inc | 182 ++++++++++++++++++++++-- target/riscv/meson.build | 1 + 6 files changed, 295 insertions(+), 13 deletions(-) create mode 100644 target/riscv/m128_helper.c 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 CPURISCVState { uint32_t misa_ext; /* current extensions */ uint32_t misa_ext_mask; /* max ext for this cpu */ + /* 128-bit helpers upper part return value */ + target_ulong retxh; + uint32_t features; #ifdef CONFIG_USER_ONLY 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(vsext_vf2_d, void, ptr, ptr, ptr, env, i32) DEF_HELPER_5(vsext_vf4_w, void, ptr, ptr, ptr, env, i32) DEF_HELPER_5(vsext_vf4_d, void, ptr, ptr, ptr, env, i32) DEF_HELPER_5(vsext_vf8_d, void, ptr, ptr, ptr, env, i32) + +/* 128-bit integer multiplication and division */ +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) 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 @@ divuw 0000001 ..... ..... 101 ..... 0111011 @r remw 0000001 ..... ..... 110 ..... 0111011 @r remuw 0000001 ..... ..... 111 ..... 0111011 @r +# *** RV128M Standard Extension (in addition to RV64M) *** +muld 0000001 ..... ..... 000 ..... 1111011 @r +divd 0000001 ..... ..... 100 ..... 1111011 @r +divud 0000001 ..... ..... 101 ..... 1111011 @r +remd 0000001 ..... ..... 110 ..... 1111011 @r +remud 0000001 ..... ..... 111 ..... 1111011 @r + # *** RV32A Standard Extension *** lr_w 00010 . . 00000 ..... 010 ..... 0101111 @atom_ld sc_w 00011 . . ..... ..... 010 ..... 0101111 @atom_st diff --git a/target/riscv/m128_helper.c b/target/riscv/m128_helper.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/target/riscv/m128_helper.c @@ -XXX,XX +XXX,XX @@ +/* + * RISC-V Emulation Helpers for QEMU. + * + * Copyright (c) 2016-2017 Sagar Karandikar, sagark@eecs.berkeley.edu + * Copyright (c) 2017-2018 SiFive, Inc. + * + * 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 "qemu/main-loop.h" +#include "exec/exec-all.h" +#include "exec/helper-proto.h" + +target_ulong HELPER(divu_i128)(CPURISCVState *env, + target_ulong ul, target_ulong uh, + target_ulong vl, target_ulong vh) +{ + target_ulong ql, qh; + Int128 q; + + if (vl == 0 && vh == 0) { /* Handle special behavior on div by zero */ + ql = ~0x0; + qh = ~0x0; + } else { + q = int128_divu(int128_make128(ul, uh), int128_make128(vl, vh)); + ql = int128_getlo(q); + qh = int128_gethi(q); + } + + env->retxh = qh; + return ql; +} + +target_ulong HELPER(remu_i128)(CPURISCVState *env, + target_ulong ul, target_ulong uh, + target_ulong vl, target_ulong vh) +{ + target_ulong rl, rh; + Int128 r; + + if (vl == 0 && vh == 0) { + rl = ul; + rh = uh; + } else { + r = int128_remu(int128_make128(ul, uh), int128_make128(vl, vh)); + rl = int128_getlo(r); + rh = int128_gethi(r); + } + + env->retxh = rh; + return rl; +} + +target_ulong HELPER(divs_i128)(CPURISCVState *env, + target_ulong ul, target_ulong uh, + target_ulong vl, target_ulong vh) +{ + target_ulong qh, ql; + Int128 q; + + if (vl == 0 && vh == 0) { /* Div by zero check */ + ql = ~0x0; + qh = ~0x0; + } else if (uh == (1ULL << (TARGET_LONG_BITS - 1)) && ul == 0 && + vh == ~0x0 && vl == ~0x0) { + /* Signed div overflow check (-2**127 / -1) */ + ql = ul; + qh = uh; + } else { + q = int128_divs(int128_make128(ul, uh), int128_make128(vl, vh)); + ql = int128_getlo(q); + qh = int128_gethi(q); + } + + env->retxh = qh; + return ql; +} + +target_ulong HELPER(rems_i128)(CPURISCVState *env, + target_ulong ul, target_ulong uh, + target_ulong vl, target_ulong vh) +{ + target_ulong rh, rl; + Int128 r; + + if (vl == 0 && vh == 0) { + rl = ul; + rh = uh; + } else { + r = int128_rems(int128_make128(ul, uh), int128_make128(vl, vh)); + rl = int128_getlo(r); + rh = int128_gethi(r); + } + + env->retxh = rh; + return rl; +} diff --git a/target/riscv/insn_trans/trans_rvm.c.inc b/target/riscv/insn_trans/trans_rvm.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvm.c.inc +++ b/target/riscv/insn_trans/trans_rvm.c.inc @@ -XXX,XX +XXX,XX @@ * this program. If not, see <http://www.gnu.org/licenses/>. */ +static void gen_mulhu_i128(TCGv r2, TCGv r3, TCGv al, TCGv ah, TCGv bl, TCGv bh) +{ + TCGv tmpl = tcg_temp_new(); + TCGv tmph = tcg_temp_new(); + TCGv r0 = tcg_temp_new(); + TCGv r1 = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_mulu2_tl(r0, r1, al, bl); + + tcg_gen_mulu2_tl(tmpl, tmph, al, bh); + tcg_gen_add2_tl(r1, r2, r1, zero, tmpl, tmph); + tcg_gen_mulu2_tl(tmpl, tmph, ah, bl); + tcg_gen_add2_tl(r1, tmph, r1, r2, tmpl, tmph); + /* Overflow detection into r3 */ + tcg_gen_setcond_tl(TCG_COND_LTU, r3, tmph, r2); + + tcg_gen_mov_tl(r2, tmph); + + tcg_gen_mulu2_tl(tmpl, tmph, ah, bh); + tcg_gen_add2_tl(r2, r3, r2, r3, tmpl, tmph); + + tcg_temp_free(tmpl); + tcg_temp_free(tmph); +} + +static void gen_mul_i128(TCGv rl, TCGv rh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + TCGv tmpl = tcg_temp_new(); + TCGv tmph = tcg_temp_new(); + TCGv tmpx = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_mulu2_tl(rl, rh, rs1l, rs2l); + tcg_gen_mulu2_tl(tmpl, tmph, rs1l, rs2h); + tcg_gen_add2_tl(rh, tmpx, rh, zero, tmpl, tmph); + tcg_gen_mulu2_tl(tmpl, tmph, rs1h, rs2l); + tcg_gen_add2_tl(rh, tmph, rh, tmpx, tmpl, tmph); + + tcg_temp_free(tmpl); + tcg_temp_free(tmph); + tcg_temp_free(tmpx); +} static bool trans_mul(DisasContext *ctx, arg_mul *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, NULL); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, gen_mul_i128); +} + +static void gen_mulh_i128(TCGv rl, TCGv rh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + TCGv t0l = tcg_temp_new(); + TCGv t0h = tcg_temp_new(); + TCGv t1l = tcg_temp_new(); + TCGv t1h = tcg_temp_new(); + + gen_mulhu_i128(rl, rh, rs1l, rs1h, rs2l, rs2h); + tcg_gen_sari_tl(t0h, rs1h, 63); + tcg_gen_and_tl(t0l, t0h, rs2l); + tcg_gen_and_tl(t0h, t0h, rs2h); + tcg_gen_sari_tl(t1h, rs2h, 63); + tcg_gen_and_tl(t1l, t1h, rs1l); + tcg_gen_and_tl(t1h, t1h, rs1h); + tcg_gen_sub2_tl(t0l, t0h, rl, rh, t0l, t0h); + tcg_gen_sub2_tl(rl, rh, t0l, t0h, t1l, t1h); + + tcg_temp_free(t0l); + tcg_temp_free(t0h); + tcg_temp_free(t1l); + tcg_temp_free(t1h); } static void gen_mulh(TCGv ret, TCGv s1, TCGv s2) @@ -XXX,XX +XXX,XX @@ static void gen_mulh_w(TCGv ret, TCGv s1, TCGv s2) static bool trans_mulh(DisasContext *ctx, arg_mulh *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith_per_ol(ctx, a, EXT_SIGN, gen_mulh, gen_mulh_w, NULL); + return gen_arith_per_ol(ctx, a, EXT_SIGN, gen_mulh, gen_mulh_w, + gen_mulh_i128); +} + +static void gen_mulhsu_i128(TCGv rl, TCGv rh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + + TCGv t0l = tcg_temp_new(); + TCGv t0h = tcg_temp_new(); + + gen_mulhu_i128(rl, rh, rs1l, rs1h, rs2l, rs2h); + tcg_gen_sari_tl(t0h, rs1h, 63); + tcg_gen_and_tl(t0l, t0h, rs2l); + tcg_gen_and_tl(t0h, t0h, rs2h); + tcg_gen_sub2_tl(rl, rh, rl, rh, t0l, t0h); + + tcg_temp_free(t0l); + tcg_temp_free(t0h); } static void gen_mulhsu(TCGv ret, TCGv arg1, TCGv arg2) @@ -XXX,XX +XXX,XX @@ static void gen_mulhsu_w(TCGv ret, TCGv arg1, TCGv arg2) static bool trans_mulhsu(DisasContext *ctx, arg_mulhsu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith_per_ol(ctx, a, EXT_NONE, gen_mulhsu, gen_mulhsu_w, NULL); + return gen_arith_per_ol(ctx, a, EXT_NONE, gen_mulhsu, gen_mulhsu_w, + gen_mulhsu_i128); } static void gen_mulhu(TCGv ret, TCGv s1, TCGv s2) @@ -XXX,XX +XXX,XX @@ static bool trans_mulhu(DisasContext *ctx, arg_mulhu *a) { REQUIRE_EXT(ctx, RVM); /* gen_mulh_w works for either sign as input. */ - return gen_arith_per_ol(ctx, a, EXT_ZERO, gen_mulhu, gen_mulh_w, NULL); + return gen_arith_per_ol(ctx, a, EXT_ZERO, gen_mulhu, gen_mulh_w, + gen_mulhu_i128); +} + +static void gen_div_i128(TCGv rdl, TCGv rdh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + gen_helper_divs_i128(rdl, cpu_env, rs1l, rs1h, rs2l, rs2h); + tcg_gen_ld_tl(rdh, cpu_env, offsetof(CPURISCVState, retxh)); } static void gen_div(TCGv ret, TCGv source1, TCGv source2) @@ -XXX,XX +XXX,XX @@ static void gen_div(TCGv ret, TCGv source1, TCGv source2) static bool trans_div(DisasContext *ctx, arg_div *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); + return gen_arith(ctx, a, EXT_SIGN, gen_div, gen_div_i128); +} + +static void gen_divu_i128(TCGv rdl, TCGv rdh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + gen_helper_divu_i128(rdl, cpu_env, rs1l, rs1h, rs2l, rs2h); + tcg_gen_ld_tl(rdh, cpu_env, offsetof(CPURISCVState, retxh)); } static void gen_divu(TCGv ret, TCGv source1, TCGv source2) @@ -XXX,XX +XXX,XX @@ static void gen_divu(TCGv ret, TCGv source1, TCGv source2) static bool trans_divu(DisasContext *ctx, arg_divu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); + return gen_arith(ctx, a, EXT_ZERO, gen_divu, gen_divu_i128); +} + +static void gen_rem_i128(TCGv rdl, TCGv rdh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + gen_helper_rems_i128(rdl, cpu_env, rs1l, rs1h, rs2l, rs2h); + tcg_gen_ld_tl(rdh, cpu_env, offsetof(CPURISCVState, retxh)); } static void gen_rem(TCGv ret, TCGv source1, TCGv source2) @@ -XXX,XX +XXX,XX @@ static void gen_rem(TCGv ret, TCGv source1, TCGv source2) static bool trans_rem(DisasContext *ctx, arg_rem *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); + return gen_arith(ctx, a, EXT_SIGN, gen_rem, gen_rem_i128); +} + +static void gen_remu_i128(TCGv rdl, TCGv rdh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + gen_helper_remu_i128(rdl, cpu_env, rs1l, rs1h, rs2l, rs2h); + tcg_gen_ld_tl(rdh, cpu_env, offsetof(CPURISCVState, retxh)); } static void gen_remu(TCGv ret, TCGv source1, TCGv source2) @@ -XXX,XX +XXX,XX @@ static void gen_remu(TCGv ret, TCGv source1, TCGv source2) static bool trans_remu(DisasContext *ctx, arg_remu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); + return gen_arith(ctx, a, EXT_ZERO, gen_remu, gen_remu_i128); } static bool trans_mulw(DisasContext *ctx, arg_mulw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, NULL); @@ -XXX,XX +XXX,XX @@ static bool trans_mulw(DisasContext *ctx, arg_mulw *a) static bool trans_divw(DisasContext *ctx, arg_divw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); @@ -XXX,XX +XXX,XX @@ static bool trans_divw(DisasContext *ctx, arg_divw *a) static bool trans_divuw(DisasContext *ctx, arg_divuw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); @@ -XXX,XX +XXX,XX @@ static bool trans_divuw(DisasContext *ctx, arg_divuw *a) static bool trans_remw(DisasContext *ctx, arg_remw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); @@ -XXX,XX +XXX,XX @@ static bool trans_remw(DisasContext *ctx, arg_remw *a) static bool trans_remuw(DisasContext *ctx, arg_remuw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); } + +static bool trans_muld(DisasContext *ctx, arg_muld *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_mul_tl, NULL); +} + +static bool trans_divd(DisasContext *ctx, arg_divd *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); +} + +static bool trans_divud(DisasContext *ctx, arg_divud *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); +} + +static bool trans_remd(DisasContext *ctx, arg_remd *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); +} + +static bool trans_remud(DisasContext *ctx, arg_remud *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); +} 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' )) riscv_softmmu_ss = ss.source_set() -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Adding the high part of a very minimal set of csr. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-16-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 4 ++++ target/riscv/machine.c | 2 ++ 2 files changed, 6 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 CPURISCVState { target_ulong hgatp; uint64_t htimedelta; + /* Upper 64-bits of 128-bit CSRs */ + uint64_t mscratchh; + uint64_t sscratchh; + /* Virtual CSRs */ /* * For RV32 this is 32-bit vsstatus and 32-bit vsstatush. 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 const VMStateDescription vmstate_rv128 = { .needed = rv128_needed, .fields = (VMStateField[]) { VMSTATE_UINTTL_ARRAY(env.gprh, RISCVCPU, 32), + VMSTATE_UINT64(env.mscratchh, RISCVCPU), + VMSTATE_UINT64(env.sscratchh, RISCVCPU), VMSTATE_END_OF_LIST() } }; -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Given the side effects they have, the csr instructions are realized as helpers. We extend this existing infrastructure for 128-bit sized csr. We return 128-bit values using the same approach as for div/rem. Theses helpers all call a unique function that is currently a fallback on the 64-bit version. The trans_csrxx functions supporting 128-bit are yet to be implemented. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-17-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 5 +++++ target/riscv/helper.h | 3 +++ target/riscv/csr.c | 17 ++++++++++++++++ target/riscv/op_helper.c | 44 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 69 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 @@ #include "exec/cpu-defs.h" #include "fpu/softfloat-types.h" #include "qom/object.h" +#include "qemu/int128.h" #include "cpu_bits.h" #define TCG_GUEST_DEFAULT_MO 0 @@ -XXX,XX +XXX,XX @@ typedef RISCVException (*riscv_csr_op_fn)(CPURISCVState *env, int csrno, target_ulong new_value, target_ulong write_mask); +RISCVException riscv_csrrw_i128(CPURISCVState *env, int csrno, + Int128 *ret_value, + Int128 new_value, Int128 write_mask); + typedef struct { const char *name; riscv_csr_predicate_fn predicate; 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_h, TCG_CALL_NO_RWG_SE, tl, i64) DEF_HELPER_2(csrr, tl, env, int) DEF_HELPER_3(csrw, void, env, int, tl) DEF_HELPER_4(csrrw, tl, env, int, tl, tl) +DEF_HELPER_2(csrr_i128, tl, env, int) +DEF_HELPER_4(csrw_i128, void, env, int, tl, tl) +DEF_HELPER_6(csrrw_i128, tl, env, int, tl, tl, tl, tl) #ifndef CONFIG_USER_ONLY DEF_HELPER_2(sret, tl, env, tl) DEF_HELPER_2(mret, tl, env, tl) 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 @@ RISCVException riscv_csrrw(CPURISCVState *env, int csrno, return RISCV_EXCP_NONE; } +RISCVException riscv_csrrw_i128(CPURISCVState *env, int csrno, + Int128 *ret_value, + Int128 new_value, Int128 write_mask) +{ + /* fall back to 64-bit version for now */ + target_ulong ret_64; + RISCVException ret = riscv_csrrw(env, csrno, &ret_64, + int128_getlo(new_value), + int128_getlo(write_mask)); + + if (ret_value) { + *ret_value = int128_make64(ret_64); + } + + return ret; +} + /* * Debugger support. If not in user mode, set env->debugger before the * riscv_csrrw call and clear it after the call. 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_csrrw(CPURISCVState *env, int csr, return val; } +target_ulong helper_csrr_i128(CPURISCVState *env, int csr) +{ + Int128 rv = int128_zero(); + RISCVException ret = riscv_csrrw_i128(env, csr, &rv, + int128_zero(), + int128_zero()); + + if (ret != RISCV_EXCP_NONE) { + riscv_raise_exception(env, ret, GETPC()); + } + + env->retxh = int128_gethi(rv); + return int128_getlo(rv); +} + +void helper_csrw_i128(CPURISCVState *env, int csr, + target_ulong srcl, target_ulong srch) +{ + RISCVException ret = riscv_csrrw_i128(env, csr, NULL, + int128_make128(srcl, srch), + UINT128_MAX); + + if (ret != RISCV_EXCP_NONE) { + riscv_raise_exception(env, ret, GETPC()); + } +} + +target_ulong helper_csrrw_i128(CPURISCVState *env, int csr, + target_ulong srcl, target_ulong srch, + target_ulong maskl, target_ulong maskh) +{ + Int128 rv = int128_zero(); + RISCVException ret = riscv_csrrw_i128(env, csr, &rv, + int128_make128(srcl, srch), + int128_make128(maskl, maskh)); + + if (ret != RISCV_EXCP_NONE) { + riscv_raise_exception(env, ret, GETPC()); + } + + env->retxh = int128_gethi(rv); + return int128_getlo(rv); +} + #ifndef CONFIG_USER_ONLY target_ulong helper_sret(CPURISCVState *env, target_ulong cpu_pc_deb) -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> As opposed to the gen_arith and gen_shift generation helpers, the csr insns do not have a common prototype, so the choice to generate 32/64 or 128-bit helper calls is done in the trans_csrxx functions. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-18-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/insn_trans/trans_rvi.c.inc | 201 +++++++++++++++++++----- 1 file changed, 158 insertions(+), 43 deletions(-) diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -XXX,XX +XXX,XX @@ static bool do_csrrw(DisasContext *ctx, int rd, int rc, TCGv src, TCGv mask) return do_csr_post(ctx); } -static bool trans_csrrw(DisasContext *ctx, arg_csrrw *a) +static bool do_csrr_i128(DisasContext *ctx, int rd, int rc) { - TCGv src = get_gpr(ctx, a->rs1, EXT_NONE); + TCGv destl = dest_gpr(ctx, rd); + TCGv desth = dest_gprh(ctx, rd); + TCGv_i32 csr = tcg_constant_i32(rc); - /* - * If rd == 0, the insn shall not read the csr, nor cause any of the - * side effects that might occur on a csr read. - */ - if (a->rd == 0) { - return do_csrw(ctx, a->csr, src); + if (tb_cflags(ctx->base.tb) & CF_USE_ICOUNT) { + gen_io_start(); } + gen_helper_csrr_i128(destl, cpu_env, csr); + tcg_gen_ld_tl(desth, cpu_env, offsetof(CPURISCVState, retxh)); + gen_set_gpr128(ctx, rd, destl, desth); + return do_csr_post(ctx); +} + +static bool do_csrw_i128(DisasContext *ctx, int rc, TCGv srcl, TCGv srch) +{ + TCGv_i32 csr = tcg_constant_i32(rc); + + if (tb_cflags(ctx->base.tb) & CF_USE_ICOUNT) { + gen_io_start(); + } + gen_helper_csrw_i128(cpu_env, csr, srcl, srch); + return do_csr_post(ctx); +} - TCGv mask = tcg_constant_tl(-1); - return do_csrrw(ctx, a->rd, a->csr, src, mask); +static bool do_csrrw_i128(DisasContext *ctx, int rd, int rc, + TCGv srcl, TCGv srch, TCGv maskl, TCGv maskh) +{ + TCGv destl = dest_gpr(ctx, rd); + TCGv desth = dest_gprh(ctx, rd); + TCGv_i32 csr = tcg_constant_i32(rc); + + if (tb_cflags(ctx->base.tb) & CF_USE_ICOUNT) { + gen_io_start(); + } + gen_helper_csrrw_i128(destl, cpu_env, csr, srcl, srch, maskl, maskh); + tcg_gen_ld_tl(desth, cpu_env, offsetof(CPURISCVState, retxh)); + gen_set_gpr128(ctx, rd, destl, desth); + return do_csr_post(ctx); +} + +static bool trans_csrrw(DisasContext *ctx, arg_csrrw *a) +{ + if (get_xl(ctx) < MXL_RV128) { + TCGv src = get_gpr(ctx, a->rs1, EXT_NONE); + + /* + * If rd == 0, the insn shall not read the csr, nor cause any of the + * side effects that might occur on a csr read. + */ + if (a->rd == 0) { + return do_csrw(ctx, a->csr, src); + } + + TCGv mask = tcg_constant_tl(-1); + return do_csrrw(ctx, a->rd, a->csr, src, mask); + } else { + TCGv srcl = get_gpr(ctx, a->rs1, EXT_NONE); + TCGv srch = get_gprh(ctx, a->rs1); + + /* + * If rd == 0, the insn shall not read the csr, nor cause any of the + * side effects that might occur on a csr read. + */ + if (a->rd == 0) { + return do_csrw_i128(ctx, a->csr, srcl, srch); + } + + TCGv mask = tcg_constant_tl(-1); + return do_csrrw_i128(ctx, a->rd, a->csr, srcl, srch, mask, mask); + } } static bool trans_csrrs(DisasContext *ctx, arg_csrrs *a) @@ -XXX,XX +XXX,XX @@ static bool trans_csrrs(DisasContext *ctx, arg_csrrs *a) * a zero value, the instruction will still attempt to write the * unmodified value back to the csr and will cause side effects. */ - if (a->rs1 == 0) { - return do_csrr(ctx, a->rd, a->csr); - } + if (get_xl(ctx) < MXL_RV128) { + if (a->rs1 == 0) { + return do_csrr(ctx, a->rd, a->csr); + } - TCGv ones = tcg_constant_tl(-1); - TCGv mask = get_gpr(ctx, a->rs1, EXT_ZERO); - return do_csrrw(ctx, a->rd, a->csr, ones, mask); + TCGv ones = tcg_constant_tl(-1); + TCGv mask = get_gpr(ctx, a->rs1, EXT_ZERO); + return do_csrrw(ctx, a->rd, a->csr, ones, mask); + } else { + if (a->rs1 == 0) { + return do_csrr_i128(ctx, a->rd, a->csr); + } + + TCGv ones = tcg_constant_tl(-1); + TCGv maskl = get_gpr(ctx, a->rs1, EXT_ZERO); + TCGv maskh = get_gprh(ctx, a->rs1); + return do_csrrw_i128(ctx, a->rd, a->csr, ones, ones, maskl, maskh); + } } static bool trans_csrrc(DisasContext *ctx, arg_csrrc *a) @@ -XXX,XX +XXX,XX @@ static bool trans_csrrc(DisasContext *ctx, arg_csrrc *a) * a zero value, the instruction will still attempt to write the * unmodified value back to the csr and will cause side effects. */ - if (a->rs1 == 0) { - return do_csrr(ctx, a->rd, a->csr); - } + if (get_xl(ctx) < MXL_RV128) { + if (a->rs1 == 0) { + return do_csrr(ctx, a->rd, a->csr); + } - TCGv mask = get_gpr(ctx, a->rs1, EXT_ZERO); - return do_csrrw(ctx, a->rd, a->csr, ctx->zero, mask); + TCGv mask = get_gpr(ctx, a->rs1, EXT_ZERO); + return do_csrrw(ctx, a->rd, a->csr, ctx->zero, mask); + } else { + if (a->rs1 == 0) { + return do_csrr_i128(ctx, a->rd, a->csr); + } + + TCGv maskl = get_gpr(ctx, a->rs1, EXT_ZERO); + TCGv maskh = get_gprh(ctx, a->rs1); + return do_csrrw_i128(ctx, a->rd, a->csr, + ctx->zero, ctx->zero, maskl, maskh); + } } static bool trans_csrrwi(DisasContext *ctx, arg_csrrwi *a) { - TCGv src = tcg_constant_tl(a->rs1); + if (get_xl(ctx) < MXL_RV128) { + TCGv src = tcg_constant_tl(a->rs1); - /* - * If rd == 0, the insn shall not read the csr, nor cause any of the - * side effects that might occur on a csr read. - */ - if (a->rd == 0) { - return do_csrw(ctx, a->csr, src); - } + /* + * If rd == 0, the insn shall not read the csr, nor cause any of the + * side effects that might occur on a csr read. + */ + if (a->rd == 0) { + return do_csrw(ctx, a->csr, src); + } - TCGv mask = tcg_constant_tl(-1); - return do_csrrw(ctx, a->rd, a->csr, src, mask); + TCGv mask = tcg_constant_tl(-1); + return do_csrrw(ctx, a->rd, a->csr, src, mask); + } else { + TCGv src = tcg_constant_tl(a->rs1); + + /* + * If rd == 0, the insn shall not read the csr, nor cause any of the + * side effects that might occur on a csr read. + */ + if (a->rd == 0) { + return do_csrw_i128(ctx, a->csr, src, ctx->zero); + } + + TCGv mask = tcg_constant_tl(-1); + return do_csrrw_i128(ctx, a->rd, a->csr, src, ctx->zero, mask, mask); + } } static bool trans_csrrsi(DisasContext *ctx, arg_csrrsi *a) @@ -XXX,XX +XXX,XX @@ static bool trans_csrrsi(DisasContext *ctx, arg_csrrsi *a) * a zero value, the instruction will still attempt to write the * unmodified value back to the csr and will cause side effects. */ - if (a->rs1 == 0) { - return do_csrr(ctx, a->rd, a->csr); - } + if (get_xl(ctx) < MXL_RV128) { + if (a->rs1 == 0) { + return do_csrr(ctx, a->rd, a->csr); + } + + TCGv ones = tcg_constant_tl(-1); + TCGv mask = tcg_constant_tl(a->rs1); + return do_csrrw(ctx, a->rd, a->csr, ones, mask); + } else { + if (a->rs1 == 0) { + return do_csrr_i128(ctx, a->rd, a->csr); + } - TCGv ones = tcg_constant_tl(-1); - TCGv mask = tcg_constant_tl(a->rs1); - return do_csrrw(ctx, a->rd, a->csr, ones, mask); + TCGv ones = tcg_constant_tl(-1); + TCGv mask = tcg_constant_tl(a->rs1); + return do_csrrw_i128(ctx, a->rd, a->csr, ones, ones, mask, ctx->zero); + } } -static bool trans_csrrci(DisasContext *ctx, arg_csrrci *a) +static bool trans_csrrci(DisasContext *ctx, arg_csrrci * a) { /* * If rs1 == 0, the insn shall not write to the csr at all, nor @@ -XXX,XX +XXX,XX @@ static bool trans_csrrci(DisasContext *ctx, arg_csrrci *a) * a zero value, the instruction will still attempt to write the * unmodified value back to the csr and will cause side effects. */ - if (a->rs1 == 0) { - return do_csrr(ctx, a->rd, a->csr); - } + if (get_xl(ctx) < MXL_RV128) { + if (a->rs1 == 0) { + return do_csrr(ctx, a->rd, a->csr); + } - TCGv mask = tcg_constant_tl(a->rs1); - return do_csrrw(ctx, a->rd, a->csr, ctx->zero, mask); + TCGv mask = tcg_constant_tl(a->rs1); + return do_csrrw(ctx, a->rd, a->csr, ctx->zero, mask); + } else { + if (a->rs1 == 0) { + return do_csrr_i128(ctx, a->rd, a->csr); + } + + TCGv mask = tcg_constant_tl(a->rs1); + return do_csrrw_i128(ctx, a->rd, a->csr, + ctx->zero, ctx->zero, mask, ctx->zero); + } } -- 2.31.1 From: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> The csrs are accessed through function pointers: we add 128-bit read operations in the table for three csrs (writes fallback to the 64-bit version as the upper 64-bit information is handled elsewhere): - misa, as mxl is needed for proper operation, - mstatus and sstatus, to return sd In addition, we also add read and write accesses to the machine and supervisor scratch registers. Signed-off-by: Frédéric Pétrot <frederic.petrot@univ-grenoble-alpes.fr> Co-authored-by: Fabien Portas <fabien.portas@grenoble-inp.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20220106210108.138226-19-frederic.petrot@univ-grenoble-alpes.fr Signed-off-by: Alistair Francis <alistair.francis@wdc.com> --- target/riscv/cpu.h | 7 ++ target/riscv/cpu_bits.h | 3 + target/riscv/csr.c | 195 +++++++++++++++++++++++++++++++++------- 3 files changed, 175 insertions(+), 30 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 @@ RISCVException riscv_csrrw_i128(CPURISCVState *env, int csrno, Int128 *ret_value, Int128 new_value, Int128 write_mask); +typedef RISCVException (*riscv_csr_read128_fn)(CPURISCVState *env, int csrno, + Int128 *ret_value); +typedef RISCVException (*riscv_csr_write128_fn)(CPURISCVState *env, int csrno, + Int128 new_value); + typedef struct { const char *name; riscv_csr_predicate_fn predicate; riscv_csr_read_fn read; riscv_csr_write_fn write; riscv_csr_op_fn op; + riscv_csr_read128_fn read128; + riscv_csr_write128_fn write128; } riscv_csr_operations; /* CSR function table constants */ 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 MSTATUS32_SD 0x80000000 #define MSTATUS64_SD 0x8000000000000000ULL +#define MSTATUSH128_SD 0x8000000000000000ULL #define MISA32_MXL 0xC0000000 #define MISA64_MXL 0xC000000000000000ULL @@ -XXX,XX +XXX,XX @@ typedef enum { #define SSTATUS_SUM 0x00040000 /* since: priv-1.10 */ #define SSTATUS_MXR 0x00080000 +#define SSTATUS64_UXL 0x0000000300000000ULL + #define SSTATUS32_SD 0x80000000 #define SSTATUS64_SD 0x8000000000000000ULL 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 const target_ulong vs_delegable_excps = DELEGABLE_EXCPS & (1ULL << (RISCV_EXCP_STORE_GUEST_AMO_ACCESS_FAULT))); static const target_ulong sstatus_v1_10_mask = SSTATUS_SIE | SSTATUS_SPIE | SSTATUS_UIE | SSTATUS_UPIE | SSTATUS_SPP | SSTATUS_FS | SSTATUS_XS | - SSTATUS_SUM | SSTATUS_MXR | SSTATUS_VS; + SSTATUS_SUM | SSTATUS_MXR | SSTATUS_VS | (target_ulong)SSTATUS64_UXL; static const target_ulong sip_writable_mask = SIP_SSIP | MIP_USIP | MIP_UEIP; static const target_ulong hip_writable_mask = MIP_VSSIP; static const target_ulong hvip_writable_mask = MIP_VSSIP | MIP_VSTIP | MIP_VSEIP; @@ -XXX,XX +XXX,XX @@ static uint64_t add_status_sd(RISCVMXL xl, uint64_t status) return status | MSTATUS32_SD; case MXL_RV64: return status | MSTATUS64_SD; + case MXL_RV128: + return MSTATUSH128_SD; default: g_assert_not_reached(); } @@ -XXX,XX +XXX,XX @@ static RISCVException write_mstatus(CPURISCVState *env, int csrno, mstatus = (mstatus & ~mask) | (val & mask); - if (riscv_cpu_mxl(env) == MXL_RV64) { + RISCVMXL xl = riscv_cpu_mxl(env); + if (xl > MXL_RV32) { /* SXL and UXL fields are for now read only */ - mstatus = set_field(mstatus, MSTATUS64_SXL, MXL_RV64); - mstatus = set_field(mstatus, MSTATUS64_UXL, MXL_RV64); + mstatus = set_field(mstatus, MSTATUS64_SXL, xl); + mstatus = set_field(mstatus, MSTATUS64_UXL, xl); } env->mstatus = mstatus; @@ -XXX,XX +XXX,XX @@ static RISCVException write_mstatush(CPURISCVState *env, int csrno, return RISCV_EXCP_NONE; } +static RISCVException read_mstatus_i128(CPURISCVState *env, int csrno, + Int128 *val) +{ + *val = int128_make128(env->mstatus, add_status_sd(MXL_RV128, env->mstatus)); + return RISCV_EXCP_NONE; +} + +static RISCVException read_misa_i128(CPURISCVState *env, int csrno, + Int128 *val) +{ + *val = int128_make128(env->misa_ext, (uint64_t)MXL_RV128 << 62); + return RISCV_EXCP_NONE; +} + static RISCVException read_misa(CPURISCVState *env, int csrno, target_ulong *val) { @@ -XXX,XX +XXX,XX @@ static RISCVException write_mcounteren(CPURISCVState *env, int csrno, } /* Machine Trap Handling */ +static RISCVException read_mscratch_i128(CPURISCVState *env, int csrno, + Int128 *val) +{ + *val = int128_make128(env->mscratch, env->mscratchh); + return RISCV_EXCP_NONE; +} + +static RISCVException write_mscratch_i128(CPURISCVState *env, int csrno, + Int128 val) +{ + env->mscratch = int128_getlo(val); + env->mscratchh = int128_gethi(val); + return RISCV_EXCP_NONE; +} + static RISCVException read_mscratch(CPURISCVState *env, int csrno, target_ulong *val) { @@ -XXX,XX +XXX,XX @@ static RISCVException rmw_mip(CPURISCVState *env, int csrno, } /* Supervisor Trap Setup */ +static RISCVException read_sstatus_i128(CPURISCVState *env, int csrno, + Int128 *val) +{ + uint64_t mask = sstatus_v1_10_mask; + uint64_t sstatus = env->mstatus & mask; + + *val = int128_make128(sstatus, add_status_sd(MXL_RV128, sstatus)); + return RISCV_EXCP_NONE; +} + static RISCVException read_sstatus(CPURISCVState *env, int csrno, target_ulong *val) { @@ -XXX,XX +XXX,XX @@ static RISCVException write_scounteren(CPURISCVState *env, int csrno, } /* Supervisor Trap Handling */ +static RISCVException read_sscratch_i128(CPURISCVState *env, int csrno, + Int128 *val) +{ + *val = int128_make128(env->sscratch, env->sscratchh); + return RISCV_EXCP_NONE; +} + +static RISCVException write_sscratch_i128(CPURISCVState *env, int csrno, + Int128 val) +{ + env->sscratch = int128_getlo(val); + env->sscratchh = int128_gethi(val); + return RISCV_EXCP_NONE; +} + static RISCVException read_sscratch(CPURISCVState *env, int csrno, target_ulong *val) { @@ -XXX,XX +XXX,XX @@ static RISCVException write_upmbase(CPURISCVState *env, int csrno, * csrrc <-> riscv_csrrw(env, csrno, ret_value, 0, value); */ -RISCVException riscv_csrrw(CPURISCVState *env, int csrno, - target_ulong *ret_value, - target_ulong new_value, target_ulong write_mask) +static inline RISCVException riscv_csrrw_check(CPURISCVState *env, + int csrno, + bool write_mask, + RISCVCPU *cpu) { - RISCVException ret; - target_ulong old_value; - RISCVCPU *cpu = env_archcpu(env); - int read_only = get_field(csrno, 0xC00) == 3; - /* check privileges and return RISCV_EXCP_ILLEGAL_INST if check fails */ + int read_only = get_field(csrno, 0xC00) == 3; #if !defined(CONFIG_USER_ONLY) int effective_priv = env->priv; @@ -XXX,XX +XXX,XX @@ RISCVException riscv_csrrw(CPURISCVState *env, int csrno, if (!csr_ops[csrno].predicate) { return RISCV_EXCP_ILLEGAL_INST; } - ret = csr_ops[csrno].predicate(env, csrno); - if (ret != RISCV_EXCP_NONE) { - return ret; - } + + return csr_ops[csrno].predicate(env, csrno); +} + +static RISCVException riscv_csrrw_do64(CPURISCVState *env, int csrno, + target_ulong *ret_value, + target_ulong new_value, + target_ulong write_mask) +{ + RISCVException ret; + target_ulong old_value; /* execute combined read/write operation if it exists */ if (csr_ops[csrno].op) { @@ -XXX,XX +XXX,XX @@ RISCVException riscv_csrrw(CPURISCVState *env, int csrno, return RISCV_EXCP_NONE; } -RISCVException riscv_csrrw_i128(CPURISCVState *env, int csrno, - Int128 *ret_value, - Int128 new_value, Int128 write_mask) +RISCVException riscv_csrrw(CPURISCVState *env, int csrno, + target_ulong *ret_value, + target_ulong new_value, target_ulong write_mask) +{ + RISCVCPU *cpu = env_archcpu(env); + + RISCVException ret = riscv_csrrw_check(env, csrno, write_mask, cpu); + if (ret != RISCV_EXCP_NONE) { + return ret; + } + + return riscv_csrrw_do64(env, csrno, ret_value, new_value, write_mask); +} + +static RISCVException riscv_csrrw_do128(CPURISCVState *env, int csrno, + Int128 *ret_value, + Int128 new_value, + Int128 write_mask) { - /* fall back to 64-bit version for now */ - target_ulong ret_64; - RISCVException ret = riscv_csrrw(env, csrno, &ret_64, - int128_getlo(new_value), - int128_getlo(write_mask)); + RISCVException ret; + Int128 old_value; + + /* read old value */ + ret = csr_ops[csrno].read128(env, csrno, &old_value); + if (ret != RISCV_EXCP_NONE) { + return ret; + } + + /* write value if writable and write mask set, otherwise drop writes */ + if (int128_nz(write_mask)) { + new_value = int128_or(int128_and(old_value, int128_not(write_mask)), + int128_and(new_value, write_mask)); + if (csr_ops[csrno].write128) { + ret = csr_ops[csrno].write128(env, csrno, new_value); + if (ret != RISCV_EXCP_NONE) { + return ret; + } + } else if (csr_ops[csrno].write) { + /* avoids having to write wrappers for all registers */ + ret = csr_ops[csrno].write(env, csrno, int128_getlo(new_value)); + if (ret != RISCV_EXCP_NONE) { + return ret; + } + } + } + /* return old value */ if (ret_value) { - *ret_value = int128_make64(ret_64); + *ret_value = old_value; + } + + return RISCV_EXCP_NONE; +} + +RISCVException riscv_csrrw_i128(CPURISCVState *env, int csrno, + Int128 *ret_value, + Int128 new_value, Int128 write_mask) +{ + RISCVException ret; + RISCVCPU *cpu = env_archcpu(env); + + ret = riscv_csrrw_check(env, csrno, int128_nz(write_mask), cpu); + if (ret != RISCV_EXCP_NONE) { + return ret; } + if (csr_ops[csrno].read128) { + return riscv_csrrw_do128(env, csrno, ret_value, new_value, write_mask); + } + + /* + * Fall back to 64-bit version for now, if the 128-bit alternative isn't + * at all defined. + * Note, some CSRs don't need to extend to MXLEN (64 upper bits non + * significant), for those, this fallback is correctly handling the accesses + */ + target_ulong old_value; + ret = riscv_csrrw_do64(env, csrno, &old_value, + int128_getlo(new_value), + int128_getlo(write_mask)); + if (ret == RISCV_EXCP_NONE && ret_value) { + *ret_value = int128_make64(old_value); + } return ret; } @@ -XXX,XX +XXX,XX @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { [CSR_MHARTID] = { "mhartid", any, read_mhartid }, /* Machine Trap Setup */ - [CSR_MSTATUS] = { "mstatus", any, read_mstatus, write_mstatus }, - [CSR_MISA] = { "misa", any, read_misa, write_misa }, + [CSR_MSTATUS] = { "mstatus", any, read_mstatus, write_mstatus, NULL, + read_mstatus_i128 }, + [CSR_MISA] = { "misa", any, read_misa, write_misa, NULL, + read_misa_i128 }, [CSR_MIDELEG] = { "mideleg", any, read_mideleg, write_mideleg }, [CSR_MEDELEG] = { "medeleg", any, read_medeleg, write_medeleg }, [CSR_MIE] = { "mie", any, read_mie, write_mie }, @@ -XXX,XX +XXX,XX @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { [CSR_MSTATUSH] = { "mstatush", any32, read_mstatush, write_mstatush }, /* Machine Trap Handling */ - [CSR_MSCRATCH] = { "mscratch", any, read_mscratch, write_mscratch }, + [CSR_MSCRATCH] = { "mscratch", any, read_mscratch, write_mscratch, NULL, + read_mscratch_i128, write_mscratch_i128 }, [CSR_MEPC] = { "mepc", any, read_mepc, write_mepc }, [CSR_MCAUSE] = { "mcause", any, read_mcause, write_mcause }, [CSR_MTVAL] = { "mtval", any, read_mtval, write_mtval }, [CSR_MIP] = { "mip", any, NULL, NULL, rmw_mip }, /* Supervisor Trap Setup */ - [CSR_SSTATUS] = { "sstatus", smode, read_sstatus, write_sstatus }, + [CSR_SSTATUS] = { "sstatus", smode, read_sstatus, write_sstatus, NULL, + read_sstatus_i128 }, [CSR_SIE] = { "sie", smode, read_sie, write_sie }, [CSR_STVEC] = { "stvec", smode, read_stvec, write_stvec }, [CSR_SCOUNTEREN] = { "scounteren", smode, read_scounteren, write_scounteren }, /* Supervisor Trap Handling */ - [CSR_SSCRATCH] = { "sscratch", smode, read_sscratch, write_sscratch }, + [CSR_SSCRATCH] = { "sscratch", smode, read_sscratch, write_sscratch, NULL, + read_sscratch_i128, write_sscratch_i128 }, [CSR_SEPC] = { "sepc", smode, read_sepc, write_sepc }, [CSR_SCAUSE] = { "scause", smode, read_scause, write_scause }, [CSR_STVAL] = { "stval", smode, read_stval, write_stval }, -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-id: 20211220064916.107241-2-alistair.francis@opensource.wdc.com --- target/riscv/translate.c | 2 ++ 1 file changed, 2 insertions(+) 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 void decode_opc(CPURISCVState *env, DisasContext *ctx, uint16_t opcode) if (!has_ext(ctx, RVC)) { gen_exception_illegal(ctx); } else { + ctx->opcode = opcode; ctx->pc_succ_insn = ctx->base.pc_next + 2; if (!decode_insn16(ctx, opcode)) { gen_exception_illegal(ctx); @@ -XXX,XX +XXX,XX @@ static void decode_opc(CPURISCVState *env, DisasContext *ctx, uint16_t opcode) opcode32 = deposit32(opcode32, 16, 16, translator_lduw(env, &ctx->base, ctx->base.pc_next + 2)); + ctx->opcode = opcode32; ctx->pc_succ_insn = ctx->base.pc_next + 4; if (!decode_insn32(ctx, opcode32)) { gen_exception_illegal(ctx); -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> In preparation for adding support for the illegal instruction address let's fixup the Hypervisor extension setting GVA logic and improve the variable names. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-id: 20211220064916.107241-3-alistair.francis@opensource.wdc.com --- target/riscv/cpu_helper.c | 21 ++++++--------------- 1 file changed, 6 insertions(+), 15 deletions(-) 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 @@ void riscv_cpu_do_interrupt(CPUState *cs) RISCVCPU *cpu = RISCV_CPU(cs); CPURISCVState *env = &cpu->env; + bool write_gva = false; uint64_t s; /* cs->exception is 32-bits wide unlike mcause which is XLEN-bits wide @@ -XXX,XX +XXX,XX @@ void riscv_cpu_do_interrupt(CPUState *cs) bool async = !!(cs->exception_index & RISCV_EXCP_INT_FLAG); target_ulong cause = cs->exception_index & RISCV_EXCP_INT_MASK; target_ulong deleg = async ? env->mideleg : env->medeleg; - bool write_tval = false; target_ulong tval = 0; target_ulong htval = 0; target_ulong mtval2 = 0; @@ -XXX,XX +XXX,XX @@ void riscv_cpu_do_interrupt(CPUState *cs) case RISCV_EXCP_INST_PAGE_FAULT: case RISCV_EXCP_LOAD_PAGE_FAULT: case RISCV_EXCP_STORE_PAGE_FAULT: - write_tval = true; + write_gva = true; tval = env->badaddr; break; default: @@ -XXX,XX +XXX,XX @@ void riscv_cpu_do_interrupt(CPUState *cs) if (riscv_has_ext(env, RVH)) { target_ulong hdeleg = async ? env->hideleg : env->hedeleg; - if (env->two_stage_lookup && write_tval) { - /* - * If we are writing a guest virtual address to stval, set - * this to 1. If we are trapping to VS we will set this to 0 - * later. - */ - env->hstatus = set_field(env->hstatus, HSTATUS_GVA, 1); - } else { - /* For other HS-mode traps, we set this to 0. */ - env->hstatus = set_field(env->hstatus, HSTATUS_GVA, 0); - } - if (riscv_cpu_virt_enabled(env) && ((hdeleg >> cause) & 1)) { /* Trap to VS mode */ /* @@ -XXX,XX +XXX,XX @@ void riscv_cpu_do_interrupt(CPUState *cs) cause == IRQ_VS_EXT) { cause = cause - 1; } - env->hstatus = set_field(env->hstatus, HSTATUS_GVA, 0); + write_gva = false; } else if (riscv_cpu_virt_enabled(env)) { /* Trap into HS mode, from virt */ riscv_cpu_swap_hypervisor_regs(env); @@ -XXX,XX +XXX,XX @@ void riscv_cpu_do_interrupt(CPUState *cs) env->hstatus = set_field(env->hstatus, HSTATUS_SPV, riscv_cpu_virt_enabled(env)); + htval = env->guest_phys_fault_addr; riscv_cpu_set_virt_enabled(env, 0); @@ -XXX,XX +XXX,XX @@ void riscv_cpu_do_interrupt(CPUState *cs) /* Trap into HS mode */ env->hstatus = set_field(env->hstatus, HSTATUS_SPV, false); htval = env->guest_phys_fault_addr; + write_gva = false; } + env->hstatus = set_field(env->hstatus, HSTATUS_GVA, write_gva); } s = env->mstatus; -- 2.31.1 From: Alistair Francis <alistair.francis@wdc.com> The stval and mtval registers can optionally contain the faulting instruction on an illegal instruction exception. This patch adds support for setting the stval and mtval registers. The RISC-V spec states that "The stval register can optionally also be used to return the faulting instruction bits on an illegal instruction exception...". In this case we are always writing the value on an illegal instruction. This doesn't match all CPUs (some CPUs won't write the data), but in QEMU let's just populate the value on illegal instructions. This won't break any guest software, but will provide more information to guests. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Message-id: 20211220064916.107241-4-alistair.francis@opensource.wdc.com --- target/riscv/cpu.h | 2 ++ target/riscv/cpu_helper.c | 3 +++ target/riscv/translate.c | 3 +++ 3 files changed, 8 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 CPURISCVState { target_ulong frm; target_ulong badaddr; + uint32_t bins; + target_ulong guest_phys_fault_addr; target_ulong priv_ver; 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 @@ void riscv_cpu_do_interrupt(CPUState *cs) write_gva = true; tval = env->badaddr; break; + case RISCV_EXCP_ILLEGAL_INST: + tval = env->bins; + break; default: break; } 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 void generate_exception_mtval(DisasContext *ctx, int excp) static void gen_exception_illegal(DisasContext *ctx) { + tcg_gen_st_i32(tcg_constant_i32(ctx->opcode), cpu_env, + offsetof(CPURISCVState, bins)); + generate_exception(ctx, RISCV_EXCP_ILLEGAL_INST); } -- 2.31.1 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