Message ID | 20210212150256.885-1-zhiwei_liu@c-sky.com |
---|---|
Headers | show |
Series | target/riscv: support packed extension v0.9.2 | expand |
ping On 2021/2/12 23:02, LIU Zhiwei wrote: > This patchset implements the packed extension for RISC-V on QEMU. > > This patchset have passed all my direct Linux user mode cases(RV64) and > bare metal cases(RV32) on X86-64 Ubuntu host machine. I will later push > these test cases to my repo(https://github.com/romanheros/qemu.git > branch:packed-upstream-v1). > > I have ported packed extension on RISU, but I didn't find a simulator or > hardware to compare with. If anyone have one, please let me know. > > Features: > * support specification packed extension v0.9.2(https://github.com/riscv/riscv-p-spec/) > * support basic packed extension. > * support Zp64. > > LIU Zhiwei (38): > target/riscv: implementation-defined constant parameters > target/riscv: Hoist vector functions > target/riscv: Fixup saturate subtract function > target/riscv: 16-bit Addition & Subtraction Instructions > target/riscv: 8-bit Addition & Subtraction Instruction > target/riscv: SIMD 16-bit Shift Instructions > target/riscv: SIMD 8-bit Shift Instructions > target/riscv: SIMD 16-bit Compare Instructions > target/riscv: SIMD 8-bit Compare Instructions > target/riscv: SIMD 16-bit Multiply Instructions > target/riscv: SIMD 8-bit Multiply Instructions > target/riscv: SIMD 16-bit Miscellaneous Instructions > target/riscv: SIMD 8-bit Miscellaneous Instructions > target/riscv: 8-bit Unpacking Instructions > target/riscv: 16-bit Packing Instructions > target/riscv: Signed MSW 32x32 Multiply and Add Instructions > target/riscv: Signed MSW 32x16 Multiply and Add Instructions > target/riscv: Signed 16-bit Multiply 32-bit Add/Subtract Instructions > target/riscv: Signed 16-bit Multiply 64-bit Add/Subtract Instructions > target/riscv: Partial-SIMD Miscellaneous Instructions > target/riscv: 8-bit Multiply with 32-bit Add Instructions > target/riscv: 64-bit Add/Subtract Instructions > target/riscv: 32-bit Multiply 64-bit Add/Subtract Instructions > target/riscv: Signed 16-bit Multiply with 64-bit Add/Subtract > Instructions > target/riscv: Non-SIMD Q15 saturation ALU Instructions > target/riscv: Non-SIMD Q31 saturation ALU Instructions > target/riscv: 32-bit Computation Instructions > target/riscv: Non-SIMD Miscellaneous Instructions > target/riscv: RV64 Only SIMD 32-bit Add/Subtract Instructions > target/riscv: RV64 Only SIMD 32-bit Shift Instructions > target/riscv: RV64 Only SIMD 32-bit Miscellaneous Instructions > target/riscv: RV64 Only SIMD Q15 saturating Multiply Instructions > target/riscv: RV64 Only 32-bit Multiply Instructions > target/riscv: RV64 Only 32-bit Multiply & Add Instructions > target/riscv: RV64 Only 32-bit Parallel Multiply & Add Instructions > target/riscv: RV64 Only Non-SIMD 32-bit Shift Instructions > target/riscv: RV64 Only 32-bit Packing Instructions > target/riscv: configure and turn on packed extension from command line > > target/riscv/cpu.c | 32 + > target/riscv/cpu.h | 6 + > target/riscv/helper.h | 332 ++ > target/riscv/insn32-64.decode | 93 +- > target/riscv/insn32.decode | 285 ++ > target/riscv/insn_trans/trans_rvp.c.inc | 1224 +++++++ > target/riscv/internals.h | 50 + > target/riscv/meson.build | 1 + > target/riscv/packed_helper.c | 3862 +++++++++++++++++++++++ > target/riscv/translate.c | 3 + > target/riscv/vector_helper.c | 90 +- > 11 files changed, 5912 insertions(+), 66 deletions(-) > create mode 100644 target/riscv/insn_trans/trans_rvp.c.inc > create mode 100644 target/riscv/packed_helper.c >
ping +1. On 2021/2/12 下午11:02, LIU Zhiwei wrote: > This patchset implements the packed extension for RISC-V on QEMU. > > This patchset have passed all my direct Linux user mode cases(RV64) and > bare metal cases(RV32) on X86-64 Ubuntu host machine. I will later push > these test cases to my repo(https://github.com/romanheros/qemu.git > branch:packed-upstream-v1). > > I have ported packed extension on RISU, but I didn't find a simulator or > hardware to compare with. If anyone have one, please let me know. > > Features: > * support specification packed extension v0.9.2(https://github.com/riscv/riscv-p-spec/) > * support basic packed extension. > * support Zp64. > > LIU Zhiwei (38): > target/riscv: implementation-defined constant parameters > target/riscv: Hoist vector functions > target/riscv: Fixup saturate subtract function > target/riscv: 16-bit Addition & Subtraction Instructions > target/riscv: 8-bit Addition & Subtraction Instruction > target/riscv: SIMD 16-bit Shift Instructions > target/riscv: SIMD 8-bit Shift Instructions > target/riscv: SIMD 16-bit Compare Instructions > target/riscv: SIMD 8-bit Compare Instructions > target/riscv: SIMD 16-bit Multiply Instructions > target/riscv: SIMD 8-bit Multiply Instructions > target/riscv: SIMD 16-bit Miscellaneous Instructions > target/riscv: SIMD 8-bit Miscellaneous Instructions > target/riscv: 8-bit Unpacking Instructions > target/riscv: 16-bit Packing Instructions > target/riscv: Signed MSW 32x32 Multiply and Add Instructions > target/riscv: Signed MSW 32x16 Multiply and Add Instructions > target/riscv: Signed 16-bit Multiply 32-bit Add/Subtract Instructions > target/riscv: Signed 16-bit Multiply 64-bit Add/Subtract Instructions > target/riscv: Partial-SIMD Miscellaneous Instructions > target/riscv: 8-bit Multiply with 32-bit Add Instructions > target/riscv: 64-bit Add/Subtract Instructions > target/riscv: 32-bit Multiply 64-bit Add/Subtract Instructions > target/riscv: Signed 16-bit Multiply with 64-bit Add/Subtract > Instructions > target/riscv: Non-SIMD Q15 saturation ALU Instructions > target/riscv: Non-SIMD Q31 saturation ALU Instructions > target/riscv: 32-bit Computation Instructions > target/riscv: Non-SIMD Miscellaneous Instructions > target/riscv: RV64 Only SIMD 32-bit Add/Subtract Instructions > target/riscv: RV64 Only SIMD 32-bit Shift Instructions > target/riscv: RV64 Only SIMD 32-bit Miscellaneous Instructions > target/riscv: RV64 Only SIMD Q15 saturating Multiply Instructions > target/riscv: RV64 Only 32-bit Multiply Instructions > target/riscv: RV64 Only 32-bit Multiply & Add Instructions > target/riscv: RV64 Only 32-bit Parallel Multiply & Add Instructions > target/riscv: RV64 Only Non-SIMD 32-bit Shift Instructions > target/riscv: RV64 Only 32-bit Packing Instructions > target/riscv: configure and turn on packed extension from command line > > target/riscv/cpu.c | 32 + > target/riscv/cpu.h | 6 + > target/riscv/helper.h | 332 ++ > target/riscv/insn32-64.decode | 93 +- > target/riscv/insn32.decode | 285 ++ > target/riscv/insn_trans/trans_rvp.c.inc | 1224 +++++++ > target/riscv/internals.h | 50 + > target/riscv/meson.build | 1 + > target/riscv/packed_helper.c | 3862 +++++++++++++++++++++++ > target/riscv/translate.c | 3 + > target/riscv/vector_helper.c | 90 +- > 11 files changed, 5912 insertions(+), 66 deletions(-) > create mode 100644 target/riscv/insn_trans/trans_rvp.c.inc > create mode 100644 target/riscv/packed_helper.c >
On Tue, Apr 13, 2021 at 1:28 PM LIU Zhiwei <zhiwei_liu@c-sky.com> wrote: > > ping +1. > > On 2021/2/12 下午11:02, LIU Zhiwei wrote: > > This patchset implements the packed extension for RISC-V on QEMU. > > > > This patchset have passed all my direct Linux user mode cases(RV64) and > > bare metal cases(RV32) on X86-64 Ubuntu host machine. I will later push > > these test cases to my repo(https://github.com/romanheros/qemu.git > > branch:packed-upstream-v1). > > > > I have ported packed extension on RISU, but I didn't find a simulator or > > hardware to compare with. If anyone have one, please let me know. > > > > Features: > > * support specification packed extension v0.9.2(https://github.com/riscv/riscv-p-spec/) > > * support basic packed extension. > > * support Zp64. > > > > LIU Zhiwei (38): > > target/riscv: implementation-defined constant parameters > > target/riscv: Hoist vector functions > > target/riscv: Fixup saturate subtract function Thanks for the patches and sorry for the long delay. I have applied patch 3 as it fixes a bug. As for the other patches they are on both my review queue and Palmer's review queue. It takes a lot of time to review these large patch series, especially as I haven't been involved with the extension development, so I have to both understand the extension and then review the code. If you would like to help speed things up you could review other patches. That way I will have more time left to review your patches. Alistair > > target/riscv: 16-bit Addition & Subtraction Instructions > > target/riscv: 8-bit Addition & Subtraction Instruction > > target/riscv: SIMD 16-bit Shift Instructions > > target/riscv: SIMD 8-bit Shift Instructions > > target/riscv: SIMD 16-bit Compare Instructions > > target/riscv: SIMD 8-bit Compare Instructions > > target/riscv: SIMD 16-bit Multiply Instructions > > target/riscv: SIMD 8-bit Multiply Instructions > > target/riscv: SIMD 16-bit Miscellaneous Instructions > > target/riscv: SIMD 8-bit Miscellaneous Instructions > > target/riscv: 8-bit Unpacking Instructions > > target/riscv: 16-bit Packing Instructions > > target/riscv: Signed MSW 32x32 Multiply and Add Instructions > > target/riscv: Signed MSW 32x16 Multiply and Add Instructions > > target/riscv: Signed 16-bit Multiply 32-bit Add/Subtract Instructions > > target/riscv: Signed 16-bit Multiply 64-bit Add/Subtract Instructions > > target/riscv: Partial-SIMD Miscellaneous Instructions > > target/riscv: 8-bit Multiply with 32-bit Add Instructions > > target/riscv: 64-bit Add/Subtract Instructions > > target/riscv: 32-bit Multiply 64-bit Add/Subtract Instructions > > target/riscv: Signed 16-bit Multiply with 64-bit Add/Subtract > > Instructions > > target/riscv: Non-SIMD Q15 saturation ALU Instructions > > target/riscv: Non-SIMD Q31 saturation ALU Instructions > > target/riscv: 32-bit Computation Instructions > > target/riscv: Non-SIMD Miscellaneous Instructions > > target/riscv: RV64 Only SIMD 32-bit Add/Subtract Instructions > > target/riscv: RV64 Only SIMD 32-bit Shift Instructions > > target/riscv: RV64 Only SIMD 32-bit Miscellaneous Instructions > > target/riscv: RV64 Only SIMD Q15 saturating Multiply Instructions > > target/riscv: RV64 Only 32-bit Multiply Instructions > > target/riscv: RV64 Only 32-bit Multiply & Add Instructions > > target/riscv: RV64 Only 32-bit Parallel Multiply & Add Instructions > > target/riscv: RV64 Only Non-SIMD 32-bit Shift Instructions > > target/riscv: RV64 Only 32-bit Packing Instructions > > target/riscv: configure and turn on packed extension from command line > > > > target/riscv/cpu.c | 32 + > > target/riscv/cpu.h | 6 + > > target/riscv/helper.h | 332 ++ > > target/riscv/insn32-64.decode | 93 +- > > target/riscv/insn32.decode | 285 ++ > > target/riscv/insn_trans/trans_rvp.c.inc | 1224 +++++++ > > target/riscv/internals.h | 50 + > > target/riscv/meson.build | 1 + > > target/riscv/packed_helper.c | 3862 +++++++++++++++++++++++ > > target/riscv/translate.c | 3 + > > target/riscv/vector_helper.c | 90 +- > > 11 files changed, 5912 insertions(+), 66 deletions(-) > > create mode 100644 target/riscv/insn_trans/trans_rvp.c.inc > > create mode 100644 target/riscv/packed_helper.c > >
On 2021/4/15 下午12:46, Alistair Francis wrote: > On Tue, Apr 13, 2021 at 1:28 PM LIU Zhiwei <zhiwei_liu@c-sky.com> wrote: >> ping +1. >> >> On 2021/2/12 下午11:02, LIU Zhiwei wrote: >>> This patchset implements the packed extension for RISC-V on QEMU. >>> >>> This patchset have passed all my direct Linux user mode cases(RV64) and >>> bare metal cases(RV32) on X86-64 Ubuntu host machine. I will later push >>> these test cases to my repo(https://github.com/romanheros/qemu.git >>> branch:packed-upstream-v1). >>> >>> I have ported packed extension on RISU, but I didn't find a simulator or >>> hardware to compare with. If anyone have one, please let me know. >>> >>> Features: >>> * support specification packed extension v0.9.2(https://github.com/riscv/riscv-p-spec/) >>> * support basic packed extension. >>> * support Zp64. >>> >>> LIU Zhiwei (38): >>> target/riscv: implementation-defined constant parameters >>> target/riscv: Hoist vector functions >>> target/riscv: Fixup saturate subtract function > Thanks for the patches and sorry for the long delay. > > I have applied patch 3 as it fixes a bug. > As for the other patches they are on both my review queue and Palmer's > review queue. It takes a lot of time to review these large patch > series, especially as I haven't been involved with the extension > development, so I have to both understand the extension and then > review the code. > > If you would like to help speed things up you could review other > patches. That way I will have more time left to review your patches. No worries. I fully understand the great efforts needed to review so many patches. Firstly, I will try to review as many as I send. Zhiwei > Alistair > >>> target/riscv: 16-bit Addition & Subtraction Instructions >>> target/riscv: 8-bit Addition & Subtraction Instruction >>> target/riscv: SIMD 16-bit Shift Instructions >>> target/riscv: SIMD 8-bit Shift Instructions >>> target/riscv: SIMD 16-bit Compare Instructions >>> target/riscv: SIMD 8-bit Compare Instructions >>> target/riscv: SIMD 16-bit Multiply Instructions >>> target/riscv: SIMD 8-bit Multiply Instructions >>> target/riscv: SIMD 16-bit Miscellaneous Instructions >>> target/riscv: SIMD 8-bit Miscellaneous Instructions >>> target/riscv: 8-bit Unpacking Instructions >>> target/riscv: 16-bit Packing Instructions >>> target/riscv: Signed MSW 32x32 Multiply and Add Instructions >>> target/riscv: Signed MSW 32x16 Multiply and Add Instructions >>> target/riscv: Signed 16-bit Multiply 32-bit Add/Subtract Instructions >>> target/riscv: Signed 16-bit Multiply 64-bit Add/Subtract Instructions >>> target/riscv: Partial-SIMD Miscellaneous Instructions >>> target/riscv: 8-bit Multiply with 32-bit Add Instructions >>> target/riscv: 64-bit Add/Subtract Instructions >>> target/riscv: 32-bit Multiply 64-bit Add/Subtract Instructions >>> target/riscv: Signed 16-bit Multiply with 64-bit Add/Subtract >>> Instructions >>> target/riscv: Non-SIMD Q15 saturation ALU Instructions >>> target/riscv: Non-SIMD Q31 saturation ALU Instructions >>> target/riscv: 32-bit Computation Instructions >>> target/riscv: Non-SIMD Miscellaneous Instructions >>> target/riscv: RV64 Only SIMD 32-bit Add/Subtract Instructions >>> target/riscv: RV64 Only SIMD 32-bit Shift Instructions >>> target/riscv: RV64 Only SIMD 32-bit Miscellaneous Instructions >>> target/riscv: RV64 Only SIMD Q15 saturating Multiply Instructions >>> target/riscv: RV64 Only 32-bit Multiply Instructions >>> target/riscv: RV64 Only 32-bit Multiply & Add Instructions >>> target/riscv: RV64 Only 32-bit Parallel Multiply & Add Instructions >>> target/riscv: RV64 Only Non-SIMD 32-bit Shift Instructions >>> target/riscv: RV64 Only 32-bit Packing Instructions >>> target/riscv: configure and turn on packed extension from command line >>> >>> target/riscv/cpu.c | 32 + >>> target/riscv/cpu.h | 6 + >>> target/riscv/helper.h | 332 ++ >>> target/riscv/insn32-64.decode | 93 +- >>> target/riscv/insn32.decode | 285 ++ >>> target/riscv/insn_trans/trans_rvp.c.inc | 1224 +++++++ >>> target/riscv/internals.h | 50 + >>> target/riscv/meson.build | 1 + >>> target/riscv/packed_helper.c | 3862 +++++++++++++++++++++++ >>> target/riscv/translate.c | 3 + >>> target/riscv/vector_helper.c | 90 +- >>> 11 files changed, 5912 insertions(+), 66 deletions(-) >>> create mode 100644 target/riscv/insn_trans/trans_rvp.c.inc >>> create mode 100644 target/riscv/packed_helper.c >>>