qemu-riscv
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RFC v4 00/70] support vector extension v1.0


From: Frank Chang
Subject: Re: [RFC v4 00/70] support vector extension v1.0
Date: Thu, 27 Aug 2020 02:12:57 +0800

On Thu, Aug 27, 2020 at 2:03 AM Alistair Francis <alistair23@gmail.com> wrote:
On Wed, Aug 26, 2020 at 10:39 AM Frank Chang <frank.chang@sifive.com> wrote:
>
> On Thu, Aug 27, 2020 at 12:56 AM Alistair Francis <alistair23@gmail.com> wrote:
>>
>> On Tue, Aug 25, 2020 at 1:29 AM Frank Chang <frank.chang@sifive.com> wrote:
>> >
>> > On Mon, Aug 17, 2020 at 4:50 PM <frank.chang@sifive.com> wrote:
>> >>
>> >> From: Frank Chang <frank.chang@sifive.com>
>> >>
>> >> This patchset implements the vector extension v1.0 for RISC-V on QEMU.
>> >>
>> >> This patchset is sent as RFC because RVV v1.0 is still in draft state.
>> >> v2 patchset was sent for RVV v0.9 and bumped to RVV v1.0 since v3 patchset.
>> >>
>> >> The port is available here:
>> >> https://github.com/sifive/qemu/tree/rvv-1.0-upstream-v4
>> >>
>> >> You can change the cpu argument: vext_spec to v1.0 (i.e. vext_spec=v1.0)
>> >> to run with RVV v1.0 instructions.
>> >>
>> >> Note: This patchset depends on two other patchsets listed in Based-on
>> >>       section below so it might not able to be built unless those two
>> >>       patchsets are applied.
>> >>
>> >> Changelog:
>> >>
>> >> v4
>> >>   * remove explicit float flmul variable in DisasContext.
>> >>   * replace floating-point calculations with shift operations to
>> >>     improve performance.
>> >>   * relax RV_VLEN_MAX to 512-bits.
>> >>
>> >> v3
>> >>   * apply nan-box helpers from Richard Henderson.
>> >>   * remove fp16 api changes as they are sent independently in another
>> >>     pathcset by Chih-Min Chao.
>> >>   * remove all tail elements clear functions as tail elements can
>> >>     retain unchanged for either VTA set to undisturbed or agnostic.
>> >>   * add fp16 nan-box check generator function.
>> >>   * add floating-point rounding mode enum.
>> >>   * replace flmul arithmetic with shifts to avoid floating-point
>> >>     conversions.
>> >>   * add Zvqmac extension.
>> >>   * replace gdbstub vector register xml files with dynamic generator.
>> >>   * bumped to RVV v1.0.
>> >>   * RVV v1.0 related changes:
>> >>     * add vl<nf>re<eew>.v and vs<nf>r.v vector whole register
>> >>       load/store instructions
>> >>     * add vrgatherei16 instruction.
>> >>     * rearranged bits in vtype to make vlmul bits into a contiguous
>> >>       field.
>> >>
>> >> v2
>> >>   * drop v0.7.1 support.
>> >>   * replace invisible return check macros with functions.
>> >>   * move mark_vs_dirty() to translators.
>> >>   * add SSTATUS_VS flag for s-mode.
>> >>   * nan-box scalar fp register for floating-point operations.
>> >>   * add gdbstub files for vector registers to allow system-mode
>> >>     debugging with GDB.
>> >>
>> >> Based-on: <20200724002807.441147-1-richard.henderson@linaro.org/>
>> >> Based-on: <1596102747-20226-1-git-send-email-chihmin.chao@sifive.com/>
>> >>
>> >> Frank Chang (62):
>> >>   target/riscv: drop vector 0.7.1 and add 1.0 support
>> >>   target/riscv: Use FIELD_EX32() to extract wd field
>> >>   target/riscv: rvv-1.0: introduce writable misa.v field
>> >>   target/riscv: rvv-1.0: remove rvv related codes from fcsr registers
>> >>   target/riscv: rvv-1.0: check MSTATUS_VS when accessing vector csr
>> >>     registers
>> >>   target/riscv: rvv-1.0: remove MLEN calculations
>> >>   target/riscv: rvv-1.0: add fractional LMUL
>> >>   target/riscv: rvv-1.0: add VMA and VTA
>> >>   target/riscv: rvv-1.0: update check functions
>> >>   target/riscv: introduce more imm value modes in translator functions
>> >>   target/riscv: rvv:1.0: add translation-time nan-box helper function
>> >>   target/riscv: rvv-1.0: configure instructions
>> >>   target/riscv: rvv-1.0: stride load and store instructions
>> >>   target/riscv: rvv-1.0: index load and store instructions
>> >>   target/riscv: rvv-1.0: fix address index overflow bug of indexed
>> >>     load/store insns
>> >>   target/riscv: rvv-1.0: fault-only-first unit stride load
>> >>   target/riscv: rvv-1.0: amo operations
>> >>   target/riscv: rvv-1.0: load/store whole register instructions
>> >>   target/riscv: rvv-1.0: update vext_max_elems() for load/store insns
>> >>   target/riscv: rvv-1.0: take fractional LMUL into vector max elements
>> >>     calculation
>> >>   target/riscv: rvv-1.0: floating-point square-root instruction
>> >>   target/riscv: rvv-1.0: floating-point classify instructions
>> >>   target/riscv: rvv-1.0: mask population count instruction
>> >>   target/riscv: rvv-1.0: find-first-set mask bit instruction
>> >>   target/riscv: rvv-1.0: set-X-first mask bit instructions
>> >>   target/riscv: rvv-1.0: iota instruction
>> >>   target/riscv: rvv-1.0: element index instruction
>> >>   target/riscv: rvv-1.0: allow load element with sign-extended
>> >>   target/riscv: rvv-1.0: register gather instructions
>> >>   target/riscv: rvv-1.0: integer scalar move instructions
>> >>   target/riscv: rvv-1.0: floating-point move instruction
>> >>   target/riscv: rvv-1.0: floating-point scalar move instructions
>> >>   target/riscv: rvv-1.0: whole register move instructions
>> >>   target/riscv: rvv-1.0: integer extension instructions
>> >>   target/riscv: rvv-1.0: single-width averaging add and subtract
>> >>     instructions
>> >>   target/riscv: rvv-1.0: single-width bit shift instructions
>> >>   target/riscv: rvv-1.0: integer add-with-carry/subtract-with-borrow
>> >>   target/riscv: rvv-1.0: narrowing integer right shift instructions
>> >>   target/riscv: rvv-1.0: widening integer multiply-add instructions
>> >>   target/riscv: rvv-1.0: add Zvqmac extension
>> >>   target/riscv: rvv-1.0: quad-widening integer multiply-add instructions
>> >>   target/riscv: rvv-1.0: single-width saturating add and subtract
>> >>     instructions
>> >>   target/riscv: rvv-1.0: integer comparison instructions
>> >>   target/riscv: use softfloat lib float16 comparison functions
>> >>   target/riscv: rvv-1.0: floating-point compare instructions
>> >>   target/riscv: rvv-1.0: mask-register logical instructions
>> >>   target/riscv: rvv-1.0: slide instructions
>> >>   target/riscv: rvv-1.0: floating-point slide instructions
>> >>   target/riscv: rvv-1.0: narrowing fixed-point clip instructions
>> >>   target/riscv: rvv-1.0: single-width floating-point reduction
>> >>   target/riscv: rvv-1.0: widening floating-point reduction instructions
>> >>   target/riscv: rvv-1.0: single-width scaling shift instructions
>> >>   target/riscv: rvv-1.0: remove widening saturating scaled multiply-add
>> >>   target/riscv: rvv-1.0: remove vmford.vv and vmford.vf
>> >>   target/riscv: rvv-1.0: remove integer extract instruction
>> >>   target/riscv: rvv-1.0: floating-point min/max instructions
>> >>   target/riscv: introduce floating-point rounding mode enum
>> >>   target/riscv: rvv-1.0: floating-point/integer type-convert
>> >>     instructions
>> >>   target/riscv: rvv-1.0: widening floating-point/integer type-convert
>> >>   target/riscv: add "set round to odd" rounding mode helper function
>> >>   target/riscv: rvv-1.0: narrowing floating-point/integer type-convert
>> >>   target/riscv: rvv-1.0: relax RV_VLEN_MAX to 512-bits
>> >>
>> >> Greentime Hu (2):
>> >>   target/riscv: rvv-1.0: add vlenb register
>> >>   target/riscv: gdb: support vector registers for rv32
>> >>
>> >> Hsiangkai Wang (2):
>> >>   target/riscv: gdb: modify gdb csr xml file to align with csr register
>> >>     map
>> >>   target/riscv: gdb: support vector registers for rv64
>> >>
>> >> LIU Zhiwei (4):
>> >>   target/riscv: rvv-1.0: add mstatus VS field
>> >>   target/riscv: rvv-1.0: add sstatus VS field
>> >>   target/riscv: rvv-1.0: add translation-time vector context status
>> >>   target/riscv: rvv-1.0: add vcsr register
>> >>
>> >>  gdb-xml/riscv-32bit-csr.xml             |   18 +-
>> >>  gdb-xml/riscv-64bit-csr.xml             |   18 +-
>> >>  target/riscv/cpu.c                      |   12 +-
>> >>  target/riscv/cpu.h                      |   97 +-
>> >>  target/riscv/cpu_bits.h                 |   10 +
>> >>  target/riscv/cpu_helper.c               |   16 +-
>> >>  target/riscv/csr.c                      |   73 +-
>> >>  target/riscv/fpu_helper.c               |   17 +-
>> >>  target/riscv/gdbstub.c                  |  126 +-
>> >>  target/riscv/helper.h                   |  523 ++--
>> >>  target/riscv/insn32-64.decode           |   18 +-
>> >>  target/riscv/insn32.decode              |  295 +-
>> >>  target/riscv/insn_trans/trans_rvv.inc.c | 2366 ++++++++++------
>> >>  target/riscv/internals.h                |   19 +-
>> >>  target/riscv/translate.c                |   68 +-
>> >>  target/riscv/vector_helper.c            | 3269 +++++++++++------------
>> >>  16 files changed, 4051 insertions(+), 2894 deletions(-)
>> >>
>> >> --
>> >> 2.17.1
>> >>
>> >
>> > ping~
>>
>> I wasn't really following too closely, but didn't Richard give comments?
>>
>> Alistair
>
>
> Yeah, they were given in v3 patchset and I've made the changes
> based on Richard's comments in this v4 patchset.

Ah ok. I missed that while I was on holidays.

Did you want to wait until the v1.0 spec is released or have the draft
extensions merged?

Alistair

>
> Frank Chang

I'm okay to wait until v1.0 spec. is released as I'm just sending
RFC patchset for now. As far as I know there are still couple of
instructions not implemented for RVV v1.0 yet (e.g. vfrsqrt7.v
and vfrece7.v). Not sure what else is going to be changed before
v1.0 spec. is ratified.

However, it would still be nice if someone can take a look of current
patches so it might speed up the process to get these patches merged
into mainline once v1.0 spec. is released.

Thanks,
Frank Chang

reply via email to

[Prev in Thread] Current Thread [Next in Thread]