From: Alistair Francis <alistair23@gmail.com>
To: LIU Zhiwei <zhiwei_liu@c-sky.com>
Cc: guoren@linux.alibaba.com,
"open list:RISC-V" <qemu-riscv@nongnu.org>,
Richard Henderson <richard.henderson@linaro.org>,
"qemu-devel@nongnu.org Developers" <qemu-devel@nongnu.org>,
wxy194768@alibaba-inc.com,
Chih-Min Chao <chihmin.chao@sifive.com>,
wenmeng_zhang@c-sky.com, Palmer Dabbelt <palmer@dabbelt.com>
Subject: Re: [PATCH v5 05/60] target/riscv: add vector stride load and store instructions
Date: Fri, 13 Mar 2020 16:38:54 -0700 [thread overview]
Message-ID: <CAKmqyKNCdFp--okyLK_bjkQwpFTQX62b_rZTnOGJhReqCvzoQA@mail.gmail.com> (raw)
In-Reply-To: <4afbf97a-ea6d-2ed3-7b5a-c58e59172c9d@c-sky.com>
On Fri, Mar 13, 2020 at 3:17 PM LIU Zhiwei <zhiwei_liu@c-sky.com> wrote:
>
>
>
> On 2020/3/14 6:05, Alistair Francis wrote:
> > On Fri, Mar 13, 2020 at 2:32 PM LIU Zhiwei <zhiwei_liu@c-sky.com> wrote:
> >>
> >>
> >> On 2020/3/14 4:38, Alistair Francis wrote:
> >>> On Thu, Mar 12, 2020 at 8:09 AM LIU Zhiwei <zhiwei_liu@c-sky.com> wrote:
> >>>> Vector strided operations access the first memory element at the base address,
> >>>> and then access subsequent elements at address increments given by the byte
> >>>> offset contained in the x register specified by rs2.
> >>>>
> >>>> Vector unit-stride operations access elements stored contiguously in memory
> >>>> starting from the base effective address. It can been seen as a special
> >>>> case of strided operations.
> >>>>
> >>>> Signed-off-by: LIU Zhiwei <zhiwei_liu@c-sky.com>
> >>>> ---
> >>>> target/riscv/cpu.h | 6 +
> >>>> target/riscv/helper.h | 105 ++++++
> >>>> target/riscv/insn32.decode | 32 ++
> >>>> target/riscv/insn_trans/trans_rvv.inc.c | 340 ++++++++++++++++++++
> >>>> target/riscv/translate.c | 7 +
> >>>> target/riscv/vector_helper.c | 406 ++++++++++++++++++++++++
> >>>> 6 files changed, 896 insertions(+)
> >>>>
> >>>> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> >>>> index 505d1a8515..b6ebb9b0eb 100644
> >>>> --- a/target/riscv/cpu.h
> >>>> +++ b/target/riscv/cpu.h
> >>>> @@ -369,6 +369,12 @@ typedef CPURISCVState CPUArchState;
> >>>> typedef RISCVCPU ArchCPU;
> >>>> #include "exec/cpu-all.h"
> >>>>
> >>>> +/* share data between vector helpers and decode code */
> >>>> +FIELD(VDATA, MLEN, 0, 8)
> >>>> +FIELD(VDATA, VM, 8, 1)
> >>>> +FIELD(VDATA, LMUL, 9, 2)
> >>>> +FIELD(VDATA, NF, 11, 4)
> >>>> +
> >>>> FIELD(TB_FLAGS, VL_EQ_VLMAX, 2, 1)
> >>>> FIELD(TB_FLAGS, LMUL, 3, 2)
> >>>> FIELD(TB_FLAGS, SEW, 5, 3)
> >>>> diff --git a/target/riscv/helper.h b/target/riscv/helper.h
> >>>> index 3c28c7e407..87dfa90609 100644
> >>>> --- a/target/riscv/helper.h
> >>>> +++ b/target/riscv/helper.h
> >>>> @@ -78,3 +78,108 @@ DEF_HELPER_1(tlb_flush, void, env)
> >>>> #endif
> >>>> /* Vector functions */
> >>>> DEF_HELPER_3(vsetvl, tl, env, tl, tl)
> >>>> +DEF_HELPER_5(vlb_v_b, void, ptr, ptr, tl, env, i32)
> >>>> +DEF_HELPER_5(vlb_v_b_mask, void, ptr, ptr, tl, env, i32)
> >>> Do you mind explaining why we have *_mask versions? I'm struggling to
> >>> understand this.
> >> When an instruction with a mask, it will only operate the active
> >> elements in vector.
> >> Whether an element is active or inactive is predicated by a mask
> >> register v0.
> >>
> >> Without mask, it will operate every element in vector in the body.
> > Doesn't the mask always apply though? Why do we need an extra helper?
> Yes, mask is always applied.
>
> As you can see, an extra helper is very special for unit stride mode.
> Other
> instructions do not have the extra helpers.
>
> That's because a more efficient implementation is possible for unit stride
> load/store with vm==1(always unmasked).
>
> It will operate a contiguous memory block, so I can probe the memory access
> and clean the tail elements more efficient.
Ah ok. I think I get what you are saying. I think this is all ok then.
I'll review the next version (after you have split it).
Alistair
>
> Zhiwei
next prev parent reply other threads:[~2020-03-13 23:40 UTC|newest]
Thread overview: 168+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-12 14:58 [PATCH v5 00/60] target/riscv: support vector extension v0.7.1 LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 01/60] target/riscv: add vector extension field in CPURISCVState LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 02/60] target/riscv: implementation-defined constant parameters LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 03/60] target/riscv: support vector extension csr LIU Zhiwei
2020-03-12 20:54 ` Alistair Francis
2020-03-14 1:11 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 04/60] target/riscv: add vector configure instruction LIU Zhiwei
2020-03-12 21:23 ` Alistair Francis
2020-03-12 22:00 ` LIU Zhiwei
2020-03-12 22:07 ` Alistair Francis
2020-03-14 1:14 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 05/60] target/riscv: add vector stride load and store instructions LIU Zhiwei
2020-03-13 20:38 ` Alistair Francis
2020-03-13 21:32 ` LIU Zhiwei
2020-03-13 22:05 ` Alistair Francis
2020-03-13 22:17 ` LIU Zhiwei
2020-03-13 23:38 ` Alistair Francis [this message]
2020-03-14 1:26 ` Richard Henderson
2020-03-14 1:49 ` LIU Zhiwei
2020-03-14 1:36 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 06/60] target/riscv: add vector index " LIU Zhiwei
2020-03-13 21:21 ` Alistair Francis
2020-03-14 1:49 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 07/60] target/riscv: add fault-only-first unit stride load LIU Zhiwei
2020-03-13 22:24 ` Alistair Francis
2020-03-13 22:41 ` LIU Zhiwei
2020-03-14 1:50 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 08/60] target/riscv: add vector amo operations LIU Zhiwei
2020-03-14 0:02 ` Alistair Francis
2020-03-14 0:36 ` LIU Zhiwei
2020-03-14 4:28 ` Richard Henderson
2020-03-14 5:07 ` LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 09/60] target/riscv: vector single-width integer add and subtract LIU Zhiwei
2020-03-14 5:25 ` Richard Henderson
2020-03-14 8:11 ` LIU Zhiwei
2020-03-23 8:10 ` LIU Zhiwei
2020-03-23 17:46 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 10/60] target/riscv: vector widening " LIU Zhiwei
2020-03-14 5:32 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 11/60] target/riscv: vector integer add-with-carry / subtract-with-borrow instructions LIU Zhiwei
2020-03-14 5:58 ` Richard Henderson
2020-03-14 6:08 ` LIU Zhiwei
2020-03-14 6:16 ` Richard Henderson
2020-03-14 6:32 ` LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 12/60] target/riscv: vector bitwise logical instructions LIU Zhiwei
2020-03-14 6:00 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 13/60] target/riscv: vector single-width bit shift instructions LIU Zhiwei
2020-03-14 6:07 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 14/60] target/riscv: vector narrowing integer right " LIU Zhiwei
2020-03-14 6:10 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 15/60] target/riscv: vector integer comparison instructions LIU Zhiwei
2020-03-14 6:33 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 16/60] target/riscv: vector integer min/max instructions LIU Zhiwei
2020-03-14 6:40 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 17/60] target/riscv: vector single-width integer multiply instructions LIU Zhiwei
2020-03-14 6:52 ` Richard Henderson
2020-03-14 7:02 ` LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 18/60] target/riscv: vector integer divide instructions LIU Zhiwei
2020-03-14 6:58 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 19/60] target/riscv: vector widening integer multiply instructions LIU Zhiwei
2020-03-14 7:06 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 20/60] target/riscv: vector single-width integer multiply-add instructions LIU Zhiwei
2020-03-14 7:10 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 21/60] target/riscv: vector widening " LIU Zhiwei
2020-03-14 7:13 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 22/60] target/riscv: vector integer merge and move instructions LIU Zhiwei
2020-03-14 7:27 ` Richard Henderson
2020-03-16 2:57 ` LIU Zhiwei
2020-03-16 5:32 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 23/60] target/riscv: vector single-width saturating add and subtract LIU Zhiwei
2020-03-14 7:52 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 24/60] target/riscv: vector single-width averaging " LIU Zhiwei
2020-03-14 8:14 ` Richard Henderson
2020-03-14 8:25 ` Richard Henderson
2020-03-14 23:12 ` LIU Zhiwei
2020-03-15 1:00 ` Richard Henderson
2020-03-15 23:23 ` LIU Zhiwei
2020-03-15 23:27 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 25/60] target/riscv: vector single-width fractional multiply with rounding and saturation LIU Zhiwei
2020-03-14 8:27 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 26/60] target/riscv: vector widening saturating scaled multiply-add LIU Zhiwei
2020-03-14 8:32 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 27/60] target/riscv: vector single-width scaling shift instructions LIU Zhiwei
2020-03-14 8:34 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 28/60] target/riscv: vector narrowing fixed-point clip instructions LIU Zhiwei
2020-03-14 8:36 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 29/60] target/riscv: vector single-width floating-point add/subtract instructions LIU Zhiwei
2020-03-14 8:40 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 30/60] target/riscv: vector widening " LIU Zhiwei
2020-03-14 8:43 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 31/60] target/riscv: vector single-width floating-point multiply/divide instructions LIU Zhiwei
2020-03-14 8:43 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 32/60] target/riscv: vector widening floating-point multiply LIU Zhiwei
2020-03-14 8:46 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 33/60] target/riscv: vector single-width floating-point fused multiply-add instructions LIU Zhiwei
2020-03-14 8:49 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 34/60] target/riscv: vector widening " LIU Zhiwei
2020-03-14 8:50 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 35/60] target/riscv: vector floating-point square-root instruction LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 36/60] target/riscv: vector floating-point min/max instructions LIU Zhiwei
2020-03-14 8:52 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 37/60] target/riscv: vector floating-point sign-injection instructions LIU Zhiwei
2020-03-14 8:57 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 38/60] target/riscv: vector floating-point compare instructions LIU Zhiwei
2020-03-14 9:08 ` Richard Henderson
2020-03-14 9:11 ` LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 39/60] target/riscv: vector floating-point classify instructions LIU Zhiwei
2020-03-14 9:10 ` Richard Henderson
2020-03-14 9:15 ` LIU Zhiwei
2020-03-14 22:06 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 40/60] target/riscv: vector floating-point merge instructions LIU Zhiwei
2020-03-14 22:47 ` Richard Henderson
2020-03-16 3:41 ` LIU Zhiwei
2020-03-16 5:37 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 41/60] target/riscv: vector floating-point/integer type-convert instructions LIU Zhiwei
2020-03-14 22:50 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 42/60] target/riscv: widening " LIU Zhiwei
2020-03-14 23:03 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 43/60] target/riscv: narrowing " LIU Zhiwei
2020-03-14 23:08 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 44/60] target/riscv: vector single-width integer reduction instructions LIU Zhiwei
2020-03-14 23:29 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 45/60] target/riscv: vector wideing " LIU Zhiwei
2020-03-14 23:34 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 46/60] target/riscv: vector single-width floating-point " LIU Zhiwei
2020-03-14 23:48 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 47/60] target/riscv: vector widening " LIU Zhiwei
2020-03-14 23:49 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 48/60] target/riscv: vector mask-register logical instructions LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 49/60] target/riscv: vector mask population count vmpopc LIU Zhiwei
2020-03-15 1:20 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 50/60] target/riscv: vmfirst find-first-set mask bit LIU Zhiwei
2020-03-15 1:36 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 51/60] target/riscv: set-X-first " LIU Zhiwei
2020-03-12 14:58 ` [PATCH v5 52/60] target/riscv: vector iota instruction LIU Zhiwei
2020-03-15 1:50 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 53/60] target/riscv: vector element index instruction LIU Zhiwei
2020-03-15 1:54 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 54/60] target/riscv: integer extract instruction LIU Zhiwei
2020-03-15 2:53 ` Richard Henderson
2020-03-15 5:15 ` LIU Zhiwei
2020-03-15 5:21 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 55/60] target/riscv: integer scalar move instruction LIU Zhiwei
2020-03-15 3:54 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 56/60] target/riscv: floating-point scalar move instructions LIU Zhiwei
2020-03-15 4:39 ` Richard Henderson
2020-03-15 6:13 ` LIU Zhiwei
2020-03-15 6:48 ` Richard Henderson
2020-03-17 6:01 ` LIU Zhiwei
2020-03-17 15:11 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 57/60] target/riscv: vector slide instructions LIU Zhiwei
2020-03-15 5:16 ` Richard Henderson
2020-03-15 6:49 ` LIU Zhiwei
2020-03-15 6:56 ` Richard Henderson
2020-03-16 8:04 ` LIU Zhiwei
2020-03-16 17:42 ` Richard Henderson
2020-03-24 10:51 ` LIU Zhiwei
2020-03-24 14:52 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 58/60] target/riscv: vector register gather instruction LIU Zhiwei
2020-03-15 5:44 ` Richard Henderson
2020-03-12 14:58 ` [PATCH v5 59/60] target/riscv: vector compress instruction LIU Zhiwei
2020-03-12 14:59 ` [PATCH v5 60/60] target/riscv: configure and turn on vector extension from command line LIU Zhiwei
2020-03-13 21:41 ` Alistair Francis
2020-03-13 21:52 ` LIU Zhiwei
2020-03-13 0:41 ` [PATCH v5 00/60] target/riscv: support vector extension v0.7.1 no-reply
2020-03-15 7:00 ` [PATCH v5 35/60] target/riscv: vector floating-point square-root instruction Richard Henderson
2020-03-15 7:26 ` [PATCH v5 51/60] target/riscv: set-X-first mask bit Richard Henderson
2020-03-15 7:34 ` [PATCH v5 59/60] target/riscv: vector compress instruction Richard Henderson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAKmqyKNCdFp--okyLK_bjkQwpFTQX62b_rZTnOGJhReqCvzoQA@mail.gmail.com \
--to=alistair23@gmail.com \
--cc=chihmin.chao@sifive.com \
--cc=guoren@linux.alibaba.com \
--cc=palmer@dabbelt.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-riscv@nongnu.org \
--cc=richard.henderson@linaro.org \
--cc=wenmeng_zhang@c-sky.com \
--cc=wxy194768@alibaba-inc.com \
--cc=zhiwei_liu@c-sky.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).