public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Michael Collison <collison@rivosinc.com>
To: "Kito Cheng" <kito.cheng@gmail.com>,
	"Jeff Law" <jeffreyalaw@gmail.com>,
	"Palmer Dabbelt" <palmer@dabbelt.com>, 钟居哲 <juzhe.zhong@rivai.ai>
Cc: gcc-patches@gcc.gnu.org
Subject: Re: [PATCH v6 0/9] RISC-V: autovec: Add autovec support
Date: Fri, 5 May 2023 13:12:40 -0400	[thread overview]
Message-ID: <6243bc21-e983-15ec-cd43-f47915e29189@rivosinc.com> (raw)
In-Reply-To: <CA+yXCZAe3-DyE09stGc-iH3fd489Rn76Pww8sKkr69+8ie8t1A@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 7495 bytes --]

Because everyone was commenting that we needed vector load/store support 
(including Juzhe). Juzhe specifically pointed me to his patch for the 
load/store patterns in his review of my code. Would you like me to 
remove the patterns?

On 5/5/23 12:34, Kito Cheng wrote:
> Errr, why you just mixed in JuZhe’s patch set into this patch set?
>
> Michael Collison <collison@rivosinc.com>於 2023年5月5日 週五,23:47寫道:
>
>     This series of patches adds foundational support for RISC-V
>     auto-vectorization support. These patches are based on the current
>     upstream rvv vector intrinsic support and is not a new
>     implementation. Most of the implementation consists of adding the
>     new vector cost model, the autovectorization patterns themselves
>     and target hooks. This implementation only provides support for
>     integer addition and subtraction as a proof of concept. This patch
>     set should not be construed to be feature complete. Based on
>     conversations with the community these patches are intended to lay
>     the groundwork for feature completion and collaboration within the
>     RISC-V community.
>
>     These patches are largely based off the work of Juzhe Zhong
>     (juzhe.zhong@rivai.ai<mailto:juzhe.zhong@rivai.ai>) of RiVAI. More
>     specifically the rvv-next branch at:
>     https://github.com/riscv-collab/riscv-gcc.git
>     <https://github.com/riscv-collab/riscv-gcc.git>is the foundation
>     of this patch set.
>
>     As discussed on this list, if these patches are approved they will
>     be merged into a "auto-vectorization" branch once gcc-13 branches
>     for release. There are two known issues related to crashes (assert
>     failures) associated with tree vectorization; one of which I have
>     sent a patch for and have received feedback.
>
>     Changes in v6:
>     - Incorporated upstream comments, added target hook for
>     TARGET_VECTORIZE_SUPPORT_VECTOR_MISALIGNMENT
>
>     Changes in v5:
>
>     - Incorporated upstream comments large to delete unnecessary code
>
>     Changes in v4:
>
>     - Added support for binary integer operations and test cases
>     - Fixed bug to support 8-bit integer vectorization
>     - Fixed several assert errors related to non-multiple of two
>     vector modes
>
>     Changes in v3:
>
>     - Removed the cost model and cost hooks based on feedback from
>     Richard Biener
>     - Used RVV_VUNDEF macro to fix failing patterns
>
>     Changes in v2
>
>     - Updated ChangeLog entry to include RiVAI contributions
>     - Fixed ChangeLog email formatting
>     - Fixed gnu formatting issues in the code
>
>     Kevin Lee (1):
>       RISC-V:autovec: This patch supports 8 bit auto-vectorization in
>     riscv.
>
>     Michael Collison (8):
>       RISC-V: Add new predicates and function prototypes
>       RISC-V: autovec: Export policy functions to global scope
>       RISC-V:autovec: Add auto-vectorization support functions
>       RISC-V:autovec: Add target vectorization hooks
>       RISC-V:autovec: Add autovectorization patterns for binary integer &
>         len_load/store
>       RISC-V:autovec: Add autovectorization tests for add & sub
>       vect: Verify that GET_MODE_NUNITS is a multiple of 2.
>       RISC-V:autovec: Add autovectorization tests for binary integer
>
>      gcc/config/riscv/riscv-opts.h                 |  10 ++
>      gcc/config/riscv/riscv-protos.h               |   9 ++
>      gcc/config/riscv/riscv-v.cc                   |  91 ++++++++++++
>      gcc/config/riscv/riscv-vector-builtins.cc     |   4 +-
>      gcc/config/riscv/riscv-vector-builtins.h      |   3 +
>      gcc/config/riscv/riscv.cc                     | 130
>     ++++++++++++++++++
>      gcc/config/riscv/riscv.md                     |   1 +
>      gcc/config/riscv/vector-auto.md               |  74 ++++++++++
>      gcc/config/riscv/vector.md                    |   4 +-
>      .../riscv/rvv/autovec/loop-add-rv32.c         |  25 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-add.c   |  25 ++++
>      .../riscv/rvv/autovec/loop-and-rv32.c         |  25 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-and.c   |  25 ++++
>      .../riscv/rvv/autovec/loop-div-rv32.c         |  27 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-div.c   |  27 ++++
>      .../riscv/rvv/autovec/loop-max-rv32.c         |  26 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-max.c   |  26 ++++
>      .../riscv/rvv/autovec/loop-min-rv32.c         |  26 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-min.c   |  26 ++++
>      .../riscv/rvv/autovec/loop-mod-rv32.c         |  27 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-mod.c   |  27 ++++
>      .../riscv/rvv/autovec/loop-mul-rv32.c         |  25 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-mul.c   |  25 ++++
>      .../riscv/rvv/autovec/loop-or-rv32.c          |  25 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-or.c    |  25 ++++
>      .../riscv/rvv/autovec/loop-sub-rv32.c         |  25 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-sub.c   |  25 ++++
>      .../riscv/rvv/autovec/loop-xor-rv32.c         |  25 ++++
>      .../gcc.target/riscv/rvv/autovec/loop-xor.c   |  25 ++++
>      gcc/testsuite/gcc.target/riscv/rvv/rvv.exp    |   4 +
>      gcc/tree-vect-slp.cc                          |   7 +-
>      31 files changed, 843 insertions(+), 6 deletions(-)
>      create mode 100644 gcc/config/riscv/vector-auto.md
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-add-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-add.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-and-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-and.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-div-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-div.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-max-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-max.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-min-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-min.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-mod-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-mod.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-mul-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-mul.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-or-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-or.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-sub-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-sub.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-xor-rv32.c
>      create mode 100644
>     gcc/testsuite/gcc.target/riscv/rvv/autovec/loop-xor.c
>
>     -- 
>     2.34.1
>

      reply	other threads:[~2023-05-05 17:12 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-05 15:45 Michael Collison
2023-05-05 15:45 ` [PATCH v6 1/9] RISC-V: autovec: Add new predicates and function prototypes Michael Collison
2023-05-06 17:01   ` Jeff Law
2023-05-05 15:46 ` [PATCH v6 2/9] RISC-V: autovec: Export policy functions to global scope Michael Collison
2023-05-06 17:20   ` Jeff Law
2023-05-05 15:46 ` [PATCH v6 3/9] RISC-V:autovec: Add auto-vectorization support functions Michael Collison
2023-05-06 17:33   ` Jeff Law
2023-05-05 15:46 ` [PATCH v6 4/9] RISC-V:autovec: Add target vectorization hooks Michael Collison
2023-05-06 18:28   ` Jeff Law
2023-05-05 15:46 ` [PATCH v6 5/9] RISC-V:autovec: Add autovectorization patterns for binary integer & len_load/store Michael Collison
2023-05-05 15:46 ` [PATCH v6 6/9] RISC-V:autovec: Add autovectorization tests for add & sub Michael Collison
2023-05-05 15:46 ` [PATCH v6 7/9] RISC-V: autovec: Verify that GET_MODE_NUNITS is a multiple of 2 Michael Collison
2023-05-06 18:39   ` Jeff Law
2023-05-05 15:46 ` [PATCH v6 8/9] RISC-V:autovec: Add autovectorization tests for binary integer Michael Collison
2023-05-05 15:46 ` [PATCH v6 9/9] RISC-V:autovec: This patch supports 8 bit auto-vectorization in riscv Michael Collison
2023-05-05 16:34 ` [PATCH v6 0/9] RISC-V: autovec: Add autovec support Kito Cheng
2023-05-05 17:12   ` Michael Collison [this message]

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=6243bc21-e983-15ec-cd43-f47915e29189@rivosinc.com \
    --to=collison@rivosinc.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=jeffreyalaw@gmail.com \
    --cc=juzhe.zhong@rivai.ai \
    --cc=kito.cheng@gmail.com \
    --cc=palmer@dabbelt.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 read-only IMAP folder(s) and NNTP newsgroup(s).