public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "crazylht at gmail dot com" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug target/99881] New: Regression compare -O2 -ftree-vectorize with -O2 on SKX/CLX Date: Fri, 02 Apr 2021 03:49:48 +0000 [thread overview] Message-ID: <bug-99881-4@http.gcc.gnu.org/bugzilla/> (raw) https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99881 Bug ID: 99881 Summary: Regression compare -O2 -ftree-vectorize with -O2 on SKX/CLX Product: gcc Version: 11.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: target Assignee: unassigned at gcc dot gnu.org Reporter: crazylht at gmail dot com Target Milestone: --- testcase is extracted from 557.xz_r void foo (int* __restrict a, int n, int c) { a[0] = n; a[1] = c; } gcc -O2 -ftree-vectorize -fvect-cost-model=very-cheap foo(int*, int, int): movd xmm0, esi movd xmm1, edx punpckldq xmm0, xmm1 movq QWORD PTR [rdi], xmm0 ret without vectorization foo(int*, int, int): mov DWORD PTR [rdi], esi mov DWORD PTR [rdi+4], edx ret cost model: scalar: 2 times scalar_store costs 24, vector: 1 times unaligned_store costs 12, vec_contruct 8 I know that the current strategy of the cost model is to enable vectorization as much as possible, but for the case above, it hurts performance. Because the throughput of punpckldq is 1 on SKX/CLX, which becomes a bottleneck (znver2 is ok). with -march=SKX, the second vmovd and unpck will be replaced by vpinsr, and it regression more since vpinsr has throught 2 on CLX/SKX. So i'm thinking to add extra cost for 2-element vec_construct to prevent the above vectorization, at the same time, try not to affect other vectorization situations.
next reply other threads:[~2021-04-02 3:49 UTC|newest] Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-04-02 3:49 crazylht at gmail dot com [this message] 2021-04-02 14:29 ` [Bug target/99881] " hjl.tools at gmail dot com 2021-04-02 19:34 ` hjl.tools at gmail dot com 2021-04-06 7:48 ` rguenth at gcc dot gnu.org 2021-04-06 10:06 ` crazylht at gmail dot com 2021-04-06 11:44 ` rguenth at gcc dot gnu.org 2021-07-28 2:48 ` cvs-commit at gcc dot gnu.org 2021-07-28 2:49 ` crazylht at gmail dot com 2021-07-28 22:47 ` jakub at gcc dot gnu.org 2021-07-29 1:09 ` crazylht at gmail dot com 2021-07-29 2:18 ` cvs-commit at gcc dot gnu.org 2021-08-19 2:32 ` crazylht at gmail dot com 2022-02-22 7:59 ` cvs-commit at gcc dot gnu.org 2022-02-22 8:00 ` rguenth at gcc dot gnu.org
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=bug-99881-4@http.gcc.gnu.org/bugzilla/ \ --to=gcc-bugzilla@gcc.gnu.org \ --cc=gcc-bugs@gcc.gnu.org \ /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: linkBe 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).