public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "amonakov at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug target/105504] New: Fails to break dependency for vcvtss2sd xmm, xmm, mem Date: Fri, 06 May 2022 09:12:40 +0000 [thread overview] Message-ID: <bug-105504-4@http.gcc.gnu.org/bugzilla/> (raw) https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105504 Bug ID: 105504 Summary: Fails to break dependency for vcvtss2sd xmm, xmm, mem Product: gcc Version: unknown Status: UNCONFIRMED Keywords: missed-optimization Severity: normal Priority: P3 Component: target Assignee: unassigned at gcc dot gnu.org Reporter: amonakov at gcc dot gnu.org Target Milestone: --- Created attachment 52933 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52933&action=edit testcase Hit by core-math team at https://gcc.gnu.org/pipermail/gcc-help/2022-May/141480.html Compile the attached testcase with -O2 -march=haswell (other AVX-capable Intel families except Alderlake are affected too) and observe that the big basic block begins with .L6: vcvtss2sd xmm1, xmm1, DWORD PTR [rsp-4] This creates a false dependency on the previous assignment into xmm1, resulting in wildly varying (and suboptimal) throughput figures depending on how long the CPU stalls waiting for the previous assignment to complete. GCC has code to emit such instructions in a manner that avoids false dependencies (see e.g. PR89071), but here it doesn't seem to work. Also there's a potentially related issue that GCC copies the initial xmm0 value to eax via stack in the beginning of the function: cr_exp10f: vmovss DWORD PTR [rsp-4], xmm0 mov eax, DWORD PTR [rsp-4] This seems wrong since xmm-reg moves on Haswell are 1 cycle afaict.
next reply other threads:[~2022-05-06 9:12 UTC|newest] Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-05-06 9:12 amonakov at gcc dot gnu.org [this message] 2022-05-07 3:02 ` [Bug target/105504] " crazylht at gmail dot com 2022-05-07 3:16 ` crazylht at gmail dot com 2022-05-07 3:36 ` crazylht at gmail dot com 2022-05-07 4:04 ` crazylht at gmail dot com 2022-05-07 8:28 ` amonakov at gcc dot gnu.org 2022-06-08 3:24 ` cvs-commit at gcc dot gnu.org 2023-08-05 15:05 ` egallager at gcc dot gnu.org 2023-08-07 1:55 ` crazylht at gmail dot com 2023-08-12 13:50 ` egallager 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-105504-4@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).