public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "tnfchris at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug tree-optimization/114151] New: [14 Regression] weird and inefficient codegen and addressing modes since g:a0b1798042d033fd2cc2c806afbb77875dd2909b Date: Wed, 28 Feb 2024 13:57:25 +0000 [thread overview] Message-ID: <bug-114151-4@http.gcc.gnu.org/bugzilla/> (raw) https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114151 Bug ID: 114151 Summary: [14 Regression] weird and inefficient codegen and addressing modes since g:a0b1798042d033fd2cc2c806afbb77875dd2909b Product: gcc Version: 14.0 Status: UNCONFIRMED Keywords: missed-optimization Severity: normal Priority: P3 Component: tree-optimization Assignee: unassigned at gcc dot gnu.org Reporter: tnfchris at gcc dot gnu.org CC: rguenth at gcc dot gnu.org Target Milestone: --- Target: aarch64* Created attachment 57559 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=57559&action=edit testcase The attached C++ testcase compiled with: -O3 -mcpu=neoverse-n2 used to compile a nice and simple loop. But after g:a0b1798042d033fd2cc2c806afbb77875dd2909b The codegen is weird and it uses horrible addressing modes. The first odd part is that it's decided to split the loop, the "main" loop has a guard after it to branch to the exit is the iteration count is 1. If not instead of just loop again it falls through the a copy of the main loop, but has destroyed addressing modes. The copy of the loop seems to have unshared the address calculations. Before we had: _128 = (void *) ivtmp.11_20; _54 = MEM <__SVFloat16_t> [(__fp16 *)_128]; _10 = MEM <__SVFloat16_t> [(__fp16 *)_128 + POLY_INT_CST [16B, 16B]]; _75 = MEM <__SVFloat16_t> [(__fp16 *)_128 + POLY_INT_CST [32B, 32B]]; etc, so all as an offset from _128. Now we have: col_i_61 = (int) ivtmp.11_100; _60 = (long unsigned int) col_i_61; _59 = _60 * 2; _58 = a_j_69 + _59; _54 = MEM <__SVFloat16_t> [(__fp16 *)_58]; _53 = _59 + POLY_INT_CST [16, 16]; _13 = a_j_69 + _53; _10 = MEM <__SVFloat16_t> [(__fp16 *)_13]; _74 = _59 + POLY_INT_CST [32, 32]; _19 = a_j_69 + _74; _75 = MEM <__SVFloat16_t> [(__fp16 *)_19]; and similarly for the stores as well. it also weirdly creates some very complicated addressing computations. Before we had: _144 = p_mat_16(D) + 6; _64 = MEM <__SVFloat16_t> [(__fp16 *)_144 + ivtmp.10_100 * 2]; _143 = p_mat_16(D) + 4; _84 = MEM <__SVFloat16_t> [(__fp16 *)_143 + ivtmp.10_100 * 2]; and after: ivtmp.23_130 = (unsigned long) p_mat_16(D); _123 = 2 - ivtmp.23_130; _124 = &MEM <__SVFloat16_t> [(__fp16 *)0B + _123 + ivtmp.12_109 * 2]; _64 = MEM <__SVFloat16_t> [(__fp16 *)_124]; _122 = -ivtmp.23_130; _120 = &MEM <__SVFloat16_t> [(__fp16 *)0B + _122 + ivtmp.12_109 * 2]; _84 = MEM <__SVFloat16_t> [(__fp16 *)_120]; This results in quite the codesize increase, and a 7-10% performance loss.
next reply other threads:[~2024-02-28 13:57 UTC|newest] Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top 2024-02-28 13:57 tnfchris at gcc dot gnu.org [this message] 2024-02-28 14:33 ` [Bug tree-optimization/114151] " rguenth at gcc dot gnu.org 2024-02-28 14:36 ` rguenth at gcc dot gnu.org 2024-02-28 16:51 ` tnfchris at gcc dot gnu.org 2024-02-29 7:19 ` rguenther at suse dot de 2024-02-29 18:15 ` amacleod at redhat dot com 2024-03-01 9:37 ` rguenth at gcc dot gnu.org 2024-03-01 15:02 ` [Bug tree-optimization/114151] [14 Regression] weird and inefficient codegen and addressing modes since r14-9193 amacleod at redhat dot com 2024-03-04 7:47 ` rguenth at gcc dot gnu.org 2024-03-06 3:37 ` amacleod at redhat dot com 2024-03-06 7:14 ` rguenth at gcc dot gnu.org 2024-03-06 7:31 ` rguenth at gcc dot gnu.org 2024-03-06 14:57 ` amacleod at redhat dot com 2024-03-06 20:05 ` amacleod at redhat dot com 2024-03-07 8:04 ` rguenth at gcc dot gnu.org 2024-03-07 15:53 ` amacleod at redhat dot com 2024-03-07 20:37 ` law at gcc dot gnu.org 2024-03-08 10:13 ` rguenth at gcc dot gnu.org 2024-03-08 10:22 ` tnfchris at gcc dot gnu.org 2024-03-08 14:10 ` rguenth at gcc dot gnu.org 2024-03-12 9:59 ` rguenth at gcc dot gnu.org 2024-03-12 10:00 ` rguenth at gcc dot gnu.org 2024-03-12 20:41 ` amacleod at redhat dot com 2024-03-13 7:38 ` rguenth at gcc dot gnu.org 2024-03-13 17:37 ` amacleod at redhat dot com 2024-03-19 12:12 ` cvs-commit at gcc dot gnu.org 2024-03-19 12:16 ` 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-114151-4@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).