public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "ebotcazou at gcc dot gnu dot org" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug rtl-optimization/42511] [4.5 Regression] bootstrap error in stage3 on alpha-linux-gnu Date: Thu, 07 Jan 2010 08:22:00 -0000 [thread overview] Message-ID: <20100107082205.8469.qmail@sourceware.org> (raw) In-Reply-To: <bug-42511-5724@http.gcc.gnu.org/bugzilla/> ------- Comment #23 from ebotcazou at gcc dot gnu dot org 2010-01-07 08:22 ------- > I'm thinking about the same situation with cse2, where constant assignment > (with its REG_EQUAL note) would match another assignment with the same > REG_EQUAL note. cse2 can equal this other assignment (through matching > REG_EQUAL notes) to the constant - actually the same thing that happens in our > problematic case. Since this transformation would happen in the same BB, it > would be perfectly valid - and disabled by removing REG_EQUAL note on > constant. This mechanism very likely results in a bug if the 2 insns are not originally in the same BB, like in the case at hand. So, assuming they are originally in the same BB and carry the same REG_EQUAL note, the constant will be propagated in both insns. > Sure, but after the assigment was moved, REG_EQUAL points to invalid > assignment, see this part from the dumps: > > 39 r78:DI=zero_extend(r154:QI) > REG_DEAD: r154:QI > REG_EQUAL: zero_extend([r152:DI]) > 583 r82:DI=0x0 << here > REG_EQUAL: zero_extend([r152:DI]) > 40 pc={(r78:DI==0x0)?L230:pc} > REG_BR_PROB: 0x1388 > > The proposed change removes REG_EQUAL note only on moved insn, (insn 538) in > our case. That's too aggressive in the general case, no need to remove a REG_EQUAL note pointing to a constant if the SRC is also function_invariant_p. An acceptable compromise could be to remove the note only if its content is also not itself function_invariant_p. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42511
next prev parent reply other threads:[~2010-01-07 8:22 UTC|newest] Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top 2009-12-26 18:01 [Bug bootstrap/42511] New: boostrap " debian-gcc at lists dot debian dot org 2009-12-26 18:01 ` [Bug bootstrap/42511] " debian-gcc at lists dot debian dot org 2009-12-31 15:19 ` rguenth at gcc dot gnu dot org 2010-01-04 17:42 ` [Bug bootstrap/42511] bootstrap " debian-gcc at lists dot debian dot org 2010-01-05 12:30 ` ubizjak at gmail dot com 2010-01-05 13:52 ` ubizjak at gmail dot com 2010-01-05 13:53 ` rguenth at gcc dot gnu dot org 2010-01-05 14:16 ` debian-gcc at lists dot debian dot org 2010-01-05 14:21 ` ubizjak at gmail dot com 2010-01-05 14:27 ` ubizjak at gmail dot com 2010-01-05 15:36 ` rguenth at gcc dot gnu dot org 2010-01-05 16:25 ` ubizjak at gmail dot com 2010-01-05 20:47 ` ubizjak at gmail dot com 2010-01-05 21:15 ` ubizjak at gmail dot com 2010-01-06 10:44 ` ubizjak at gmail dot com 2010-01-06 10:45 ` ubizjak at gmail dot com 2010-01-06 11:32 ` [Bug rtl-optimization/42511] [4.5 Regression] " rguenth at gcc dot gnu dot org 2010-01-06 15:17 ` ubizjak at gmail dot com 2010-01-06 18:14 ` ubizjak at gmail dot com 2010-01-06 23:16 ` ubizjak at gmail dot com 2010-01-07 0:00 ` ebotcazou at gcc dot gnu dot org 2010-01-07 7:29 ` ubizjak at gmail dot com 2010-01-07 7:39 ` ebotcazou at gcc dot gnu dot org 2010-01-07 8:02 ` ubizjak at gmail dot com 2010-01-07 8:22 ` ebotcazou at gcc dot gnu dot org [this message] 2010-01-07 8:50 ` ubizjak at gmail dot com 2010-01-07 9:18 ` ubizjak at gmail dot com 2010-01-07 9:23 ` ubizjak at gmail dot com 2010-01-07 11:22 ` ubizjak at gmail dot com 2010-01-07 11:55 ` debian-gcc at lists dot debian dot org 2010-01-07 13:03 ` uros at gcc dot gnu dot org 2010-01-07 14:45 ` rguenth at gcc dot gnu dot org 2010-01-07 15:00 ` uros at gcc dot gnu dot org 2010-01-07 17:32 ` uros at gcc dot gnu dot org 2010-01-08 7:33 ` ubizjak at gmail dot com
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=20100107082205.8469.qmail@sourceware.org \ --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).