public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "curdeius at gmail dot com" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug c++/105577] New: [12 Regression] ICE in delete_unmarked_insns, at dce.cc:653 Date: Thu, 12 May 2022 08:52:13 +0000 [thread overview] Message-ID: <bug-105577-4@http.gcc.gnu.org/bugzilla/> (raw) https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105577 Bug ID: 105577 Summary: [12 Regression] ICE in delete_unmarked_insns, at dce.cc:653 Product: gcc Version: 12.1.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ Assignee: unassigned at gcc dot gnu.org Reporter: curdeius at gmail dot com Target Milestone: --- Similar to bug https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90082. GCC 12.1.0 ICEs when compiling this code with: ``` g++ -DROCKSDB_PLATFORM_POSIX -isystem rocksdb-cloud -isystem rocksdb-cloud/include -O3 -march=haswell -fnon-call-exceptions -c rocksdb-cloud/db/db_impl/db_impl_compaction_flush.cc ``` All the three flags are important, as the ICE doesn't happen with -O2, nor without -march, nor with -march=skylake, but it does happen with microarchs older than haswell. ICE doesn't happen without -fnon-call-exceptions either. Version: ``` $ g++ --version g++ (GCC) 12.1.0 Copyright (C) 2022 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. ``` The minimized repro (couldn't do better for the moment, preprocessed source attached of both minimal repro and the full file attached): ``` #include "db/db_impl/db_impl.h" namespace ROCKSDB_NAMESPACE { void DBImpl::InstallSuperVersionAndScheduleWork( ColumnFamilyData* cfd, SuperVersionContext* sv_context, const MutableCFOptions& mutable_cf_options) { if (UNLIKELY(sv_context->new_superversion == nullptr)) { sv_context->NewSuperVersion(); } bottommost_files_mark_threshold_ = kMaxSequenceNumber; for (auto* my_cfd : *versions_->GetColumnFamilySet()) { bottommost_files_mark_threshold_ = std::min( bottommost_files_mark_threshold_, my_cfd->current()->storage_info()->bottommost_files_mark_threshold()); } } } // namespace ROCKSDB_NAMESPACE ```
next reply other threads:[~2022-05-12 8:52 UTC|newest] Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-05-12 8:52 curdeius at gmail dot com [this message] 2022-05-12 8:53 ` [Bug c++/105577] " curdeius at gmail dot com 2022-05-12 8:54 ` curdeius at gmail dot com 2022-05-12 9:04 ` marxin at gcc dot gnu.org 2022-05-12 9:29 ` [Bug rtl-optimization/105577] [12/13 " pinskia at gcc dot gnu.org 2022-05-12 9:47 ` curdeius at gmail dot com 2022-05-12 10:18 ` marxin at gcc dot gnu.org 2022-05-12 10:19 ` [Bug rtl-optimization/105577] [12/13 Regression] ICE in delete_unmarked_insns, at dce.cc:653 since r12-248-gb58dc0b803057c0e marxin at gcc dot gnu.org 2022-05-12 10:25 ` pinskia at gcc dot gnu.org 2022-05-12 11:53 ` rguenth at gcc dot gnu.org 2022-05-12 12:03 ` rguenth at gcc dot gnu.org 2022-05-12 13:05 ` cvs-commit at gcc dot gnu.org 2022-05-12 13:06 ` [Bug rtl-optimization/105577] [12 " rguenth at gcc dot gnu.org 2022-05-12 14:25 ` marxin at gcc dot gnu.org 2022-05-16 10:08 ` cvs-commit at gcc dot gnu.org 2022-05-19 12:47 ` cvs-commit at gcc dot gnu.org 2022-05-19 12:47 ` cvs-commit at gcc dot gnu.org 2022-05-19 12:49 ` rguenth at gcc dot gnu.org 2022-06-01 15:18 ` curdeius 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=bug-105577-4@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).