Summary: | sys-devel/gcc: ICE when building sci-mathematics/form-4.3.0-r1 (canonical types differ for identical types ‘UBYTE [7]’ {aka ‘unsigned char [7]’} and ‘unsigned char [7]’) | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Toralf Förster <toralf> |
Component: | Current packages | Assignee: | Gentoo Toolchain Maintainers <toolchain> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | sam, sci-mathematics |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
See Also: | https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109521 | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | |||
Bug Blocks: | 865117 | ||
Attachments: |
emerge-info.txt
emerge-history.txt environment etc.clang.tar.bz2 etc.portage.tar.bz2 logs.tar.bz2 sci-mathematics:form-4.3.0-r1:20230414-193854.log temp.tar.bz2 |
Description
Toralf Förster
![]() Created attachment 860103 [details]
emerge-info.txt
Created attachment 860104 [details]
emerge-history.txt
Created attachment 860105 [details]
environment
Created attachment 860106 [details]
etc.clang.tar.bz2
Created attachment 860107 [details]
etc.portage.tar.bz2
Created attachment 860108 [details]
logs.tar.bz2
Created attachment 860109 [details]
sci-mathematics:form-4.3.0-r1:20230414-193854.log
Created attachment 860110 [details]
temp.tar.bz2
Looks like a gcc bug. There are a number of 'canonical types differ for identical types' in https://gcc.gnu.org/bugzilla/ (In reply to Andrey Grozin from comment #9) > Looks like a gcc bug. There are a number of 'canonical types differ for > identical types' in https://gcc.gnu.org/bugzilla/ Please assign to toolchain (or CC us at least) if you suspect a GCC bug. Thanks, I'll look into it now. Can't reproduce it yet... toralf, is this reproducible for you? (In reply to Sam James from comment #12) > toralf, is this reproducible for you? Can you run: arch=x86-64-v3; for t in param target; do cmd="gcc -Q -O2 -march=$arch --help=$t"; diff -U0 <(LANG=C $cmd) <(LANG=C $cmd -march=x86-64-v4); done arch=znver3; for t in param target; do cmd="gcc -Q -O2 -march=$arch --help=$t"; diff -U0 <(LANG=C $cmd) <(LANG=C $cmd -march=x86-64-v4); done arch=znver2; for t in param target; do cmd="gcc -Q -O2 -march=$arch --help=$t"; diff -U0 <(LANG=C $cmd) <(LANG=C $cmd -march=x86-64-v4); done nevermind, I got it: [08:13:50] <@sam_> it's a checking ICE plus I needed some of (not sure which yet): "-fno-stack-protector -fcf-protection=none -U_GLIBCXX_ASSERTIONS -U_FORTIFY_SOURCE" *** Bug 905313 has been marked as a duplicate of this bug. *** The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=32bf1dfde8a72f36a664ce986076509778202f93 commit 32bf1dfde8a72f36a664ce986076509778202f93 Author: Sam James <sam@gentoo.org> AuthorDate: 2023-04-30 00:19:38 +0000 Commit: Sam James <sam@gentoo.org> CommitDate: 2023-04-30 00:19:38 +0000 sci-mathematics/form: workaround GCC -fchecking ICE w/ -Wno-uninitialized Reported upstream at https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109521 but let's apply a workaround for now to avoid dupes. It's "only" an error in diagnostics anyway. Closes: https://bugs.gentoo.org/904339 Signed-off-by: Sam James <sam@gentoo.org> sci-mathematics/form/form-4.3.0-r1.ebuild | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) |