https://blogs.gentoo.org/ago/2020/07/04/gentoo-tinderbox/ Issue: dev-perl/Math-Pari-2.10.809.0-r2 fails to compile (lto). Discovered on: amd64 (internal ref: lto_tinderbox) NOTE: This machine uses lto with CFLAGS=-flto -Werror=odr -Werror=lto-type-mismatch -Werror=strict-aliasing
Created attachment 789869 [details] build.log build log and emerge --info
Error(s) that match a know pattern in addition to what has been reported in the summary: sh: line 1: gp: command not found ../../pari-2.3.5/src/headers/paricast.h:22:50: error: dereferencing type-punned pointer will break strict-aliasing rules [-Werror=strict-aliasing[https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wstrict-aliasing]]
Here is a bit of explanation: -Werror=lto-type-mismatch: User to find possible runtime issues in packages. It likely means the package is unsafe to build & use with LTO. For projects using the same identifier but with different types across different files, they must be fixed to be consistent across the codebase. -Werror=odr: Used to find possible runtime issues in packages. These bugs are a problem anyway but may be even worse when combined with LTO. C++ code must comply with the One Definition Rule (ODR) - see https://en.cppreference.com/w/cpp/language/definition#One_Definition_Rule. -Werror=strict-aliasing: Used to find possible runtime issues in packages. These bugs are a problem anyway but may be even worse when combined with LTO. Workarounds: - If upstream is friendly and still active, file a bug upstream. For emulators, codecs, games, or multimedia packages, it may be worth just applying a workaround instead, as upstreams sometimes aren't receptive to these bugs (VALID FOR ALL). - Use the new 'filter-lto' from flag-o-matic.eclass as it's likely to be unsafe with LTO (VALID FOR lto-type-mismatch - odr). - Fix it yourself if interested, of course (VALID FOR ALL). - Append-flags -fno-strict-aliasing (VALID FOR strict-aliasing). - Use memcpy() but a union is sometimes suitable too (VALID FOR strict-aliasing). - -fstrict-aliasing is implied by -O2, so this must be addressed in some form (VALID FOR strict-aliasing). See also: https://marc.info/?l=gentoo-dev&m=165639574126280&w=2
x86_64-pc-linux-gnu-gcc -c -I ../pari-2.3.5/src -I ../pari-2.3.5/src/headers -I ../pari-2.3.5/src -I ./libPARI -I ../../pari-2.3.5/src -I ../../pari-2.3.5/src/headers -I ../../pari-2.3.5/src/graph -I . -D_REENTRANT -D_GNU_SOURCE -O2 -pipe -march=x86-64 -frecord-gcc-switches -fno-diagnostics-color -fmessage-length=0 -Werror=odr -Werror=lto-type-mismatch -Werror=strict-aliasing -fwrapv -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -pipe -march=x86-64 -frecord-gcc-switches -fno-diagnostics-color -fmessage-length=0 -Werror=odr -Werror=lto-type-mismatch -Werror=strict-aliasing -DVERSION=\"\" -DXS_VERSION=\"\" -fPIC -Derr=pari_err -DHAS_STAT -DHAS_OPENDIR -DASMINLINE -DGCC_INLINE -DDYNAMIC_PLOTTING -o arith1.o ../../pari-2.3.5/src/basemath/arith1.c In file included from ../../pari-2.3.5/src/headers/pari.h:69, from ../../pari-2.3.5/src/basemath/arith1.c:22: ../../pari-2.3.5/src/basemath/arith1.c: In function ‘classno’: ../../pari-2.3.5/src/headers/paricast.h:22:50: error: dereferencing type-punned pointer will break strict-aliasing rules [-Werror=strict-aliasing[https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wstrict-aliasing]] 22 | #define gmael1(m,x1) (((GEN*) (m))[x1]) | ^ ../../pari-2.3.5/src/headers/paricast.h:28:15: note: in expansion of macro ‘gmael1’ 28 | #define gel gmael1 | ^~~~~~ ../../pari-2.3.5/src/basemath/arith1.c:2920:11: note: in expansion of macro ‘gel’ 2920 | gel(forms,0) = f; | ^~~
So this is very confusing to me, because if I build the same version, I get this: x86_64-pc-linux-gnu-gcc -c -I ../pari-2.3.5/src -I ../pari-2.3.5/src/headers -I ../pari-2.3.5/src -I ./libPARI -I ../../pari-2.3.5/src -I ../../pari-2.3.5/src/headers -I ../../pari-2.3.5/src/graph -I . -O2 -pipe -march=x86-64-v3 -fno-strict-aliasing -DNO_PERL_RAND_SEED -fwrapv -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -pipe -march=x86-64-v3 -fno-strict-aliasing -DVERSION=\"\" -DXS_VERSION=\"\" -fPIC -Derr=pari_err -DHAS_STAT -DHAS_OPENDIR -DASMINLINE -DGCC_INLINE -DDYNAMIC_PLOTTING -o arith1.o ../../pari-2.3.5/src/basemath/arith1.c It *looks* like maybe this only respects whichever flags perl itself was built with, which makes it hard to know how to resolve this bug... I guess the answer is basically that due to bug 877659 the entire perl world handily opts out of strict-aliasing?