https://bugs.kde.org/show_bug.cgi?id=401742
--- Comment #8 from Philippe Waroquiers <philippe.waroqui...@skynet.be> --- (In reply to Bernhard M. Wiedemann from comment #6) > The patch only disables LTO for .a file creation where it does not make > sense to have it. Can you give more explanations about what you mean with 'disables LTO ... where it does not make sense to have it' ? Something like: It does not make sense to have LTO for these .a files because ..... Otherwise, with reference to comment 7: I effectively think that this makes lto almost useless : very few .o, most of the code in in the coregrind and VEX .a. See e.g. the below link command that builds memcheck: ../coregrind/link_tool_exe_linux 0x58000000 gcc -o memcheck-amd64-linux -flto -flto-partition=one -fuse-linker-plugin -m64 -O2 -finline-functions -g -Wall -Wmissing-prototypes -Wshadow -Wpointer-arith -Wstrict-prototypes -Wmissing-declarations -Wcast-align -Wcast-qual -Wwrite-strings -Wempty-body -Wformat -Wformat-security -Wignored-qualifiers -Wmissing-parameter-type -Wlogical-op -Wold-style-declaration -fno-stack-protector -fno-strict-aliasing -fno-builtin -fomit-frame-pointer -O2 -static -nodefaultlibs -nostartfiles -u _start -m64 memcheck_amd64_linux-mc_leakcheck.o memcheck_amd64_linux-mc_malloc_wrappers.o memcheck_amd64_linux-mc_main.o memcheck_amd64_linux-mc_main_asm.o memcheck_amd64_linux-mc_translate.o memcheck_amd64_linux-mc_machine.o memcheck_amd64_linux-mc_errors.o ../coregrind/libcoregrind-amd64-linux.a ../VEX/libvex-amd64-linux.a -lgcc I am also wondering why -flto option makes something not reproducable : if given twice the same compiler commands with the same args and same sources, why is gcc producing different things when given -flto ? That looks like a 'non reproducible result' produced by gcc, and gcc should be fixed to make -flto builds reproducible ? -- You are receiving this mail because: You are watching all bug changes.
