https://sourceware.org/bugzilla/show_bug.cgi?id=32641
--- Comment #3 from 孙文举 <swj22 at mails dot tsinghua.edu.cn> --- (In reply to Nick Clifton from comment #2) > Fixed. > > This was another case where the linker's code was assuming that relocations > and symbols were well formed and not checking for possible corruption. > > Note - even with this patch applied, I still saw a sanitizer failure: > > ================================================================= > ==3041740==ERROR: AddressSanitizer: requested allocation size 0x490000000030 > (0x490000001030 after adjustments for alignment, red zones etc.) exceeds > maximum supported size of 0x10000000000 (thread T0) > #0 0x14d83aac2897 in malloc (/lib64/libasan.so.8+0xc2897) (BuildId: > 0505b45e5a5d9a6c8ecb1d529aaaf13cd21fbe4e) > #1 0x967506 in bfd_malloc > /work/sources/binutils/upstream/current/bfd/libbfd.c:291 > #2 0xa9af43 in bfd_elf_final_link > /work/sources/binutils/upstream/current/bfd/elflink.c:13030 > #3 0x45d86a in ldwrite > /work/sources/binutils/upstream/current/ld/ldwrite.c:550 > #4 0x45662f in main > /work/sources/binutils/upstream/current/ld/ldmain.c:560 > #5 0x14d83a74f247 in __libc_start_call_main (/lib64/libc.so.6+0x3247) > (BuildId: 515c33a35f41020661fea8ac4eb995e26ccd6b00) > #6 0x14d83a74f30a in __libc_start_main@GLIBC_2.2.5 > (/lib64/libc.so.6+0x330a) (BuildId: 515c33a35f41020661fea8ac4eb995e26ccd6b00) > #7 0x401c34 in _start > > This however is a false positive, since it is reporting that the requested > output size of the linked binary is too big. The linker code does handle > this situation correctly however, so I added "allocator_may_return_null=1" > to my sanitizer options in order to stop this error. With that option > enabled the link no longer triggers any sanitization issues. Thans for your quickly fixing, Got it! -- You are receiving this mail because: You are on the CC list for the bug.
