Le lun. 20 sept. 2021 à 10:39, Ondrej Zary <ond...@zary.sk> a écrit : > > > Ok now try to run the whole thing against valgrind... > Seems that valgrind does not work with asan: > > $ LD_PRELOAD=/usr/lib/i386-linux-gnu/libasan.so.5.0.0 valgrind yarnpkg install > ==752== Memcheck, a memory error detector > ==752== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. > ==752== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info > ==752== Command: /usr/bin/yarnpkg install > ==752== > ==752==ASan runtime does not come first in initial library list; you should > either link runtime to your application or manually preload it with > LD_PRELOAD. > ==752== > ==752== HEAP SUMMARY: > ==752== in use at exit: 0 bytes in 0 blocks > ==752== total heap usage: 0 allocs, 0 frees, 0 bytes allocated > ==752== > ==752== All heap blocks were freed -- no leaks are possible > ==752== > ==752== For counts of detected and suppressed errors, rerun with: -v > ==752== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0) > > valgrind with clean libuv1 (no asan): > runuser -u gitlab -- sh -c 'valgrind --trace-children=yes yarnpkg install' > ==3163== Memcheck, a memory error detector > ==3163== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. > ==3163== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info > ==3163== Command: /usr/bin/yarnpkg install > ==3163== > ==3163== Memcheck, a memory error detector > ==3163== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. > ==3163== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info > ==3163== Command: /usr/bin/node /usr/bin/yarnpkg install > ==3163== > yarn install v1.13.0 > [1/5] Validating package.json... > [2/5] Resolving packages... > [3/5] Fetching packages... > [---------------------------------------------------------------------------------------------------------------------------------------------------] > 0/520==3163== Invalid read of size 4 > ==3163== at 0x4556B5B: node::fs::FSReqWrap::~FSReqWrap() (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x4547A42: node::fs::FSReqAfterScope::~FSReqAfterScope() (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x45484FD: node::fs::AfterInteger(uv_fs_s*) (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x556170F: uv__work_done (in > /usr/lib/i386-linux-gnu/libuv.so.1.0.0) > ==3163== by 0x55657FD: ??? (in /usr/lib/i386-linux-gnu/libuv.so.1.0.0) > ==3163== by 0x5575527: uv__io_poll (in > /usr/lib/i386-linux-gnu/libuv.so.1.0.0) > ==3163== by 0x55661C5: uv_run (in /usr/lib/i386-linux-gnu/libuv.so.1.0.0) > ==3163== by 0x4515C75: node::Start(v8::Isolate*, node::IsolateData*, > std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, > std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, > std::char_traits<char>, std::allocator<char> > > > const&, > std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, > std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, > std::char_traits<char>, std::allocator<char> > > > const&) (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x4513C96: node::Start(int, char**) (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x8049157: main (in /usr/bin/node) > ==3163== Address 0x1085 is not stack'd, malloc'd or (recently) free'd > ==3163== > ==3163== > ==3163== Process terminating with default action of signal 11 (SIGSEGV) > ==3163== Access not within mapped region at address 0x1085 > ==3163== at 0x4556B5B: node::fs::FSReqWrap::~FSReqWrap() (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x4547A42: node::fs::FSReqAfterScope::~FSReqAfterScope() (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x45484FD: node::fs::AfterInteger(uv_fs_s*) (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x556170F: uv__work_done (in > /usr/lib/i386-linux-gnu/libuv.so.1.0.0) > ==3163== by 0x55657FD: ??? (in /usr/lib/i386-linux-gnu/libuv.so.1.0.0) > ==3163== by 0x5575527: uv__io_poll (in > /usr/lib/i386-linux-gnu/libuv.so.1.0.0) > ==3163== by 0x55661C5: uv_run (in /usr/lib/i386-linux-gnu/libuv.so.1.0.0) > ==3163== by 0x4515C75: node::Start(v8::Isolate*, node::IsolateData*, > std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, > std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, > std::char_traits<char>, std::allocator<char> > > > const&, > std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, > std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, > std::char_traits<char>, std::allocator<char> > > > const&) (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x4513C96: node::Start(int, char**) (in > /usr/lib/i386-linux-gnu/libnode.so.64) > ==3163== by 0x8049157: main (in /usr/bin/node) > ==3163== If you believe this happened as a result of a stack > ==3163== overflow in your program's main thread (unlikely but > ==3163== possible), you can try to increase the size of the > ==3163== main thread stack using the --main-stacksize= flag. > ==3163== The main thread stack size used in this run was 8388608. > ==3163== Invalid read of size 1 > ==3163== at 0x786A6A4: check_free (dlerror.c:189) > ==3163== by 0x786ABD8: free_key_mem (dlerror.c:221) > ==3163== by 0x786ABD8: __dlerror_main_freeres (dlerror.c:239) > ==3163== by 0x7CA4667: __libc_freeres (in > /usr/lib/i386-linux-gnu/libc-2.28.so) > ==3163== by 0x402D1DE: _vgnU_freeres (in > /usr/lib/i386-linux-gnu/valgrind/vgpreload_core-x86-linux.so) > ==3163== Address 0xedbde is not stack'd, malloc'd or (recently) free'd > ==3163== > ==3163== > ==3163== Process terminating with default action of signal 11 (SIGSEGV) > ==3163== Access not within mapped region at address 0xEDBDE > ==3163== at 0x786A6A4: check_free (dlerror.c:189) > ==3163== by 0x786ABD8: free_key_mem (dlerror.c:221) > ==3163== by 0x786ABD8: __dlerror_main_freeres (dlerror.c:239) > ==3163== by 0x7CA4667: __libc_freeres (in > /usr/lib/i386-linux-gnu/libc-2.28.so) > ==3163== by 0x402D1DE: _vgnU_freeres (in > /usr/lib/i386-linux-gnu/valgrind/vgpreload_core-x86-linux.so) > ==3163== If you believe this happened as a result of a stack > ==3163== overflow in your program's main thread (unlikely but > ==3163== possible), you can try to increase the size of the > ==3163== main thread stack using the --main-stacksize= flag. > ==3163== The main thread stack size used in this run was 8388608. > ==3163== > ==3163== HEAP SUMMARY: > ==3163== in use at exit: 1,908,342 bytes in 19,155 blocks > ==3163== total heap usage: 743,018 allocs, 723,863 frees, 572,337,823 bytes > allocated > ==3163== > ==3163== LEAK SUMMARY: > ==3163== definitely lost: 78 bytes in 1 blocks > ==3163== indirectly lost: 0 bytes in 0 blocks > ==3163== possibly lost: 81,312 bytes in 14 blocks > ==3163== still reachable: 1,826,952 bytes in 19,140 blocks > ==3163== of which reachable via heuristic: > ==3163== newarray : 50,384 bytes in 37 > blocks > ==3163== multipleinheritance: 32 bytes in 1 blocks > ==3163== suppressed: 0 bytes in 0 blocks > ==3163== Rerun with --leak-check=full to see details of leaked memory > ==3163== > ==3163== For counts of detected and suppressed errors, rerun with: -v > ==3163== ERROR SUMMARY: 2 errors from 2 contexts (suppressed: 0 from 0) > Segmentation fault add --track-origins=yes to valgrind
And try to rebuild the whole libuv and nodejs with -fstack-protector-all Bastien > > -- > Ondrej Zary > > -- > Pkg-javascript-devel mailing list > pkg-javascript-de...@alioth-lists.debian.net > https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/pkg-javascript-devel
