Uh oh!
There was an error while loading. Please reload this page.
src: remove ERR prefix in WebCryptoKeyExportStatus - #35639
Conversation
This commit suggests removing the ERR prefix in the WebCryptoKeyExportStatus enum. The motivation for this is that I think it improves the readability of the code. For example, the following line had me look twice to see what was going on: case WebCryptoKeyExportStatus::ERR_OK: // Success!
nodejs-github-bot
commented
Oct 14, 2020
Review requested:
|
nodejs-github-bot
commented
Oct 14, 2020
jasnell
left a comment
There was a problem hiding this comment.
I prefer the prefix but not enough to block ;)
I'm looking into the cctest output$ out/Release/cctest --gtest_filter=EnvironmentTest.BufferWithFreeCallbackIsDetachedRunning main() from ../test/cctest/gtest/gtest_main.ccNote: Google Test filter = EnvironmentTest.BufferWithFreeCallbackIsDetached[==========] Running 1 test from 1 test suite.[----------] Global test environment set-up.[----------] 1 test from EnvironmentTest[ RUN ] EnvironmentTest.BufferWithFreeCallbackIsDetached6===================================================================2773765==ERROR: AddressSanitizer: new-delete-type-mismatch on 0x60400001af50 in thread T0: object passed to delete has wrong type: size of the allocated type: 48 bytes; size of the deallocated type: 1 bytes. #0 0x7f0a3a306175 in operator delete(void*, unsigned long) (/lib64/libasan.so.5+0x111175) #1 0x38328c2 in v8::internal::JSArrayBuffer::Detach(bool) (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x38328c2) #2 0x29bf88a in v8::ArrayBuffer::Detach() (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x29bf88a) #3 0x151adae in node::Buffer::(anonymous namespace)::CallbackInfo::CleanupHook(void*) (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x151adae) #4 0x144a5ff in node::Environment::RunCleanup() (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x144a5ff) #5 0x131a6f7 in node::FreeEnvironment(node::Environment*) (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x131a6f7) #6 0x11d7730 in EnvironmentTest_BufferWithFreeCallbackIsDetached_Test::TestBody() (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x11d7730) #7 0x1114cfd in testing::Test::Run() [clone .part.0] (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x1114cfd) #8 0x1115af0 in testing::TestInfo::Run() [clone .part.0] (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x1115af0) #9 0x11162c9 in testing::TestSuite::Run() [clone .part.0] (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x11162c9) #10 0x1137831 in testing::internal::UnitTestImpl::RunAllTests() (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x1137831) #11 0x11383bf in testing::UnitTest::Run() (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x11383bf) #12 0xa4d5b3 in main (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0xa4d5b3) #13 0x7f0a39ccf1a2 in __libc_start_main (/lib64/libc.so.6+0x271a2) #14 0xa7f09d in _start (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0xa7f09d)0x60400001af50 is located 0 bytes inside of 48-byte region [0x60400001af50,0x60400001af80)allocated by thread T0 here: #0 0x7f0a3a304a97 in operator new(unsigned long) (/lib64/libasan.so.5+0x10fa97) #1 0x34cb1a8 in v8::internal::BackingStore::WrapAllocation(void*, unsigned long, void (*)(void*, unsigned long, void*), void*, v8::internal::SharedFlag) (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x34cb1a8) #2 0x29c18d5 in v8::ArrayBuffer::NewBackingStore(void*, unsigned long, void (*)(void*, unsigned long, void*), void*) (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x29c18d5) #3 0x1547193 in node::Buffer::New(node::Environment*, char*, unsigned long, void (*)(char*, void*), void*) (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x1547193) #4 0x1548e95 in node::Buffer::New(v8::Isolate*, char*, unsigned long, void (*)(char*, void*), void*) (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x1548e95) #5 0x11d7536 in EnvironmentTest_BufferWithFreeCallbackIsDetached_Test::TestBody() (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x11d7536) #6 0x1114cfd in testing::Test::Run() [clone .part.0] (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x1114cfd) #7 0x1115af0 in testing::TestInfo::Run() [clone .part.0] (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x1115af0) #8 0x11162c9 in testing::TestSuite::Run() [clone .part.0] (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x11162c9) #9 0x1137831 in testing::internal::UnitTestImpl::RunAllTests() (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x1137831) #10 0x11383bf in testing::UnitTest::Run() (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0x11383bf) #11 0xa4d5b3 in main (/home/danielbevenius/work/nodejs/node/out/Release/cctest+0xa4d5b3) #12 0x7f0a39ccf1a2 in __libc_start_main (/lib64/libc.so.6+0x271a2)SUMMARY: AddressSanitizer: new-delete-type-mismatch (/lib64/libasan.so.5+0x111175) in operator delete(void*, unsigned long)==2773765==HINT: if you don't care about these errors you may set ASAN_OPTIONS=new_delete_type_mismatch=0==2773765==ABORTING |
jasnell
commented
Oct 15, 2020
@danbev I get those asan failures from time to time also. They definitely seem unrelated. |
Thanks, I'm able to reproduce this consistently on my local machine so I'll take a closer look at it today. I've collected some notes and might have a patch for this, but have run out of time today (and tomorrow is a Red Hat Recharge day so I won't be working). I'll continue with this when I get back on Monday. I've opened #35669 to track this. |
This commit suggests removing the ERR prefix in the WebCryptoKeyExportStatus enum. The motivation for this is that I think it improves the readability of the code. For example, the following line had me look twice to see what was going on: case WebCryptoKeyExportStatus::ERR_OK: // Success! PR-URL: #35639 Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Rich Trott <rtrott@gmail.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Tobias Nießen <tniessen@tnie.de>
danbev
commented
Oct 28, 2020
Landed in 923f76d. |
This commit suggests removing the ERR prefix in the WebCryptoKeyExportStatus enum. The motivation for this is that I think it improves the readability of the code. For example, the following line had me look twice to see what was going on: case WebCryptoKeyExportStatus::ERR_OK: // Success! PR-URL: #35639 Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Rich Trott <rtrott@gmail.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Tobias Nießen <tniessen@tnie.de>
This commit suggests removing the
ERRprefix in theWebCryptoKeyExportStatusenum.The motivation for this is that I think it improves the readability of
the code. For example, the following line had me look twice to see what
was going on:
Checklist
make -j4 test(UNIX), orvcbuild test(Windows) passes