Update compiler-rt to llvm-12#14280
Conversation
sbc100
force-pushed
the
update_compiler_rt
branch
from
8f6f87d to
e3a88be
Compare
sbc100
force-pushed
the
update_compiler_rt
branch
from
e3a88be to
2045373
Compare
|
To see the diff from upstream you can use this like: |
kripken
left a comment
kripken
left a comment
There was a problem hiding this comment.
Looking at the diff to upstream:
- Is the diff in
ubsan_checks.incnecessary? - The diff in
ubsan_handlers.cpplooks odd - no ifdefing for emscripten, just a pure change to upstream?
| thrd_detach: 'emscripten_builtin_pthread_detach', | ||
| #else | ||
| thrd_detach: 'pthread_detach', | ||
| #endif |
There was a problem hiding this comment.
this could be on a single line like the {{{ }}} stuff above.
There was a problem hiding this comment.
I couldn't make it work when the string is on the right of the ":" for some reason. It was giving me JS compiler errors. Plus I think this syntax is more readable in the end.
There was a problem hiding this comment.
But it works just several lines above in the diff? I don't understand. It's just adding a prefix to a string, isn't it?
There was a problem hiding this comment.
Somehow it works in the key position but not in the value position. Oh maybe its because the keys are not in quotes.. Let me try again.
These are part of the existing 10.0.0 changes. I don't really know why they were removed, but they were. Maybe something to do with the way emscripten handles NULL vs native platforms. See the existing 10.0.0 changes:
Again, these are just carried over from the current changes: We can look at cleaning them or and/removing as we look into upstreaming perhaps? |
Followup to #14280 Changes from upstream can be seen at: llvm/llvm-project@llvmorg-12.0.0...emscripten-core:emscripten-libs-12.0.0 Fixes: #14244
Followup to #14280 Changes from upstream can be seen at: llvm/llvm-project@llvmorg-12.0.0...emscripten-core:emscripten-libs-12.0.0 Fixes: #14244
Followup to #14280 Changes from upstream can be seen at: llvm/llvm-project@llvmorg-12.0.0...emscripten-core:emscripten-libs-12.0.0 Fixes: #14244
Followup to #14280 Changes from upstream can be seen at: llvm/llvm-project@llvmorg-12.0.0...emscripten-core:emscripten-libs-12.0.0 Fixes: #14244
|
It looks several asan/lsan tests started to break after this. Starting commit: https://ci.chromium.org/ui/p/emscripten-releases/builders/ci/linux-test-suites/b8846238787694172272/overview |
|
Ah yes dropped the ball on that.. thanks for reminding me. |
This test leaks allocates the asyncify stack during the exit handlers and its never freed. Interestingly enough this leak was not caught prior to #14280. I did some investigation and found that prior to #14280 this chunk was bring detected a reachable via the conservative stack scan here: https://github.com/emscripten-core/emscripten/blob/98a8cbf5032f8d369be636a59180a59f1eb8833f/system/lib/compiler-rt/lib/lsan/lsan_common_emscripten.cpp#L141-L153 If I comment out this stack scan then this leak will be detected even before #14280 Also, add `--profiling` to asan and lsan builds to make error reports more useful. See #14360
This test leaks allocates the asyncify stack during the exit handlers and its never freed. Interestingly enough this leak was not caught prior to #14280. I did some investigation and found that prior to #14280 this chunk was bring detected a reachable via the conservative stack scan here: https://github.com/emscripten-core/emscripten/blob/98a8cbf5032f8d369be636a59180a59f1eb8833f/system/lib/compiler-rt/lib/lsan/lsan_common_emscripten.cpp#L141-L153 If I comment out this stack scan then this leak will be detected even before #14280 Also, add `--profiling` to asan and lsan builds to make error reports more useful. See #14360
3 participants
No description provided.