-
Notifications
You must be signed in to change notification settings - Fork 4.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
JIT: Improve forwardsub for small types #94631
Conversation
Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch Issue Detailsnull
|
63197bd
to
fb4d103
Compare
@jakobbotsch @SingleAccretion thanks for the suggestion for a better fix! Is there anything else needed here then? This should fix some of the regressions I hit in #94621, but I want to land it separately to make that one zero-diff (ideally). example: cc @dotnet/jit-contrib |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM -- please fix the PR title and commit message
Should probably run jitstress and Fuzzlyn too. |
/azp run runtime-coreclr outerloop, runtime-coreclr jitstress, fuzzlyn |
Azure Pipelines successfully started running 3 pipeline(s). |
@jakobbotsch have you seen Fuzzlyn failures like this one https://dev.azure.com/dnceng-public/public/_build/results?buildId=466925&view=ms.vss-build-web.run-extensions-tab ? |
No, I haven't seen that before. I checked and it is hitting an assert with and without this change, so it is not related to this PR. Assert failure(PID 15296 [0x00003bc0], Thread: 13572 [0x3504]): Assertion failed 'codeSize <= MAX_PROLOG_SIZE_BYTES' in 'Program:.cctor()' during 'Emit code' (IL size 10191; hash 0x02c770ad; FullOpts)
File: C:\dev\dotnet\runtime\src\coreclr\jit\unwindarmarch.cpp Line: 834
Image: H:\dotnet\runtime\artifacts\tests\coreclr\windows.arm64.Checked\Tests\Core_Root_base\corerun.exe (Looks like Fuzzlyn should be looking at inner exceptions when it sees |
Regression in |
Diffs