-
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
[wasm] rewrite crypto-worker to typescript #73073
Conversation
Tagging subscribers to 'arch-wasm': @lewing Issue Details
Fixes #72941
|
/azp run runtime-wasm |
Azure Pipelines successfully started running 1 pipeline(s). |
/azp run runtime-wasm |
Azure Pipelines successfully started running 1 pipeline(s). |
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.
thanks for converting the crypto worker to TS!
I wrote and then spoke to @javiercn this morning about how to locate the url of the I added new resource type @radical I know you guys also spoke about it. |
/azp run runtime-wasm |
Azure Pipelines successfully started running 1 pipeline(s). |
because slice makes copy, the constructor make a view. Unless I made some mistake. |
Ah, that's a difference between JS and .NET. Looks like what we want is |
I'm not familiar with how the assets stuff works... Emscripten just does this: allocateUnusedWorker: function() {
if (!Module["locateFile"]) {
PThread.unusedWorkers.push(new Worker(new URL("dotnet.worker.js", import.meta.url)));
return;
}
var pthreadMainJs = locateFile("dotnet.worker.js");
PThread.unusedWorkers.push(new Worker(pthreadMainJs));
}, but I don't see any changes to Update I suppose we could replace |
Right, different PR |
/azp run runtime-wasm |
Azure Pipelines successfully started running 1 pipeline(s). |
Azure Pipelines successfully started running 1 pipeline(s). |
/azp run runtime-wasm |
Azure Pipelines successfully started running 1 pipeline(s). |
No - locally on my codespaces VM they take around 70-90 seconds. |
On my windows, when it works the it's 2 minutes. But I saw it deadlock multiple times, now I'm trying to reproduce it on main, with just the logging fixed here #73468 |
Yeah, one concern I have with this PR is that you are trying to change too much at once. It is super hard to know what went wrong when making multiple changes in a single PR. |
I was able to reproduce the deadlock on this branch locally on my windows
|
0830d25
to
1013b63
Compare
/azp run runtime-wasm |
Azure Pipelines successfully started running 1 pipeline(s). |
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 - thanks for splitting this up!
@eerhardt are you guys able to have look at the deadlock ? do you need more info? |
I haven't been able to look at the deadlock. Does it happen in |
Probably on main as well, but the fix to logging will help you debugging it without XHarness. Otherwise your chrome will crash with OOM. |
I haven't seen a test failure report in a CI, and the tests have been running for over a month. |
…ccess and preventing the deadlock. This is RC1 stop-gap.
Returned some console writes (which I commented out 10 days ago) in hope they prevented the deadlock. I think they were serializing the access to shared buffer. Perhaps we could find better solution later. |
/azp run runtime-wasm |
Azure Pipelines successfully started running 1 pipeline(s). |
InitCryptoMessageData
config.diagnosticTracing
Moved some changes from this PR to #73468 and #73073