[BUG] SharedArrayBuffer should work in WebKit

[yury-s]

This popped up in #13976, SharedArrayBuffer was reenabled in Safari https://webkit.org/blog/12140/new-webkit-features-in-safari-15-2/ but we disabled COOP handling in #9185 as it broke some other functionality in WebKit. We should reenable COOP to match Safari and make SharedArrayBuffer work.

[tadd]

Hi all, thanks for the great tool.

Why has this issue stopped for almost half a year? I don't think there is a major barrier since the fix is to undo one line, making it behave the same as the current real Safari.

My problem is not with SharedArrayBuffer but with performance.now; however, I think that the root cause is the same and lies in this issue.

Since developing a cross-browser JS library, I have been using Playwright and performance.now to detect performance degradation with all multiple engines. (This is not a serious performance measurement but a part of CI.) However, the time resolution is currently fixed at 1 second with plain Playwright. Then I have been waiting several months, with almost no possibility of checking for degradation with WebKit.

I have confirmed the problem was mitigated by the proper use of COOP/COEP on two other engines. I wish to be corrected to the same status as the others.

[tadd]

After a while, I realized that my measurement approach was fundamentally wrong.

I had called performance.now() within page.evaluate(() => {...}). But as a policy, to begin with, I think it was a mistake to make the accuracy of the data dependent on the observed objects. The responsibility should be on the observer, i.e., my code on the Node.js/Playwright side.

Through some trial and error, my problem is solved because I have found several ways independent of the browser's time resolution. Therefore, I am no longer in a hurry about this issue.

In any case, I think it is not desirable to behave differently from plain Safari, so I would appreciate it if you continued to consider fixing this problem.

[tadd]

A while ago, I realized I was fundamentally wrong about another thing. Not only had I checked the behavior at hand, but I also believed that the patch in browser_patches/webkit was up to date when I wrote my first comment here .

That was a mistake. The browser patch has been closed/proprietary since last August. #16845 #17556

Here is what I want to believe next.

"Microsoft loves open source"
Judge us by the actions we have taken in the recent past, our actions today and in the future
https://pulse.microsoft.com/nl-nl/transform-nl-nl/na/fa1-microsoft-loves-open-source/

[SamVerschueren]

Are there any plans to re-enabling SABs in WebKit? I'm looking at migrating from Puppeteer to Playwright because Playwright supports WebKit, and it doesn't look like WebKit support is coming anytime soon in Puppeteer. But at StackBlitz, we heavily rely on SABs.

[eatsjobs]

Hi! Just wanna ask if there's any chance to have this sooner or later. :) @yury-s

[swissspidy]

Is there any update on this?

[swissspidy]

Starting with Playwright 1.43.0, WebKit now seems to immediately crash when opening a page with cross-origin isolation enabled. Before that, the page could at least be opened and was functioning.

[IbrahimCSAE]

is there any updates on this? we are considering using playwright over at OHIF and CornerstoneJS.