[Bug 46252] StarCitizen has bad performance

wine-bugs at winehq.org wine-bugs at winehq.org
Fri Dec 7 19:23:52 CST 2018


https://bugs.winehq.org/show_bug.cgi?id=46252

Zebediah Figura <z.figura12 at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |z.figura12 at gmail.com

--- Comment #3 from Zebediah Figura <z.figura12 at gmail.com> ---
(In reply to Austin English from comment #1)
> Is this a regression caused by 6060d2703c2e1382f076ad40ea43980781eaca2c? If
> I'm reading the bug correctly, it didn't work before, so I don't see how it
> would be a regression.

Kind of, though not really. It's more of a continuation of that bug's
regression from 7c430f5b3—I suspect that all the overhead here is coming from
wineserver interaction (though certainly the CS doesn't help).

(In reply to rawfox from comment #2)
> What ive read on this topic is, Windows uses WaitOnAddress() while Linux
> uses kernel futexes. So to my understanding, a fast thread organisation is
> done in the linux Kernel while it seems to be in the ntdll on Windows.

These would have to go into the kernel on Windows too, otherwise the thread
couldn't schedule. The overhead in Wine comes from going through the server.
Futexes go to the kernel too (well, sometimes), but they only do it once and we
don't block on waiting for the server.

-- 
Do not reply to this email, post in Bugzilla using the
above URL to reply.
You are receiving this mail because:
You are watching all bug changes.


More information about the wine-bugs mailing list