Re: [patch] futex requeueing feature, futex-requeue-2.5.69-D3

Rusty Russell (rusty@rustcorp.com.au)
Tue, 20 May 2003 10:04:31 +1000


In message <Pine.LNX.4.44.0305191103500.5653-100000@localhost.localdomain> you write:
>
> the attached patch addresses a futex related SMP scalability problem of
> glibc. A number of regressions have been reported to the NTPL mailing list
> when going to many CPUs, for applications that use condition variables and
> the pthread_cond_broadcast() API call. Using this functionality, testcode
> shows a slowdown from 0.12 seconds runtime to over 237 seconds (!)
> runtime, on 4-CPU systems.

I gave feedback on this before, but didn't get a response.

1) Overload the last futex arg (change from timeval * to void *),
don't add YA arg at the end.

2) Use __alignof__(u32) not sizeof(u32). Sure, they're the same, but
you really mean __alignof__ here.

I'm glad you finally put your name at the top of the file...

Thanks!
Rusty.

--
  Anyone who quotes me in their sig is an idiot. -- Rusty Russell.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/