On Mon, 2005-05-30 at 10:31 -0700, Linus Torvalds wrote:
Also, pipes are a bit special from a scheduling standpoint because they
use the magic "synchronous wakeup" thing, and it might be worthwhile
trying to just change the two calls to "wake_up_interruptible_sync()" in
fs/pipe.c to the non-sync version (ie just remove the "_sync" part).
Btw, I was looking at this yesterday but noticed there are no
__wake_up_sync() calls in the oprofile report. So I don't think we're
hitting the sync wakeup paths at all.
I also read Crossover sources a bit and I think the setup is something like
this:
wineserver:
- Polls a 'reply' pipe to see if there's something to read.
- If the pipe has something for us, read it.
- Do something and the write results to another pipe.
wine-preloader:
- Write to the 'reply' pipe.
- Read from the another pipe.
Looks if the processes keep on waking up each other and thus eat up all CPU
time. (Even more so if Crossover uses RT priority, have to check that.)
P.S. I can also verify that it is indeed the 64 KB change that breaks
Crossover. I changed PIPE_BUFFERS to 1 and could not reproduce the hang.
Increasing it to 8 makes the problem come up again.
Pekka
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]