lkml.org 
[lkml]   [2003]   [Jan]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: lots of calls to __write/read_lock_failed
On Thu, Jan 16, 2003 at 04:47:30AM +0000, William Lee Irwin III wrote:
> On Wed, Jan 15, 2003 at 08:18:13PM -0800, Dave Hansen wrote:
> > file_table:_raw_read_lock() 3300000
> > Call Trace:
> > [<c0152469>] fget+0x9d/0xa0
> > [<c0152b27>] sys_fsync+0x21/0xbe
> > [<c0151b53>] sys_writev+0x47/0x56
> > [<c010931f>] syscall_call+0x7/0xb
>
> read_lock(&file->files_lock);

You mean read_lock(&files->file_lock); :)

Dave, does your webserver benchmark clone() tasks with CLONE_FILES ? Unless
the fd table is shared, can't see why there would be contention on this.
If it is indeed necessary to share fd table, then there is a somewhat
unmaintained lockfree fget() patch (files_struct_rcu) that you might want
to try.

>
> On Wed, Jan 15, 2003 at 08:18:13PM -0800, Dave Hansen wrote:
> > time:_raw_write_lock() 1350000
> > Call Trace:
> > [<c010f321>] timer_interrupt+0x99/0x9c
> > [<c010b150>] handle_IRQ_event+0x38/0x5c
>
> read_lock_irqsave(&xtime_lock, flags)
> or
> write_lock_irq(&xtime_lock);

ISTR a patch from Stephen Hemminger at OSDL that used Andrea's
sequence number trick based rwlock (frlock) to implement do_gettimeofday.
It might be relevant here.

Thanks
Dipankar
-
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/

\
 
 \ /
  Last update: 2005-03-22 13:32    [W:0.339 / U:0.380 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site