On Wed, 23 May 2007, Peter Zijlstra wrote:
> On Wed, 2007-05-23 at 12:11 -0400, Jason Baron wrote:
> > On Wed, 23 May 2007, Peter Zijlstra wrote:
> >
> > > On Wed, 2007-05-23 at 10:40 -0400, Jason Baron wrote:
> > > > On Wed, 23 May 2007, Peter Zijlstra wrote:
> > > >
> > > > > Count lock
On Wed, 2007-05-23 at 12:11 -0400, Jason Baron wrote:
> On Wed, 23 May 2007, Peter Zijlstra wrote:
>
> > On Wed, 2007-05-23 at 10:40 -0400, Jason Baron wrote:
> > > On Wed, 23 May 2007, Peter Zijlstra wrote:
> > >
> > > > Count lock contention events per lock class. Additionally track the
> > >
On Wed, 23 May 2007, Peter Zijlstra wrote:
> On Wed, 2007-05-23 at 10:40 -0400, Jason Baron wrote:
> > On Wed, 23 May 2007, Peter Zijlstra wrote:
> >
> > > Count lock contention events per lock class. Additionally track the first
> > > four
> > > callsites that resulted in the contention.
> >
On Wed, 2007-05-23 at 10:40 -0400, Jason Baron wrote:
> On Wed, 23 May 2007, Peter Zijlstra wrote:
>
> > Count lock contention events per lock class. Additionally track the first
> > four
> > callsites that resulted in the contention.
> >
>
> I think that we need the total number of locking
On Wed, 23 May 2007, Peter Zijlstra wrote:
> Count lock contention events per lock class. Additionally track the first four
> callsites that resulted in the contention.
>
I think that we need the total number of locking calls, not just the total
number of *contended* locking calls, in order
Count lock contention events per lock class. Additionally track the first four
callsites that resulted in the contention.
/proc/lock_contentions - gives a single line per lock class:
<(write) contentions>
Followed by another three columns per contention point:
writing a '0' to the
Count lock contention events per lock class. Additionally track the first four
callsites that resulted in the contention.
/proc/lock_contentions - gives a single line per lock class:
class name (write) contentions read contentions
Followed by another three columns per contention point:
On Wed, 23 May 2007, Peter Zijlstra wrote:
Count lock contention events per lock class. Additionally track the first four
callsites that resulted in the contention.
I think that we need the total number of locking calls, not just the total
number of *contended* locking calls, in order to
On Wed, 2007-05-23 at 10:40 -0400, Jason Baron wrote:
On Wed, 23 May 2007, Peter Zijlstra wrote:
Count lock contention events per lock class. Additionally track the first
four
callsites that resulted in the contention.
I think that we need the total number of locking calls, not
On Wed, 23 May 2007, Peter Zijlstra wrote:
On Wed, 2007-05-23 at 10:40 -0400, Jason Baron wrote:
On Wed, 23 May 2007, Peter Zijlstra wrote:
Count lock contention events per lock class. Additionally track the first
four
callsites that resulted in the contention.
I think
On Wed, 2007-05-23 at 12:11 -0400, Jason Baron wrote:
On Wed, 23 May 2007, Peter Zijlstra wrote:
On Wed, 2007-05-23 at 10:40 -0400, Jason Baron wrote:
On Wed, 23 May 2007, Peter Zijlstra wrote:
Count lock contention events per lock class. Additionally track the
first four
On Wed, 23 May 2007, Peter Zijlstra wrote:
On Wed, 2007-05-23 at 12:11 -0400, Jason Baron wrote:
On Wed, 23 May 2007, Peter Zijlstra wrote:
On Wed, 2007-05-23 at 10:40 -0400, Jason Baron wrote:
On Wed, 23 May 2007, Peter Zijlstra wrote:
Count lock contention events per
12 matches
Mail list logo