On Wed, Mar 08, 2006 at 03:15:16PM -0800, Russ Blaine wrote:
> mechanism itself is also really convoluted
Sure, it's a Pentium 4 :)
> clean, responsible API. You do raise a good point, though, that perhaps
> on Pentium IV we could use PEBS underneath the current proposed APIs to
> deliver rich
John Levon wrote:
On Fri, Feb 24, 2006 at 04:37:13PM -0800, Russ Blaine wrote:
CPC in-kernel buffering adds a counter buffer such that the user can
request the kernel to throw an overflow interrupt when the counter
buffer overflows.
Has thought been given to the PEBS support in the Pentium I
On Fri, Feb 24, 2006 at 04:37:13PM -0800, Russ Blaine wrote:
> CPC in-kernel buffering adds a counter buffer such that the user can
> request the kernel to throw an overflow interrupt when the counter
> buffer overflows.
Has thought been given to the PEBS support in the Pentium IV? The
hardware
Marty Itzkowitz of the Sun Studio team had this to say:
"We would not use the buffering, because of the restriction to only accumulate
PC's from the events, and we want full callstacks. But it seems like
a good idea.
I'd suggest extending the API for counters that refer to memory
operations, to
On Fri, Feb 24, 2006 at 04:37:13PM -0800, Russ Blaine wrote:
> Here is a paper written by Willian Chen of SunLabs. He proposes a change
> to the CPC (CPU Performance Counters) subsystem whereby Solaris will
> buffer overflow events in the kernel.
>
> In the interests of soliciting feedback from
Here is a paper written by Willian Chen of SunLabs. He proposes a change
to the CPC (CPU Performance Counters) subsystem whereby Solaris will
buffer overflow events in the kernel.
In the interests of soliciting feedback from community members and being
more public about the types of changes we
