On Thu, 2012-08-09 at 22:05 -0400, Mathieu Desnoyers wrote:
> In your example, is the attribute attached to the if() or the following
> basic block ? Attaching it to the basic block allows a nice level of
> genericity:
Yes, I meant the block, not the if().
Heck, any block could be done this way,
* Steven Rostedt (rost...@goodmis.org) wrote:
> On Thu, 2012-08-09 at 16:54 -0700, David Daney wrote:
> > On 08/09/2012 04:16 PM, H. Peter Anvin wrote:
> > > On 08/09/2012 03:25 PM, Steven Rostedt wrote:
> > >>>
> > >>> It might be better to improve gcc to move really cold branches out of
> > >>> l
On Thu, 2012-08-09 at 16:54 -0700, David Daney wrote:
> On 08/09/2012 04:16 PM, H. Peter Anvin wrote:
> > On 08/09/2012 03:25 PM, Steven Rostedt wrote:
> >>>
> >>> It might be better to improve gcc to move really cold branches out of
> >>> line (really, really far away), and use the compiler to do
On 08/09/2012 04:16 PM, H. Peter Anvin wrote:
On 08/09/2012 03:25 PM, Steven Rostedt wrote:
It might be better to improve gcc to move really cold branches out of
line (really, really far away), and use the compiler to do this, rather
than to use an extra indirection that adds bloat and complexi
On 08/09/2012 03:25 PM, Steven Rostedt wrote:
>>
>> It might be better to improve gcc to move really cold branches out of
>> line (really, really far away), and use the compiler to do this, rather
>> than to use an extra indirection that adds bloat and complexity to the
>> kernel.
>
> I think modi
On Thu, 2012-08-09 at 16:50 -0400, Mathieu Desnoyers wrote:
> I guess the very first thing to do would be to benchmark this patch
> thoroughly to see if it brings significant performance improvements to
> the "tracing built-in, not enabled" case. If it does bring a significant
> improvement, then
* Steven Rostedt (rost...@goodmis.org) wrote:
> With the recent talk of performance regressions, I decided to take a
> look a tracepoints. For those that are not familiar with them, they are
> static locations scattered throughout the kernel that let developers
> trace their code when needed, with
With the recent talk of performance regressions, I decided to take a
look a tracepoints. For those that are not familiar with them, they are
static locations scattered throughout the kernel that let developers
trace their code when needed, with very little overhead when not needed
but configured in
8 matches
Mail list logo
