* Linus Torvalds ([EMAIL PROTECTED]) wrote:
>
>
> On Mon, 14 Jan 2008, Mathieu Desnoyers wrote:
> >
> > We would have to figure out if enabling -freorder-blocks-and-partition
> > makes sense kernel-wide.
>
> Last I saw, it generates crappy code, with lots more jumps back and forth,
> and the i
On Mon, 14 Jan 2008, Mathieu Desnoyers wrote:
>
> We would have to figure out if enabling -freorder-blocks-and-partition
> makes sense kernel-wide.
Last I saw, it generates crappy code, with lots more jumps back and forth,
and the image just blows up.
There's a reason we use -Os, and that's t
* Pavel Machek ([EMAIL PROTECTED]) wrote:
> On Mon 2008-01-07 13:59:54, Mathieu Desnoyers wrote:
> > * Ingo Molnar ([EMAIL PROTECTED]) wrote:
> > >
> > > * Frank Ch. Eigler <[EMAIL PROTECTED]> wrote:
> > >
> > > > > [...] this is a general policy matter. It is _so much easier_ to add
> > > > > m
On Mon 2008-01-07 13:59:54, Mathieu Desnoyers wrote:
> * Ingo Molnar ([EMAIL PROTECTED]) wrote:
> >
> > * Frank Ch. Eigler <[EMAIL PROTECTED]> wrote:
> >
> > > > [...] this is a general policy matter. It is _so much easier_ to add
> > > > markers if they _can_ have near-zero overhead (as in 1-2
This patch converts the tracing mechanism of Preempt RCU boosting into
markers. The handler functions for these markers are included inside
rcupreempt_trace.c and will be included only when PREEMPT_RCU_BOOST is
chosen.
Signed-off-by: K.Prasad <[EMAIL PROTECTED]>
---
include/linux/rcupreempt_trac
On Sat, Jan 05, 2008 at 07:46:32AM -0500, Mathieu Desnoyers wrote:
> * Gautham R Shenoy ([EMAIL PROTECTED]) wrote:
> > On Thu, Jan 03, 2008 at 11:30:55AM -0500, Mathieu Desnoyers wrote:
> > > * K. Prasad ([EMAIL PROTECTED]) wrote:
> > > > This patch converts the tracing mechanism of Preempt RCU boo
* Ingo Molnar ([EMAIL PROTECTED]) wrote:
>
> * Frank Ch. Eigler <[EMAIL PROTECTED]> wrote:
>
> > > [...] this is a general policy matter. It is _so much easier_ to add
> > > markers if they _can_ have near-zero overhead (as in 1-2
> > > instructions). Otherwise we'll keep arguing about it, espe
* Gautham R Shenoy ([EMAIL PROTECTED]) wrote:
> On Thu, Jan 03, 2008 at 11:30:55AM -0500, Mathieu Desnoyers wrote:
> > * K. Prasad ([EMAIL PROTECTED]) wrote:
> > > This patch converts the tracing mechanism of Preempt RCU boosting into
> > > markers. The handler functions for these markers are inclu
On Thu, Jan 03, 2008 at 11:30:55AM -0500, Mathieu Desnoyers wrote:
> * K. Prasad ([EMAIL PROTECTED]) wrote:
> > This patch converts the tracing mechanism of Preempt RCU boosting into
> > markers. The handler functions for these markers are included inside
> > rcupreempt_trace.c and will be included
* Ingo Molnar ([EMAIL PROTECTED]) wrote:
>
> * K. Prasad <[EMAIL PROTECTED]> wrote:
>
> > @@ -486,12 +309,12 @@ void rcu_unboost_readers(void)
> >
> > spin_lock_irqsave(&rcu_boost_wake_lock, flags);
> >
> > - rcu_trace_boost_try_unboost_readers(RCU_BOOST_ME);
> > + trace_mark(try_unbo
* K. Prasad ([EMAIL PROTECTED]) wrote:
> This patch converts the tracing mechanism of Preempt RCU boosting into
> markers. The handler functions for these markers are included inside
> rcupreempt_trace.c and will be included only when PREEMPT_RCU_BOOST is
> chosen.
>
> Signed-off-by: K.Prasad <[E
On Wed, 2008-01-02 at 11:33 -0500, Frank Ch. Eigler wrote:
> Hi -
>
> On Wed, Jan 02, 2008 at 01:47:34PM +0100, Ingo Molnar wrote:
> > [...]
> > > FWIW, I'm not keen about the format strings either, but they don't
> > > constitute a performance hit beyond an additional parameter. It does
> > >
* Frank Ch. Eigler <[EMAIL PROTECTED]> wrote:
> Hi -
>
> On Wed, Jan 02, 2008 at 06:01:57PM +0100, Ingo Molnar wrote:
> > [...]
> > well, -freorder-blocks seems to be default-enabled at -O2 on gcc 4.2, so
> > we should already be getting that, right?
>
> Right.
>
> > [...] So it would be nic
Hi -
On Wed, Jan 02, 2008 at 06:01:57PM +0100, Ingo Molnar wrote:
> [...]
> well, -freorder-blocks seems to be default-enabled at -O2 on gcc 4.2, so
> we should already be getting that, right?
Right.
> [...] So it would be nice if we could collect all this offline code
> and stuff it away into
* Frank Ch. Eigler <[EMAIL PROTECTED]> wrote:
> > [...] this is a general policy matter. It is _so much easier_ to add
> > markers if they _can_ have near-zero overhead (as in 1-2
> > instructions). Otherwise we'll keep arguing about it, especially if
> > any is added to performance-critical c
Hi -
On Wed, Jan 02, 2008 at 01:47:34PM +0100, Ingo Molnar wrote:
> [...]
> > FWIW, I'm not keen about the format strings either, but they don't
> > constitute a performance hit beyond an additional parameter. It does
> > not need to actually get parsed at run time.
>
> "only" an additional pa
* Frank Ch. Eigler <[EMAIL PROTECTED]> wrote:
> Ingo Molnar <[EMAIL PROTECTED]> writes:
>
> > [...] Firstly, why on earth does a full format string have to be
> > passed in for something as simple as a CPU id? This way we basically
> > codify it forever that tracing _has_ to be expensive when
>
Ingo Molnar <[EMAIL PROTECTED]> writes:
> [...] Firstly, why on earth does a full format string have to be
> passed in for something as simple as a CPU id? This way we basically
> codify it forever that tracing _has_ to be expensive when
> enabled. [...]
FWIW, I'm not keen about the format strin
* K. Prasad <[EMAIL PROTECTED]> wrote:
> @@ -486,12 +309,12 @@ void rcu_unboost_readers(void)
>
> spin_lock_irqsave(&rcu_boost_wake_lock, flags);
>
> - rcu_trace_boost_try_unboost_readers(RCU_BOOST_ME);
> + trace_mark(try_unboost_readers, "%u", smp_processor_id());
this isnt di
This patch converts the tracing mechanism of Preempt RCU boosting into
markers. The handler functions for these markers are included inside
rcupreempt_trace.c and will be included only when PREEMPT_RCU_BOOST is
chosen.
Signed-off-by: K.Prasad <[EMAIL PROTECTED]>
---
include/linux/rcupreempt_trac
This patch converts the tracing mechanism of Preempt RCU boosting into
markers. The handler functions for these markers are included inside
rcupreempt_trace.c and will be included only when PREEMPT_RCU_BOOST is
chosen.
Signed-off-by: K.Prasad <[EMAIL PROTECTED]>
---
include/linux/rcupreempt_trace
21 matches
Mail list logo
