On Sat, 2007-09-08 at 01:22 -0700, David Miller wrote:
From: Daniel Walker [EMAIL PROTECTED]
Date: Fri, 07 Sep 2007 13:26:25 -0700
This driver uses the inet_lro facilities , but it doesn't force it
to be enabled .. Someone would have to know to enable inet_lro if
they select the driver
On Fri, 2007-09-07 at 19:59 -0400, Jeff Garzik wrote:
> >
> > commit 9fd380e892e078b582920325357292c07cc9
> > Author: David S. Miller <[EMAIL PROTECTED](none)>
> > Date: Thu Sep 6 21:44:36 2007 +0100
> >
> > [MYRI10GE]: Need to select INET_LRO.
> >
> > Signed-off-by: David S.
This driver uses the inet_lro facilities , but it doesn't force it
to be enabled .. Someone would have to know to enable inet_lro if
they select the driver ..
Instead, just force INET_LRO if this driver is selected..
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
drivers/net/K
On Fri, 2007-09-07 at 18:30 +0100, Denys Vlasenko wrote:
> On Friday 07 September 2007 17:31, Daniel Walker wrote:
> > On Thu, 2007-09-06 at 18:07 +0100, Denys Vlasenko wrote:
> > > A bit extended version:
> > >
> > > In the process in making it wor
On Fri, 2007-09-07 at 19:24 +0200, Sam Ravnborg wrote:
> Hi Daniel.
>
> > > I did that before I posted patches to lkml.
> > > IOW: posted patches are not broken versus module loading.
> >
> > Ok, this is more like the explanation I was looking for..
> >
> > During this thread you seemed to
On Thu, 2007-09-06 at 18:07 +0100, Denys Vlasenko wrote:
> A bit extended version:
>
> In the process in making it work I saw ~10% vmlinux size reductions
> (which basically matches what Marcelo says) when I wasn't retaining
> sections needed for EXPORT_SYMBOLs, but module loading didn't work.
>
On Thu, 2007-09-06 at 18:07 +0100, Denys Vlasenko wrote:
A bit extended version:
In the process in making it work I saw ~10% vmlinux size reductions
(which basically matches what Marcelo says) when I wasn't retaining
sections needed for EXPORT_SYMBOLs, but module loading didn't work.
On Fri, 2007-09-07 at 19:24 +0200, Sam Ravnborg wrote:
Hi Daniel.
I did that before I posted patches to lkml.
IOW: posted patches are not broken versus module loading.
Ok, this is more like the explanation I was looking for..
During this thread you seemed to indicate the patches
On Fri, 2007-09-07 at 18:30 +0100, Denys Vlasenko wrote:
On Friday 07 September 2007 17:31, Daniel Walker wrote:
On Thu, 2007-09-06 at 18:07 +0100, Denys Vlasenko wrote:
A bit extended version:
In the process in making it work I saw ~10% vmlinux size reductions
(which basically
This driver uses the inet_lro facilities , but it doesn't force it
to be enabled .. Someone would have to know to enable inet_lro if
they select the driver ..
Instead, just force INET_LRO if this driver is selected..
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
drivers/net/Kconfig
On Fri, 2007-09-07 at 19:59 -0400, Jeff Garzik wrote:
commit 9fd380e892e078b582920325357292c07cc9
Author: David S. Miller [EMAIL PROTECTED](none)
Date: Thu Sep 6 21:44:36 2007 +0100
[MYRI10GE]: Need to select INET_LRO.
Signed-off-by: David S. Miller [EMAIL
On Thu, 2007-09-06 at 11:57 +0100, Denys Vlasenko wrote:
> On Wednesday 05 September 2007 20:46, Daniel Walker wrote:
> > On Wed, 2007-09-05 at 20:49 +0100, Denys Vlasenko wrote:
> >
> > > What does "it" stand for in this sentence?
> >
> > &quo
On Thu, 2007-09-06 at 11:57 +0100, Denys Vlasenko wrote:
On Wednesday 05 September 2007 20:46, Daniel Walker wrote:
On Wed, 2007-09-05 at 20:49 +0100, Denys Vlasenko wrote:
What does it stand for in this sentence?
it is your patches, and I think we got to bottom of it .. it (i.e
On Wed, 2007-09-05 at 20:49 +0100, Denys Vlasenko wrote:
> What does "it" stand for in this sentence?
"it" is your patches, and I think we got to bottom of it .. "it" (i.e.
your patches) don't actually work with modules, which is what you
originally contended ..
> My patch was tested to work in
On Wed, 2007-09-05 at 21:31 +0200, Adrian Bunk wrote:
> On Wed, Sep 05, 2007 at 08:14:12PM +0100, Denys Vlasenko wrote:
> > On Wednesday 05 September 2007 19:38, Daniel Walker wrote:
> > > > > You version doesn't work with CONFIG_MODULES right?
> > > >
On Wed, 2007-09-05 at 20:14 +0100, Denys Vlasenko wrote:
> On Wednesday 05 September 2007 19:38, Daniel Walker wrote:
> > > > You version doesn't work with CONFIG_MODULES right?
> > >
> > > It works with CONFIG_MODULES.
> >
> > Really? Take a look a
On Wed, 2007-09-05 at 19:37 +0100, Denys Vlasenko wrote:
> On Wednesday 05 September 2007 17:29, Daniel Walker wrote:
> > On Wed, 2007-09-05 at 14:43 +0100, Denys Vlasenko wrote:
> > > Build system: section garbage collection for vmlinux
> > >
> > >
> >
On Wed, 2007-09-05 at 14:43 +0100, Denys Vlasenko wrote:
> Build system: section garbage collection for vmlinux
>
>
> Newer gcc and binutils can do dead code and data removal
> at link time. It is achieved using combination of
> -ffunction-sections -fdata-sections options for gcc and
>
On Wed, 2007-09-05 at 14:43 +0100, Denys Vlasenko wrote:
Build system: section garbage collection for vmlinux
Newer gcc and binutils can do dead code and data removal
at link time. It is achieved using combination of
-ffunction-sections -fdata-sections options for gcc and
--gc-sections
On Wed, 2007-09-05 at 19:37 +0100, Denys Vlasenko wrote:
On Wednesday 05 September 2007 17:29, Daniel Walker wrote:
On Wed, 2007-09-05 at 14:43 +0100, Denys Vlasenko wrote:
Build system: section garbage collection for vmlinux
Newer gcc and binutils can do dead code and data
On Wed, 2007-09-05 at 20:14 +0100, Denys Vlasenko wrote:
On Wednesday 05 September 2007 19:38, Daniel Walker wrote:
You version doesn't work with CONFIG_MODULES right?
It works with CONFIG_MODULES.
Really? Take a look at this version,
http://lkml.org/lkml/2006/6/4/169
On Wed, 2007-09-05 at 21:31 +0200, Adrian Bunk wrote:
On Wed, Sep 05, 2007 at 08:14:12PM +0100, Denys Vlasenko wrote:
On Wednesday 05 September 2007 19:38, Daniel Walker wrote:
You version doesn't work with CONFIG_MODULES right?
It works with CONFIG_MODULES.
Really? Take
On Wed, 2007-09-05 at 20:49 +0100, Denys Vlasenko wrote:
What does it stand for in this sentence?
it is your patches, and I think we got to bottom of it .. it (i.e.
your patches) don't actually work with modules, which is what you
originally contended ..
My patch was tested to work in my
On Tue, 2007-09-04 at 17:12 -0700, Fernando Lopez-Lezcano wrote:
> Hi Ingo... I'm getting reports from some of my Planet CCRMA users (which
> I confirmed) that the latest rt kernel I released has broken suspend
> (tested on fc6 & fc7, stock Fedora kernel works fine - the rt
> configuration files
On Tue, 2007-09-04 at 17:12 -0700, Fernando Lopez-Lezcano wrote:
Hi Ingo... I'm getting reports from some of my Planet CCRMA users (which
I confirmed) that the latest rt kernel I released has broken suspend
(tested on fc6 fc7, stock Fedora kernel works fine - the rt
configuration files are
On Mon, 2007-09-03 at 20:20 +0200, Roman Zippel wrote:
> Basically that's it and I hope that explains the basic math a bit easier. :-)
>
It helps a tiny bit .. However, I appreciate that you took the time to
write this .. Thanks you.
Daniel
-
To unsubscribe from this list: send the line
On Mon, 2007-09-03 at 20:20 +0200, Roman Zippel wrote:
Basically that's it and I hope that explains the basic math a bit easier. :-)
It helps a tiny bit .. However, I appreciate that you took the time to
write this .. Thanks you.
Daniel
-
To unsubscribe from this list: send the line
On Sun, 2007-09-02 at 16:47 +0200, Roman Zippel wrote:
> > > (1) time = sum_{t}^{T}(time_{t})
> > > (2) weight_sum = sum_{t}^{T}(weight_{t})
> >
> > I read your description, but I was distracted by this latex style
> > notation .. Could you walk through in english what these two
On Sun, 2007-09-02 at 16:47 +0200, Roman Zippel wrote:
(1) time = sum_{t}^{T}(time_{t})
(2) weight_sum = sum_{t}^{T}(weight_{t})
I read your description, but I was distracted by this latex style
notation .. Could you walk through in english what these two equation
are
On Sat, 2007-09-01 at 22:46 -0400, Matthew Wilcox wrote:
> */
> if (task == current) return 0;
>
> - if ((task->state != TASK_STOPPED) && (task->state !=
> TASK_TRACED)) {
> + if (!is_task_stopped_or_traced(task->state)) {
> DPRINT(("cannot attach to
On Fri, 2007-08-31 at 04:05 +0200, Roman Zippel wrote:
> Hi,
>
> I'm glad to announce a working prototype of the basic algorithm I
> already suggested last time.
> As I already tried to explain previously CFS has a considerable
> algorithmic and computational complexity. This patch should now
On Sat, 2007-09-01 at 23:45 +0200, Andi Kleen wrote:
> On Saturday 01 September 2007 22:54:17 Daniel Walker wrote:
> > In the case when an nmi gets stucks the endflag stays equal to zero. This
> > causes the busy looping on other cpus to continue, even tho the nmi test
> > is
On Sat, 2007-09-01 at 14:33 -0700, Stephane Eranian wrote:
> Daniel,
>
> Thanks for your help tracking down this bug. Maybe we can close
> the bugzilla report now.
No problem , and thanks for taking the time to dig into it ..
I'll close the bugzilla report.
Daniel
-
To unsubscribe from this
to checking
if the test was successful or not.
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
arch/i386/kernel/nmi.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Index: linux-2.6.22/arch/i386/kernel/nmi.c
===
---
On Sat, 2007-09-01 at 22:32 +0200, Andi Kleen wrote:
> On Saturday 01 September 2007 21:51:42 Stephane Eranian wrote:
> > Hello,
> >
> > Here is a patch to fix the NMI watchdog problem on CoreDuo processor.
> > I think we still need Daniel's patch to make the error path (when you
> > are stuck)
On Sat, 2007-09-01 at 08:51 -0700, Daniel Walker wrote:
> The warnings,
>
> drivers/char/hpet.c:72: warning: 'clocksource_hpet' defined but not used
> drivers/char/hpet.c:81: warning: 'hpet_clocksource' defined but not used
>
> from,
>
> commit 3b2b64fd311c92f2137eb7cee
p
test this?
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
drivers/char/hpet.c | 32
1 file changed, 20 insertions(+), 12 deletions(-)
Index: linux-2.6.22/drivers/char/hpet.c
===
--- linux
On Fri, 2007-08-31 at 22:59 +0200, Thomas Gleixner wrote:
> - loadavgrt fix (Luis Claudio R. Goncalves)
Sven caught that this patch was broken, I think there is an updated
one.. It's also breaking !PREEMPT_RT compiles..
dwalker2/140 failed
kernel/timer.c: In function 'calc_load':
On Sat, 2007-09-01 at 09:27 +0200, Thomas Gleixner wrote:
>
> Yeah, I know. Still I wanted to go through it myself.
>
Ok,
Here are some other trivial fixes,
http://lkml.org/lkml/2007/8/28/331
http://lkml.org/lkml/2007/8/28/334
Daniel
-
To unsubscribe from this list: send the line
On Sat, 2007-09-01 at 09:27 +0200, Thomas Gleixner wrote:
Yeah, I know. Still I wanted to go through it myself.
Ok,
Here are some other trivial fixes,
http://lkml.org/lkml/2007/8/28/331
http://lkml.org/lkml/2007/8/28/334
Daniel
-
To unsubscribe from this list: send the line unsubscribe
On Fri, 2007-08-31 at 22:59 +0200, Thomas Gleixner wrote:
- loadavgrt fix (Luis Claudio R. Goncalves)
Sven caught that this patch was broken, I think there is an updated
one.. It's also breaking !PREEMPT_RT compiles..
dwalker2/140 failed
kernel/timer.c: In function 'calc_load':
this?
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
drivers/char/hpet.c | 32
1 file changed, 20 insertions(+), 12 deletions(-)
Index: linux-2.6.22/drivers/char/hpet.c
===
--- linux-2.6.22.orig/drivers
On Sat, 2007-09-01 at 08:51 -0700, Daniel Walker wrote:
The warnings,
drivers/char/hpet.c:72: warning: 'clocksource_hpet' defined but not used
drivers/char/hpet.c:81: warning: 'hpet_clocksource' defined but not used
from,
commit 3b2b64fd311c92f2137eb7cee7025794cd854057
Author: Linus
On Sat, 2007-09-01 at 22:32 +0200, Andi Kleen wrote:
On Saturday 01 September 2007 21:51:42 Stephane Eranian wrote:
Hello,
Here is a patch to fix the NMI watchdog problem on CoreDuo processor.
I think we still need Daniel's patch to make the error path (when you
are stuck) do the right
to checking
if the test was successful or not.
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
arch/i386/kernel/nmi.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Index: linux-2.6.22/arch/i386/kernel/nmi.c
===
--- linux
On Sat, 2007-09-01 at 14:33 -0700, Stephane Eranian wrote:
Daniel,
Thanks for your help tracking down this bug. Maybe we can close
the bugzilla report now.
No problem , and thanks for taking the time to dig into it ..
I'll close the bugzilla report.
Daniel
-
To unsubscribe from this list:
On Sat, 2007-09-01 at 23:45 +0200, Andi Kleen wrote:
On Saturday 01 September 2007 22:54:17 Daniel Walker wrote:
In the case when an nmi gets stucks the endflag stays equal to zero. This
causes the busy looping on other cpus to continue, even tho the nmi test
is done.
On my machine
On Fri, 2007-08-31 at 04:05 +0200, Roman Zippel wrote:
Hi,
I'm glad to announce a working prototype of the basic algorithm I
already suggested last time.
As I already tried to explain previously CFS has a considerable
algorithmic and computational complexity. This patch should now make it
On Sat, 2007-09-01 at 22:46 -0400, Matthew Wilcox wrote:
*/
if (task == current) return 0;
- if ((task-state != TASK_STOPPED) (task-state !=
TASK_TRACED)) {
+ if (!is_task_stopped_or_traced(task-state)) {
DPRINT((cannot attach to non-stopped
On Sat, 2007-09-01 at 03:00 +0200, Björn Steinbrink wrote:
> On 2007.08.31 17:24:46 -0700, Daniel Walker wrote:
> > On Fri, 2007-08-31 at 20:06 +0200, Björn Steinbrink wrote:
> >
> >
> > > > something to do with the nmi hertz adjustment that happens
On Fri, 2007-08-31 at 20:06 +0200, Björn Steinbrink wrote:
> > something to do with the nmi hertz adjustment that happens after
> > check_nmi_watchdog() ..
>
> Hm hm, does the same thing (watchdog stuck after check) happen with
> older kernels, ie. those before Stephane's changeset that made it
oper integration
> - latency tracer changes (Daniel Walker): needs review
> - PICK_OP changes (Daniel Walker): needs review
The PICK_OP changes got reviewed by Ingo , as of,
http://marc.info/?l=linux-rt-users=118638506125380=2
They do need one small fix tho .. Below ..
Signed-off-by: Danie
On Fri, 2007-08-31 at 09:21 -0700, Stephane Eranian wrote:
> Daniel,
>
> On Fri, Aug 31, 2007 at 07:43:20AM -0700, Daniel Walker wrote:
> > On Thu, 2007-08-30 at 14:05 -0700, Stephane Eranian wrote:
> > > Daniel,
> >
> > > Yes, I realized I missed
On Thu, 2007-08-30 at 14:05 -0700, Stephane Eranian wrote:
> Daniel,
> Yes, I realized I missed a small detail in the switch statement.
> Could you try the new version?
This patch still has the stuck NMI .. Essentially the same thing that
happened without the patch..
Feel free to keep sending
On Thu, 2007-08-30 at 14:05 -0700, Stephane Eranian wrote:
Daniel,
Yes, I realized I missed a small detail in the switch statement.
Could you try the new version?
This patch still has the stuck NMI .. Essentially the same thing that
happened without the patch..
Feel free to keep sending
On Fri, 2007-08-31 at 09:21 -0700, Stephane Eranian wrote:
Daniel,
On Fri, Aug 31, 2007 at 07:43:20AM -0700, Daniel Walker wrote:
On Thu, 2007-08-30 at 14:05 -0700, Stephane Eranian wrote:
Daniel,
Yes, I realized I missed a small detail in the switch statement.
Could you try
get_monotonic_cycles() changes are racy .. It might be a little
premature to include them .. It at least fouls latency tracing on my
test machine.
- simple_irq change (Kevin Hilman): needs more thought
- RCU updates (Paul McKenney): needs proper integration
- latency tracer changes (Daniel Walker): needs review
On Fri, 2007-08-31 at 20:06 +0200, Björn Steinbrink wrote:
something to do with the nmi hertz adjustment that happens after
check_nmi_watchdog() ..
Hm hm, does the same thing (watchdog stuck after check) happen with
older kernels, ie. those before Stephane's changeset that made it use
On Sat, 2007-09-01 at 03:00 +0200, Björn Steinbrink wrote:
On 2007.08.31 17:24:46 -0700, Daniel Walker wrote:
On Fri, 2007-08-31 at 20:06 +0200, Björn Steinbrink wrote:
something to do with the nmi hertz adjustment that happens after
check_nmi_watchdog() ..
Hm hm, does
On Wed, 2007-08-29 at 14:24 -0700, Stephane Eranian wrote:
> Now on Core Duo, there is no PEBS anyway, so it is okay to use counter 0
> for NMI. The problem is that the detection code in perfctr-watchdog.c
> treats a Core Duo and a Core 2 Duo the same way as they both have the
>
On Fri, 2007-08-10 at 08:14 -0600, Gregory Haskins wrote:
> ---
>
> include/linux/sched.h |7 +--
> kernel/latency_trace.c | 18 +++---
> 2 files changed, 16 insertions(+), 9 deletions(-)
>
> diff --git a/include/linux/sched.h b/include/linux/sched.h
> index
I've updated to 2.6.23-rc4 .. This is an "Ingo appears to be busy.."
release .. I've been waiting for a -rt update , but there hasn't been
one for a while. So my tree is the best everyone will get for now.
ftp://source.mvista.com/pub/dwalker/rt/patch-2.6.23-rc4-dw1
I've updated to 2.6.23-rc4 .. This is an Ingo appears to be busy..
release .. I've been waiting for a -rt update , but there hasn't been
one for a while. So my tree is the best everyone will get for now.
ftp://source.mvista.com/pub/dwalker/rt/patch-2.6.23-rc4-dw1
On Fri, 2007-08-10 at 08:14 -0600, Gregory Haskins wrote:
---
include/linux/sched.h |7 +--
kernel/latency_trace.c | 18 +++---
2 files changed, 16 insertions(+), 9 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 8ebb43c..233d26c
On Wed, 2007-08-29 at 14:24 -0700, Stephane Eranian wrote:
Now on Core Duo, there is no PEBS anyway, so it is okay to use counter 0
for NMI. The problem is that the detection code in perfctr-watchdog.c
treats a Core Duo and a Core 2 Duo the same way as they both have the
On Tue, 2007-08-28 at 22:18 -0400, Theodore Tso wrote:
> On Mon, Aug 27, 2007 at 02:12:56PM +0200, Jes Sorensen wrote:
> > Yes, as well as 12 committee members, of which 5 didn't even comply with
> > their own git commit requirement last time I checked.
>
> Note that the git commit metric is not
On Wed, 2007-08-29 at 09:44 +1000, Nick Piggin wrote:
> Daniel Walker wrote:
> > PICK_FUNCTION() is similar to the other PICK_OP style macros, and was
> > created to replace them all. I used variable argument macros to handle
> > PICK_FUNC_2ARG/PICK_FUNC_1ARG. Otherwise t
Reaplace old PICK_OP style macros with the new PICK_FUNCTION macro.
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
include/linux/sched.h| 13 -
include/linux/spinlock.h | 345 ++-
kernel/rtmutex.c |2
lib/dec_and_
. Although the seqlock changes are larger than the originals
I think over all the patchset is worth while.
Incorporated peterz's suggestion to not require TYPE_EQUAL() to only
use pointers.
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
include/linux/pickop.h | 36 +
include
just used num_possible_cpus() . I also drop
a string which only hold "CPU" and just inserted it where it was used.
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
include/linux/latency_hist.h |1
kernel/latency_hist.c| 119 +
Replace the old PICK_OP style macros with PICK_FUNCTION. Although,
seqlocks has some alien code, which I also replaced as can be seen
from the line count below.
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
include/linux/pickop.h |4
include/linux/seqlock.h
without this the idle routine still gets traced.. This is done already
for ACPI idle , but it should also be done for other idle routines.
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
arch/i386/kernel/process.c |9 +
arch/x86_64/kernel/process.c | 10 ++
2
This enables the /proc/preempt_max_latency facility for timing modes,
even if event tracing is disabled. Wakeup latency was the only one
that had this feature in the past.
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
kernel/sysctl.c |2 +-
1 file changed, 1 insertion(+), 1 de
Just get_monotonic_cycles() switched to now() ..
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
kernel/latency_trace.c | 18 +-
1 file changed, 9 insertions(+), 9 deletions(-)
Index: linux-2.6.22/kernel/latency_t
Lines are too long..
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
---
kernel/fork.c |6 --
1 file changed, 4 insertions(+), 2 deletions(-)
Index: linux-2.6.22/kernel/fork.c
===
--- linux-2.6.22.orig/kernel/
Here's a simpler patch that fixes the boot hang ..
We have to call off the IPI looping regardless of the check_nmi_watchdog
outcome..
Signed-off-by: Daniel Walker <[EMAIL PROTECTED]>
Index: linux-2.6.22/arch/i386/kernel
s the hang (not the stuck NMI) .. Not totally sure
why, but the cpus are stuck in a loop waiting for the endflag which
never comes .. This also plays with the nmi hz which might do
something.. /proc/interrupt doesn't show any nmi's either..
Daniel
Signed-off-by: Daniel Walker <[EMAI
On Tue, 2007-08-28 at 10:05 -0700, Stephane Eranian wrote:
> Daniel,
>
> On Tue, Aug 28, 2007 at 07:34:44AM -0700, Daniel Walker wrote:
> > On Tue, 2007-08-28 at 02:12 -0700, Stephane Eranian wrote:
> > > Daniel,
> > >
> > > On Mon, Aug 27, 200
On Tue, 2007-08-28 at 02:12 -0700, Stephane Eranian wrote:
> Daniel,
>
> On Mon, Aug 27, 2007 at 04:07:54PM -0700, Daniel Walker wrote:
> > On Mon, 2007-08-27 at 15:55 -0700, Stephane Eranian wrote:
> >
> > > Yet the model name looks strange. So we need to run o
On Tue, 2007-08-28 at 12:39 +0200, John Sigler wrote:
> Why does pdflush kick in to ruin my party? :-)
> The expected latency is ~600 µs.
> http://linux.kernel.free.fr/latency/pdflush.trace
> Does ide_inb mean I'm reading from the disk?
Does your real time application lock it's memory, or allow
On Tue, 2007-08-28 at 12:39 +0200, John Sigler wrote:
Why does pdflush kick in to ruin my party? :-)
The expected latency is ~600 µs.
http://linux.kernel.free.fr/latency/pdflush.trace
Does ide_inb mean I'm reading from the disk?
Does your real time application lock it's memory, or allow
On Tue, 2007-08-28 at 02:12 -0700, Stephane Eranian wrote:
Daniel,
On Mon, Aug 27, 2007 at 04:07:54PM -0700, Daniel Walker wrote:
On Mon, 2007-08-27 at 15:55 -0700, Stephane Eranian wrote:
Yet the model name looks strange. So we need to run one more test,
as the fam/model
On Tue, 2007-08-28 at 10:05 -0700, Stephane Eranian wrote:
Daniel,
On Tue, Aug 28, 2007 at 07:34:44AM -0700, Daniel Walker wrote:
On Tue, 2007-08-28 at 02:12 -0700, Stephane Eranian wrote:
Daniel,
On Mon, Aug 27, 2007 at 04:07:54PM -0700, Daniel Walker wrote:
On Mon, 2007-08-27
in a loop waiting for the endflag which
never comes .. This also plays with the nmi hz which might do
something.. /proc/interrupt doesn't show any nmi's either..
Daniel
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
Index: linux-2.6.22/arch/i386/kernel/nmi.c
Here's a simpler patch that fixes the boot hang ..
We have to call off the IPI looping regardless of the check_nmi_watchdog
outcome..
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
Index: linux-2.6.22/arch/i386/kernel/nmi.c
Lines are too long..
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
kernel/fork.c |6 --
1 file changed, 4 insertions(+), 2 deletions(-)
Index: linux-2.6.22/kernel/fork.c
===
--- linux-2.6.22.orig/kernel/fork.c
Just get_monotonic_cycles() switched to now() ..
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
kernel/latency_trace.c | 18 +-
1 file changed, 9 insertions(+), 9 deletions(-)
Index: linux-2.6.22/kernel/latency_trace.c
without this the idle routine still gets traced.. This is done already
for ACPI idle , but it should also be done for other idle routines.
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
arch/i386/kernel/process.c |9 +
arch/x86_64/kernel/process.c | 10 ++
2 files
This enables the /proc/preempt_max_latency facility for timing modes,
even if event tracing is disabled. Wakeup latency was the only one
that had this feature in the past.
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
kernel/sysctl.c |2 +-
1 file changed, 1 insertion(+), 1 deletion
Replace the old PICK_OP style macros with PICK_FUNCTION. Although,
seqlocks has some alien code, which I also replaced as can be seen
from the line count below.
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
include/linux/pickop.h |4
include/linux/seqlock.h | 235
Reaplace old PICK_OP style macros with the new PICK_FUNCTION macro.
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
include/linux/sched.h| 13 -
include/linux/spinlock.h | 345 ++-
kernel/rtmutex.c |2
lib/dec_and_lock.c
. Although the seqlock changes are larger than the originals
I think over all the patchset is worth while.
Incorporated peterz's suggestion to not require TYPE_EQUAL() to only
use pointers.
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
include/linux/pickop.h | 36 +
include/linux
num_possible_cpus() . I also drop
a string which only hold CPU and just inserted it where it was used.
Signed-off-by: Daniel Walker [EMAIL PROTECTED]
---
include/linux/latency_hist.h |1
kernel/latency_hist.c| 119 ---
kernel/latency_trace.c
On Wed, 2007-08-29 at 09:44 +1000, Nick Piggin wrote:
Daniel Walker wrote:
PICK_FUNCTION() is similar to the other PICK_OP style macros, and was
created to replace them all. I used variable argument macros to handle
PICK_FUNC_2ARG/PICK_FUNC_1ARG. Otherwise the marcos are similar
On Tue, 2007-08-28 at 22:18 -0400, Theodore Tso wrote:
On Mon, Aug 27, 2007 at 02:12:56PM +0200, Jes Sorensen wrote:
Yes, as well as 12 committee members, of which 5 didn't even comply with
their own git commit requirement last time I checked.
Note that the git commit metric is not a
On Mon, 2007-08-27 at 15:55 -0700, Stephane Eranian wrote:
> Yet the model name looks strange. So we need to run one more test,
> as the fam/model is not enough. What we need to check is whether or
> not this processor implements architectural perfmon or not.
>
> Could you please compile and run
On Mon, 2007-08-27 at 10:54 -0700, Stephane Eranian wrote:
> Daniel,
>
> You mentioned you have a development board with a dual Pentium-M.
>
> What are the exact family/model numbers?
>
> What Bjorn points you at below is a fix that applies to
> Intel Core Duo/Solo (Yonah) and Intel Core 2
On Mon, 2007-08-27 at 19:02 +0200, Michal Piotrowski wrote:
> Indeed, now we need a predefined search "show regressions in 2.6.x" :)
>
> So the plan is simple:
> - copy all regressions into bugzilla after each release
> - make sure that all regressions reported on lkml after realase hit bugzilla
On Mon, 2007-08-27 at 09:44 -0700, Andrew Morton wrote:
> It's a hassle when someone doesn't have a bugzilla account. But there are
> humans sitting behind bugzilla handling stuff (fsvo "human"). I've already
> forwarded your bugzilla report to Stephane pointing out that he doesn't
> have an
On Mon, 2007-08-27 at 00:51 -0700, Andrew Morton wrote:
> On Mon, 27 Aug 2007 02:45:02 +0200 Michal Piotrowski <[EMAIL PROTECTED]>
> wrote:
>
> > Daniel Walker pisze:
> > [snip]
> > > Have you considered maintaining all the lists in Bugzilla?
> >
> &
501 - 600 of 1531 matches
Mail list logo