compatible = "arm,ccn-502", },
> { .compatible = "arm,ccn-504", },
> + { .compatible = "arm,ccn-512", },
> {},
> };
> MODULE_DEVICE_TABLE(of, arm_ccn_match);
Acked-by: Pawel Moll
Thanks!
Pawel
On Wed, 2017-10-11 at 22:33 +0100, Kim Phillips wrote:
> Unregistering the driver before calling cpuhp_remove_multi_state()
> removes
> any remaining hotplug cpu instances so
> __cpuhp_remove_state_cpuslocked()
> doesn't emit this warning:
Cool, thanks!
I'm getting together a series of fixes now,
On Wed, 2017-10-04 at 13:32 +0100, Marc Zyngier wrote:
>
> > Acked-by: Pawel Moll
> >
> > I assume you'll get this merged yourself? Or do you want me to
> relay
> > the CCN one (I've got a couple of other small changes to the driver
> in
> > the
istration, making sure that we don't race against a CPU down
> operation.
>
> Signed-off-by: Marc Zyngier
Acked-by: Pawel Moll
I assume you'll get this merged yourself? Or do you want me to relay
the CCN one (I've got a couple of other small changes to the driver in
the queue
On Mon, 2017-07-03 at 13:01 +0530, Arvind Yadav wrote:
> attribute_groups are not supposed to change at runtime. All functions
> working with attribute_groups provided by work with const
> attribute_group. So mark the non-const structs as const.
>
> File size before:
> text data b
On Fri, 2017-06-16 at 13:39 -0700, Scott Branden wrote:
> Arnd,
>
> Should this patchset go through the ARM maintainers?
Unless something changed in the last months, these patches should go
through arm-soc. Arnd is probably just waiting for me to send a pull
request. And indeed shame on me that I
On Wed, 2017-04-19 at 09:29 +0200, SF Markus Elfring wrote:
> From: Markus Elfring
> Date: Wed, 19 Apr 2017 09:12:43 +0200
>
> Multiplications for the size determination of memory allocations
> indicated that array data structures should be processed.
> Thus use the corresponding function "devm_k
On Tue, 2017-02-14 at 17:48 +, Mark Rutland wrote:
> On Fri, Feb 10, 2017 at 12:42:47PM -0800, Velibor Markovski wrote:
> >
> > Add CCN-502 to the list of supported devices by ARM CCN PMU driver.
> >
> > Signed-off-by: Velibor Markovski
> Acked-by: Mark Rutland
>
> I assume Pawel will take
On Fri, 2017-02-03 at 18:31 -0300, Javier Martinez Canillas wrote:
> On 01/02/2017 10:02 AM, Javier Martinez Canillas wrote:
> >
> > If the driver is built as a module, autoload won't work because the module
> > alias information is not filled. So user-space can't match the registered
> > device w
On Fri, 2017-02-03 at 18:31 -0300, Javier Martinez Canillas wrote:
> On 01/02/2017 10:02 AM, Javier Martinez Canillas wrote:
> >
> > If the driver is built as a module, autoload won't work because the module
> > alias information is not filled. So user-space can't match the registered
> > device wi
: Arnd Bergmann
> Cc: Dan Carpenter
> Cc: Linus Torvalds
> Cc: Pawel Moll
> Cc: Peter Zijlstra
> Cc: Suzuki K Poulose
> Cc: Thomas Gleixner
> Signed-off-by: Anna-Maria Gleixner
> @@ -1316,7 +1309,6 @@ static void arm_ccn_pmu_cleanup(struct a
>
On Mon, 2016-07-11 at 12:28 +, Anna-Maria Gleixner wrote:
> @@ -1270,9 +1262,10 @@ static int arm_ccn_pmu_init(struct arm_c
>* ... and change the selection when it goes offline.
> Priority is
>* picked to have a chance to migrate events before perf is
> notified.
>*/
> -
On Mon, 2016-07-11 at 12:28 +, Anna-Maria Gleixner wrote:
> From: Sebastian Andrzej Siewior
>
> Install the callbacks via the state machine and let the core invoke
> the callbacks on the already online CPUs.
>
> Signed-off-by: Sebastian Andrzej Siewior
> Cc: Pawel
On Tue, 2016-03-29 at 13:30 +0100, Will Deacon wrote:
> On Tue, Mar 29, 2016 at 12:53:06PM +0100, Pawel Moll wrote:
> > Dnia 2016-03-29, Tue o godzinie 12:45 +0100, Will Deacon pisze:
> > > I'd much rather fix the driver to build as a module, if at all
> > > possibl
Dnia 2016-03-29, Tue o godzinie 12:45 +0100, Will Deacon pisze:
> I'd much rather fix the driver to build as a module, if at all
> possible.
> Suzuki (CC'd) is taking a look at that, so please drop this patch for
> now.
There's no problem with building arm-ccn.c as a module - all it's
really doing
On Thu, 2015-10-15 at 16:42 +0100, Mark Rutland wrote:
> On Thu, Oct 15, 2015 at 04:20:37PM +0100, Pawel Moll wrote:
> > On Thu, 2015-10-15 at 08:15 -0700, Drew Richardson wrote:
> > > On Thu, Oct 15, 2015 at 02:21:12PM +0100, Will Deacon wrote:
> > > > On Tue, Oc
On Thu, 2015-10-15 at 08:15 -0700, Drew Richardson wrote:
> On Thu, Oct 15, 2015 at 02:21:12PM +0100, Will Deacon wrote:
> > On Tue, Oct 13, 2015 at 08:36:45AM -0700, Drew Richardson wrote:
> > > Add additional information about the ARM architected hardware events
> > > to make counters self descri
>
> [1]: http://permalink.gmane.org/gmane.linux.kernel/2038576
>
> Signed-off-by: Andrzej Hajda
This has been already reported and discussed:
http://thread.gmane.org/gmane.linux.kernel.janitors/34058
but apparently slipped through cracks :-( therefore:
Acked-by: Pawel Moll
Thanks for b
On Wed, 2015-08-05 at 18:56 +0100, Stephen Boyd wrote:
> On 08/05/2015 03:29 AM, Pawel Moll wrote:
> >
> > Right, that's probably my fault. The sp810 nodes appeared in the VE
> > trees very early (in terms of the "DT era"), when we weren't so strict
>
On Mon, 2015-08-03 at 18:55 +0100, Stephen Boyd wrote:
> On 08/03/2015 07:01 AM, Pawel Moll wrote:
> > On Sat, 2015-08-01 at 00:44 +0100, Stephen Boyd wrote:
> >> @@ -179,6 +124,15 @@ static void __init clk_sp810_of_setup(struct
> >> device_node *node)
> >>
On Sat, 2015-08-01 at 00:44 +0100, Stephen Boyd wrote:
> @@ -179,6 +124,15 @@ static void __init clk_sp810_of_setup(struct device_node
> *node)
> sp810->timerclken[i].channel = i;
> sp810->timerclken[i].hw.init = &init;
>
> + /*
> + * If DT is
On Fri, 2015-07-31 at 18:04 +0100, Stephen Boyd wrote:
> We're removing struct clk from the clk provider API, so switch
> this code to using the clk_hw based provider APIs.
>
> Cc: Pawel Moll
> Cc: Linus Walleij
> Signed-off-by: Stephen Boyd
> ---
> drivers/
Commit-ID: 0927beeca5f9d1a7978f8da9c9d28647859816d3
Gitweb: http://git.kernel.org/tip/0927beeca5f9d1a7978f8da9c9d28647859816d3
Author: Pawel Moll
AuthorDate: Tue, 28 Jul 2015 15:10:13 +0100
Committer: Arnaldo Carvalho de Melo
CommitDate: Tue, 28 Jul 2015 13:03:49 -0300
perf tools: Fix
ir string
using make's abspath function.
Signed-off-by: Pawel Moll
---
tools/perf/config/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index b4a9c29..e8d987a 100644
--- a/tools/perf/config/Makefile
+++ b/t
On Tue, 2015-07-28 at 13:40 +0100, Pawel Moll wrote:
> Interesting. Building from a perf-tar-src-pkg tarball works:
>
> $ rm -rf /tmp/krava/ && make install prefix=/tmp/krava
> [...]
> LINK perf
> LINK libperf-gtk.so
> INSTALL GTK UI
> INST
On Mon, 2015-07-27 at 19:36 +0100, Jiri Olsa wrote:
> > @@ -150,7 +150,7 @@ ifndef NO_LIBPYTHON
> >PYTHON2 := $(if $(call get-executable,python2),python2,python)
> >override PYTHON := $(call get-executable-or-default,PYTHON,$(PYTHON2))
> >PYTHON2_CONFIG := \
> > -$(if $(call
> > ge
On Mon, 2015-07-27 at 19:24 +0100, Jiri Olsa wrote:
> I'm getting install error with your patch:
>
> [jolsa@krava perf]$ rm -rf /tmp/krava/ && make install prefix=/tmp/krava
>
> SNIP
>
> LINK perf
> INSTALL binaries
> INSTALL tests
> install: cannot create directory ‘’: No such file
r string
using make's realpath function.
Signed-off-by: Pawel Moll
---
tools/perf/config/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index b4a9c29..59e8376 100644
--- a/tools/perf/config/Makefile
+++ b/t
1: command: Illegal option -c
Although harmless, the message can be quite confusing and hard to track.
Signed-off-by: Pawel Moll
---
tools/perf/config/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index 094
On Fri, 2015-06-12 at 14:28 +0100, Pawel Moll wrote:
> On Fri, 2015-06-12 at 14:15 +0100, Adrian Hunter wrote:
> > >>>> all 3 are already part of sample_id.
> > >>>
> > >>> You have to decide whether you expect to be able to use an event without
&
On Fri, 2015-06-12 at 14:15 +0100, Adrian Hunter wrote:
> all 3 are already part of sample_id.
> >>>
> >>> You have to decide whether you expect to be able to use an event without
> >>> sample_id. MMAP and MMAP2 both have pid, tid which are in sample_id, LOST
> >>> has id, EXIT and FORK have t
k/versatile/clk-sp810.c:138:13: warning: symbol 'clk_sp810_of_setup'
> was not declared. Should it be static?
>
> Cc: Pawel Moll
> Signed-off-by: Stephen Boyd
Acked-by: Pawel Moll
> @@ -156,7 +156,7 @@ void __init clk_sp810_of_setup(struct device_node
On Thu, 2015-04-23 at 16:28 +0100, Pawel Moll wrote:
> On Wed, 2015-04-22 at 16:36 +0100, David Ahern wrote:
> > On 4/22/15 8:47 AM, Arnaldo Carvalho de Melo wrote:
> > > Em Wed, Apr 22, 2015 at 08:53:14AM -0400, Steven Rostedt escreveu:
> > >> >On Tue, 21 Ap
On Wed, 2015-04-22 at 16:36 +0100, David Ahern wrote:
> On 4/22/15 8:47 AM, Arnaldo Carvalho de Melo wrote:
> > Em Wed, Apr 22, 2015 at 08:53:14AM -0400, Steven Rostedt escreveu:
> >> >On Tue, 21 Apr 2015 21:24:51 -0500
> >> >Ron Rechenmacher wrote:
> >>> > >I've looked at the above reference brie
On Mon, 2015-03-30 at 18:37 +0100, Michael S. Tsirkin wrote:
> virtio_device_is_legacy_only is always false now,
> drop the test from virtio mmio.
>
> Signed-off-by: Michael S. Tsirkin
Slightly ironic ack ;-) after all the battle you fought for this:
Acked-by: Pawel Moll
Th
; /* Platform driver */
>
> -static struct of_device_id virtio_mmio_match[] = {
> +static const struct of_device_id virtio_mmio_match[] = {
> { .compatible = "virtio,mmio", },
> {},
> };
Acked-by: Pawel Moll
Thanks!
Pawel
--
To unsubscribe from this l
ng we can do for version 1, so return
> > generation id 0.
> >
> > Signed-off-by: Michael S. Tsirkin
> > ---
> >
> > Pawel, you mentioned you have a working virtio 1.0
> > hypervisor, can you pls confirm this works correctly?
Again, belated
Acked-by: Pawe
gt; this immediately (with your updated commit message) to my
> pending-rebases and give him the weekend to Ack.
Not that it matters any more, but a belated
Acked-by: Pawel Moll
Thanks, Michael, for fixing this! (although, I must admit, I haven't
tested it yet - all in right time :-)
Pa
On Wed, 2015-03-04 at 00:37 +, Arun Ramamurthy wrote:
> > That way you're precisely describing the way the hardware is wired up.
> > And the driver simply tries to get clcdclk first, if it's defined -
> > cool, set clksel to 1, if not - try hclk and set clksel to 0. If neither
> > of them is pr
On Tue, 2015-03-03 at 10:02 +, Pawel Moll wrote:
> On Mon, 2015-03-02 at 19:09 +, Arun Ramamurthy wrote:
> > > The existing bindings intentionally avoided quoting internal registers -
> > > they are supposed to describe how the hardware is wired up...
> > >
On Mon, 2015-03-02 at 19:09 +, Arun Ramamurthy wrote:
> > The existing bindings intentionally avoided quoting internal registers -
> > they are supposed to describe how the hardware is wired up...
> >
> > So how about something like "arm,pl11x,tft-invert-clac"? Then the driver
> > sets the bit
On Mon, 2015-03-02 at 19:09 +, Arun Ramamurthy wrote:
> On 15-03-02 08:00 AM, Pawel Moll wrote:
> > On Wed, 2015-02-25 at 21:01 +, Arun Ramamurthy wrote:
> >> Added ioctl and interrupt handler functions to support FBIO_WAITFORVSYNC
> >> Also corrected documenta
On Wed, 2015-02-25 at 21:01 +, Arun Ramamurthy wrote:
> Added code based on linaro tree:
> http://git.linaro.org/kernel/linux-linaro-stable.git
> with commit id:6846e7822c4cab5a84672baace3b768c2d0db142
> at drivers/video/amba-clcd.c. This lets the driver set
> certain tim2 register bits after r
On Wed, 2015-02-25 at 21:01 +, Arun Ramamurthy wrote:
> Added code to support FBIOPAN_DISPLAY. Also added yres_virtual
> parameter to device tree to set the virtual y resolution
>
> Reviewed-by: Ray Jui
> Reviewed-by: Scott Branden
> Signed-off-by: Arun Ramamurthy
> ---
> .../devicetree/bi
On Wed, 2015-02-25 at 21:01 +, Arun Ramamurthy wrote:
> Added ioctl and interrupt handler functions to support FBIO_WAITFORVSYNC
> Also corrected documentation to make interrupts and interrupt-names
> optional as they are not required properties.
You may not be aware of this fact, but its the
On Fri, 2015-02-20 at 14:29 +, Peter Zijlstra wrote:
> The below patch makes the distinction between these two cases by
> adding perf_event_clock() which is used for the second case. It
> further makes this configurable on a per-event basis, but adds a few
> sanity checks such that we cannot co
On Thu, 2015-02-19 at 17:58 +, Pawel Moll wrote:
> and what Adrian did, explicitly
> defining possible timestamps at perf_event_attr level instead of
> relating them to to posix clock ids is the way to go. No strong opinion
> here.
One note here: I'd rather make it "
On Thu, 2015-02-19 at 17:50 +, John Stultz wrote:
> On Thu, Feb 19, 2015 at 9:40 AM, Adrian Hunter
> wrote:
> > On 19/02/2015 7:24 p.m., John Stultz wrote:
> >>
> >> On Thu, Feb 19, 2015 at 4:11 AM, Adrian Hunter
> >> wrote:
> >>>
> >>> Hi
> >>>
> >>> With the advent of switching perf_clock
On Fri, 2015-02-13 at 02:52 +, Rusty Russell wrote:
> "Michael S. Tsirkin" writes:
> > On Tue, Feb 10, 2015 at 12:02:37PM +1030, Rusty Russell wrote:
> >> Jason Wang writes:
> >> > This patch enables the interrupt coalescing setting through ethtool.
> >>
> >> The problem is that there's noth
On Tue, 2015-02-03 at 08:30 +, ajh mls wrote:
> There is still
>
> http://marc.info/?l=linux-kernel&m=142141223902303
Uh. I have no idea why, but I haven't got this mail at all :-(
Thanks for pointing it out!
Pawel
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel"
Afternoon, Peter,
On Wed, 2015-01-21 at 20:27 +, Pawel Moll wrote:
> Until now, perf framework never defined the meaning of the timestamps
> captured as PERF_SAMPLE_TIME sample type. The values were obtaining
> from local (sched) clock, which is unavailable in userspace. This
On Mon, 2015-01-05 at 13:45 +, Peter Zijlstra wrote:
> Also, one would expect something like:
>
> default: {
> struct k_clock *kc = clockid_to_kclock(event->attr.clock);
> struct timespec ts;
> if (kc) {
> kc->clock_get(even
proc/sys/kernel/perf_sample_time_clk_id") which can be used by the
user to obtain the clk_id to be used with POSIX clock API (eg.
clock_gettime()) to obtain a time value comparable with perf samples.
Old behaviour can be restored by using "perf_use_local_clock" kernel
paramet
On Wed, 2015-01-21 at 19:48 +, Pawel Moll wrote:
> On Wed, 2015-01-21 at 15:52 +0000, Pawel Moll wrote:
> > On Mon, 2015-01-05 at 13:00 +, Peter Zijlstra wrote:
> > > On Thu, Nov 06, 2014 at 04:51:56PM +, Pawel Moll wrote:
> > > > Documentat
On Wed, 2015-01-21 at 15:52 +, Pawel Moll wrote:
> On Mon, 2015-01-05 at 13:00 +, Peter Zijlstra wrote:
> > On Thu, Nov 06, 2014 at 04:51:56PM +0000, Pawel Moll wrote:
> > > Documentation/kernel-parameters.txt | 9 +
> > > kernel/events
On Wed, 2015-01-21 at 17:44 +, John Stultz wrote:
> That said, there is the dynamic posix clockids. I'm not sure if it
> would make sense, but even if we don't bump MAX_CLOCKS, might there
> be some case where someone wants to use a dynamic posix clock for the
> perf reference?
If I remember c
On Mon, 2015-01-05 at 13:45 +, Peter Zijlstra wrote:
> On Thu, Nov 06, 2014 at 04:51:58PM +0000, Pawel Moll wrote:
> > Currently three clocks are implemented: CLOCK_REALITME = 0,
> > CLOCK_MONOTONIC = 1 and CLOCK_MONOTONIC_RAW = 2. The clock field is
> > 5 bits wid
On Mon, 2015-01-05 at 13:12 +, Peter Zijlstra wrote:
> On Thu, Nov 06, 2014 at 04:51:57PM +0000, Pawel Moll wrote:
> > This patch adds a PR_TASK_PERF_UEVENT prctl call which can be used by
> > any process to inject custom data into perf data stream as a new
> > PERF_REC
On Mon, 2015-01-05 at 13:00 +, Peter Zijlstra wrote:
> On Thu, Nov 06, 2014 at 04:51:56PM +0000, Pawel Moll wrote:
> > Documentation/kernel-parameters.txt | 9 +
> > kernel/events/core.c| 37
> > +
>
On Mon, 2015-01-05 at 13:01 +, Peter Zijlstra wrote:
> On Thu, Dec 11, 2014 at 01:39:13PM +0000, Pawel Moll wrote:
> > On Thu, 2014-11-27 at 15:05 +0000, Pawel Moll wrote:
> > > It's been 3 weeks without any negative feedback (no feedback at all, but
> >
Now, as all VE platforms have to be booted with DT,
the code handling non-DT case can be removed.
Signed-off-by: Pawel Moll
---
drivers/mfd/vexpress-sysreg.c | 71 +--
1 file changed, 15 insertions(+), 56 deletions(-)
diff --git a/drivers/mfd/vexpress
Now, as all VE platforms have to be booted with DT,
the code handling non-DT case can be removed.
Signed-off-by: Pawel Moll
---
drivers/power/reset/vexpress-poweroff.c | 17 +++--
1 file changed, 3 insertions(+), 14 deletions(-)
diff --git a/drivers/power/reset/vexpress-poweroff.c
On Thu, 2014-11-27 at 15:05 +, Pawel Moll wrote:
> On Thu, 2014-11-06 at 16:51 +0000, Pawel Moll wrote:
> > Until now, perf framework never defined the meaning of the timestamps
> > captured as PERF_SAMPLE_TIME sample type. The values were obtaining
> > from local (s
On Fri, 2014-11-28 at 17:56 +, Joe Perches wrote:
> On Wed, 2014-11-26 at 12:13 +0000, Pawel Moll wrote:
> > This patch adds a separate section for the ARM
> > Versatile Express platform maintainers, listing
> > all different bits and bobs used by it.
> >
> >
On Fri, 2014-11-28 at 16:31 +, Arnd Bergmann wrote:
> I would suggest applying your patch from
> http://thread.gmane.org/gmane.linux.ports.arm.kernel/223426/focus=224128
Yes, please :-)
Do you want me to resend?
Pawel
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel
On Fri, 2014-11-28 at 15:30 +, Russell King - ARM Linux wrote:
> mmci-pl18x 10005000.mmci: no support for card's volts
> mmc0: error -22 whilst initialising SD card
Just tried it again:
--8<
ARM Versatile Express Boot Monitor
Version:V5.2.1
Build Date: Apr 4 2
On Thu, 2014-11-06 at 16:51 +, Pawel Moll wrote:
> Until now, perf framework never defined the meaning of the timestamps
> captured as PERF_SAMPLE_TIME sample type. The values were obtaining
> from local (sched) clock, which is unavailable in userspace. This made
> it impossible
On Wed, 2014-11-26 at 12:17 +, Arnd Bergmann wrote:
> On Wednesday 26 November 2014 12:04:09 Pawel Moll wrote:
> >
> > Right, of course. Whole directory.
> >
> > > F:*/*/vexpress*
> >
> > Cool, didn't think about multiple wildcards.
&
This patch adds a separate section for the ARM
Versatile Express platform maintainers, listing
all different bits and bobs used by it.
Acked-by: Liviu Dudau
Acked-by: Sudeep Holla
Acked-by: Lorenzo Pieralisi
Signed-off-by: Pawel Moll
Acked-by: Michael Turquette
---
MAINTAINERS | 12
On Wed, 2014-11-26 at 01:59 +, Joe Perches wrote:
> On Tue, 2014-11-25 at 17:54 -0800, Mike Turquette wrote:
> > Quoting Pawel Moll (2014-11-25 10:17:35)
> > > This patch adds a separate section for the ARM
> > > Versatile Express platform maintainers, listing
> &
This patch adds a separate section for the ARM
Versatile Express platform maintainers, listing
all different bits and bobs used by it.
Acked-by: Liviu Dudau
Acked-by: Sudeep Holla
Acked-by: Lorenzo Pieralisi
Signed-off-by: Pawel Moll
---
MAINTAINERS | 19 +++
1 file changed
Turquette
Signed-off-by: Pawel Moll
---
Changes since v1:
- Removed MFD and power/reset changes from the patch;
will be posted separately after this is merged.
arch/arm/Kconfig | 2 +
arch/arm/mach-vexpress/Kconfig| 3 -
arch/arm/mac
a/drivers/video/fbdev/amba-clcd.c b/drivers/video/fbdev/amba-clcd.c
> index 6ad23bd..32c0b6b 100644
> --- a/drivers/video/fbdev/amba-clcd.c
> +++ b/drivers/video/fbdev/amba-clcd.c
> @@ -27,7 +27,6 @@
> #include
> #include
> #include
> -#include
> #include
&g
On Thu, 2014-11-13 at 09:39 +, Shannon Zhao wrote:
> When we use only virtio-mmio or vhost-net without irqfd, the device uses
> qemu_set_irq(within qemu)
> to inject interrupt and at the same time qemu update
> "VIRTIO_MMIO_INTERRUPT_STATUS" to tell guest
> driver whom this interrupt to. All
On Wed, 2014-11-12 at 08:32 +, Shannon Zhao wrote:
> On 2014/11/11 23:11, Pawel Moll wrote:
> > On Tue, 2014-11-04 at 09:35 +, Shannon Zhao wrote:
> >> As the current virtio-mmio only support single irq,
> >> so some advanced features such as vhost-net with i
On Tue, 2014-11-04 at 09:35 +, Shannon Zhao wrote:
> As the current virtio-mmio only support single irq,
> so some advanced features such as vhost-net with irqfd
> are not supported. And the net performance is not
> the best without vhost-net and irqfd supporting.
Could you, please, help under
to obtain the clk_id to be used with POSIX clock API (eg.
clock_gettime()) to obtain a time value comparable with perf samples.
Old behaviour can be restored by using "perf_use_local_clock" kernel
parameter.
Signed-off-by: Pawel Moll
---
Ingo, I remember your comments about this appro
correct debugging information.
Signed-off-by: Pawel Moll
---
Changes since v3:
- none
include/linux/perf_event.h | 4 +++
include/uapi/linux/perf_event.h | 23 -
include/uapi/linux/prctl.h | 10 ++
kernel/events/core.c| 71 +
OCK_REALITME = 0,
CLOCK_MONOTONIC = 1 and CLOCK_MONOTONIC_RAW = 2. The clock field is
5 bits wide to allow for future extension to custom, non-POSIX clock
sources(MAX_CLOCK for those is 16, see include/uapi/linux/time.h) like
ARM CoreSight (hardware trace) timestamp generator.
Signed-off-by: Pawel Moll
---
ead.gmane.org/gmane.linux.kernel/1790231
- v2: http://thread.gmane.org/gmane.linux.kernel/1793272
- v3: http://thread.gmane.org/gmane.linux.kernel.api/5681
Pawel Moll (3):
perf: Use monotonic clock as a source for timestamps
perf: Userspace event
perf: Sample additional clock value
Documenta
On Tue, 2014-11-04 at 06:33 +, Namhyung Kim wrote:
> Hi Pawel,
>
> On Tue, 4 Nov 2014 00:28:37 +0000, Pawel Moll wrote:
> > + /*
> > +* Data in userspace event record is transparent for the kernel
> > +*
> > +* Userspace perf tool code mai
On Tue, 2014-11-04 at 09:24 +, Masami Hiramatsu wrote:
> What I'd like to do is the binary version of ftrace-marker, the text
> version is already supported by qemu (see below).
> https://lists.gnu.org/archive/html/qemu-devel/2013-04/msg00505.html
>
> But since that is just a string data (not
On Tue, 2014-11-04 at 07:23 +, Peter Zijlstra wrote:
> On Tue, Nov 04, 2014 at 12:28:36AM +0000, Pawel Moll wrote:
>
> > +int sysctl_perf_sample_time_clk_id = CLOCK_MONOTONIC;
>
> const ?
Sure (unless we have to change it as mentioned below)
> > /*
> > *
On Tue, 2014-11-04 at 01:25 +, Andy Lutomirski wrote:
> >> If you're going to add double-stamped packets, can you also add a
> >> syscall to read multiple clocks at once, atomically? Or can you
> >> otherwise add a non-perf mechanism to get at this data?
> >
> > I've got some thoughts on what
From: Pawel Moll
This patch adds an option to sample value of an additional clock with
any perf event, with the the aim of allowing time correlation between
data coming from perf and other sources like hardware trace which is
timestamped with an external clock.
The idea is to generate periodic
From: Pawel Moll
This patch adds a PR_TASK_PERF_UEVENT prctl call which can be used by
any process to inject custom data into perf data stream as a new
PERF_RECORD_UEVENT record, if such process is being observed or if it
is running on a CPU being observed by the perf framework.
The prctl call
From: Pawel Moll
Hello again,
Back to the subject, this time with a slightly different angle...
I've organised a session on the subject during the tracing
minisummit at LPC 2014 in Dusseldorf. Notes taken from the discussion
taken by Steven Rostedt (thanks Steve!)
to obtain the clk_id to be used with POSIX clock API (eg.
clock_gettime()) to obtain a time value comparable with perf samples.
Signed-off-by: Pawel Moll
---
Ingo, I remember your comments about this approach in the past, but
during discussions at LPC Thomas was convinced that it's the ri
On Mon, 2014-11-03 at 14:48 +, Tomeu Vizoso wrote:
> On 29 September 2014 17:53, Pawel Moll wrote:
> > On Mon, 2014-09-29 at 16:32 +0100, Peter Zijlstra wrote:
> >> Also none of the many words above describe
> >> PERF_SAMPLE_USERSPACE_EVENT(), wth is that about?
&g
On Mon, 2014-09-29 at 16:32 +0100, Peter Zijlstra wrote:
> Also none of the many words above describe
> PERF_SAMPLE_USERSPACE_EVENT(), wth is that about?
Hopefully description of the v2 makes better job in this:
http://thread.gmane.org/gmane.linux.kernel/1793272/focus=4813
where it's already cal
On Mon, 2014-09-29 at 16:28 +0100, Peter Zijlstra wrote:
> On Thu, Sep 18, 2014 at 03:34:32PM +0100, Pawel Moll wrote:
> > @@ -4456,6 +4459,13 @@ static void __perf_event_header__init_id(struct
> > perf_event_header *header,
> > data->cpu_entry.cpu
On Sat, 2014-09-27 at 18:14 +0100, Frederic Weisbecker wrote:
> 2014-09-25 20:33 GMT+02:00 Ingo Molnar :
> >
> > * Pawel Moll wrote:
> >
> >> On Wed, 2014-09-24 at 08:49 +0100, Ingo Molnar wrote:
> >> > * Pawel Moll wrote:
> >> >
> &
On Fri, 2014-09-26 at 20:25 +0100, David Ahern wrote:
> On 9/26/14, 9:05 AM, Pawel Moll wrote:
> > To do the correlation you need both timestamps to be "taken"
> > simultaneously:
> >
> > perf event user event
> > -O
On Fri, 2014-09-26 at 15:38 +0100, Namhyung Kim wrote:
> > Then I have loads of normal normal samples, timestamped with sched clock
> > only, and every now and then one with both timestamps which then I can
> > use for time correlation. The whole point is that the frequency of such
> > "synchronisa
On Fri, 2014-09-26 at 12:23 +0100, Ingo Molnar wrote:
> > As in: allows *all* tasks to inject the data? Are you sure we
> > don't want more fine-grained control, in particular per task?
>
> Yeah. If the profiler allows it, then any task that is being
> traced can inject data.
The "that is being
On Fri, 2014-09-26 at 07:21 +0100, Namhyung Kim wrote:
> It looks like what trace-marker in ftrace does.. We might connect
> output of the trace marker into a perf event somehow.
I can probably trace_marker's write handler do the same as the new
prctl() would do. But this means that we really wan
On Fri, 2014-09-26 at 07:16 +0100, Namhyung Kim wrote:
> > It would be doable, I guess, but what
> > if someone *wants* to have sched clock as the timestamps source (because
> > it's cheap) but still be able to correlate them with userspace? In this
> > case two separate timestamps are required to
On Thu, 2014-09-25 at 19:33 +0100, Ingo Molnar wrote:
> > How would we select tasks that can write to a given buffer? Maybe an
> > ioctl() on a perf fd? Something like this?
> >
> > ioctl(perf_fd, PERF_EVENT_IOC_ENABLE_UEVENT, pid);
> > ioctl(perf_fd, PERF_EVENT_IOC_DISABLE_UEVENT, pid);
>
On Wed, 2014-09-24 at 08:49 +0100, Ingo Molnar wrote:
> * Pawel Moll wrote:
>
> > On Thu, 2014-09-18 at 15:34 +0100, Pawel Moll wrote:
> > > This patch adds a PERF_COUNT_SW_USERSPACE_EVENT type,
> > > which can be generated by user with PERF_EVENT_IOC_ENTRY
> &g
On Wed, 2014-09-24 at 07:07 +0100, Namhyung Kim wrote:
> On Tue, 23 Sep 2014 18:03:07 +0100, Pawel Moll wrote:
> > This patch adds a new PERF_COUNT_SW_UEVENT software event
> > and a related PERF_SAMPLE_UEVENT sample. User can now
> > write to the the perf file descriptor, inj
1 - 100 of 374 matches
Mail list logo