* Borislav Petkov wrote:
> > > A problem here is that mmap'ed buffer size (number of pages) must be
> > > be equal to the pre-existing buffer size and thus to be known
> > > somehow.
> >
> > What about also exporting the buffer size via sysfs pmu directory?
>
> Yes, we've been discussing buf
On Thu, Jun 27, 2013 at 02:46:36PM +0900, Namhyung Kim wrote:
> How about using 2 bits for perfsistent flag, 1 for connecting to an
> existing one, 2 for creating new one.
No need since persistent events don't need to be duplicated. Think of a
tracepoint: the samples you get there are the same, no
Hi Robert,
On Wed, 26 Jun 2013 14:44:24 +0200, Robert Richter wrote:
> On 26.06.13 13:45:38, Ingo Molnar wrote:
>> [ How about to additional logic: attr.persistent=1 && attr.config==0 means
>> a new persistent event is created straight away - no ioctl is needed to
>> detach it explicitly. ]
On 26.06.13 13:45:38, Ingo Molnar wrote:
> * Robert Richter wrote:
> > Creating a persistent event from userspace:
> >
> > * A process opens a system-wide event with the syscall and gets a fd.
>
> Should this really be limited to system-wide events?
It must not necessarily be restricted to sys
* Ingo Molnar wrote:
> Note, for tracing the PERF_FLAG_FD_OUTPUT method of multiplexing
> multiple events onto a single mmap buffers is probably useful (also
> usable via the PERF_EVENT_IOC_SET_OUTPUT ioctl()), so please make sure
> the scheme works naturally with that model as well, not just
* Robert Richter wrote:
> On 26.06.13 10:24:08, Borislav Petkov wrote:
> > On Wed, Jun 26, 2013 at 10:12:23AM +0200, Robert Richter wrote:
> > > We get a new fd by opening the persistent event with the syscall.
> > > There would be 2 new ioctls:
> > >
> > > ioctl(fd, PERF_EVENT_IOC_DETACH, 0);
On 26.06.13 10:24:08, Borislav Petkov wrote:
> On Wed, Jun 26, 2013 at 10:12:23AM +0200, Robert Richter wrote:
> > We get a new fd by opening the persistent event with the syscall.
> > There would be 2 new ioctls:
> >
> > ioctl(fd, PERF_EVENT_IOC_DETACH, 0);
> > ioctl(fd, PERF_EVENT_IOC_ATTACH,
On Wed, Jun 26, 2013 at 11:46:34AM +0200, Ingo Molnar wrote:
> If already existing persistent events show up somewhere in sysfs
> (or in a separate pseudofilesystem) then an open() of them [given
> sufficient privileges of the caller, etc.] could attach to them.
Yep, I think we want to say sys_per
* Borislav Petkov wrote:
> On Wed, Jun 26, 2013 at 10:12:23AM +0200, Robert Richter wrote:
> > We get a new fd by opening the persistent event with the syscall.
> > There would be 2 new ioctls:
> >
> > ioctl(fd, PERF_EVENT_IOC_DETACH, 0);
> > ioctl(fd, PERF_EVENT_IOC_ATTACH, 0);
> >
> > This
On Wed, Jun 26, 2013 at 10:12:23AM +0200, Robert Richter wrote:
> We get a new fd by opening the persistent event with the syscall.
> There would be 2 new ioctls:
>
> ioctl(fd, PERF_EVENT_IOC_DETACH, 0);
> ioctl(fd, PERF_EVENT_IOC_ATTACH, 0);
>
> This would be fine and reuses existing infrastru
On 25.06.13 21:16:54, Borislav Petkov wrote:
> On Tue, Jun 25, 2013 at 07:57:29PM +0200, Robert Richter wrote:
> > But what options there are to detach the event from all processes and
> > make it persistent?
>
> Something like this:
>
> ioctl(fd, PERF_EVENT_IOC_DETACH, 0);
>
> I guess thi
On Tue, Jun 25, 2013 at 07:57:29PM +0200, Robert Richter wrote:
> But what options there are to detach the event from all processes and
> make it persistent?
Something like this:
ioctl(fd, PERF_EVENT_IOC_DETACH, 0);
I guess this could simply set the persistent flag so that the rest of
th
On 24.06.13 21:45:11, Ingo Molnar wrote:
>
> * Peter Zijlstra wrote:
>
> > Oh and what Boris and Ingo said; persistent events should 'persist' and
> > not be tied to particular processes.
Fine with me too. But we need an answer how to create/release
persistent events.
It's easy to add them as
On 24.06.13 12:22:00, Peter Zijlstra wrote:
> On Tue, Jun 11, 2013 at 06:42:26PM +0200, Robert Richter wrote:
> > Note that perf tools need to support the 'attr' syntax that is
> > added in a separate patch set. With it we are able to run perf tool
> > commands to read persistent events, e.g.:
>
>
On 24.06.13 12:08:19, Peter Zijlstra wrote:
> > git://git.kernel.org/pub/scm/linux/kernel/git/rric/oprofile.git
> > persistent-v2
> >
>
> OK, so I gave up on reading the patches :/ and went and looked at the git
> tree. There's just too much needless churn in the patches.
Ok, we will rework th
* Peter Zijlstra wrote:
> On Tue, Jun 11, 2013 at 06:42:26PM +0200, Robert Richter wrote:
>
> Oh and what Boris and Ingo said; persistent events should 'persist' and
> not be tied to particular processes. I'm not sure about the entire
> eventfs thing; but the proposed sysfs thing should definit
On Tue, Jun 11, 2013 at 06:42:26PM +0200, Robert Richter wrote:
Oh and what Boris and Ingo said; persistent events should 'persist' and
not be tied to particular processes. I'm not sure about the entire
eventfs thing; but the proposed sysfs thing should definitely work for
now.
--
To unsubscribe f
On Tue, Jun 11, 2013 at 06:42:26PM +0200, Robert Richter wrote:
> The basic concept is to use a pmu as an event container for persistent
> events. The pmu registers events in sysfs and provides format and
> event information for the userland. The persistent event framework
> requires to add events
On Tue, Jun 11, 2013 at 06:42:26PM +0200, Robert Richter wrote:
> Note that perf tools need to support the 'attr' syntax that is
> added in a separate patch set. With it we are able to run perf tool
> commands to read persistent events, e.g.:
where is this patch? I can't find it.
I also find attr
On Tue, Jun 11, 2013 at 06:42:26PM +0200, Robert Richter wrote:
> This patch set implements out of the box support of perf tools for
> persistent events. For this the kernel must provide necessary
> information about existing persistent events via sysfs to userland.
> Persistent events are provided
This patch set implements out of the box support of perf tools for
persistent events. For this the kernel must provide necessary
information about existing persistent events via sysfs to userland.
Persistent events are provided by the kernel with readonly event
buffers. To allow the independent usa
21 matches
Mail list logo
