When a kretprobe is installed on a kernel function, there is a maximum
limit of how many calls in parallel it can catch (aka "maxactive"). A
kernel module could call register_kretprobe() and initialize maxactive
(see example in samples/kprobes/kretprobe_example.c).
But that is not exposed to users
2017-03-27 15:34+0200, Alexander Graf:
> On 15/03/2017 22:22, Michael S. Tsirkin wrote:
>> Guests running Mac OS 5, 6, and 7 (Leopard through Lion) have a problem:
>> unless explicitly provided with kernel command line argument
>> "idlehalt=0" they'd implicitly assume MONITOR and MWAIT availability
"[PATCH v1]" I like your confidence, or lack of, that there isn't going
to be a v2 or v3 ;-)
Masami, what do you think of this?
-- Steve
On Tue, 28 Mar 2017 15:52:22 +0200
Alban Crequy wrote:
> When a kretprobe is installed on a kernel function, there is a maximum
> limit of how many calls i
On Tue, 28 Mar 2017 15:52:22 +0200
Alban Crequy wrote:
> When a kretprobe is installed on a kernel function, there is a maximum
> limit of how many calls in parallel it can catch (aka "maxactive"). A
> kernel module could call register_kretprobe() and initialize maxactive
> (see example in sample
On Wed, 29 Mar 2017 00:23:35 +0900
Masami Hiramatsu wrote:
> > @@ -598,8 +601,10 @@ static int create_trace_kprobe(int argc, char **argv)
> > {
> > /*
> > * Argument syntax:
> > -* - Add kprobe: p[:[GRP/]EVENT] [MOD:]KSYM[+OFFS]|KADDR [FETCHARGS]
> > -* - Add kretprobe: r[:[GR
Thanks for the review,
On Tue, Mar 28, 2017 at 5:23 PM, Masami Hiramatsu wrote:
> On Tue, 28 Mar 2017 15:52:22 +0200
> Alban Crequy wrote:
>
>> When a kretprobe is installed on a kernel function, there is a maximum
>> limit of how many calls in parallel it can catch (aka "maxactive"). A
>> kerne
Hi Leo,
On Sun, Mar 26, 2017 at 02:23:14AM +0800, Leo Yan wrote:
> Coresight includes debug module and usually the module connects with CPU
> debug logic. ARMv8 architecture reference manual (ARM DDI 0487A.k) has
> description for related info in "Part H: External Debug".
>
> Chapter H7 "The Samp
On Tue, Mar 28, 2017 at 7:28 AM, Radim Krčmář wrote:
> 2017-03-27 15:34+0200, Alexander Graf:
>> On 15/03/2017 22:22, Michael S. Tsirkin wrote:
>>> Guests running Mac OS 5, 6, and 7 (Leopard through Lion) have a problem:
>>> unless explicitly provided with kernel command line argument
>>> "idlehal
On Tue, 28 Mar 2017 11:34:07 -0400
Steven Rostedt wrote:
> On Wed, 29 Mar 2017 00:23:35 +0900
> Masami Hiramatsu wrote:
>
> > > @@ -598,8 +601,10 @@ static int create_trace_kprobe(int argc, char **argv)
> > > {
> > > /*
> > >* Argument syntax:
> > > - * - Add kprobe: p[:[GRP/]EVENT] [M
On Tue, 28 Mar 2017 18:08:16 +0200
Alban Crequy wrote:
> Thanks for the review,
>
> On Tue, Mar 28, 2017 at 5:23 PM, Masami Hiramatsu wrote:
> > On Tue, 28 Mar 2017 15:52:22 +0200
> > Alban Crequy wrote:
> >
> >> When a kretprobe is installed on a kernel function, there is a maximum
> >> limit
On Wed, Mar 22, 2017 at 05:59:49AM +, Amit Kama IL wrote:
> Add initial SX3000b platform related documentation to document tree:
> - Vendor prefix
> - Platform binding documentation
> - Interrupt Controller Unit binding documentation.
Probably should be 3 patches. Preferred subject prefix i
Hi Mathieu,
On Tue, Mar 28, 2017 at 10:50:10AM -0600, Mathieu Poirier wrote:
> On Sun, Mar 26, 2017 at 02:23:14AM +0800, Leo Yan wrote:
[...]
> > +static void debug_force_cpu_powered_up(struct debug_drvdata *drvdata)
> > +{
> > + int timeout = DEBUG_WAIT_TIMEOUT;
> > +
> > + drvdata->edprsr
Hi Suzuki,
On Mon, Mar 27, 2017 at 05:34:57PM +0100, Suzuki K Poulose wrote:
> On 25/03/17 18:23, Leo Yan wrote:
[...]
> Leo,
>
> Thanks a lot for the quick rework. I don't fully understand (yet!) why we
> need the
> idle_constraint. I will leave it for Sudeep to comment on it, as he is the
>
Here is a correction of patches to introduce kretprobe_instance
dynamic allocation for avoiding kretprobe silently miss-hits.
Original issue was reported by Lukasz on linux-trace ml.
https://www.spinics.net/lists/linux-trace/msg00448.html
Also Alban is working on kprobe-tracer side because of
io
Show sum of probe and retprobe nmissed count in
kprobe_profile, since retprobe can be missed even
if the kprobe itself succeeeded.
This explains user why their return probe didn't hit
sometimes.
Signed-off-by: Masami Hiramatsu
---
kernel/trace/trace_kprobe.c |2 +-
1 file changed, 1 insertio
Try to allocate kretprobe instance by GFP_ATOMIC if kretprobe's
free list is empty. This can prevent kretprobe miss-hit on the
function which can be heavily invoked and slept inside (like
locking syscall entries.)
NOTE: This may easily cause nested kprobe call which will be
just skipped (but nmiss
Limit kretprobe maximum instance up to MAXACTIVE_ALLOC.
Without this limit, kretprobe user can specify huge number
(e.g. forget to zero-fill struct kretprobe) to maxactive
and may cause out-of-memory.
Signed-off-by: Masami Hiramatsu
---
kernel/kprobes.c |3 +++
1 file changed, 3 insertions(+
* Masami Hiramatsu wrote:
> @@ -1824,6 +1823,30 @@ void unregister_jprobes(struct jprobe **jps, int num)
> EXPORT_SYMBOL_GPL(unregister_jprobes);
>
> #ifdef CONFIG_KRETPROBES
> +
> +/* Try to use free instance first, if failed, try to allocate new instance */
> +struct kretprobe_instance *kr
18 matches
Mail list logo
