All of lore.kernel.org
 help / color / mirror / Atom feed
* Re: syzkaller/syzbot query?
       [not found] <f4894658cfcb94263c859a57f016a06d7fb32347.camel@sipsolutions.net>
@ 2020-10-09  9:08 ` Dmitry Vyukov
  2020-10-09  9:14   ` Nikolai Kondrashov
  2020-10-09  9:35   ` Johannes Berg
  0 siblings, 2 replies; 4+ messages in thread
From: Dmitry Vyukov @ 2020-10-09  9:08 UTC (permalink / raw)
  To: Johannes Berg; +Cc: syzkaller, kernelci

On Fri, Oct 9, 2020 at 10:50 AM Johannes Berg <johannes@sipsolutions.net> wrote:
>
> Hi,
>
> Since I'm sort of drowning in syzbot reports (just recorded 23 I should
> look at), and I'm afraid I'd probably miss some.
>
> Is there a way to go filter
>
> https://syzkaller.appspot.com/upstream#open
>
> by ... something? I mean, clearly syzbot _knows_ something about where
> to send the emails, so it'd be useful if I could - in addition to having
> the emails - query that list as well, so nothing gets missed?
>
> I don't see anything obvious there, hence the question.

Hi Johannes,

Currently there is no way to cross open bugs with a particular
recipient/subsystem.
There are some outstanding related feature requests for this, but this
is not implemented yet:
https://github.com/google/syzkaller/issues/544
https://github.com/google/syzkaller/issues/550
https://github.com/google/syzkaller/issues/608

The best I can suggest is searches on lore:
https://lore.kernel.org/linux-wireless/?q=f%3Asyzbot
but this obviously does not cross it with "open" status. Lore can
filter based on dates, receivers, etc.

We recently started exporting info into KCIDB and it may solve the
problem long term, but probably not yet as well:
https://staging.kernelci.org:3000/d/home/home?orgId=1&var-origin=syzbot&var-dataset=kernelci04&from=now-90d&to=now&refresh=30m

In the end the problem is not syzbot-specific. How would you do this
if the bugs were reported by somebody else? ;)

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: syzkaller/syzbot query?
  2020-10-09  9:08 ` syzkaller/syzbot query? Dmitry Vyukov
@ 2020-10-09  9:14   ` Nikolai Kondrashov
  2020-10-09  9:35   ` Johannes Berg
  1 sibling, 0 replies; 4+ messages in thread
From: Nikolai Kondrashov @ 2020-10-09  9:14 UTC (permalink / raw)
  To: kernelci, dvyukov, Johannes Berg; +Cc: syzkaller

On 10/9/20 12:08 PM, Dmitry Vyukov via groups.io wrote:
 > On Fri, Oct 9, 2020 at 10:50 AM Johannes Berg <johannes@sipsolutions.net> wrote:
 >> Is there a way to go filter
 >>
 >> https://syzkaller.appspot.com/upstream#open
 >>
 >> by ... something? I mean, clearly syzbot _knows_ something about where
 >> to send the emails, so it'd be useful if I could - in addition to having
 >> the emails - query that list as well, so nothing gets missed?
 > We recently started exporting info into KCIDB and it may solve the
 > problem long term, but probably not yet as well:
 > https://staging.kernelci.org:3000/d/home/home?orgId=1&var-origin=syzbot&var-dataset=kernelci04&from=now-90d&to=now&refresh=30m

The ultimate plan with KCIDB is to let developers manage their own
subscriptions (what results/tests/trees/architectures/etc. they're interested
in), and naturally it then should then be possible to get a list of objects
(results) matching the subscription.

There's a lot of work left to reach that, though.

Nick

On 10/9/20 12:08 PM, Dmitry Vyukov via groups.io wrote:
 > On Fri, Oct 9, 2020 at 10:50 AM Johannes Berg <johannes@sipsolutions.net> wrote:
 >>
 >> Hi,
 >>
 >> Since I'm sort of drowning in syzbot reports (just recorded 23 I should
 >> look at), and I'm afraid I'd probably miss some.
 >>
 >> Is there a way to go filter
 >>
 >> https://syzkaller.appspot.com/upstream#open
 >>
 >> by ... something? I mean, clearly syzbot _knows_ something about where
 >> to send the emails, so it'd be useful if I could - in addition to having
 >> the emails - query that list as well, so nothing gets missed?
 >>
 >> I don't see anything obvious there, hence the question.
 >
 > Hi Johannes,
 >
 > Currently there is no way to cross open bugs with a particular
 > recipient/subsystem.
 > There are some outstanding related feature requests for this, but this
 > is not implemented yet:
 > https://github.com/google/syzkaller/issues/544
 > https://github.com/google/syzkaller/issues/550
 > https://github.com/google/syzkaller/issues/608
 >
 > The best I can suggest is searches on lore:
 > https://lore.kernel.org/linux-wireless/?q=f%3Asyzbot
 > but this obviously does not cross it with "open" status. Lore can
 > filter based on dates, receivers, etc.
 >
 > We recently started exporting info into KCIDB and it may solve the
 > problem long term, but probably not yet as well:
 > https://staging.kernelci.org:3000/d/home/home?orgId=1&var-origin=syzbot&var-dataset=kernelci04&from=now-90d&to=now&refresh=30m
 >
 > In the end the problem is not syzbot-specific. How would you do this
 > if the bugs were reported by somebody else? ;)
 >
 >
 > 
 >
 >


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: syzkaller/syzbot query?
  2020-10-09  9:08 ` syzkaller/syzbot query? Dmitry Vyukov
  2020-10-09  9:14   ` Nikolai Kondrashov
@ 2020-10-09  9:35   ` Johannes Berg
  2020-10-09 10:07     ` Dmitry Vyukov
  1 sibling, 1 reply; 4+ messages in thread
From: Johannes Berg @ 2020-10-09  9:35 UTC (permalink / raw)
  To: Dmitry Vyukov; +Cc: syzkaller, kernelci

Hi Dmitry,

> Currently there is no way to cross open bugs with a particular
> recipient/subsystem.
> There are some outstanding related feature requests for this, but this
> is not implemented yet:
> https://github.com/google/syzkaller/issues/544
> https://github.com/google/syzkaller/issues/550
> https://github.com/google/syzkaller/issues/608

OK, thanks.

> The best I can suggest is searches on lore:
> https://lore.kernel.org/linux-wireless/?q=f%3Asyzbot
> but this obviously does not cross it with "open" status. Lore can
> filter based on dates, receivers, etc.

Right. I guess I could cross-reference that with the syzbot ID from the
"open" list on syzkaller.appspot.com, but it seemed a bit odd to have to
do that :)

> We recently started exporting info into KCIDB and it may solve the
> problem long term, but probably not yet as well:
> https://staging.kernelci.org:3000/d/home/home?orgId=1&var-origin=syzbot&var-dataset=kernelci04&from=now-90d&to=now&refresh=30m

:)

> In the end the problem is not syzbot-specific. How would you do this
> if the bugs were reported by somebody else? ;)

Why, we ask them to report to bugzilla.kernel.org :)

johannes


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: syzkaller/syzbot query?
  2020-10-09  9:35   ` Johannes Berg
@ 2020-10-09 10:07     ` Dmitry Vyukov
  0 siblings, 0 replies; 4+ messages in thread
From: Dmitry Vyukov @ 2020-10-09 10:07 UTC (permalink / raw)
  To: Johannes Berg; +Cc: syzkaller, kernelci

On Fri, Oct 9, 2020 at 11:35 AM Johannes Berg <johannes@sipsolutions.net> wrote:
>
> Hi Dmitry,
>
> > Currently there is no way to cross open bugs with a particular
> > recipient/subsystem.
> > There are some outstanding related feature requests for this, but this
> > is not implemented yet:
> > https://github.com/google/syzkaller/issues/544
> > https://github.com/google/syzkaller/issues/550
> > https://github.com/google/syzkaller/issues/608
>
> OK, thanks.
>
> > The best I can suggest is searches on lore:
> > https://lore.kernel.org/linux-wireless/?q=f%3Asyzbot
> > but this obviously does not cross it with "open" status. Lore can
> > filter based on dates, receivers, etc.
>
> Right. I guess I could cross-reference that with the syzbot ID from the
> "open" list on syzkaller.appspot.com, but it seemed a bit odd to have to
> do that :)

I totally agree.

> > We recently started exporting info into KCIDB and it may solve the
> > problem long term, but probably not yet as well:
> > https://staging.kernelci.org:3000/d/home/home?orgId=1&var-origin=syzbot&var-dataset=kernelci04&from=now-90d&to=now&refresh=30m
>
> :)
>
> > In the end the problem is not syzbot-specific. How would you do this
> > if the bugs were reported by somebody else? ;)
>
> Why, we ask them to report to bugzilla.kernel.org :)

+1
Well, the kernel project creates interesting situation for maintainers
and testing systems by rejecting to create common infrastructure,
rules and tooling... That's claimed to be the right positive thing but
every reporter cannot adopt and implement custom rules of each of 2000
kernel subsystems :)
So here we are. Reporting to mailing lists looked like the main way to
report bugs for the kernel and early on Greg KH confirmed that that's
the way to go.

Okay. I am getting carried away.
Grepping the syzbot page for "80211" and maybe a few other function
prefixes ("sta") may be another reasonably practical solution for now.

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2020-10-09 10:07 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <f4894658cfcb94263c859a57f016a06d7fb32347.camel@sipsolutions.net>
2020-10-09  9:08 ` syzkaller/syzbot query? Dmitry Vyukov
2020-10-09  9:14   ` Nikolai Kondrashov
2020-10-09  9:35   ` Johannes Berg
2020-10-09 10:07     ` Dmitry Vyukov

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.