On Tue, 8 Sep 2009, Jesse Barnes wrote:
> > This regression is almost two months old, and apparently the Intel
> > graphics people DID ABSOLUTELY NOTHING about it during those two
> > months, because they couldn't be bothered to look at it.
>
> Yeah sorry, this is the first I've seen of it... I usually troll the
> regressions lists but I must have missed this one.
Hmm. We must have screwed up something, because this was bisected to the
intel DRI commits back in July. See
http://bugzilla.kernel.org/show_bug.cgi?id=13819#c4
and while there was some confusion about exactly which commit caused
it - probably because the irq thing obviously depends on timing -
Reinette had a list of three commits that he used to be able to revert to
get things going:
drm/i915: Don't update display FIFO watermark on IGDNG
drm/i915: add FIFO watermark support
drm/i915: enable error detection & state collection
So Andrew assigned it to DRI, and Rafael has had both Eric and Ma Ling on
the cc for his regression reports because of the bisection. And that has
been going on for a long time, I just checked:
Date: Sun, 26 Jul 2009 22:28:26 +0200 (CEST)
From: Rafael J. Wysocki <[email protected]>
To: Linux Kernel Mailing List <[email protected]>
Cc: Kernel Testers List <[email protected]>, Eric Anholt
<[email protected]>, "[email protected]" <[email protected]>,
Linus Torvalds <[email protected]>, Ma Ling
<[email protected]>, Reinette Chatre <[email protected]>
Subject: [Bug #13819] system freeze when switching to console
If you didn't see it, then that means that we have screw-ups with the
bugzilla thing. You're actually listed as a "Reviewed-by" on the commit
that the fixed-up bisection blamed - And I get the feeling that Rafael's
bugzilla "bugme" scripts may only pick up "Signed-off-by:" lines.
The point is: this bug has been in bisected in bugzilla for a month and a
half, and had at least two Intel DRI people cc'd on the weekly reminder
reports, along with being
Assigned To: [email protected]
We have other bugs on the regression list that are even older (no, I'm not
proud of them):
http://bugzilla.kernel.org/show_bug.cgi?id=13809
http://bugzilla.kernel.org/show_bug.cgi?id=13740
http://bugzilla.kernel.org/show_bug.cgi?id=13733
http://bugzilla.kernel.org/show_bug.cgi?id=13645
but they aren't bisected and it's not nearly as clear what is going on
there. The last one in particular I don't know if it even happens any
more and the first one seems to be fixed in -rc5, or at least the
reporter couldn't reproduce it any more..
Linus
--
To unsubscribe from this list: send the line "unsubscribe kernel-testers" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
