Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.kernel > #67989 > unrolled thread

Bug#964839: Still happens on 5.7.17

Back to article view | Back to linux.debian.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Bug#964839: Still happens on 5.7.17 Felix Dörre <debian@felixdoerre.de> - 2020-09-05 11:00 +0200
    Bug#964839: Still happens on 5.7.17 Salvatore Bonaccorso <carnil@debian.org> - 2020-09-06 14:50 +0200

#67989 — Bug#964839: Still happens on 5.7.17

The behavior on:
Linux chinchilla 5.7.0-3-amd64 #1 SMP Debian 5.7.17-1 (2020-08-23) 
x86_64 GNU/Linux
is still the same: for the first suspend you need to enable the graphics 
card manually. On any subsequent suspend the mechanic works.

Could someone please look into this bug? It lies around since nearly a 
month now with no reaction at all.

-- 
Kind regards,
Felix Dörre

[toc] | [next] | [standalone]

#68009

Hi Felix,

On Sat, Sep 05, 2020 at 10:53:59AM +0200, Felix Dörre wrote:
> The behavior on:
> Linux chinchilla 5.7.0-3-amd64 #1 SMP Debian 5.7.17-1 (2020-08-23) x86_64
> GNU/Linux
> is still the same: for the first suspend you need to enable the graphics
> card manually. On any subsequent suspend the mechanic works.
> 
> Could someone please look into this bug? It lies around since nearly a month
> now with no reaction at all.

There was an update for 5.8.7-1 yesterday to unstable (although the
signed packages are not yet there). Could you test that one as well?

Now if the issue is still present in 5.8.7 then if I understand you
correctly then it is okay with 5.6.14-2, but is an issue first with
5.7.6-1. 

What you can try to do is, check if to bisect between the two stable
upstream versions 5.6.14 and 5.7.6 and try to find the commit
introducing the issue. If it is convicable enough that it introduced a
bug we are probably on the lucky side (saying that because if you see
the issue only with bbswitch, and a tainted kernel upstream might ask
to reproduce the issue with an untainted kernel, or it is convicable a
bug otherwise).

Can you try those two things? So first check if 5.8.7 has the issue,
if possible then confirm as well for the current mainline, and
secondly if the issue is still present in 5.8.7, try to isolate the
commit introducing the issue beween 5.6.14 and 5.7.6.

Some help on how to do the bisect can be found in
https://wiki.debian.org/DebianKernelReportingBugs
https://wiki.debian.org/DebianKernel/GitBisect

Regards,
Salvatore

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.kernel

csiph-web