On Fri, 2020-04-10 at 12:42 -0400, Alexei Podtelezhnikov wrote:
> On Fri, Apr 10, 2020 at 11:14 AM Olivier Fourdan wrote:
> > On Fri, Apr 10, 2020 at 4:38 PM Adam Jackson wrote:
> > > On Fri, 2020-04-10 at 15:35 +0200, Olivier Fourdan wrote:
> > > > [...]
> > > > Adam, want me to launch an
On Fri, Apr 10, 2020 at 11:14 AM Olivier Fourdan wrote:
>
> On Fri, Apr 10, 2020 at 4:38 PM Adam Jackson wrote:
>>
>> On Fri, 2020-04-10 at 15:35 +0200, Olivier Fourdan wrote:
>> > [...]
>> > Adam, want me to launch an official build, just to rebuild?
>>
>> Yes, please!
>
> Oops, looks like I've
On Fri, Apr 10, 2020 at 4:38 PM Adam Jackson wrote:
> On Fri, 2020-04-10 at 15:35 +0200, Olivier Fourdan wrote:
> > [...]
> > Adam, want me to launch an official build, just to rebuild?
>
> Yes, please!
>
Oops, looks like I've been presumptuous on that one, I don't have the
rights.
So I've
On Fri, 2020-04-10 at 15:35 +0200, Olivier Fourdan wrote:
> On Fri, Apr 10, 2020 at 3:24 PM Alexei Podtelezhnikov
> wrote:
> > >>
> > >> Is koji still fc31? My problematic rebuilds are obviously fc32.
> > >
> > >
> > > All the scratch builds I spawned for this issue are for F32 of course.
> >
>
On Fri, Apr 10, 2020 at 3:24 PM Alexei Podtelezhnikov
wrote:
> >>
> >> Is koji still fc31? My problematic rebuilds are obviously fc32.
> >
> >
> > All the scratch builds I spawned for this issue are for F32 of course.
>
> I am pondering a compiler bug. Your scratch builds are fine so it
> seems.
On Fri, Apr 10, 2020 at 3:24 PM Alexei Podtelezhnikov
wrote:
> >>
> >> Is koji still fc31? My problematic rebuilds are obviously fc32.
> >
> >
> > All the scratch builds I spawned for this issue are for F32 of course.
>
> I am pondering a compiler bug. Your scratch builds are fine so it
> seems.
>>
>> Is koji still fc31? My problematic rebuilds are obviously fc32.
>
>
> All the scratch builds I spawned for this issue are for F32 of course.
I am pondering a compiler bug. Your scratch builds are fine so it
seems. What if you rebuild without koji on your desktop using shiny
Fedora 32.
On Fri, Apr 10, 2020 at 2:57 PM Alexei Podtelezhnikov
wrote:
> On Fri, Apr 10, 2020 at 8:48 AM Alexei Podtelezhnikov
> wrote:
> >
> > On Fri, Apr 10, 2020 at 7:55 AM Olivier Fourdan
> wrote:
> > > The backtrace I get with the current Fedora build is the same every
> time, but does not involve
On Fri, Apr 10, 2020 at 8:48 AM Alexei Podtelezhnikov
wrote:
>
> On Fri, Apr 10, 2020 at 7:55 AM Olivier Fourdan wrote:
> > The backtrace I get with the current Fedora build is the same every time,
> > but does not involve `sna_accel_flush()` (it crashes in
> > `kgem_buffer_release()` down
On Fri, Apr 10, 2020 at 7:55 AM Olivier Fourdan wrote:
> The backtrace I get with the current Fedora build is the same every time, but
> does not involve `sna_accel_flush()` (it crashes in `kgem_buffer_release()`
> down from `sna_accel_block()`)
That is because the debug build triggers the
Hi Alexei,
On Fri, Apr 10, 2020 at 1:54 PM Olivier Fourdan wrote:
> On Fri, Apr 10, 2020 at 1:14 PM Alexei Podtelezhnikov
> wrote:
>
[...]
>
I rebuild the driver several times recently with and without full
> debugging. Nothing is fixed yet, except that I fixed
> --enable-debug=full and found
Hi Alexei
On Fri, Apr 10, 2020 at 1:14 PM Alexei Podtelezhnikov
wrote:
> > As a test, I just rebuilt the current package (no change in the code)
> and it works fine here, no more crash:
> >
> > https://koji.fedoraproject.org/koji/taskinfo?taskID=43188723
>
> I rebuild the driver several times
> As a test, I just rebuilt the current package (no change in the code) and it
> works fine here, no more crash:
>
> https://koji.fedoraproject.org/koji/taskinfo?taskID=43188723
I rebuild the driver several times recently with and without full
debugging. Nothing is fixed yet, except that I fixed
Hi,
On Fri, Apr 10, 2020 at 10:16 AM Olivier Fourdan
wrote:
> On Fri, Apr 10, 2020 at 9:29 AM Olivier Fourdan
> wrote:
>
>> On Fri, Apr 10, 2020 at 9:09 AM Adam Williamson <
>> adamw...@fedoraproject.org> wrote:
>>
>>>
>>> I think these do all look like the same problem. If we can verify a fix
On Fri, Apr 10, 2020 at 9:29 AM Olivier Fourdan wrote:
> On Fri, Apr 10, 2020 at 9:09 AM Adam Williamson <
> adamw...@fedoraproject.org> wrote:
>
>>
>> I think these do all look like the same problem. If we can verify a fix
>> for this, an FE is probably reasonable...
>>
>>
> Looks like I can
Hi,
On Fri, Apr 10, 2020 at 9:09 AM Adam Williamson
wrote:
> On Fri, 2020-04-10 at 00:19 -0400, Alexei Podtelezhnikov wrote:
> > On Mon, Apr 6, 2020 at 12:57 PM Adam Jackson wrote:
> > > And this mostly doesn't matter because we default to a different driver
> > > for most Intel hardware
On Fri, 2020-04-10 at 00:19 -0400, Alexei Podtelezhnikov wrote:
> On Mon, Apr 6, 2020 at 12:57 PM Adam Jackson wrote:
> > And this mostly doesn't matter because we default to a different driver
> > for most Intel hardware released since about 2006, with the notable
> > exception of your
On Mon, Apr 6, 2020 at 12:57 PM Adam Jackson wrote:
> And this mostly doesn't matter because we default to a different driver
> for most Intel hardware released since about 2006, with the notable
> exception of your (Alexei's) machine which was new as of about 2010.
Please consider Freeze
On Tue, 2020-04-07 at 14:01 +0200, Nicolas Mailhot via devel wrote:
> Le mardi 07 avril 2020 à 06:31 -0400, Paul Dufresne via devel a écrit :
> > Le 20-04-06 à 21 h 01, Paul Dufresne a écrit :
> > > BTW, thanks I was searching for an example of package using a git
> > version rather than a
Le mardi 07 avril 2020 à 06:31 -0400, Paul Dufresne via devel a écrit :
> Le 20-04-06 à 21 h 01, Paul Dufresne a écrit :
> > BTW, thanks I was searching for an example of package using a git
> version rather than a released archive!
> Just for the record, *I think* the current package is a bad
|Le 20-04-06 à 21 h 01, Paul Dufresne a écrit :> BTW, thanks I was
searching for an example of package using a git||version rather than a
released archive!||
|
|Just for the record, *I think* the current package is a bad example of
a package using a forge like git.|||
||
|Current
Le 20-04-06 à 15 h 34, Adam Jackson a écrit :
On Mon, 2020-04-06 at 13:46 -0400, Alexei Podtelezhnikov wrote:
I think we have a fix for that. Let me poke at things a bit.
Have read:
https://www.phoronix.com/scan.php?page=news_item=Mesa-i915-OpenGL-2-Drop
In the link Alexei posted to Arch,
On Mon, Apr 6, 2020 at 6:03 PM Adam Jackson wrote:
> On Mon, 2020-04-06 at 16:28 -0400, Paul Dufresne via devel wrote:
> > Le 20-04-06 à 15 h 34, Adam Jackson a écrit :
> > > On Mon, 2020-04-06 at 13:46 -0400, Alexei Podtelezhnikov wrote:
> > > > Xorg does not start without xorg-x11-drv-intel on
On Mon, 2020-04-06 at 16:28 -0400, Paul Dufresne via devel wrote:
> Le 20-04-06 à 15 h 34, Adam Jackson a écrit :
> > On Mon, 2020-04-06 at 13:46 -0400, Alexei Podtelezhnikov wrote:
> > > Xorg does not start without xorg-x11-drv-intel on GMA 3150. OpenGL
> > > 2.1 used to be enabled on this
Le 20-04-06 à 15 h 34, Adam Jackson a écrit :
On Mon, 2020-04-06 at 13:46 -0400, Alexei Podtelezhnikov wrote:
Xorg does not start without xorg-x11-drv-intel on GMA 3150. OpenGL
2.1 used to be enabled on this hardware, but got dropped
I think we have a fix for that. Let me poke at things a
On Mon, 2020-04-06 at 13:46 -0400, Alexei Podtelezhnikov wrote:
>
>
> On Mon, Apr 6, 2020 at 12:57 PM Adam Jackson wrote:
> > But the driver we default to on all the newer Intel chips should work
> > on yours too, though I'm not entirely sure whether you'll end up with
> > working 3D
On Mon, 2020-04-06 at 12:13 -0400, Alexei Podtelezhnikov wrote:
> > "a011" is precisely the last entry found in
> > https://src.fedoraproject.org/rpms/xorg-x11-server/blob/HEAD/f/06_use-intel-only-on-pre-gen4.diff
> > which explains why Xorg picks "intel" as seen in the Xorg logs:
> >
> > [
On Mon, Apr 6, 2020 at 12:57 PM Adam Jackson wrote:
>
> But the driver we default to on all the newer Intel chips should work
> on yours too, though I'm not entirely sure whether you'll end up with
> working 3D acceleration. And if you uninstall xorg-x11-drv-intel,
> you'll fall back to the
On Mon, 2020-04-06 at 09:27 -0400, Alexei Podtelezhnikov wrote:
> Hi All,
>
> Please urgently downgrade xorg-x11-drv-intel before shipping Fedora 32 and
> spare users some pain. At least two very recent crash/segfault reports are
> fixed by downgrading to the fc31 version of xor
>
> "a011" is precisely the last entry found in
> https://src.fedoraproject.org/rpms/xorg-x11-server/blob/HEAD/f/06_use-intel-only-on-pre-gen4.diff
> which explains why Xorg picks "intel" as seen in the Xorg logs:
>
> [ 7821.917] (==) Matched intel as autoconfigured driver 0
> [ 7821.917] (==)
> > >Are they? We stopped using intel driver around F26, we use
> > > modesetting for Intel GPUs now. The above is clearly non-default
> > > configuratiob.
> >
> > I am far from an expert on the subject, but I am pretty sure we did not stop
> > using intel driver when we use kernel mode
Hi,
On Mon, Apr 6, 2020 at 4:57 PM Tomasz Torcz wrote:
> Yes, that's the kernel side part. But in userland,
> we are NOT using xorg-x11-drv-intel for intel gen4 and up
> (if I understand
>
> https://src.fedoraproject.org/rpms/xorg-x11-server/blob/HEAD/f/06_use-intel-only-on-pre-gen4.diff
>
On Mon, Apr 06, 2020 at 10:40:59AM -0400, Paul Dufresne via devel wrote:
> Le 20-04-06 à 10 h 04, Tomasz Torcz a écrit :
> > On Mon, Apr 06, 2020 at 09:27:58AM -0400, Alexei Podtelezhnikov wrote:
> >Are they? We stopped using intel driver around F26, we use
> > modesetting for Intel GPUs now.
Le 20-04-06 à 10 h 04, Tomasz Torcz a écrit :
On Mon, Apr 06, 2020 at 09:27:58AM -0400, Alexei Podtelezhnikov wrote:
Are they? We stopped using intel driver around F26, we use
modesetting for Intel GPUs now. The above is clearly non-default
configuratiob.
I am far from an expert on the
On Mon, Apr 6, 2020 at 10:20 AM Peter Robinson wrote:
> > > https://bugzilla.redhat.com/show_bug.cgi?id=1820815
>
> >
> > Are they? We stopped using intel driver around F26, we use
> > modesetting for Intel GPUs now. The above is clearly non-default
> > configuratiob.
>
> I thought that was
> > Hi All,
> >
> > Please urgently downgrade xorg-x11-drv-intel before shipping Fedora 32 and
> > spare users some pain. At least two very recent crash/segfault reports are
> > fixed by downgrading to the fc31 version of xorg-x11-drv-intel.
> >
> > h
On Mon, Apr 06, 2020 at 09:27:58AM -0400, Alexei Podtelezhnikov wrote:
> Hi All,
>
> Please urgently downgrade xorg-x11-drv-intel before shipping Fedora 32 and
> spare users some pain. At least two very recent crash/segfault reports are
> fixed by downgrading to the fc31 version
Hi All,
Please urgently downgrade xorg-x11-drv-intel before shipping Fedora 32 and
spare users some pain. At least two very recent crash/segfault reports are
fixed by downgrading to the fc31 version of xorg-x11-drv-intel.
https://bugzilla.redhat.com/show_bug.cgi?id=1820815
https
38 matches
Mail list logo
