Re: [Nouveau] NVAC "No Signal"

2017-01-20 Thread poma
On 09.01.2017 19:20, Roy Spliet wrote:
> Op 09-01-17 om 00:24 schreef Ben Skeggs:
>> On 12/24/2016 07:48 PM, Roy Spliet wrote:
>>> I've observed this regression on my NVAC board; a 1920x1080 TV on HDMI
>>> (single monitor set-up) gets no signal with Fedora kernels from 4.8.
>>> Trace sent to the mmio dumps mailbox. A brief scan already revealed that
>>> register 0xe840 is never touched, so it appears that NVIDIA does
>>> something different.
>>> VBIOS for this board is in the usual place. Commenting out 0xac from the
>>> workaround seems to solve the problem, as tested on a Fedora 4.9 kernel.
>>> I hope that helps you get a little further with this issue. Cheers, and
>>> happy holidays!
>> Thanks Roy,
>>
>> NVIDIA told me that it applied to these boards, but, I can't see NVIDIA
>> attempting the workaround in your trace, so until there's evidence to
>> the contrary, I've disabled it for MCP7x for now.
>>
>> Ben.
> Thanks. Since your patch is the exact modification I made to verify the 
> workaround was bugging me, consider it:
> 
> Tested-by: Roy Spliet 
> 
> Given "no display on HDMI since 4.8" is quite a serious regression 
> (albeit for a small userbase), please consider submitting this to 4.10 
> as well as a "back-port" to the upstream 4.8 and 4.9 trees.
> Thanks again! Cheers,
> 
> Roy
> 


Thanks Roy

Hello Dave, Greg

If it is not already, please push for:
- stable 4.9.6
- mainline 4.10-rc5

Also:
Reported-by: poma 
Tested-by: poma 

Ref.
https://lists.freedesktop.org/archives/nouveau/2016-October/026242.html
https://github.com/skeggsb/nouveau/commit/2e5fba2.patch

___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2017-01-09 Thread Roy Spliet

Op 09-01-17 om 00:24 schreef Ben Skeggs:

On 12/24/2016 07:48 PM, Roy Spliet wrote:

I've observed this regression on my NVAC board; a 1920x1080 TV on HDMI
(single monitor set-up) gets no signal with Fedora kernels from 4.8.
Trace sent to the mmio dumps mailbox. A brief scan already revealed that
register 0xe840 is never touched, so it appears that NVIDIA does
something different.
VBIOS for this board is in the usual place. Commenting out 0xac from the
workaround seems to solve the problem, as tested on a Fedora 4.9 kernel.
I hope that helps you get a little further with this issue. Cheers, and
happy holidays!

Thanks Roy,

NVIDIA told me that it applied to these boards, but, I can't see NVIDIA
attempting the workaround in your trace, so until there's evidence to
the contrary, I've disabled it for MCP7x for now.

Ben.
Thanks. Since your patch is the exact modification I made to verify the 
workaround was bugging me, consider it:


Tested-by: Roy Spliet 

Given "no display on HDMI since 4.8" is quite a serious regression 
(albeit for a small userbase), please consider submitting this to 4.10 
as well as a "back-port" to the upstream 4.8 and 4.9 trees.

Thanks again! Cheers,

Roy




Roy.

Op 22-11-16 om 15:04 schreef poma:

On 20.10.2016 00:46, Ben Skeggs wrote:
[...]

I'd like to see a mmiotrace of the NVIDIA binary driver on a system
where this WAR breaks things.  I applied it to all the GPUs that NVIDIA
told me required it.

Ben.


Still broken, more than four months,
tested with mainline 4.9-rc6.

According to Ben Skeggs,
since this is "told" by NVIDIA i.e. by you, or perhaps some other of your 
colleagues,
what's more that I don't see progress in this regard,
would you guys from NVIDIA R mind help him to finally solve this?

Thanks.


Ref.
https://lists.freedesktop.org/archives/nouveau/2016-October/026242.html
http://goo.gl/Gm4ffO
mmiotrace-nouveau/
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau



___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2017-01-08 Thread Ben Skeggs
On 12/24/2016 07:48 PM, Roy Spliet wrote:
> I've observed this regression on my NVAC board; a 1920x1080 TV on HDMI
> (single monitor set-up) gets no signal with Fedora kernels from 4.8.
> Trace sent to the mmio dumps mailbox. A brief scan already revealed that
> register 0xe840 is never touched, so it appears that NVIDIA does
> something different.
> VBIOS for this board is in the usual place. Commenting out 0xac from the
> workaround seems to solve the problem, as tested on a Fedora 4.9 kernel.
> I hope that helps you get a little further with this issue. Cheers, and
> happy holidays!
Thanks Roy,

NVIDIA told me that it applied to these boards, but, I can't see NVIDIA
attempting the workaround in your trace, so until there's evidence to
the contrary, I've disabled it for MCP7x for now.

Ben.

> 
> Roy.
> 
> Op 22-11-16 om 15:04 schreef poma:
>> On 20.10.2016 00:46, Ben Skeggs wrote:
>> [...]
>>> I'd like to see a mmiotrace of the NVIDIA binary driver on a system
>>> where this WAR breaks things.  I applied it to all the GPUs that NVIDIA
>>> told me required it.
>>>
>>> Ben.
>>>
>> Still broken, more than four months,
>> tested with mainline 4.9-rc6.
>>
>> According to Ben Skeggs,
>> since this is "told" by NVIDIA i.e. by you, or perhaps some other of your 
>> colleagues,
>> what's more that I don't see progress in this regard,
>> would you guys from NVIDIA R mind help him to finally solve this?
>>
>> Thanks.
>>
>>
>> Ref.
>> https://lists.freedesktop.org/archives/nouveau/2016-October/026242.html
>> http://goo.gl/Gm4ffO
>> mmiotrace-nouveau/
>> ___
>> Nouveau mailing list
>> Nouveau@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/nouveau
> 



signature.asc
Description: OpenPGP digital signature
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-12-28 Thread Jeff Backus

Hi all,

My apologies - I'm trying to reply to a thread and the web interface 
doesn't play well with Mutt. I tried to hack it. Hopefully I didn't goof 
it up too much!


I am also getting a "No Signal" on the HDMI port when the resolution 
changes on boot. This is on an ION-based desktop. VGA is fine when the 
monitor is plugged in on boot, but is disabled if I plug it in 
afterwards. When the issue is observed, the X server appears to start up 
without issue. I can SSH in from another machine, set my DISPLAY 
accordingly, and then query and change resolution as if the monitor is 
behaving normally. However, monitor indicates no signal. Additionally, 
this affects both the kernel and the text console.


This is on a Fedora install. Issue was observed with kernels 4.8.4, 
4.8.13, and 4.8.15. I verified that kernels 4.5.5 and 4.7.9 work.


Additionally, this issue was observed with:
- Nouveau, version 1.0.12, I believe
- Version 340.98 of the nVidia binary driver found in the RPMFusion repo 
 (xorg-x11-drv-nvidia-340xx)
- Version 340.101 of the official nVidia binary driver, downloaded on 
 12/27/2016.


Re: information from nVidia - given that this affects both Nouveau and 
the official binary, I suspect that there is some confusion on the 
nVidia end. Perhaps a case of "mostly correct" documentation?


It looks like this bug is also related?
https://bugs.freedesktop.org/show_bug.cgi?id=98405

Have there been any updates on this since Peirre's Nov. 9th update?

Thanks!

Regards,
Jeff

--
Jeff Backus
jeff.bac...@gmail.com
http://github.com/jsbackus

___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-12-24 Thread Roy Spliet
I've observed this regression on my NVAC board; a 1920x1080 TV on HDMI 
(single monitor set-up) gets no signal with Fedora kernels from 4.8. 
Trace sent to the mmio dumps mailbox. A brief scan already revealed that 
register 0xe840 is never touched, so it appears that NVIDIA does 
something different.
VBIOS for this board is in the usual place. Commenting out 0xac from the 
workaround seems to solve the problem, as tested on a Fedora 4.9 kernel. 
I hope that helps you get a little further with this issue. Cheers, and 
happy holidays!


Roy.

Op 22-11-16 om 15:04 schreef poma:

On 20.10.2016 00:46, Ben Skeggs wrote:
[...]

I'd like to see a mmiotrace of the NVIDIA binary driver on a system
where this WAR breaks things.  I applied it to all the GPUs that NVIDIA
told me required it.

Ben.


Still broken, more than four months,
tested with mainline 4.9-rc6.

According to Ben Skeggs,
since this is "told" by NVIDIA i.e. by you, or perhaps some other of your 
colleagues,
what's more that I don't see progress in this regard,
would you guys from NVIDIA R mind help him to finally solve this?

Thanks.


Ref.
https://lists.freedesktop.org/archives/nouveau/2016-October/026242.html
http://goo.gl/Gm4ffO
mmiotrace-nouveau/
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-11-22 Thread poma
On 20.10.2016 00:46, Ben Skeggs wrote:
[...]
> I'd like to see a mmiotrace of the NVIDIA binary driver on a system
> where this WAR breaks things.  I applied it to all the GPUs that NVIDIA
> told me required it.
> 
> Ben.
> 

Still broken, more than four months,
tested with mainline 4.9-rc6.

According to Ben Skeggs,
since this is "told" by NVIDIA i.e. by you, or perhaps some other of your 
colleagues,
what's more that I don't see progress in this regard,
would you guys from NVIDIA R mind help him to finally solve this?

Thanks.


Ref.
https://lists.freedesktop.org/archives/nouveau/2016-October/026242.html
http://goo.gl/Gm4ffO
mmiotrace-nouveau/
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-11-09 Thread Pierre Moreau
I retested with both 4.8.4, Ben’s linux-4.9 and linux-4.10 branches (so only
the linux-4.10 branch had the atomic+MST patches).
For all branches, if I bootplug the external screen, I get nothing on the
external screen, and the laptop screen is frozen with some mix of images (but
the laptop itself is not locked up), nothing in dmesg. If I hotplug the screen,
the laptop screen is not frozen, but starting with linux-4.10, I get an EVO
timeout when turning off the external monitor. Adding your patch did not change
anything for me.
Looks like I’ll need to spend some more time testing and bisecting those
issues…

Pierre

On 06:12 am - Nov 08 2016, poma wrote:
> On 21.10.2016 10:56, Pierre Moreau wrote:
> > On 01:15 am - Oct 21 2016, Lukas Wunner wrote:
> >> On Thu, Oct 20, 2016 at 10:08:28AM +0200, Lukas Wunner wrote:
> >>> On Wed, Oct 19, 2016 at 07:58:06PM +0200, Pierre Moreau wrote:
>  For example, my laptop (which also has an NVAC) has been triggering the
>  no-signal message on external monitors way before Ben???s patch landed,
>  but only for some adapters. I haven???t tried Ben???s patch yet, nor
>  yours, but I will certainly do it, and see what effect each of them has.
> >>>
> >>> The external DP port on your MBP5,3 is switchable between GPUs and
> >>> the apple-gmux driver switches it in unison with the panel.  Thus
> >>> the NVAC cannot drive external displays when gmux is switched to
> >>> the MCP79.  (You probably were aware of this, just wanted to mention
> >>   ^
> >> I meant G96, sorry I mixed it up.
> >>
> >> Lukas
> > 
> > Yes, that bit had stayed in my memory, that switching between the two GPUs
> > would not only switch them for the laptop screen, but for the external ones 
> > as
> > well. IIRC, I am getting the no signal in both cases, but I need to retest.
> > 
> > Pierre
> 
> Any news related on your side?
> 
> 
> 


signature.asc
Description: PGP signature
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-11-07 Thread poma
On 21.10.2016 10:56, Pierre Moreau wrote:
> On 01:15 am - Oct 21 2016, Lukas Wunner wrote:
>> On Thu, Oct 20, 2016 at 10:08:28AM +0200, Lukas Wunner wrote:
>>> On Wed, Oct 19, 2016 at 07:58:06PM +0200, Pierre Moreau wrote:
 For example, my laptop (which also has an NVAC) has been triggering the
 no-signal message on external monitors way before Ben???s patch landed,
 but only for some adapters. I haven???t tried Ben???s patch yet, nor
 yours, but I will certainly do it, and see what effect each of them has.
>>>
>>> The external DP port on your MBP5,3 is switchable between GPUs and
>>> the apple-gmux driver switches it in unison with the panel.  Thus
>>> the NVAC cannot drive external displays when gmux is switched to
>>> the MCP79.  (You probably were aware of this, just wanted to mention
>>   ^
>> I meant G96, sorry I mixed it up.
>>
>> Lukas
> 
> Yes, that bit had stayed in my memory, that switching between the two GPUs
> would not only switch them for the laptop screen, but for the external ones as
> well. IIRC, I am getting the no signal in both cases, but I need to retest.
> 
> Pierre

Any news related on your side?



___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-11-07 Thread poma
On 06.11.2016 18:02, poma wrote:
> 
> http://goo.gl/Gm4ffO
> mmiotrace-nouveau/
> 

$ uname -r
4.9.0-0.rc4.git0.1.fc26.x86_64+debug

$ dmesg -t | grep -P '(?=.*nouveau)(?=.*MMIO)'
nouveau :01:00.0: bus: MMIO write of 8015 FAULT at 61a804
nouveau :01:00.0: bus: MMIO read of 0010 FAULT at 64
nouveau :01:00.0: bus: MMIO read of 0007 FAULT at 61a804
nouveau :01:00.0: bus: MMIO write of 8015 FAULT at 61a804
nouveau :01:00.0: bus: MMIO read of 07ff FAULT at 61a804
nouveau :01:00.0: bus: MMIO write of 00100180 FAULT at 61a80c
nouveau :01:00.0: bus: MMIO read of 8055 FAULT at 61a804
nouveau :01:00.0: bus: MMIO read of 0007 FAULT at 61a804
nouveau :01:00.0: bus: MMIO write of 8015 FAULT at 61a804
nouveau :01:00.0: bus: MMIO read of 0001 FAULT at 61a804
nouveau :01:00.0: bus: MMIO write of 00100180 FAULT at 61a80c
nouveau :01:00.0: bus: MMIO read of  FAULT at 61a804
nouveau :01:00.0: bus: MMIO write of fb08d1ff FAULT at 61a804
nouveau :01:00.0: bus: MMIO read of 0001 FAULT at 641000
nouveau :01:00.0: bus: MMIO write of 0020 FAULT at 641000
nouveau :01:00.0: bus: MMIO read of 0001 FAULT at 64
nouveau :01:00.0: bus: MMIO read of 0030 FAULT at 64
nouveau :01:00.0: bus: MMIO write of 0014 FAULT at 64
nouveau :01:00.0: bus: MMIO write of 006c FAULT at 641000
nouveau :01:00.0: bus: MMIO read of 0007 FAULT at 64
nouveau :01:00.0: bus: MMIO write of 0010 FAULT at 64
nouveau :01:00.0: bus: MMIO write of  FAULT at 647080
nouveau :01:00.0: bus: MMIO write of 0008 FAULT at 64
nouveau :01:00.0: bus: MMIO read of  FAULT at 64
nouveau :01:00.0: bus: MMIO read of 0014 FAULT at 64
nouveau :01:00.0: bus: MMIO write of  FAULT at 647080
nouveau :01:00.0: bus: MMIO read of  FAULT at 64
nouveau :01:00.0: bus: MMIO write of  FAULT at 647080
nouveau :01:00.0: bus: MMIO read of  FAULT at 64
nouveau :01:00.0: bus: MMIO read of 0007 FAULT at 64
nouveau :01:00.0: bus: MMIO write of  FAULT at 647080
nouveau :01:00.0: bus: MMIO read of  FAULT at 64
nouveau :01:00.0: bus: MMIO write of  FAULT at 647080
nouveau :01:00.0: bus: MMIO read of  FAULT at 64
nouveau :01:00.0: bus: MMIO read of 0014 FAULT at 64
nouveau :01:00.0: bus: MMIO write of  FAULT at 647080


___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-11-06 Thread poma

http://goo.gl/Gm4ffO
mmiotrace-nouveau/


___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-21 Thread Pierre Moreau
On 01:15 am - Oct 21 2016, Lukas Wunner wrote:
> On Thu, Oct 20, 2016 at 10:08:28AM +0200, Lukas Wunner wrote:
> > On Wed, Oct 19, 2016 at 07:58:06PM +0200, Pierre Moreau wrote:
> > > For example, my laptop (which also has an NVAC) has been triggering the
> > > no-signal message on external monitors way before Ben???s patch landed,
> > > but only for some adapters. I haven???t tried Ben???s patch yet, nor
> > > yours, but I will certainly do it, and see what effect each of them has.
> > 
> > The external DP port on your MBP5,3 is switchable between GPUs and
> > the apple-gmux driver switches it in unison with the panel.  Thus
> > the NVAC cannot drive external displays when gmux is switched to
> > the MCP79.  (You probably were aware of this, just wanted to mention
>   ^
> I meant G96, sorry I mixed it up.
> 
> Lukas

Yes, that bit had stayed in my memory, that switching between the two GPUs
would not only switch them for the laptop screen, but for the external ones as
well. IIRC, I am getting the no signal in both cases, but I need to retest.

Pierre

> 
> > it in case you weren't.)  (In theory we could change vga_switcheroo
> > and apple-gmux to support switching the panel and the external port
> > separately rather than in unison if there is demand.  It's only
> > supported on MBPs introduced from 2008 to 2010 though, the 2011+
> > models can only drive external ports with the discrete GPU.)


signature.asc
Description: PGP signature
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-20 Thread Lukas Wunner
On Thu, Oct 20, 2016 at 10:08:28AM +0200, Lukas Wunner wrote:
> On Wed, Oct 19, 2016 at 07:58:06PM +0200, Pierre Moreau wrote:
> > For example, my laptop (which also has an NVAC) has been triggering the
> > no-signal message on external monitors way before Ben???s patch landed,
> > but only for some adapters. I haven???t tried Ben???s patch yet, nor
> > yours, but I will certainly do it, and see what effect each of them has.
> 
> The external DP port on your MBP5,3 is switchable between GPUs and
> the apple-gmux driver switches it in unison with the panel.  Thus
> the NVAC cannot drive external displays when gmux is switched to
> the MCP79.  (You probably were aware of this, just wanted to mention
  ^
I meant G96, sorry I mixed it up.

Lukas

> it in case you weren't.)  (In theory we could change vga_switcheroo
> and apple-gmux to support switching the panel and the external port
> separately rather than in unison if there is demand.  It's only
> supported on MBPs introduced from 2008 to 2010 though, the 2011+
> models can only drive external ports with the discrete GPU.)
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-20 Thread Lukas Wunner
On Wed, Oct 19, 2016 at 07:58:06PM +0200, Pierre Moreau wrote:
> For example, my laptop (which also has an NVAC) has been triggering the
> no-signal message on external monitors way before Ben???s patch landed,
> but only for some adapters. I haven???t tried Ben???s patch yet, nor
> yours, but I will certainly do it, and see what effect each of them has.

The external DP port on your MBP5,3 is switchable between GPUs and
the apple-gmux driver switches it in unison with the panel.  Thus
the NVAC cannot drive external displays when gmux is switched to
the MCP79.  (You probably were aware of this, just wanted to mention
it in case you weren't.)  (In theory we could change vga_switcheroo
and apple-gmux to support switching the panel and the external port
separately rather than in unison if there is demand.  It's only
supported on MBPs introduced from 2008 to 2010 though, the 2011+
models can only drive external ports with the discrete GPU.)

Best regards,

Lukas
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-19 Thread Ben Skeggs


On 10/20/2016 03:58 AM, Pierre Moreau wrote:
> Hello,
> 
> On 07:37 pm - Oct 19 2016, poma wrote:
>> On 19.10.2016 17:03, Karol Herbst wrote:
>>
>>> You don't get why I try to say. We have to actually find out when to
>>> apply this workaround, not to create some silly whitelist/blacklist.
>>> It's the last option, we never want to actually use.
>>>
>>
>> Well if you do not say, who can understand!? :)
>> Besides, you can mock with "silly" whitelist/blacklist", however there is 
>> nothing wrong with the method as such, it is used practically everywhere.
>>
>>> And even if we would have to create such lists, who tells us, that if
>>> affects every GPU with your device id? Usually quirks are applied
>>> depending on the sub-vendor-id and sub-device-id if actually required.
>>>
>>> In the end we need something like this: If byte X in table Y is set in
>>> the vbios or if bits A-B in reg Z in the MMIO space are set to
>>> whatever, we have to apply that workaround.
>>>
>>> In the end we should also wait until Ben replies, because he might
>>> know the exact reasons why this workaround was actually needed.
I'd like to see a mmiotrace of the NVIDIA binary driver on a system
where this WAR breaks things.  I applied it to all the GPUs that NVIDIA
told me required it.

Ben.

>>>
>>
>> If you eager to leave it broken even more than three months that have 
>> already been passed since the original commit ...
>>
>>> We might have a GPU with the same chipset like yours and we might be
>>> able to verify the issue
>>>
>>
>> Ah, I see.
>> You do not have confidence in my test results, good to know.
> 
> Testing on another GPU with the same chipset does not mean he does not trust
> your results. For example, my laptop (which also has an NVAC) has been
> triggering the no-signal message on external monitors way before Ben’s patch
> landed, but only for some adapters. I haven’t tried Ben’s patch yet, nor 
> yours,
> but I will certainly do it, and see what effect each of them has.
> 
> Regards,
> Pierre
> 
>>
>> Oh! Carol
>>
>>
>> ___
>> Nouveau mailing list
>> Nouveau@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/nouveau
>>
>>
>> ___
>> Nouveau mailing list
>> Nouveau@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/nouveau



signature.asc
Description: OpenPGP digital signature
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-19 Thread Pierre Moreau
Hello,

On 07:37 pm - Oct 19 2016, poma wrote:
> On 19.10.2016 17:03, Karol Herbst wrote:
> 
> > You don't get why I try to say. We have to actually find out when to
> > apply this workaround, not to create some silly whitelist/blacklist.
> > It's the last option, we never want to actually use.
> > 
> 
> Well if you do not say, who can understand!? :)
> Besides, you can mock with "silly" whitelist/blacklist", however there is 
> nothing wrong with the method as such, it is used practically everywhere.
> 
> > And even if we would have to create such lists, who tells us, that if
> > affects every GPU with your device id? Usually quirks are applied
> > depending on the sub-vendor-id and sub-device-id if actually required.
> > 
> > In the end we need something like this: If byte X in table Y is set in
> > the vbios or if bits A-B in reg Z in the MMIO space are set to
> > whatever, we have to apply that workaround.
> > 
> > In the end we should also wait until Ben replies, because he might
> > know the exact reasons why this workaround was actually needed.
> > 
> 
> If you eager to leave it broken even more than three months that have already 
> been passed since the original commit ...
> 
> > We might have a GPU with the same chipset like yours and we might be
> > able to verify the issue
> > 
> 
> Ah, I see.
> You do not have confidence in my test results, good to know.

Testing on another GPU with the same chipset does not mean he does not trust
your results. For example, my laptop (which also has an NVAC) has been
triggering the no-signal message on external monitors way before Ben’s patch
landed, but only for some adapters. I haven’t tried Ben’s patch yet, nor yours,
but I will certainly do it, and see what effect each of them has.

Regards,
Pierre

> 
> Oh! Carol
> 
> 
> ___
> Nouveau mailing list
> Nouveau@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/nouveau


signature.asc
Description: PGP signature
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-19 Thread poma
On 19.10.2016 17:03, Karol Herbst wrote:

> You don't get why I try to say. We have to actually find out when to
> apply this workaround, not to create some silly whitelist/blacklist.
> It's the last option, we never want to actually use.
> 

Well if you do not say, who can understand!? :)
Besides, you can mock with "silly" whitelist/blacklist", however there is 
nothing wrong with the method as such, it is used practically everywhere.

> And even if we would have to create such lists, who tells us, that if
> affects every GPU with your device id? Usually quirks are applied
> depending on the sub-vendor-id and sub-device-id if actually required.
> 
> In the end we need something like this: If byte X in table Y is set in
> the vbios or if bits A-B in reg Z in the MMIO space are set to
> whatever, we have to apply that workaround.
> 
> In the end we should also wait until Ben replies, because he might
> know the exact reasons why this workaround was actually needed.
> 

If you eager to leave it broken even more than three months that have already 
been passed since the original commit ...

> We might have a GPU with the same chipset like yours and we might be
> able to verify the issue
> 

Ah, I see.
You do not have confidence in my test results, good to know.

Oh! Carol


___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-19 Thread Karol Herbst
2016-10-19 16:47 GMT+02:00 poma :
> On 18.10.2016 16:02, Karol Herbst wrote:
>> well, I just don't want that this fix breaks the same thing for other
>> users, that's why I am asking.
>>
>
> Affected device ID:
> https://github.com/skeggsb/nouveau/blob/master/drm/nouveau/nvkm/engine/device/pci.c#L1229
> can it be excluded from device->chipset case 0xac ?
>
> Care to create a patch?
> I'll test it.
>
>

You don't get why I try to say. We have to actually find out when to
apply this workaround, not to create some silly whitelist/blacklist.
It's the last option, we never want to actually use.

And even if we would have to create such lists, who tells us, that if
affects every GPU with your device id? Usually quirks are applied
depending on the sub-vendor-id and sub-device-id if actually required.

In the end we need something like this: If byte X in table Y is set in
the vbios or if bits A-B in reg Z in the MMIO space are set to
whatever, we have to apply that workaround.

In the end we should also wait until Ben replies, because he might
know the exact reasons why this workaround was actually needed.

We might have a GPU with the same chipset like yours and we might be
able to verify the issue
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-19 Thread poma
On 18.10.2016 16:02, Karol Herbst wrote:
> well, I just don't want that this fix breaks the same thing for other
> users, that's why I am asking.
> 

Affected device ID:
https://github.com/skeggsb/nouveau/blob/master/drm/nouveau/nvkm/engine/device/pci.c#L1229
can it be excluded from device->chipset case 0xac ?

Care to create a patch?
I'll test it.


___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-18 Thread Karol Herbst
well, I just don't want that this fix breaks the same thing for other
users, that's why I am asking.

2016-10-18 13:56 GMT+02:00 poma :
> On 18.10.2016 09:35, Karol Herbst wrote:
>> how sure are you, that this is needed for _every_ nvac?
>>
>
> Thank you for asking.
>
> If you consider, as relevant,
> referring to the original commit:
> "drm/nouveau/disp/g94: implement workaround for dvi issue on fx380"
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=2a4bd8a
> 
> Fixes the second DVI output on Quadro FX380.
> Thanks to NVIDIA for providing the details on the full workaround.
>
> [...]
> +   switch (device->chipset) {
> +   case 0x94:
> +   case 0x96:
> +   case 0x98:
> +   case 0xaa:
> +   case 0xac:
> +   return true;
> [...]
> 
>
> and to Quadro FX380 as defined:
>
> 1. https://nouveau.freedesktop.org/wiki/CodeNames/#NV50
>NV96 (G96) ...
>
> 2. https://en.wikipedia.org/wiki/Nvidia_Quadro
>G96 ... GeForce 9400 based
>
> 3. 
> https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units#Quadro_FX_.28x800.29_series
>G96 ...
>
>
> The right question would be,
> for you Karol, Ben and perhaps the ones from the NVIDIA - those to which Ben 
> refers,
>
> whether device->chipset:
> +   case 0x94:
> +   case 0x98:
> +   case 0xaa:
> +   case 0xac:
> are redundant, in the first place?
>
> Moreover, even if case 0x96 applies only,
> how sure are -you-, that this is needed for _every_ nv96?
>
> And given that I am here only the user, who is only caring for my hardware,
> I can only appreciate your sense of humor. ;)
>
>
>
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-18 Thread poma
On 18.10.2016 09:35, Karol Herbst wrote:
> how sure are you, that this is needed for _every_ nvac?
> 

Thank you for asking.

If you consider, as relevant,
referring to the original commit:
"drm/nouveau/disp/g94: implement workaround for dvi issue on fx380"
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=2a4bd8a

Fixes the second DVI output on Quadro FX380.
Thanks to NVIDIA for providing the details on the full workaround. 

[...]
+   switch (device->chipset) {
+   case 0x94:
+   case 0x96:
+   case 0x98:
+   case 0xaa:
+   case 0xac:
+   return true;
[...]


and to Quadro FX380 as defined:

1. https://nouveau.freedesktop.org/wiki/CodeNames/#NV50
   NV96 (G96) ...

2. https://en.wikipedia.org/wiki/Nvidia_Quadro
   G96 ... GeForce 9400 based

3. 
https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units#Quadro_FX_.28x800.29_series
   G96 ...


The right question would be,
for you Karol, Ben and perhaps the ones from the NVIDIA - those to which Ben 
refers,

whether device->chipset:
+   case 0x94:
+   case 0x98:
+   case 0xaa:
+   case 0xac:
are redundant, in the first place?

Moreover, even if case 0x96 applies only,
how sure are -you-, that this is needed for _every_ nv96?

And given that I am here only the user, who is only caring for my hardware,
I can only appreciate your sense of humor. ;)



___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


Re: [Nouveau] NVAC "No Signal"

2016-10-18 Thread Karol Herbst
how sure are you, that this is needed for _every_ nvac?

2016-10-18 5:55 GMT+02:00 poma :
> Fixes "No Signal" via HDMI from NVIDIA Corporation ION VGA (rev b1)
>
> Ref.
> "drm/nouveau/disp/g94: implement workaround for dvi issue on fx380"
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=2a4bd8a
>
> The last working Fedora kernel 4.8.0-0.rc0.git3.1.fc25
>
> Patched and tested with:
> $ modinfo -n nouveau
> /lib/modules/4.8.2-300.fc25.x86_64/updates/nouveau.ko
>
> Tested-by: poma 
> ---
>  drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c 
> b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
> index fbb8c7d..c9e40e7 100644
> --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
> +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
> @@ -434,7 +434,8 @@ nv50_disp_dptmds_war(struct nvkm_device *device)
> case 0x96:
> case 0x98:
> case 0xaa:
> -   case 0xac:
> +/* NVIDIA MCP79/MCP7A "No Signal" */
> +/* case 0xac:*/
> return true;
> default:
> break;
> --
> 2.7.4
> ___
> Nouveau mailing list
> Nouveau@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/nouveau
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau


[Nouveau] NVAC "No Signal"

2016-10-17 Thread poma
Fixes "No Signal" via HDMI from NVIDIA Corporation ION VGA (rev b1)

Ref.
"drm/nouveau/disp/g94: implement workaround for dvi issue on fx380"
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=2a4bd8a

The last working Fedora kernel 4.8.0-0.rc0.git3.1.fc25

Patched and tested with:
$ modinfo -n nouveau
/lib/modules/4.8.2-300.fc25.x86_64/updates/nouveau.ko

Tested-by: poma 
---
 drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c 
b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
index fbb8c7d..c9e40e7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.c
@@ -434,7 +434,8 @@ nv50_disp_dptmds_war(struct nvkm_device *device)
case 0x96:
case 0x98:
case 0xaa:
-   case 0xac:
+/* NVIDIA MCP79/MCP7A "No Signal" */
+/* case 0xac:*/
return true;
default:
break;
-- 
2.7.4
___
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau