[OpenIndiana-discuss] nVIDIA again

[Apostolos Syropoulos via openindiana-discuss]

Hello,

Some weeks ago I reported my problems with the nVIDIA driver.

Sometimes when I booted my system, the screen got black and nothing 

happened. Then I had to reset the system. Here is what /var/adm/messages 

reported just before the manual reboot:


Sep 27 15:38:42 adalind nvidia: [ID 702911 kern.notice] NOTICE: NVRM: GPU at 
PCI::01:00: GPU-88af19ba-ec73-7f91-09dd-448d8abf45ce
Sep 27 15:38:42 adalind nvidia: [ID 702911 kern.notice] NVRM: Xid 
(PCI::01:00): 56, CMDre  0080  0005 0005
Sep 27 15:38:46 adalind last message repeated 2 times
Sep 27 15:38:46 adalind console-kit-daemon[745]: [ID 702911 daemon.warning] 
GLib-GObject-WARNING: g_object_set_property: construct property "seat-id" for 
object 'CkSession' can't be set after construction
Sep 27 15:38:47 adalind gnome-session[1905]: [ID 702911 daemon.warning] 
WARNING: Failed to run gconf-sanity-check-2: Failed to execute child process "" 
(No such file or directory)
Sep 27 15:38:47 adalind gnome-session[1905]: [ID 702911 daemon.warning] 
WARNING: IceListenForConnections returned 2 non-local listeners: 
inet/adalind:55155,inet6/adalind:39567
Sep 27 15:38:49 adalind nvidia: [ID 702911 kern.notice] NVRM: Xid 
(PCI::01:00): 56, CMDre  0080  0005 0005
Sep 27 15:38:51 adalind syslog[1923]: [ID 702911 daemon.warning] Gtk-WARNING: 
gtkwidget.c:5709: widget not within a GtkWindow

According to 

https://docs.nvidia.com/deploy/xid-errors/

the error message 56 is a Display Engine error. This means that it is either a 
hardware error or a driver error.

I replaced the GeForce GT-730 by MSI with a GT-730 from Gigabyte, which I 
bought last year, and now the system
is stable. No X server crashes and no black screens, So it was a hardware 
problem. I was not very fond of MSI
products but now I will never buy anything else from MSI.

A.S.

--
Apostolos Syropoulos
Xanthi, Greece

___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss

Re: [OpenIndiana-discuss] NVIDIA again

[Alan Coopersmith]

On 09/13/16 03:38 PM, Lionel Cons wrote:

On 14 September 2016 at 00:34, Alan Coopersmith
 wrote:

On 08/17/16 11:57 AM, Alan Coopersmith wrote:


Note that on Solarish OS'es, the backtrace will sometimes warn you about
the
last linker lookup failure, even when it's not relevant - it could be the
driver looked for that symbol, didn't find it because your Xorg is an
older
release, and went on about it's business only to crash later.

(I think, but have not had time to confirm that this is because the code I
 stuck in osinit.c years ago to catch actual errors here fails to check in
 the signal handler what signal was received so prints it for every
signal:

https://cgit.freedesktop.org/xorg/xserver/commit?id=98f4179156391752e6688339487458ad7828abf4

 If correct, the fix would be to wrap the first chunk of that in
 OsSigHandler inside an "if (signo == SIGQUIT)" block.)



Confirmed and fixed upstream since I finally got tired of explaining my bug
to
everyone:

https://cgit.freedesktop.org/xorg/xserver/commit/?id=75c1d04650f63464263c159d2e95364482be724f

-alan-


Mind fixing the patch and print the siginfo information to the error
message to have more diagnostics?


Sorry, I don't understand what you're asking for.

-alan-

___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss

Re: [OpenIndiana-discuss] NVIDIA again

[Lionel Cons]

On 14 September 2016 at 00:34, Alan Coopersmith
 wrote:
> On 08/17/16 11:57 AM, Alan Coopersmith wrote:
>>
>> Note that on Solarish OS'es, the backtrace will sometimes warn you about
>> the
>> last linker lookup failure, even when it's not relevant - it could be the
>> driver looked for that symbol, didn't find it because your Xorg is an
>> older
>> release, and went on about it's business only to crash later.
>>
>> (I think, but have not had time to confirm that this is because the code I
>>  stuck in osinit.c years ago to catch actual errors here fails to check in
>>  the signal handler what signal was received so prints it for every
>> signal:
>>
>> https://cgit.freedesktop.org/xorg/xserver/commit?id=98f4179156391752e6688339487458ad7828abf4
>>
>>  If correct, the fix would be to wrap the first chunk of that in
>>  OsSigHandler inside an "if (signo == SIGQUIT)" block.)
>
>
> Confirmed and fixed upstream since I finally got tired of explaining my bug
> to
> everyone:
>
> https://cgit.freedesktop.org/xorg/xserver/commit/?id=75c1d04650f63464263c159d2e95364482be724f
>
> -alan-

Mind fixing the patch and print the siginfo information to the error
message to have more diagnostics?

Lionel

___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss

Re: [OpenIndiana-discuss] NVIDIA again

[Alan Coopersmith]

On 08/17/16 11:57 AM, Alan Coopersmith wrote:

Note that on Solarish OS'es, the backtrace will sometimes warn you about the
last linker lookup failure, even when it's not relevant - it could be the
driver looked for that symbol, didn't find it because your Xorg is an older
release, and went on about it's business only to crash later.

(I think, but have not had time to confirm that this is because the code I
 stuck in osinit.c years ago to catch actual errors here fails to check in
 the signal handler what signal was received so prints it for every signal:
https://cgit.freedesktop.org/xorg/xserver/commit?id=98f4179156391752e6688339487458ad7828abf4

 If correct, the fix would be to wrap the first chunk of that in
 OsSigHandler inside an "if (signo == SIGQUIT)" block.)


Confirmed and fixed upstream since I finally got tired of explaining my bug to
everyone:

https://cgit.freedesktop.org/xorg/xserver/commit/?id=75c1d04650f63464263c159d2e95364482be724f

-alan-

___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss

Re: [OpenIndiana-discuss] NVIDIA again

[Apostolos Syropoulos via openindiana-discuss]

>
>You can set it to enableIndirectGLX= false in xorg.conf.
>~ Ken
>

I did it:

$ cat /etc/X11/xorg.conf

..
Section "Device"
Identifier "Device0"
Driver "nvidia"
VendorName "NVIDIA Corporation"
EndSection

Section "Screen"
Identifier "Screen0"
Device "Device0"
Monitor"Monitor0"
DefaultDepth24
SubSection "Display"
Depth   24
EndSubSection
Option "AllowIndirectGLXProtocol" "off"
EndSection



A.S.
 
--
Apostolos Syropoulos
Xanthi, Greece

___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss

Re: [OpenIndiana-discuss] NVIDIA again

[ken mays via openindiana-discuss]

You can set it to enableIndirectGLX= false in xorg.conf.~ Ken
 

On Wednesday, August 17, 2016 11:58 AM, Alan Coopersmith 
 wrote:
 

 On 08/17/16 11:31 AM, Alexander Pyhalov wrote:
> Apostolos Syropoulos via openindiana-discuss писал 17.08.2016 21:03:
>> Hello,
>>
>> I reported my problem to the NVIDIA people who responed:
>>
>> https://devtalk.nvidia.com/default/topic/955405/solaris/
>>
>> It seems that something is wrong with the Xorg binary.
>>
>
> Hello.
> They could just managed to build it against newer Xorg. However, I don't
> understand, how in this case it works at all.
> As you can see, the mentioned symbol ( enableIndirectGLX ) is defined in Xorg
> 1.17 headers
> (https://cgit.freedesktop.org/xorg/xserver/tree/include/opaque.h?h=server-1.17-branch),
>
> but not in 1.14 headers
> (https://cgit.freedesktop.org/xorg/xserver/tree/include/opaque.h?h=server-1.14-branch),
> which we ship.
> So, our Xorg couldn't know about it.
>
> If you run elfdump on nvidia libraries, do you see it there?

Note that on Solarish OS'es, the backtrace will sometimes warn you about the
last linker lookup failure, even when it's not relevant - it could be the
driver looked for that symbol, didn't find it because your Xorg is an older
release, and went on about it's business only to crash later.

(I think, but have not had time to confirm that this is because the code I
  stuck in osinit.c years ago to catch actual errors here fails to check in
  the signal handler what signal was received so prints it for every signal:
https://cgit.freedesktop.org/xorg/xserver/commit?id=98f4179156391752e6688339487458ad7828abf4
  If correct, the fix would be to wrap the first chunk of that in
  OsSigHandler inside an "if (signo == SIGQUIT)" block.)

Still, that backtrace looks like some invalid function pointer got put
into the list of callbacks for the wakeup routine - either directly or
by something else overwriting a crucial bit of memory with 0x32.

-- 
    -Alan Coopersmith-              alan.coopersm...@oracle.com
    Oracle Solaris Engineering - http://blogs.oracle.com/alanc

___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss


  
___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss

Re: [OpenIndiana-discuss] NVIDIA again

[Alan Coopersmith]

On 08/17/16 11:31 AM, Alexander Pyhalov wrote:

Apostolos Syropoulos via openindiana-discuss писал 17.08.2016 21:03:

Hello,

I reported my problem to the NVIDIA people who responed:

https://devtalk.nvidia.com/default/topic/955405/solaris/

It seems that something is wrong with the Xorg binary.



Hello.
They could just managed to build it against newer Xorg. However, I don't
understand, how in this case it works at all.
As you can see, the mentioned symbol ( enableIndirectGLX ) is defined in Xorg
1.17 headers
(https://cgit.freedesktop.org/xorg/xserver/tree/include/opaque.h?h=server-1.17-branch),

but not in 1.14 headers
(https://cgit.freedesktop.org/xorg/xserver/tree/include/opaque.h?h=server-1.14-branch),
which we ship.
So, our Xorg couldn't know about it.

If you run elfdump on nvidia libraries, do you see it there?


Note that on Solarish OS'es, the backtrace will sometimes warn you about the
last linker lookup failure, even when it's not relevant - it could be the
driver looked for that symbol, didn't find it because your Xorg is an older
release, and went on about it's business only to crash later.

(I think, but have not had time to confirm that this is because the code I
 stuck in osinit.c years ago to catch actual errors here fails to check in
 the signal handler what signal was received so prints it for every signal:
https://cgit.freedesktop.org/xorg/xserver/commit?id=98f4179156391752e6688339487458ad7828abf4
 If correct, the fix would be to wrap the first chunk of that in
 OsSigHandler inside an "if (signo == SIGQUIT)" block.)

Still, that backtrace looks like some invalid function pointer got put
into the list of callbacks for the wakeup routine - either directly or
by something else overwriting a crucial bit of memory with 0x32.

--
-Alan Coopersmith-  alan.coopersm...@oracle.com
 Oracle Solaris Engineering - http://blogs.oracle.com/alanc

___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss

Re: [OpenIndiana-discuss] NVIDIA again

[Alexander Pyhalov]

Apostolos Syropoulos via openindiana-discuss писал 17.08.2016 21:03:

Hello,

I reported my problem to the NVIDIA people who responed:

https://devtalk.nvidia.com/default/topic/955405/solaris/

It seems that something is wrong with the Xorg binary.



Hello.
They could just managed to build it against newer Xorg. However, I don't 
understand, how in this case it works at all.
As you can see, the mentioned symbol ( enableIndirectGLX ) is defined in 
Xorg 1.17 headers 
(https://cgit.freedesktop.org/xorg/xserver/tree/include/opaque.h?h=server-1.17-branch),
but not in 1.14 headers 
(https://cgit.freedesktop.org/xorg/xserver/tree/include/opaque.h?h=server-1.14-branch), 
which we ship.

So, our Xorg couldn't know about it.

If you run elfdump on nvidia libraries, do you see it there?

---
System Administrator of Southern Federal University Computer Center

___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss

[OpenIndiana-discuss] NVIDIA again

[Apostolos Syropoulos via openindiana-discuss]

Hello,

I reported my problem to the NVIDIA people who responed:

https://devtalk.nvidia.com/default/topic/955405/solaris/

It seems that something is wrong with the Xorg binary.

A.S.

 --
Apostolos Syropoulos
Xanthi, Greece

___
openindiana-discuss mailing list
openindiana-discuss@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss