[oi-dev] Change in video driver ABI
Aurélien Larcher
aurelien.larcher at gmail.com
Mon Jun 5 10:55:29 UTC 2017
On Mon, Jun 5, 2017 at 12:47 PM, Jean-Pierre André <
jean-pierre.andre at wanadoo.fr> wrote:
> Aurélien Larcher wrote:
>
>>
>>
>> On Mon, Jun 5, 2017 at 11:32 AM, Jean-Pierre André
>> <jean-pierre.andre at wanadoo.fr <mailto:jean-pierre.andre at wanadoo.fr>>
>> wrote:
>>
>> Aurélien Larcher wrote:
>>
>>
>>
>> On Mon, Jun 5, 2017 at 9:48 AM, Jean-Pierre André
>> <jean-pierre.andre at wanadoo.fr
>> <mailto:jean-pierre.andre at wanadoo.fr>
>> <mailto:jean-pierre.andre at wanadoo.fr
>> <mailto:jean-pierre.andre at wanadoo.fr>>> wrote:
>>
>> Aurélien Larcher wrote:
>>
>>
>>
>> On Mon, Jun 5, 2017 at 9:35 AM, Jean-Pierre André
>> <jean-pierre.andre at wanadoo.fr
>> <mailto:jean-pierre.andre at wanadoo.fr>
>> <mailto:jean-pierre.andre at wanadoo.fr
>> <mailto:jean-pierre.andre at wanadoo.fr>>
>> <mailto:jean-pierre.andre at wanadoo.fr
>> <mailto:jean-pierre.andre at wanadoo.fr>
>> <mailto:jean-pierre.andre at wanadoo.fr
>> <mailto:jean-pierre.andre at wanadoo.fr>>>> wrote:
>>
>> Jean-Pierre André wrote:
>>
>>
>> [...]
>>
>>
>> Update : I see the following comment for
>> NVIDIA-Solaris-x86-304.135
>> "Added support for X.Org xserver ABI 23
>> (xorg-server 1.19)"
>>
>> Is that the new ABI used by Xorg on OpenIndiana ?
>>
>>
>> One always needs to update to a driver version
>> supported by the
>> current
>> Xorg.
>> We updated to 1.18 (ABI == 20) and soon we will update
>> to 1.19
>> (ABI ==
>> 23), so you should install the driver you mention.
>>
>>
>> Ok, thanks. In the Nvidia changelog, I see that
>> NVIDIA-Solaris-x86-304.131 has support for ABI 20,
>> so I am to try this one.
>>
>>
>> The support is backward compatible, so the latest will work as
>> well.
>>
>>
>> So I tried :
>> - NVIDIA-Solaris-x86-304.131
>> - NVIDIA-Solaris-x86-304.132
>> - NVIDIA-Solaris-x86-304.135
>>
>> The first couple of them use ABI 20, and the last one
>> uses ABI 23.
>>
>> They run successfully with the initial BE, and they all
>> fail the same way with the updated BE. When switching to
>> graphics mode, I get a black screen with a winking cursor
>> on the top left. The keyboard and mouse are probably not
>> working (but I can connect through ssh to find out what
>> is going on).
>>
>> There are a lot of differences in the Xorg.log. Below are
>> the main the differences for NVIDIA-Solaris-x86-304.135
>> running on the working BE and on the updated BE (so the
>> driver is the same, but Xorg is different).
>>
>> I have put all the differences on
>> http://jp-andre.pagesperso-orange.fr/Xorg.5jun.zip
>> <http://jp-andre.pagesperso-orange.fr/Xorg.5jun.zip>
>>
>> A lot of missing initializations, such as
>>
>> < [] Initializing built-in extension Generic Event Extension
>> < [] Initializing built-in extension SHAPE
>> < [] Initializing built-in extension MIT-SHM
>> < [] Initializing built-in extension XInputExtension
>> < [] Initializing built-in extension XTEST
>> (and 22 more similar lines)
>>
>> Missing report about module tsol (it was expected to fail,
>> but the point here is the loading was not even tried).
>>
>> < [] Loading extension SolarisIA
>> < [] (II) LoadModule: "xtsol"
>> < [] (WW) Warning, couldn't open module xtsol
>> < [] (II) UnloadModule: "xtsol"
>> < [] (II) Unloading xtsol
>> < [] (EE) Failed to load module "xtsol" (module does not exist, 0)
>>
>> EDID information not taken into account :
>>
>> < [] (**) NVIDIA(0): Using HorizSync/VertRefresh ranges from the
>> EDID for display
>> < [] (**) NVIDIA(0): device IQT L70S (CRT-0) (Using EDID
>> frequencies has been
>> < [] (**) NVIDIA(0): enabled on all display devices.)
>>
>> This looks much like some package is missing.
>>
>
> I was saying that because several xorg functions were
> not activated (initializations, processing of EDID), which
> could means they were not called.
>
> Anyway there are the same list of xorg packages in the
> working BE and the updated one.
>
>
>> Since the driver is binary, the only thing I can offer for now I to
>> provide a build of Xorg 1.19 that you could try out, ... but maybe it is
>> just that nvidia drivers are broken for your model.
>>
>
> I would not blame the driver, as all versions tried behave
> properly with the initial BE. Also I can imagine the
> interface between the driver and xorg has been tested on
> other configurations, so the issue must be elsewhere.
>
But the first BE is the one with Xorg 1.14.7 right?
>
> I am keeping the configuration as is until somebody has
> something else to suggest or you have an updated Xorg.
> No need to rush it out, I will use the working BE in the
> meantime.
>
OK. I will keep you posted when I have something ready.
>
> Thank you for your help.
>
You are welcome.
>
> Jean-Pierre
>
>
>
>
>
> _______________________________________________
> oi-dev mailing list
> oi-dev at openindiana.org
> https://openindiana.org/mailman/listinfo/oi-dev
>
--
---
Praise the Caffeine embeddings
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://openindiana.org/pipermail/oi-dev/attachments/20170605/658f8447/attachment-0005.html>
More information about the oi-dev
mailing list