[Mesa-dev] GLX extension for vendor name lookup in libglvnd

[Martin Peres]

On 14/03/16 19:59, Kyle Brenneman wrote:
> On 03/14/2016 11:43 AM, Martin Peres wrote:
>> On 14/03/16 19:16, Kyle Brenneman wrote:
>>> On 03/11/2016 05:25 PM, Martin Peres wrote:
>>>>
>>>> On 10/03/16 20:07, Adam Jackson wrote:
>>>>> On Thu, 2016-03-10 at 10:53 -0700, Kyle Brenneman wrote:
>>>>>> On 03/10/2016 10:47 AM, Martin Peres wrote:
>>>>>>> That could be a hacky way of handling the case where multiple 3D
>>>>>>> drivers could be used to drive the same GPU. This may be
>>>>>>> necessary in
>>>>>>> the future if two mesa drivers support the same GPU but one is
>>>>>>> considered better than the other. We can also imagine a case where a
>>>>>>> proprietary driver would need to be co-installable with an open
>>>>>>> source
>>>>>>> one and would still use the same DDX. Isn't that what AMD is
>>>>>>> going to
>>>>>>> do soon? Did anyone think about this case?
>>>>>> That case is the reason for allowing multiple vendor names. For a
>>>>>> case
>>>>>> like AMD's driver, it would hand back two names. The order would
>>>>>> be up
>>>>>> to the driver implementation, but I would guess that it would list
>>>>>> the
>>>>>> proprietary driver first and the open source driver second. If the
>>>>>> proprietary one is installed, then the client would use it, and if
>>>>>> not,
>>>>>> the client would use the open source one.
>>>> Very good! That could be worth mentioning in the spec. To make it
>>>> clear that it is the intended goal and to help implementers understand
>>>> the logic behind this proposal.
>>> That's what I meant to convey with the description of multiple
>>> client-side drivers that work with the same server-side driver. But if
>>> that wasn't clear I can add a more specific example. Would something
>>> like this help?
>>>
>>>
>>>       For example, some vendors may have both a proprietary client-side
>>>       vendor library and an open source vendor library that work with the
>>>       same server-side driver. In that case, the server would return the
>>>       names for both of the vendor libraries. The client would then be
>>> able
>>>       to select one of those vendor libraries, depending on which of
>>> them is
>>>       installed.
>> This is definitely a nice addition, however, I propose to reword it to
>> add information about priorities:
>>
>>
>>    For example, some vendors may have both a proprietary client-side
>>    vendor library and an open source vendor library that work with the
>>    same server-side driver. In that case, the server would return the
>>    names for both of the vendor libraries, in the order of preference.
>>    The client would then try to open them sequentially and select the
>>    first one that is present and got loaded successfully.
>>
>> How does this sound?
> Issue #3 addresses the question of how the vendor names are ordered.
> Instead of duplicating the description, maybe change the last sentence
> to reference that issue:
>
> "The client would then try loading each vendor library as described
> below in Issue 3."

Sounds great!