[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Public WebGL] Re: extensions status



On Tue, Jan 27, 2015 at 2:22 AM, Kenneth Russell <kbr@google.com> wrote:
On Fri, Jan 23, 2015 at 10:19 PM, Florian Bösch <pyalot@gmail.com> wrote:
> On Wed, Jan 14, 2015 at 9:42 AM, Florian Bösch <pyalot@gmail.com> wrote:
>>
>> EXT_disjoint_timer_query (draft)
>>
>> Is there intent to implement it?
>> If not, should it be dropped from the registry?
>
> Any input on EXT_disjoint_timer_query?

There's intent to implement it.
Splendid :)
 
>> WEBGL_shared_resources (draft)
>>
>> Is there intent to implement it?
>> If not, should it be dropped from the registry?
>
> Proposing to reject this extension, any objections?
> https://github.com/KhronosGroup/WebGL/pull/826

Not ready to reject this. Some extension or core functionality will
ultimately be needed in this area to improve parallelism. Even though
it hasn't reached the top of the priority list I'd like to leave it as
is.
 Far as I understand it workers aren't ready to use WebGL contexts, and most of the usecase could be covered if a WebGL context could render to different canvases. I'm a bit concerned about this extension because in case somebody isn't using it from a worker, the sole intent of it would be to render to different canvases, but since JS is single threaded, acquisition/release of shared objects would be unnecessarily complex and cumbersome for that usecase. If, however, somebody is using workers, then the acquisition/release logic will almost certainly be very difficult to handle (deadlocks, race conditions, what have you).