Re: [PATCH 0/9] drm/syncobj: Add full-featured wait support (v2)

Am 21.08.2017 um 23:42 schrieb Jason Ekstrand: On Wed, Aug 16, 2017 at 1:10 PM, Jason Ekstrand > wrote: On Wed, Aug 16, 2017 at 9:53 AM, Christian König mailto:christian.koe...@amd.com>> wrote: [SNIP] See a wait_queue is a callba

Re: [PATCH 0/9] drm/syncobj: Add full-featured wait support (v2)

On Wed, Aug 16, 2017 at 1:10 PM, Jason Ekstrand wrote: > On Wed, Aug 16, 2017 at 9:53 AM, Christian König > wrote: > >> [SNIP] >> >>> See a wait_queue is a callback mechanism anyway, so you are wrapping a callback mechanism inside another callback mechanism and that makes not really mu

Re: [PATCH 0/9] drm/syncobj: Add full-featured wait support (v2)

[SNIP] See a wait_queue is a callback mechanism anyway, so you are wrapping a callback mechanism inside another callback mechanism and that makes not really much sense. Fair enough. There is one little snag though: We need to wait on sync objects and fences at

Re: [PATCH 0/9] drm/syncobj: Add full-featured wait support (v2)

On Mon, Aug 14, 2017 at 12:36 AM, Christian König wrote: > Am 14.08.2017 um 01:14 schrieb Jason Ekstrand: > >> On August 13, 2017 8:52:21 AM Christian König >> wrote: >> >> Am 13.08.2017 um 17:26 schrieb Jason Ekstrand: >>> On August 13, 2017 6:19:53 AM Christian König wrote: >>>

Re: [PATCH 0/9] drm/syncobj: Add full-featured wait support (v2)

On August 13, 2017 4:14:45 PM Jason Ekstrand wrote: On August 13, 2017 8:52:21 AM Christian König wrote: Am 13.08.2017 um 17:26 schrieb Jason Ekstrand: On August 13, 2017 6:19:53 AM Christian König wrote: Patches #1-#4 are Acked-by: Christian König . Patch #5: NAK, that will break radeo

Re: [PATCH 0/9] drm/syncobj: Add full-featured wait support (v2)

On August 13, 2017 6:19:53 AM Christian König wrote: Patches #1-#4 are Acked-by: Christian König . Patch #5: NAK, that will break radeon. On radeon we need the non-default wait or otherwise we can run into a situation where we never signal a fence. The general question is why do you need thi

[PATCH 0/9] drm/syncobj: Add full-featured wait support (v2)

This series does the same thing as my earlier series in that it adds a sync object wait interface complete with WAIT_FOR_SUBMIT flag. While the uapi remains unchanged, the guts look a bit different. Instead of adding a callback mechanism to drm_syncobj that fired whenever replace_fence was called