Re: [RFC v3 06/14] drm/sched: Implement RR via FIFO

2025-04-05 Thread Michel Dänzer
On 2025-04-02 14:00, Philipp Stanner wrote: > On Wed, 2025-04-02 at 12:58 +0200, Michel Dänzer wrote: >> On 2025-04-02 12:46, Philipp Stanner wrote: >>> On Mon, 2025-03-31 at 21:16 +0100, Tvrtko Ursulin wrote: Round-robin being the non-default policy and unclear how much it is used,

Re: [RFC v3 06/14] drm/sched: Implement RR via FIFO

2025-04-04 Thread Philipp Stanner
On Wed, 2025-04-02 at 12:58 +0200, Michel Dänzer wrote: > On 2025-04-02 12:46, Philipp Stanner wrote: > > On Mon, 2025-03-31 at 21:16 +0100, Tvrtko Ursulin wrote: > > > Round-robin being the non-default policy and unclear how much it > > > is > > > used, > > > we can notice that it can be implement

Re: [RFC v3 06/14] drm/sched: Implement RR via FIFO

2025-04-04 Thread Michel Dänzer
On 2025-04-02 12:46, Philipp Stanner wrote: > On Mon, 2025-03-31 at 21:16 +0100, Tvrtko Ursulin wrote: >> Round-robin being the non-default policy and unclear how much it is >> used, >> we can notice that it can be implemented using the FIFO data >> structures if >> we only invent a fake submit tim

Re: [RFC v3 06/14] drm/sched: Implement RR via FIFO

2025-04-04 Thread Tvrtko Ursulin
On 02/04/2025 14:37, Christian König wrote: Am 02.04.25 um 15:22 schrieb Michel Dänzer: On 2025-04-02 14:00, Philipp Stanner wrote: On Wed, 2025-04-02 at 12:58 +0200, Michel Dänzer wrote: On 2025-04-02 12:46, Philipp Stanner wrote: On Mon, 2025-03-31 at 21:16 +0100, Tvrtko Ursulin wrote: R

Re: [RFC v3 06/14] drm/sched: Implement RR via FIFO

2025-04-02 Thread Tvrtko Ursulin
On 02/04/2025 11:46, Philipp Stanner wrote: On Mon, 2025-03-31 at 21:16 +0100, Tvrtko Ursulin wrote: Round-robin being the non-default policy and unclear how much it is used, we can notice that it can be implemented using the FIFO data structures if we only invent a fake submit timestamp which

Re: [RFC v3 06/14] drm/sched: Implement RR via FIFO

2025-04-02 Thread Christian König
Am 02.04.25 um 15:22 schrieb Michel Dänzer: > On 2025-04-02 14:00, Philipp Stanner wrote: >> On Wed, 2025-04-02 at 12:58 +0200, Michel Dänzer wrote: >>> On 2025-04-02 12:46, Philipp Stanner wrote: On Mon, 2025-03-31 at 21:16 +0100, Tvrtko Ursulin wrote: > Round-robin being the non-default

Re: [RFC v3 06/14] drm/sched: Implement RR via FIFO

2025-04-02 Thread Philipp Stanner
On Mon, 2025-03-31 at 21:16 +0100, Tvrtko Ursulin wrote: > Round-robin being the non-default policy and unclear how much it is > used, > we can notice that it can be implemented using the FIFO data > structures if > we only invent a fake submit timestamp which is monotonically > increasing > inside

[RFC v3 06/14] drm/sched: Implement RR via FIFO

2025-03-31 Thread Tvrtko Ursulin
Round-robin being the non-default policy and unclear how much it is used, we can notice that it can be implemented using the FIFO data structures if we only invent a fake submit timestamp which is monotonically increasing inside drm_sched_rq instances. So instead of remembering which was the last