Re: [PATCH 0/1] Always record job cycle and timestamp information

2024-02-28 Thread Daniel Vetter
On Wed, Feb 21, 2024 at 03:13:41PM +, Adrián Larumbe wrote: > > On 21.02.2024 14:34, Tvrtko Ursulin wrote: > > > > On 21/02/2024 09:40, Adrián Larumbe wrote: > > > Hi, > > > > > > I just wanted to make sure we're on the same page on this matter. So in > > > Panfrost, and I guess in almost eve

Re: [PATCH 0/1] Always record job cycle and timestamp information

2024-02-21 Thread Adrián Larumbe
> On 21.02.2024 14:34, Tvrtko Ursulin wrote: > > On 21/02/2024 09:40, Adrián Larumbe wrote: > > Hi, > > > > I just wanted to make sure we're on the same page on this matter. So in > > Panfrost, and I guess in almost every other single driver out there, HW perf > > counters and their uapi interfac

Re: [PATCH 0/1] Always record job cycle and timestamp information

2024-02-21 Thread Tvrtko Ursulin
On 21/02/2024 09:40, Adrián Larumbe wrote: Hi, I just wanted to make sure we're on the same page on this matter. So in Panfrost, and I guess in almost every other single driver out there, HW perf counters and their uapi interface are orthogonal to fdinfo's reporting on drm engine utilisation.

Re: [PATCH 0/1] Always record job cycle and timestamp information

2024-02-21 Thread Adrián Larumbe
Hi, I just wanted to make sure we're on the same page on this matter. So in Panfrost, and I guess in almost every other single driver out there, HW perf counters and their uapi interface are orthogonal to fdinfo's reporting on drm engine utilisation. At the moment it seems like HW perfcounters an

Re: [PATCH 0/1] Always record job cycle and timestamp information

2024-02-16 Thread Tvrtko Ursulin
On 16/02/2024 16:57, Daniel Vetter wrote: On Wed, Feb 14, 2024 at 01:52:05PM +, Steven Price wrote: Hi Adrián, On 14/02/2024 12:14, Adrián Larumbe wrote: A driver user expressed interest in being able to access engine usage stats through fdinfo when debugfs is not built into their kernel

Re: [PATCH 0/1] Always record job cycle and timestamp information

2024-02-16 Thread Daniel Vetter
On Wed, Feb 14, 2024 at 01:52:05PM +, Steven Price wrote: > Hi Adrián, > > On 14/02/2024 12:14, Adrián Larumbe wrote: > > A driver user expressed interest in being able to access engine usage stats > > through fdinfo when debugfs is not built into their kernel. In the current > > implementation

Re: [PATCH 0/1] Always record job cycle and timestamp information

2024-02-14 Thread Steven Price
Hi Adrián, On 14/02/2024 12:14, Adrián Larumbe wrote: > A driver user expressed interest in being able to access engine usage stats > through fdinfo when debugfs is not built into their kernel. In the current > implementation, this wasn't possible, because it was assumed even for > inflight jobs e

[PATCH 0/1] Always record job cycle and timestamp information

2024-02-14 Thread Adrián Larumbe
A driver user expressed interest in being able to access engine usage stats through fdinfo when debugfs is not built into their kernel. In the current implementation, this wasn't possible, because it was assumed even for inflight jobs enabling the cycle counter and timestamp registers would incur i