On Wed, 28 Aug 2024 18:37:41 +0100
Mihail Atanassov wrote:
> On 28/08/2024 18:27, Boris Brezillon wrote:
> > On Wed, 28 Aug 2024 18:07:03 +0200
> > Boris Brezillon wrote:
> >
> >> On Wed, 28 Aug 2024 14:22:51 +0100
> >> Mihail Atanassov wrote:
> >>
> >>> Hi Boris,
> >>>
> >>> On 28/08/2024
On Wed, 28 Aug 2024 18:07:03 +0200
Boris Brezillon wrote:
> On Wed, 28 Aug 2024 14:22:51 +0100
> Mihail Atanassov wrote:
>
> > Hi Boris,
> >
> > On 28/08/2024 13:09, Boris Brezillon wrote:
> > > Hi Mihail,
> > >
> > > On Thu, 8 Aug 2024 12:41:05 +0300
> > > Mihail Atanassov wrote:
> > >
On 28/08/2024 18:27, Boris Brezillon wrote:
On Wed, 28 Aug 2024 18:07:03 +0200
Boris Brezillon wrote:
On Wed, 28 Aug 2024 14:22:51 +0100
Mihail Atanassov wrote:
Hi Boris,
On 28/08/2024 13:09, Boris Brezillon wrote:
Hi Mihail,
On Thu, 8 Aug 2024 12:41:05 +0300
Mihail Atanassov wrote:
On Wed, 28 Aug 2024 14:22:51 +0100
Mihail Atanassov wrote:
> Hi Boris,
>
> On 28/08/2024 13:09, Boris Brezillon wrote:
> > Hi Mihail,
> >
> > On Thu, 8 Aug 2024 12:41:05 +0300
> > Mihail Atanassov wrote:
> >
> >>>
> >>> +/** + * struct drm_panthor_timestamp_info - Timestamp information +
>
Hi Boris,
On 28/08/2024 13:09, Boris Brezillon wrote:
Hi Mihail,
On Thu, 8 Aug 2024 12:41:05 +0300
Mihail Atanassov wrote:
+/** + * struct drm_panthor_timestamp_info - Timestamp information +
* + * Structure grouping all queryable information relating to the
GPU timestamp. + */ +struct drm_
Hi Mihail,
On Thu, 8 Aug 2024 12:41:05 +0300
Mihail Atanassov wrote:
> >
> > +/** + * struct drm_panthor_timestamp_info - Timestamp information +
> > * + * Structure grouping all queryable information relating to the
> > GPU timestamp. + */ +struct drm_panthor_timestamp_info { + /**
> > @times
On 08/08/2024 13:24, Mary Guillemard wrote:
As it stands, this query has nothing to do with the actual GPU so
doesn't really belong here.
It'd be more valuable, and can maybe give better calibration results
than querying the system timestamp separately in userspace, if you
reported all of:
> As it stands, this query has nothing to do with the actual GPU so
> doesn't really belong here.
>
> It'd be more valuable, and can maybe give better calibration results
> than querying the system timestamp separately in userspace, if you
> reported all of:
> * the system timer value
> * the sy
Sorry about the accidental re-wrap, I got fat fingers...
On 08/08/2024 10:41, Mihail Atanassov wrote:
Hi Mary,
Thanks for your patch.
On 07/08/2024 16:35, Mary Guillemard wrote:
Expose system timestamp and frequency supported by the GPU with a
new device query.
[... snip ...]
base-commit:
Hi Mary,
Thanks for your patch.
On 07/08/2024 16:35, Mary Guillemard wrote:
Expose system timestamp and frequency supported by the GPU with a
new device query.
Mali uses the generic arch timer as GPU system time so we currently
wire cntvct_el0 and cntfrq_el0 respectively to a new device query
Expose system timestamp and frequency supported by the GPU with a new
device query.
Mali uses the generic arch timer as GPU system time so we currently
wire cntvct_el0 and cntfrq_el0 respectively to a new device query.
We could have directly read those values from userland but handling
this here a
11 matches
Mail list logo
