[PATCH v5 2/2] drm/xe/bo: add GPU memory trace points

Juston Li justonli at chromium.org
Wed Jul 9 16:18:58 UTC 2025


On Tue, 2025-07-08 at 15:29 -0500, Lucas De Marchi wrote:
> On Mon, Jul 07, 2025 at 01:38:23PM -0700, Juston Li wrote:
> > Add TRACE_GPU_MEM tracepoints for tracking global GPU memory usage.
> > 
> > These are required by VSR on Android 12+ for reporting GPU driver
> > memory
> > allocations.
> > 
> > v5:
> > - Drop process_mem tracking
> > - Set the gpu_id field to dev->primary->index (Lucas, Tvrtko)
> > - Formatting cleanup under 80 columns
> > 
> > v3:
> > - Use now configurable CONFIG_TRACE_GPU_MEM instead of adding a
> >   per-driver Kconfig (Lucas)
> > 
> > v2:
> > - Use u64 as preferred by checkpatch (Tvrtko)
> > - Fix errors in comments/Kconfig description (Tvrtko)
> > - drop redundant "CONFIG" in Kconfig
> > 
> > Signed-off-by: Juston Li <justonli at chromium.org>
> > Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin at igalia.com>
> > ---
> > drivers/gpu/drm/xe/xe_bo.c           | 22 ++++++++++++++++++++++
> > drivers/gpu/drm/xe/xe_device_types.h |  8 ++++++++
> > 2 files changed, 30 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/xe/xe_bo.c
> > b/drivers/gpu/drm/xe/xe_bo.c
> > index 4e39188a021ab..950eef514c11c 100644
> > --- a/drivers/gpu/drm/xe/xe_bo.c
> > +++ b/drivers/gpu/drm/xe/xe_bo.c
> > @@ -19,6 +19,8 @@
> > 
> > #include <kunit/static_stub.h>
> > 
> > +#include <trace/events/gpu_mem.h>
> > +
> > #include "xe_device.h"
> > #include "xe_dma_buf.h"
> > #include "xe_drm_client.h"
> > @@ -418,6 +420,24 @@ static void xe_ttm_tt_account_subtract(struct
> > xe_device *xe, struct ttm_tt *tt)
> > 		xe_shrinker_mod_pages(xe->mem.shrinker, -(long)tt-
> > >num_pages, 0);
> > }
> > 
> > +#if IS_ENABLED(CONFIG_TRACE_GPU_MEM)
> > +static void update_global_total_pages(struct ttm_device *ttm_dev,
> > +				      long num_pages)
> > +{
> 
> wouldn't it be better to move the ifdef inside the function?
> Any compiler would remove the empty call regardless of the inline
> annotation, so I think this would make it less error prone if we
> change
> function signature.
> 
> No need to send a new version, I can move that while applying
> if agreed.

SGTM, thanks!

Juston

> Lucas De Marchi


More information about the Intel-xe mailing list