[Intel-gfx] [PATCH] drm/i915/lmem: debugfs for LMEM details

Ramalingam C ramalingam.c at intel.com
Fri Dec 20 15:08:25 UTC 2019


On 2019-12-20 at 13:34:40 +0000, Chris Wilson wrote:
> Quoting Ramalingam C (2019-12-20 12:51:16)
> > From: Lukasz Fiedorowicz <lukasz.fiedorowicz at intel.com>
> > 
> > Debugfs i915_gem_object is extended to enable the IGTs to
> > detect the LMEM's availability and the total size of LMEM.
> > 
> > Signed-off-by: Lukasz Fiedorowicz <lukasz.fiedorowicz at intel.com>
> > Signed-off-by: Matthew Auld <matthew.auld at intel.com>
> > Signed-off-by: Stuart Summers <stuart.summers at intel.com>
> > Signed-off-by: Ramalingam C <ramalingam.c at intel.com>
> > Cc: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>
> > cc: Chris Wilson <chris at chris-wilson.co.uk>
> > ---
> >  drivers/gpu/drm/i915/i915_debugfs.c        | 6 +++++-
> >  drivers/gpu/drm/i915/intel_memory_region.c | 5 ++++-
> >  drivers/gpu/drm/i915/intel_memory_region.h | 3 +++
> >  3 files changed, 12 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
> > index d28468eaed57..856ded8cd332 100644
> > --- a/drivers/gpu/drm/i915/i915_debugfs.c
> > +++ b/drivers/gpu/drm/i915/i915_debugfs.c
> > @@ -373,7 +373,11 @@ static int i915_gem_object_info(struct seq_file *m, void *data)
> >                    atomic_read(&i915->mm.free_count),
> >                    i915->mm.shrink_memory);
> >  
> > -       seq_putc(m, '\n');
> > +       if (HAS_LMEM(i915)) {
> > +               seq_printf(m, "LMEM total: %llu bytes, available %llu bytes\n",
> > +                          (u64)i915->mm.regions[INTEL_REGION_LMEM]->total,
> > +                          (u64)i915->mm.regions[INTEL_REGION_LMEM]->avail);
> 
> %pa for resource_size_t
> 
with READ_ONCE(i915->mm.regions[INTEL_REGION_LMEM]->total) %pa is not
going well. Getting -Werror=format=. %llu is compiling without any warnings.

> Use READ_ONCE() to indicate to the reader these are being accessed
> outside of the mem->lock.
> 
> > +       }
> >  
> >         print_context_stats(m, i915);
> >  
> > diff --git a/drivers/gpu/drm/i915/intel_memory_region.c b/drivers/gpu/drm/i915/intel_memory_region.c
> > index e24c280e5930..15e539de0a82 100644
> > --- a/drivers/gpu/drm/i915/intel_memory_region.c
> > +++ b/drivers/gpu/drm/i915/intel_memory_region.c
> > @@ -37,7 +37,7 @@ __intel_memory_region_put_pages_buddy(struct intel_memory_region *mem,
> >                                       struct list_head *blocks)
> >  {
> >         mutex_lock(&mem->mm_lock);
> > -       intel_memory_region_free_pages(mem, blocks);
> > +       mem->avail += intel_memory_region_free_pages(mem, blocks);
> >         mutex_unlock(&mem->mm_lock);
> >  }
> >  
> > @@ -106,6 +106,7 @@ __intel_memory_region_get_pages_buddy(struct intel_memory_region *mem,
> >                         break;
> >         } while (1);
> >  
> > +       mem->avail -= size;
> >         mutex_unlock(&mem->mm_lock);
> >         return 0;
> 
> These two look nice and symmetrical.
> 
> >  
> > @@ -164,6 +165,8 @@ intel_memory_region_create(struct drm_i915_private *i915,
> >         mem->io_start = io_start;
> >         mem->min_page_size = min_page_size;
> >         mem->ops = ops;
> > +       mem->total = size;
> > +       mem->avail = mem->total;
> >  
> >         mutex_init(&mem->objects.lock);
> >         INIT_LIST_HEAD(&mem->objects.list);
> > diff --git a/drivers/gpu/drm/i915/intel_memory_region.h b/drivers/gpu/drm/i915/intel_memory_region.h
> > index 238722009677..da56d8ff1b01 100644
> > --- a/drivers/gpu/drm/i915/intel_memory_region.h
> > +++ b/drivers/gpu/drm/i915/intel_memory_region.h
> > @@ -94,6 +94,9 @@ struct intel_memory_region {
> >                 struct list_head list;
> >                 struct list_head purgeable;
> >         } objects;
> > +
> > +       resource_size_t total;
> > +       resource_size_t avail;
> 
> Sensible placement? There'll be one less hole if you put these next to
> the other resource_size_t
done.
> 
> Fix the nits, and
> Reviewed-by: Chris Wilson <chris at chris-wilson.co.uk>

Thank you.

-Ram
> -Chris


More information about the Intel-gfx mailing list