Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Subject: Re: [Intel-gfx] [PATCH] drm/i915: Implement fdinfo memory stats printing
Date: Tue, 6 Jun 2023 12:09:24 +0100	[thread overview]
Message-ID: <2ae9d24d-d1be-a072-90ec-6168a72778b9@linux.intel.com> (raw)
In-Reply-To: <ee3ee473-6800-d4b4-292c-135caa9571b3@linux.intel.com>


On 06/06/2023 10:47, Tvrtko Ursulin wrote:
> 
> On 05/06/2023 23:22, Andi Shyti wrote:
>> Hi Tvrtko,
>>
>> On Mon, Jun 05, 2023 at 03:37:20PM +0100, Tvrtko Ursulin wrote:
>>> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>
>>> Use the newly added drm_print_memory_stats helper to show memory
>>> utilisation of our objects in drm/driver specific fdinfo output.
>>>
>>> To collect the stats we walk the per memory regions object lists
>>> and accumulate object size into the respective drm_memory_stats
>>> categories.
>>>
>>> Objects with multiple possible placements are reported in multiple
>>> regions for total and shared sizes, while other categories are
>>> counted only for the currently active region.
>>>
>>> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>> Cc: Rob Clark <robdclark@gmail.com>
>>> ---
>>>   drivers/gpu/drm/i915/i915_drm_client.c | 66 ++++++++++++++++++++++++++
>>>   1 file changed, 66 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/i915/i915_drm_client.c 
>>> b/drivers/gpu/drm/i915/i915_drm_client.c
>>> index 2a44b3876cb5..2a40f763f8f6 100644
>>> --- a/drivers/gpu/drm/i915/i915_drm_client.c
>>> +++ b/drivers/gpu/drm/i915/i915_drm_client.c
>>> @@ -41,6 +41,70 @@ void __i915_drm_client_free(struct kref *kref)
>>>   }
>>>   #ifdef CONFIG_PROC_FS
>>> +static void
>>> +add_obj_meminfo(struct drm_i915_gem_object *obj,
>>> +        struct intel_memory_region *mr,
>>> +        struct drm_memory_stats stats[INTEL_REGION_UNKNOWN])
>>> +{
>>> +    u64 sz = obj->base.size;
>>> +    enum intel_region_id id;
>>> +    unsigned int i;
>>> +
>>> +    id = mr->id;
>>> +    if (obj->base.handle_count > 1)
>>> +        stats[id].shared += sz;
>>> +    else
>>> +        stats[id].private += sz;
>>> +
>>> +    if (i915_gem_object_has_pages(obj)) {
>>> +        stats[id].resident += sz;
>>> +
>>> +        if (!dma_resv_test_signaled(obj->base.resv,
>>> +                        dma_resv_usage_rw(true)))
>>> +            stats[id].active += sz;
>>> +        else if (i915_gem_object_is_shrinkable(obj) &&
>>> +             obj->mm.madv == I915_MADV_DONTNEED)
>>> +            stats[id].purgeable += sz;
>>
>> this is a bit off... otherwise:
>>
>> Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
> 
> Thanks!
> 
> I will sit on this until I can find time to add some tests to drm_fdinfo.c.

Actually.. Aravind pointed off list that my brain wasn't switched on 
when writing this. Fdinfo stats need to be per client and this patch is 
therefore almost completely bogus.

Regards,

Tvrtko

>>
>> Andi
>>
>>> +    }
>>> +
>>> +    /* Attribute size and shared to all possible object memory 
>>> regions. */
>>> +    for (i = 0; i < obj->mm.n_placements; i++) {
>>> +        if (obj->mm.placements[i] == mr)
>>> +            continue;
>>> +
>>> +        id = obj->mm.placements[i]->id;
>>> +        if (obj->base.handle_count > 1)
>>> +            stats[id].shared += sz;
>>> +        else
>>> +            stats[id].private += sz;
>>> +    }
>>> +}
>>> +
>>> +static void show_meminfo(struct drm_printer *p, struct drm_file *file)
>>> +{
>>> +    struct drm_i915_file_private *file_priv = file->driver_priv;
>>> +    struct drm_memory_stats stats[INTEL_REGION_UNKNOWN] = {};
>>> +    struct drm_i915_private *i915 = file_priv->i915;
>>> +    struct intel_memory_region *mr;
>>> +    unsigned int id;
>>> +
>>> +    for_each_memory_region(mr, i915, id) {
>>> +        struct drm_i915_gem_object *obj;
>>> +
>>> +        mutex_lock(&mr->objects.lock);
>>> +        list_for_each_entry(obj, &mr->objects.list, mm.region_link)
>>> +            add_obj_meminfo(obj, mr, stats);
>>> +        mutex_unlock(&mr->objects.lock);
>>> +    }
>>> +
>>> +    for_each_memory_region(mr, i915, id)
>>> +        drm_print_memory_stats(p,
>>> +                       &stats[id],
>>> +                       DRM_GEM_OBJECT_RESIDENT |
>>> +                       DRM_GEM_OBJECT_PURGEABLE,
>>> +                       mr->name);
>>> +}
>>> +
>>>   static const char * const uabi_class_names[] = {
>>>       [I915_ENGINE_CLASS_RENDER] = "render",
>>>       [I915_ENGINE_CLASS_COPY] = "copy",
>>> @@ -102,6 +166,8 @@ void i915_drm_client_fdinfo(struct drm_printer 
>>> *p, struct drm_file *file)
>>>        * 
>>> ******************************************************************
>>>        */
>>> +    show_meminfo(p, file);
>>> +
>>>       if (GRAPHICS_VER(i915) < 8)
>>>           return;
>>> -- 
>>> 2.39.2

      reply	other threads:[~2023-06-06 11:09 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-05 14:37 [Intel-gfx] [PATCH] drm/i915: Implement fdinfo memory stats printing Tvrtko Ursulin
2023-06-05 21:24 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for " Patchwork
2023-06-05 22:22 ` [Intel-gfx] [PATCH] " Andi Shyti
2023-06-06  9:47   ` Tvrtko Ursulin
2023-06-06 11:09     ` Tvrtko Ursulin [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=2ae9d24d-d1be-a072-90ec-6168a72778b9@linux.intel.com \
    --to=tvrtko.ursulin@linux.intel.com \
    --cc=Intel-gfx@lists.freedesktop.org \
    --cc=andi.shyti@linux.intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox