From: Boris Brezillon <boris.brezillon@collabora.com>
To: "Adrián Larumbe" <adrian.larumbe@collabora.com>
Cc: maarten.lankhorst@linux.intel.com, mripard@kernel.org,
tzimmermann@suse.de, airlied@gmail.com, daniel@ffwll.ch,
robdclark@gmail.com, quic_abhinavk@quicinc.com,
dmitry.baryshkov@linaro.org, sean@poorly.run,
marijn.suijten@somainline.org, robh@kernel.org,
steven.price@arm.com, linux-arm-msm@vger.kernel.org,
linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
healych@amazon.com, kernel@collabora.com,
freedreno@lists.freedesktop.org
Subject: Re: [PATCH v2 4/6] drm/drm_file: Add DRM obj's RSS reporting function for fdinfo
Date: Wed, 30 Aug 2023 12:34:38 +0200 [thread overview]
Message-ID: <20230830123438.782a5e42@collabora.com> (raw)
In-Reply-To: <20230824013604.466224-5-adrian.larumbe@collabora.com>
On Thu, 24 Aug 2023 02:34:47 +0100
Adrián Larumbe <adrian.larumbe@collabora.com> wrote:
> Some BO's might be mapped onto physical memory chunkwise and on demand,
> like Panfrost's tiler heap. In this case, even though the
> drm_gem_shmem_object page array might already be allocated, only a very
> small fraction of the BO is currently backed by system memory, but
> drm_show_memory_stats will then proceed to add its entire virtual size to
> the file's total resident size regardless.
>
> This led to very unrealistic RSS sizes being reckoned for Panfrost, where
> said tiler heap buffer is initially allocated with a virtual size of 128
> MiB, but only a small part of it will eventually be backed by system memory
> after successive GPU page faults.
>
> Provide a new DRM object generic function that would allow drivers to
> return a more accurate RSS size for their BOs.
>
> Signed-off-by: Adrián Larumbe <adrian.larumbe@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
> ---
> drivers/gpu/drm/drm_file.c | 5 ++++-
> include/drm/drm_gem.h | 9 +++++++++
> 2 files changed, 13 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c
> index 883d83bc0e3d..762965e3d503 100644
> --- a/drivers/gpu/drm/drm_file.c
> +++ b/drivers/gpu/drm/drm_file.c
> @@ -944,7 +944,10 @@ void drm_show_memory_stats(struct drm_printer *p, struct drm_file *file)
> }
>
> if (s & DRM_GEM_OBJECT_RESIDENT) {
> - status.resident += obj->size;
> + if (obj->funcs && obj->funcs->rss)
> + status.resident += obj->funcs->rss(obj);
> + else
> + status.resident += obj->size;
> } else {
> /* If already purged or not yet backed by pages, don't
> * count it as purgeable:
> diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h
> index c0b13c43b459..78ed9fab6044 100644
> --- a/include/drm/drm_gem.h
> +++ b/include/drm/drm_gem.h
> @@ -208,6 +208,15 @@ struct drm_gem_object_funcs {
> */
> enum drm_gem_object_status (*status)(struct drm_gem_object *obj);
>
> + /**
> + * @rss:
> + *
> + * Return resident size of the object in physical memory.
> + *
> + * Called by drm_show_memory_stats().
> + */
> + size_t (*rss)(struct drm_gem_object *obj);
> +
> /**
> * @vm_ops:
> *
WARNING: multiple messages have this Message-ID (diff)
From: Boris Brezillon <boris.brezillon@collabora.com>
To: "Adrián Larumbe" <adrian.larumbe@collabora.com>
Cc: tzimmermann@suse.de, sean@poorly.run, quic_abhinavk@quicinc.com,
mripard@kernel.org, steven.price@arm.com,
freedreno@lists.freedesktop.org, healych@amazon.com,
dri-devel@lists.freedesktop.org, linux-arm-msm@vger.kernel.org,
dmitry.baryshkov@linaro.org, marijn.suijten@somainline.org,
kernel@collabora.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 4/6] drm/drm_file: Add DRM obj's RSS reporting function for fdinfo
Date: Wed, 30 Aug 2023 12:34:38 +0200 [thread overview]
Message-ID: <20230830123438.782a5e42@collabora.com> (raw)
In-Reply-To: <20230824013604.466224-5-adrian.larumbe@collabora.com>
On Thu, 24 Aug 2023 02:34:47 +0100
Adrián Larumbe <adrian.larumbe@collabora.com> wrote:
> Some BO's might be mapped onto physical memory chunkwise and on demand,
> like Panfrost's tiler heap. In this case, even though the
> drm_gem_shmem_object page array might already be allocated, only a very
> small fraction of the BO is currently backed by system memory, but
> drm_show_memory_stats will then proceed to add its entire virtual size to
> the file's total resident size regardless.
>
> This led to very unrealistic RSS sizes being reckoned for Panfrost, where
> said tiler heap buffer is initially allocated with a virtual size of 128
> MiB, but only a small part of it will eventually be backed by system memory
> after successive GPU page faults.
>
> Provide a new DRM object generic function that would allow drivers to
> return a more accurate RSS size for their BOs.
>
> Signed-off-by: Adrián Larumbe <adrian.larumbe@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
> ---
> drivers/gpu/drm/drm_file.c | 5 ++++-
> include/drm/drm_gem.h | 9 +++++++++
> 2 files changed, 13 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c
> index 883d83bc0e3d..762965e3d503 100644
> --- a/drivers/gpu/drm/drm_file.c
> +++ b/drivers/gpu/drm/drm_file.c
> @@ -944,7 +944,10 @@ void drm_show_memory_stats(struct drm_printer *p, struct drm_file *file)
> }
>
> if (s & DRM_GEM_OBJECT_RESIDENT) {
> - status.resident += obj->size;
> + if (obj->funcs && obj->funcs->rss)
> + status.resident += obj->funcs->rss(obj);
> + else
> + status.resident += obj->size;
> } else {
> /* If already purged or not yet backed by pages, don't
> * count it as purgeable:
> diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h
> index c0b13c43b459..78ed9fab6044 100644
> --- a/include/drm/drm_gem.h
> +++ b/include/drm/drm_gem.h
> @@ -208,6 +208,15 @@ struct drm_gem_object_funcs {
> */
> enum drm_gem_object_status (*status)(struct drm_gem_object *obj);
>
> + /**
> + * @rss:
> + *
> + * Return resident size of the object in physical memory.
> + *
> + * Called by drm_show_memory_stats().
> + */
> + size_t (*rss)(struct drm_gem_object *obj);
> +
> /**
> * @vm_ops:
> *
next prev parent reply other threads:[~2023-08-30 18:32 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-24 1:34 [PATCH v2 0/6] Add fdinfo support to Panfrost Adrián Larumbe
2023-08-24 1:34 ` Adrián Larumbe
2023-08-24 1:34 ` [PATCH v2 1/6] drm/panfrost: Add cycle count GPU register definitions Adrián Larumbe
2023-08-24 1:34 ` Adrián Larumbe
2023-08-30 10:35 ` Boris Brezillon
2023-08-30 10:35 ` Boris Brezillon
2023-08-31 15:54 ` Steven Price
2023-08-31 15:54 ` Steven Price
2023-08-24 1:34 ` [PATCH v2 2/6] drm/panfrost: Add fdinfo support GPU load metrics Adrián Larumbe
2023-08-24 1:34 ` Adrián Larumbe
2023-08-24 4:12 ` kernel test robot
2023-08-24 4:12 ` kernel test robot
2023-08-30 10:17 ` Boris Brezillon
2023-08-30 10:17 ` Boris Brezillon
2023-08-31 23:23 ` Adrián Larumbe
2023-08-31 23:23 ` Adrián Larumbe
2023-08-31 15:54 ` Steven Price
2023-08-31 15:54 ` Steven Price
2023-08-31 21:34 ` Adrián Larumbe
2023-08-31 21:34 ` Adrián Larumbe
2023-09-04 8:22 ` Steven Price
2023-09-04 8:22 ` Steven Price
2023-09-02 3:20 ` kernel test robot
2023-09-02 3:20 ` kernel test robot
2023-08-24 1:34 ` [PATCH v2 3/6] drm/panfrost: Add fdinfo support for memory stats Adrián Larumbe
2023-08-24 1:34 ` Adrián Larumbe
2023-08-30 10:31 ` Boris Brezillon
2023-08-30 10:31 ` Boris Brezillon
2023-08-31 23:07 ` Adrián Larumbe
2023-08-31 23:07 ` Adrián Larumbe
2023-08-24 1:34 ` [PATCH v2 4/6] drm/drm_file: Add DRM obj's RSS reporting function for fdinfo Adrián Larumbe
2023-08-24 1:34 ` Adrián Larumbe
2023-08-30 10:34 ` Boris Brezillon [this message]
2023-08-30 10:34 ` Boris Brezillon
2023-08-24 1:34 ` [PATCH v2 5/6] drm/panfrost: Implement generic DRM object RSS reporting function Adrián Larumbe
2023-08-24 1:34 ` Adrián Larumbe
2023-08-24 11:13 ` kernel test robot
2023-08-24 11:13 ` kernel test robot
2023-08-30 10:52 ` Boris Brezillon
2023-08-30 10:52 ` Boris Brezillon
2023-09-01 0:03 ` Adrián Larumbe
2023-09-01 0:03 ` Adrián Larumbe
2023-09-01 6:44 ` Boris Brezillon
2023-09-01 6:44 ` Boris Brezillon
2023-08-24 1:34 ` [PATCH v2 6/6] drm/drm-file: Allow size unit selection in drm_show_memory_stats Adrián Larumbe
2023-08-24 1:34 ` Adrián Larumbe
2023-08-24 6:49 ` kernel test robot
2023-08-24 6:49 ` kernel test robot
2023-08-28 15:00 ` Rob Clark
2023-08-28 15:00 ` Rob Clark
2023-08-30 15:51 ` Adrián Larumbe
2023-08-30 15:51 ` Adrián Larumbe
2023-09-05 22:23 ` Rob Clark
2023-09-05 22:23 ` Rob Clark
2023-09-01 22:18 ` kernel test robot
2023-09-01 22:18 ` kernel test robot
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=20230830123438.782a5e42@collabora.com \
--to=boris.brezillon@collabora.com \
--cc=adrian.larumbe@collabora.com \
--cc=airlied@gmail.com \
--cc=daniel@ffwll.ch \
--cc=dmitry.baryshkov@linaro.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=freedreno@lists.freedesktop.org \
--cc=healych@amazon.com \
--cc=kernel@collabora.com \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=maarten.lankhorst@linux.intel.com \
--cc=marijn.suijten@somainline.org \
--cc=mripard@kernel.org \
--cc=quic_abhinavk@quicinc.com \
--cc=robdclark@gmail.com \
--cc=robh@kernel.org \
--cc=sean@poorly.run \
--cc=steven.price@arm.com \
--cc=tzimmermann@suse.de \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.