From: Maxime Ripard <mripard@kernel.org>
To: "Andrew Morton" <akpm@linux-foundation.org>,
"Marek Szyprowski" <m.szyprowski@samsung.com>,
"Robin Murphy" <robin.murphy@arm.com>,
"Sumit Semwal" <sumit.semwal@linaro.org>,
"Christian König" <christian.koenig@amd.com>,
"Benjamin Gaignard" <benjamin.gaignard@collabora.com>,
"Brian Starkey" <Brian.Starkey@arm.com>,
"John Stultz" <jstultz@google.com>,
"T.J. Mercier" <tjmercier@google.com>,
"Maarten Lankhorst" <maarten.lankhorst@linux.intel.com>,
"Thomas Zimmermann" <tzimmermann@suse.de>,
"David Airlie" <airlied@gmail.com>,
"Simona Vetter" <simona@ffwll.ch>,
"Tomasz Figa" <tfiga@chromium.org>,
"Mauro Carvalho Chehab" <mchehab@kernel.org>
Cc: Hans Verkuil <hverkuil@xs4all.nl>,
Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
iommu@lists.linux.dev, linux-media@vger.kernel.org,
dri-devel@lists.freedesktop.org, linaro-mm-sig@lists.linaro.org,
Maxime Ripard <mripard@kernel.org>
Subject: [PATCH RFC 12/12] media: videobuf2: Track buffer allocations through the dmem cgroup
Date: Mon, 10 Mar 2025 13:06:18 +0100 [thread overview]
Message-ID: <20250310-dmem-cgroups-v1-12-2984c1bc9312@kernel.org> (raw)
In-Reply-To: <20250310-dmem-cgroups-v1-0-2984c1bc9312@kernel.org>
The dmem cgroup allows to track any DMA memory allocation made by the
userspace. Let's charge our allocations in videobuf2 to enable proper
memory tracking.
Signed-off-by: Maxime Ripard <mripard@kernel.org>
---
drivers/media/common/videobuf2/videobuf2-dma-contig.c | 19 +++++++++++++++++++
1 file changed, 19 insertions(+)
diff --git a/drivers/media/common/videobuf2/videobuf2-dma-contig.c b/drivers/media/common/videobuf2/videobuf2-dma-contig.c
index a13ec569c82f6da2d977222b94af32e74c6c6c82..48384e18030812f4f89f1c225c38def2ac6aa3ca 100644
--- a/drivers/media/common/videobuf2/videobuf2-dma-contig.c
+++ b/drivers/media/common/videobuf2/videobuf2-dma-contig.c
@@ -8,10 +8,11 @@
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation.
*/
+#include <linux/cgroup_dmem.h>
#include <linux/dma-buf.h>
#include <linux/module.h>
#include <linux/refcount.h>
#include <linux/scatterlist.h>
#include <linux/sched.h>
@@ -40,10 +41,14 @@ struct vb2_dc_buf {
struct sg_table *sgt_base;
/* DMABUF related */
struct dma_buf_attachment *db_attach;
+#ifdef CONFIG_CGROUP_DMEM
+ struct dmem_cgroup_pool_state *cgroup_pool_state;
+#endif
+
struct vb2_buffer *vb;
bool non_coherent_mem;
};
/*********************************************/
@@ -169,10 +174,14 @@ static void vb2_dc_put(void *buf_priv)
struct vb2_dc_buf *buf = buf_priv;
if (!refcount_dec_and_test(&buf->refcount))
return;
+#ifdef CONFIG_CGROUP_DMEM
+ dmem_cgroup_uncharge(buf->cgroup_pool_state, buf->size);
+#endif
+
if (buf->non_coherent_mem) {
if (buf->vaddr)
dma_vunmap_noncontiguous(buf->dev, buf->vaddr);
dma_free_noncontiguous(buf->dev, buf->size,
buf->dma_sgt, buf->dma_dir);
@@ -230,10 +239,11 @@ static int vb2_dc_alloc_non_coherent(struct vb2_dc_buf *buf)
static void *vb2_dc_alloc(struct vb2_buffer *vb,
struct device *dev,
unsigned long size)
{
+ struct dmem_cgroup_pool_state *pool;
struct vb2_dc_buf *buf;
int ret;
if (WARN_ON(!dev))
return ERR_PTR(-EINVAL);
@@ -249,25 +259,34 @@ static void *vb2_dc_alloc(struct vb2_buffer *vb,
buf->size = size;
/* Prevent the device from being released while the buffer is used */
buf->dev = get_device(dev);
+ ret = dmem_cgroup_try_charge(dma_get_dmem_cgroup_region(dev), size, &pool, NULL);
+ if (ret)
+ return ret;
+
if (buf->non_coherent_mem)
ret = vb2_dc_alloc_non_coherent(buf);
else
ret = vb2_dc_alloc_coherent(buf);
if (ret) {
dev_err(dev, "dma alloc of size %lu failed\n", size);
+ dmem_cgroup_uncharge(pool, size);
kfree(buf);
return ERR_PTR(-ENOMEM);
}
buf->handler.refcount = &buf->refcount;
buf->handler.put = vb2_dc_put;
buf->handler.arg = buf;
+#ifdef CONFIG_CGROUP_DMEM
+ buf->cgroup_pool_state = pool;
+#endif
+
refcount_set(&buf->refcount, 1);
return buf;
}
--
2.48.1
next prev parent reply other threads:[~2025-03-10 12:06 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-10 12:06 [PATCH RFC 00/12] dma: Enable dmem cgroup tracking Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 01/12] cma: Register dmem region for each cma region Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 02/12] cma: Provide accessor to cma dmem region Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 03/12] dma: coherent: Register dmem region for each coherent region Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 04/12] dma: coherent: Provide accessor to dmem region Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 05/12] dma: contiguous: " Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 06/12] dma: direct: " Maxime Ripard
2025-03-10 14:56 ` Robin Murphy
2025-03-10 16:28 ` Maxime Ripard
2025-03-10 18:44 ` Robin Murphy
2025-03-13 18:16 ` Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 07/12] dma: Create default dmem region for DMA allocations Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 08/12] dma: Provide accessor to dmem region Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 09/12] dma-buf: Clear cgroup accounting on release Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 10/12] dma-buf: cma: Account for allocations in dmem cgroup Maxime Ripard
2025-03-10 12:06 ` [PATCH RFC 11/12] drm/gem: Add cgroup memory accounting Maxime Ripard
2025-03-10 15:06 ` Robin Murphy
2025-03-10 12:06 ` Maxime Ripard [this message]
2025-03-10 12:15 ` [PATCH RFC 00/12] dma: Enable dmem cgroup tracking Maxime Ripard
2025-03-10 14:16 ` Christian König
2025-03-10 14:26 ` Maxime Ripard
2025-03-31 20:43 ` Dave Airlie
2025-04-01 11:03 ` Christian König
2025-04-03 6:07 ` Dave Airlie
2025-04-03 7:39 ` Christian König
2025-04-03 15:47 ` Maxime Ripard
2025-04-04 8:47 ` Christian König
2025-04-05 1:57 ` T.J. Mercier
2025-04-07 11:46 ` Christian König
2025-04-08 1:03 ` T.J. Mercier
2025-04-03 8:27 ` Simona Vetter
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=20250310-dmem-cgroups-v1-12-2984c1bc9312@kernel.org \
--to=mripard@kernel.org \
--cc=Brian.Starkey@arm.com \
--cc=airlied@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=benjamin.gaignard@collabora.com \
--cc=christian.koenig@amd.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=hverkuil@xs4all.nl \
--cc=iommu@lists.linux.dev \
--cc=jstultz@google.com \
--cc=laurent.pinchart+renesas@ideasonboard.com \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=m.szyprowski@samsung.com \
--cc=maarten.lankhorst@linux.intel.com \
--cc=mchehab@kernel.org \
--cc=robin.murphy@arm.com \
--cc=simona@ffwll.ch \
--cc=sumit.semwal@linaro.org \
--cc=tfiga@chromium.org \
--cc=tjmercier@google.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox