From: Rob Clark <robin.clark@oss.qualcomm.com>
To: dri-devel@lists.freedesktop.org
Cc: freedreno@lists.freedesktop.org, linux-arm-msm@vger.kernel.org,
Connor Abbott <cwabbott0@gmail.com>,
Rob Clark <robin.clark@oss.qualcomm.com>,
Rob Clark <robdclark@gmail.com>,
Abhinav Kumar <quic_abhinavk@quicinc.com>,
Dmitry Baryshkov <lumag@kernel.org>, Sean Paul <sean@poorly.run>,
Marijn Suijten <marijn.suijten@somainline.org>,
David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>,
linux-kernel@vger.kernel.org (open list)
Subject: [PATCH v6 10/40] drm/msm: Refcount framebuffer pins
Date: Thu, 5 Jun 2025 11:28:55 -0700 [thread overview]
Message-ID: <20250605183111.163594-11-robin.clark@oss.qualcomm.com> (raw)
In-Reply-To: <20250605183111.163594-1-robin.clark@oss.qualcomm.com>
We were already keeping a refcount of # of prepares (pins), to clear the
iova array. Use that to avoid unpinning the iova until the last cleanup
(unpin). This way, when msm_gem_unpin_iova() actually tears down the
mapping, we won't have problems if the fb is being scanned out on
another display (for example).
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
drivers/gpu/drm/msm/msm_fb.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/msm/msm_fb.c b/drivers/gpu/drm/msm/msm_fb.c
index 8a3b88130f4d..3b17d83f6673 100644
--- a/drivers/gpu/drm/msm/msm_fb.c
+++ b/drivers/gpu/drm/msm/msm_fb.c
@@ -85,7 +85,8 @@ int msm_framebuffer_prepare(struct drm_framebuffer *fb, bool needs_dirtyfb)
if (needs_dirtyfb)
refcount_inc(&msm_fb->dirtyfb);
- atomic_inc(&msm_fb->prepare_count);
+ if (atomic_inc_return(&msm_fb->prepare_count) > 1)
+ return 0;
for (i = 0; i < n; i++) {
ret = msm_gem_get_and_pin_iova(fb->obj[i], vm, &msm_fb->iova[i]);
@@ -108,11 +109,13 @@ void msm_framebuffer_cleanup(struct drm_framebuffer *fb, bool needed_dirtyfb)
if (needed_dirtyfb)
refcount_dec(&msm_fb->dirtyfb);
+ if (atomic_dec_return(&msm_fb->prepare_count))
+ return;
+
+ memset(msm_fb->iova, 0, sizeof(msm_fb->iova));
+
for (i = 0; i < n; i++)
msm_gem_unpin_iova(fb->obj[i], vm);
-
- if (!atomic_dec_return(&msm_fb->prepare_count))
- memset(msm_fb->iova, 0, sizeof(msm_fb->iova));
}
uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb, int plane)
--
2.49.0
next prev parent reply other threads:[~2025-06-05 18:32 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-05 18:28 [PATCH v6 00/40] drm/msm: sparse / "VM_BIND" support Rob Clark
2025-06-05 18:28 ` [PATCH v6 01/40] drm/gem: Add ww_acquire_ctx support to drm_gem_lru_scan() Rob Clark
2025-06-05 18:28 ` [PATCH v6 02/40] drm/msm: Rename msm_file_private -> msm_context Rob Clark
2025-06-05 18:28 ` [PATCH v6 03/40] drm/msm: Improve msm_context comments Rob Clark
2025-06-05 18:28 ` [PATCH v6 04/40] drm/msm: Rename msm_gem_address_space -> msm_gem_vm Rob Clark
2025-06-05 18:28 ` [PATCH v6 05/40] drm/msm: Remove vram carveout support Rob Clark
2025-06-05 18:28 ` [PATCH v6 06/40] drm/msm: Collapse vma allocation and initialization Rob Clark
2025-06-05 18:28 ` [PATCH v6 07/40] drm/msm: Collapse vma close and delete Rob Clark
2025-06-05 18:28 ` [PATCH v6 08/40] drm/msm: Don't close VMAs on purge Rob Clark
2025-06-05 18:28 ` [PATCH v6 09/40] drm/msm: Stop passing vm to msm_framebuffer Rob Clark
2025-06-05 18:28 ` Rob Clark [this message]
2025-06-05 18:28 ` [PATCH v6 11/40] drm/msm: drm_gpuvm conversion Rob Clark
2025-06-05 18:28 ` [PATCH v6 12/40] drm/msm: Convert vm locking Rob Clark
2025-06-05 18:28 ` [PATCH v6 13/40] drm/msm: Use drm_gpuvm types more Rob Clark
2025-06-05 18:28 ` [PATCH v6 14/40] drm/msm: Split out helper to get iommu prot flags Rob Clark
2025-06-05 18:29 ` [PATCH v6 15/40] drm/msm: Add mmu support for non-zero offset Rob Clark
2025-06-05 18:29 ` [PATCH v6 16/40] drm/msm: Add PRR support Rob Clark
2025-06-05 18:29 ` [PATCH v6 17/40] drm/msm: Rename msm_gem_vma_purge() -> _unmap() Rob Clark
2025-06-05 18:29 ` [PATCH v6 18/40] drm/msm: Drop queued submits on lastclose() Rob Clark
2025-06-05 18:29 ` [PATCH v6 19/40] drm/msm: Lazily create context VM Rob Clark
2025-06-05 18:29 ` [PATCH v6 20/40] drm/msm: Add opt-in for VM_BIND Rob Clark
2025-06-05 18:29 ` [PATCH v6 21/40] drm/msm: Mark VM as unusable on GPU hangs Rob Clark
2025-06-05 18:29 ` [PATCH v6 22/40] drm/msm: Add _NO_SHARE flag Rob Clark
2025-06-05 18:29 ` [PATCH v6 23/40] drm/msm: Crashdump prep for sparse mappings Rob Clark
2025-06-05 18:29 ` [PATCH v6 24/40] drm/msm: rd dumping " Rob Clark
2025-06-05 18:29 ` [PATCH v6 25/40] drm/msm: Crashdump support for sparse Rob Clark
2025-06-05 18:29 ` [PATCH v6 26/40] drm/msm: rd dumping " Rob Clark
2025-06-05 18:29 ` [PATCH v6 27/40] drm/msm: Extract out syncobj helpers Rob Clark
2025-06-05 18:29 ` [PATCH v6 28/40] drm/msm: Use DMA_RESV_USAGE_BOOKKEEP/KERNEL Rob Clark
2025-06-05 18:29 ` [PATCH v6 29/40] drm/msm: Add VM_BIND submitqueue Rob Clark
2025-06-05 18:29 ` [PATCH v6 30/40] drm/msm: Support IO_PGTABLE_QUIRK_NO_WARN_ON Rob Clark
2025-06-05 18:29 ` [PATCH v6 31/40] drm/msm: Support pgtable preallocation Rob Clark
2025-06-05 18:29 ` [PATCH v6 32/40] drm/msm: Split out map/unmap ops Rob Clark
2025-06-05 18:29 ` [PATCH v6 33/40] drm/msm: Add VM_BIND ioctl Rob Clark
2025-06-05 18:29 ` [PATCH v6 34/40] drm/msm: Add VM logging for VM_BIND updates Rob Clark
2025-06-05 18:29 ` [PATCH v6 35/40] drm/msm: Add VMA unmap reason Rob Clark
2025-06-05 18:29 ` [PATCH v6 36/40] drm/msm: Add mmu prealloc tracepoint Rob Clark
2025-06-05 18:29 ` [PATCH v6 37/40] drm/msm: use trylock for debugfs Rob Clark
2025-06-05 18:29 ` [PATCH v6 38/40] drm/msm: Bump UAPI version Rob Clark
2025-06-05 18:29 ` [PATCH v6 39/40] drm/msm: Defer VMA unmap for fb unpins Rob Clark
2025-06-05 18:29 ` [PATCH v6 40/40] drm/msm: Add VM_BIND throttling Rob Clark
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=20250605183111.163594-11-robin.clark@oss.qualcomm.com \
--to=robin.clark@oss.qualcomm.com \
--cc=airlied@gmail.com \
--cc=cwabbott0@gmail.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=freedreno@lists.freedesktop.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lumag@kernel.org \
--cc=marijn.suijten@somainline.org \
--cc=quic_abhinavk@quicinc.com \
--cc=robdclark@gmail.com \
--cc=sean@poorly.run \
--cc=simona@ffwll.ch \
/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
all inboxes | Powered by JetHome®