FAILED: patch "[PATCH] drm/v3d: Skip CSD when it has zeroed workgroups" failed to apply to 6.1-stable tree

FAILED: patch "[PATCH] drm/v3d: Skip CSD when it has zeroed workgroups" failed to apply to 6.1-stable tree

[gregkh]

The patch below does not apply to the 6.1-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable@vger.kernel.org>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y
git checkout FETCH_HEAD
git cherry-pick -x 7f93fad5ea0affc9e1505dd0f7596c0fdb496213
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<stable@vger.kernel.org>' --in-reply-to '2026061526-coil-arbitrary-316e@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^..

Possible dependencies:



thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 7f93fad5ea0affc9e1505dd0f7596c0fdb496213 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 2 Jun 2026 14:50:15 -0300
Subject: [PATCH] drm/v3d: Skip CSD when it has zeroed workgroups
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A compute shader dispatch encodes its workgroup counts in the CFG0..CFG2
registers. Kicking off a dispatch with a zero count in any of the three
dimensions is invalid. First, the hardware will process 0 as 65536,
while the user-space driver exposes a maximum of 65535. Over that, a
submission with a zeroed workgroup dimension should be a no-op.

These zeroed counts can reach the dispatch path through an indirect CSD
job, whose workgroup counts are only known once the indirect buffer is
read and may legitimately be zero, but such scenario should only result in
a no-op.

Overwrite the indirect CSD job workgroup counts with the indirect BO
ones, even if they are zeroed, and don't submit the job to the hardware
when any of the workgroup counts is zero, so the job completes immediately
instead of running the shader.

Cc: stable@vger.kernel.org
Fixes: d223f98f0209 ("drm/v3d: Add support for compute shader dispatch.")
Suggested-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://patch.msgid.link/20260602-v3d-fix-indirect-csd-v4-2-654309e32bc0@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>

diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index 47f83936cd73..8a635a9ec046 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -352,6 +352,16 @@ v3d_csd_job_run(struct drm_sched_job *sched_job)
 		return NULL;
 	}
 
+	/* The HW interprets a workgroup size of 0 as 65536; however, the
+	 * user-space driver exposes a maximum of 65535. Therefore, a 0 in
+	 * any dimension means that we have no workgroups and the compute
+	 * shader should not be dispatched.
+	 */
+	if (!V3D_GET_FIELD(job->args.cfg[0], V3D_CSD_QUEUED_CFG0_NUM_WGS_X) ||
+	    !V3D_GET_FIELD(job->args.cfg[1], V3D_CSD_QUEUED_CFG1_NUM_WGS_Y) ||
+	    !V3D_GET_FIELD(job->args.cfg[2], V3D_CSD_QUEUED_CFG2_NUM_WGS_Z))
+		return NULL;
+
 	v3d->queue[V3D_CSD].active_job = &job->base;
 
 	v3d_invalidate_caches(v3d);
@@ -402,13 +412,13 @@ v3d_rewrite_csd_job_wg_counts_from_indirect(struct v3d_cpu_job *job)
 
 	wg_counts = (uint32_t *)(bo->vaddr + indirect_csd->offset);
 
-	if (wg_counts[0] == 0 || wg_counts[1] == 0 || wg_counts[2] == 0)
-		goto unmap_bo;
-
 	args->cfg[0] = wg_counts[0] << V3D_CSD_CFG012_WG_COUNT_SHIFT;
 	args->cfg[1] = wg_counts[1] << V3D_CSD_CFG012_WG_COUNT_SHIFT;
 	args->cfg[2] = wg_counts[2] << V3D_CSD_CFG012_WG_COUNT_SHIFT;
 
+	if (wg_counts[0] == 0 || wg_counts[1] == 0 || wg_counts[2] == 0)
+		goto unmap_bo;
+
 	num_batches = DIV_ROUND_UP(indirect_csd->wg_size, 16) *
 		      (wg_counts[0] * wg_counts[1] * wg_counts[2]);
 

[PATCH 6.1.y 1/2] drm/v3d: Store the active job inside the queue's state

[Sasha Levin]

From: Maíra Canal <mcanal@igalia.com>

[ Upstream commit 0d3768826d38c0ac740f8b45cd13346630535f2b ]

Instead of storing the queue's active job in four different variables,
store the active job inside the queue's state. This way, it's possible
to access all active jobs using an index based in `enum v3d_queue`.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-2-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Stable-dep-of: 7f93fad5ea0a ("drm/v3d: Skip CSD when it has zeroed workgroups")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/v3d/v3d_drv.h   |  8 +++-----
 drivers/gpu/drm/v3d/v3d_gem.c   |  5 +++--
 drivers/gpu/drm/v3d/v3d_irq.c   | 24 ++++++++++++++----------
 drivers/gpu/drm/v3d/v3d_sched.c | 26 ++++++++++++++++++--------
 4 files changed, 38 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
index a366ea208787df..c1e389635de816 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.h
+++ b/drivers/gpu/drm/v3d/v3d_drv.h
@@ -26,6 +26,9 @@ struct v3d_queue_state {
 
 	u64 fence_context;
 	u64 emit_seqno;
+
+	/* Currently active job for this queue */
+	struct v3d_job *active_job;
 };
 
 /* Performance monitor object. The perform lifetime is controlled by userspace
@@ -110,11 +113,6 @@ struct v3d_dev {
 
 	struct work_struct overflow_mem_work;
 
-	struct v3d_bin_job *bin_job;
-	struct v3d_render_job *render_job;
-	struct v3d_tfu_job *tfu_job;
-	struct v3d_csd_job *csd_job;
-
 	struct v3d_queue_state queue[V3D_MAX_QUEUES];
 
 	/* Spinlock used to synchronize the overflow memory
diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c
index 8b6450a96ebc39..6d5fa39423b24a 100644
--- a/drivers/gpu/drm/v3d/v3d_gem.c
+++ b/drivers/gpu/drm/v3d/v3d_gem.c
@@ -1115,14 +1115,15 @@ void
 v3d_gem_destroy(struct drm_device *dev)
 {
 	struct v3d_dev *v3d = to_v3d_dev(dev);
+	enum v3d_queue q;
 
 	v3d_sched_fini(v3d);
 
 	/* Waiting for jobs to finish would need to be done before
 	 * unregistering V3D.
 	 */
-	WARN_ON(v3d->bin_job);
-	WARN_ON(v3d->render_job);
+	for (q = 0; q < V3D_MAX_QUEUES; q++)
+		WARN_ON(v3d->queue[q].active_job);
 
 	drm_mm_takedown(&v3d->mm);
 
diff --git a/drivers/gpu/drm/v3d/v3d_irq.c b/drivers/gpu/drm/v3d/v3d_irq.c
index 641315dbee8b29..1717504742a665 100644
--- a/drivers/gpu/drm/v3d/v3d_irq.c
+++ b/drivers/gpu/drm/v3d/v3d_irq.c
@@ -40,6 +40,8 @@ v3d_overflow_mem_work(struct work_struct *work)
 		container_of(work, struct v3d_dev, overflow_mem_work);
 	struct drm_device *dev = &v3d->drm;
 	struct v3d_bo *bo = v3d_bo_create(dev, NULL /* XXX: GMP */, 256 * 1024);
+	struct v3d_queue_state *queue = &v3d->queue[V3D_BIN];
+	struct v3d_bin_job *bin_job;
 	struct drm_gem_object *obj;
 	unsigned long irqflags;
 
@@ -59,13 +61,15 @@ v3d_overflow_mem_work(struct work_struct *work)
 	 * some binner pool anyway.
 	 */
 	spin_lock_irqsave(&v3d->job_lock, irqflags);
-	if (!v3d->bin_job) {
+	bin_job = (struct v3d_bin_job *)queue->active_job;
+
+	if (!bin_job) {
 		spin_unlock_irqrestore(&v3d->job_lock, irqflags);
 		goto out;
 	}
 
 	drm_gem_object_get(obj);
-	list_add_tail(&bo->unref_head, &v3d->bin_job->render->unref_list);
+	list_add_tail(&bo->unref_head, &bin_job->render->unref_list);
 	spin_unlock_irqrestore(&v3d->job_lock, irqflags);
 
 	V3D_CORE_WRITE(0, V3D_PTB_BPOA, bo->node.start << PAGE_SHIFT);
@@ -99,11 +103,11 @@ v3d_irq(int irq, void *arg)
 
 	if (intsts & V3D_INT_FLDONE) {
 		struct v3d_fence *fence =
-			to_v3d_fence(v3d->bin_job->base.irq_fence);
+			to_v3d_fence(v3d->queue[V3D_BIN].active_job->irq_fence);
 
 		trace_v3d_bcl_irq(&v3d->drm, fence->seqno);
 
-		v3d->bin_job = NULL;
+		v3d->queue[V3D_BIN].active_job = NULL;
 		dma_fence_signal(&fence->base);
 
 		status = IRQ_HANDLED;
@@ -111,11 +115,11 @@ v3d_irq(int irq, void *arg)
 
 	if (intsts & V3D_INT_FRDONE) {
 		struct v3d_fence *fence =
-			to_v3d_fence(v3d->render_job->base.irq_fence);
+			to_v3d_fence(v3d->queue[V3D_RENDER].active_job->irq_fence);
 
 		trace_v3d_rcl_irq(&v3d->drm, fence->seqno);
 
-		v3d->render_job = NULL;
+		v3d->queue[V3D_RENDER].active_job = NULL;
 		dma_fence_signal(&fence->base);
 
 		status = IRQ_HANDLED;
@@ -123,11 +127,11 @@ v3d_irq(int irq, void *arg)
 
 	if (intsts & V3D_INT_CSDDONE) {
 		struct v3d_fence *fence =
-			to_v3d_fence(v3d->csd_job->base.irq_fence);
+			to_v3d_fence(v3d->queue[V3D_CSD].active_job->irq_fence);
 
 		trace_v3d_csd_irq(&v3d->drm, fence->seqno);
 
-		v3d->csd_job = NULL;
+		v3d->queue[V3D_CSD].active_job = NULL;
 		dma_fence_signal(&fence->base);
 
 		status = IRQ_HANDLED;
@@ -162,11 +166,11 @@ v3d_hub_irq(int irq, void *arg)
 
 	if (intsts & V3D_HUB_INT_TFUC) {
 		struct v3d_fence *fence =
-			to_v3d_fence(v3d->tfu_job->base.irq_fence);
+			to_v3d_fence(v3d->queue[V3D_TFU].active_job->irq_fence);
 
 		trace_v3d_tfu_irq(&v3d->drm, fence->seqno);
 
-		v3d->tfu_job = NULL;
+		v3d->queue[V3D_TFU].active_job = NULL;
 		dma_fence_signal(&fence->base);
 
 		status = IRQ_HANDLED;
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index 41493cf3d03b81..ff91cdb75bb912 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -80,14 +80,18 @@ static struct dma_fence *v3d_bin_job_run(struct drm_sched_job *sched_job)
 	struct dma_fence *fence;
 	unsigned long irqflags;
 
-	if (unlikely(job->base.base.s_fence->finished.error))
+	if (unlikely(job->base.base.s_fence->finished.error)) {
+		spin_lock_irqsave(&v3d->job_lock, irqflags);
+		v3d->queue[V3D_BIN].active_job = NULL;
+		spin_unlock_irqrestore(&v3d->job_lock, irqflags);
 		return NULL;
+	}
 
 	/* Lock required around bin_job update vs
 	 * v3d_overflow_mem_work().
 	 */
 	spin_lock_irqsave(&v3d->job_lock, irqflags);
-	v3d->bin_job = job;
+	v3d->queue[V3D_BIN].active_job = &job->base;
 	/* Clear out the overflow allocation, so we don't
 	 * reuse the overflow attached to a previous job.
 	 */
@@ -134,10 +138,12 @@ static struct dma_fence *v3d_render_job_run(struct drm_sched_job *sched_job)
 	struct drm_device *dev = &v3d->drm;
 	struct dma_fence *fence;
 
-	if (unlikely(job->base.base.s_fence->finished.error))
+	if (unlikely(job->base.base.s_fence->finished.error)) {
+		v3d->queue[V3D_RENDER].active_job = NULL;
 		return NULL;
+	}
 
-	v3d->render_job = job;
+	v3d->queue[V3D_RENDER].active_job = &job->base;
 
 	/* Can we avoid this flush?  We need to be careful of
 	 * scheduling, though -- imagine job0 rendering to texture and
@@ -179,10 +185,12 @@ v3d_tfu_job_run(struct drm_sched_job *sched_job)
 	struct drm_device *dev = &v3d->drm;
 	struct dma_fence *fence;
 
-	if (unlikely(job->base.base.s_fence->finished.error))
+	if (unlikely(job->base.base.s_fence->finished.error)) {
+		v3d->queue[V3D_TFU].active_job = NULL;
 		return NULL;
+	}
 
-	v3d->tfu_job = job;
+	v3d->queue[V3D_TFU].active_job = &job->base;
 
 	fence = v3d_fence_create(v3d, V3D_TFU);
 	if (IS_ERR(fence))
@@ -221,10 +229,12 @@ v3d_csd_job_run(struct drm_sched_job *sched_job)
 	struct dma_fence *fence;
 	int i;
 
-	if (unlikely(job->base.base.s_fence->finished.error))
+	if (unlikely(job->base.base.s_fence->finished.error)) {
+		v3d->queue[V3D_CSD].active_job = NULL;
 		return NULL;
+	}
 
-	v3d->csd_job = job;
+	v3d->queue[V3D_CSD].active_job = &job->base;
 
 	v3d_invalidate_caches(v3d);
 
-- 
2.53.0

[PATCH 6.1.y 2/2] drm/v3d: Skip CSD when it has zeroed workgroups

[Sasha Levin]

From: Maíra Canal <mcanal@igalia.com>

[ Upstream commit 7f93fad5ea0affc9e1505dd0f7596c0fdb496213 ]

A compute shader dispatch encodes its workgroup counts in the CFG0..CFG2
registers. Kicking off a dispatch with a zero count in any of the three
dimensions is invalid. First, the hardware will process 0 as 65536,
while the user-space driver exposes a maximum of 65535. Over that, a
submission with a zeroed workgroup dimension should be a no-op.

These zeroed counts can reach the dispatch path through an indirect CSD
job, whose workgroup counts are only known once the indirect buffer is
read and may legitimately be zero, but such scenario should only result in
a no-op.

Overwrite the indirect CSD job workgroup counts with the indirect BO
ones, even if they are zeroed, and don't submit the job to the hardware
when any of the workgroup counts is zero, so the job completes immediately
instead of running the shader.

Cc: stable@vger.kernel.org
Fixes: d223f98f0209 ("drm/v3d: Add support for compute shader dispatch.")
Suggested-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://patch.msgid.link/20260602-v3d-fix-indirect-csd-v4-2-654309e32bc0@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/v3d/v3d_sched.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index ff91cdb75bb912..ab872bc818004a 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -234,6 +234,16 @@ v3d_csd_job_run(struct drm_sched_job *sched_job)
 		return NULL;
 	}
 
+	/* The HW interprets a workgroup size of 0 as 65536; however, the
+	 * user-space driver exposes a maximum of 65535. Therefore, a 0 in
+	 * any dimension means that we have no workgroups and the compute
+	 * shader should not be dispatched.
+	 */
+	if (!V3D_GET_FIELD(job->args.cfg[0], V3D_CSD_QUEUED_CFG0_NUM_WGS_X) ||
+	    !V3D_GET_FIELD(job->args.cfg[1], V3D_CSD_QUEUED_CFG1_NUM_WGS_Y) ||
+	    !V3D_GET_FIELD(job->args.cfg[2], V3D_CSD_QUEUED_CFG2_NUM_WGS_Z))
+		return NULL;
+
 	v3d->queue[V3D_CSD].active_job = &job->base;
 
 	v3d_invalidate_caches(v3d);
-- 
2.53.0