[RFC,15/24] drm/msm: Mark VM as unusable on faults

Commit Message

Rob Clark Dec. 7, 2024, 4:15 p.m. UTC

From: Rob Clark <[e-mail]>

If userspace has opted-in to VM_BIND, then GPU faults and VM_BIND errors
will mark the VM as unusable.

Signed-off-by: Rob Clark <[e-mail]>
---
 drivers/gpu/drm/msm/msm_gem.h        | 17 +++++++++++++++++
 drivers/gpu/drm/msm/msm_gem_submit.c |  3 +++
 drivers/gpu/drm/msm/msm_gpu.c        | 16 ++++++++++++++--
 3 files changed, 34 insertions(+), 2 deletions(-)

Patch

diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h
index acb976722580..7cb720137548 100644
--- a/drivers/gpu/drm/msm/msm_gem.h
+++ b/drivers/gpu/drm/msm/msm_gem.h
@@ -82,6 +82,23 @@  struct msm_gem_vm {
 
 	/** @managed: is this a kernel managed VM? */
 	bool managed;
+
+	/**
+	 * @unusable: True if the VM has turned unusable because something
+	 * bad happened during an asynchronous request.
+	 *
+	 * We don't try to recover from such failures, because this implies
+	 * informing userspace about the specific operation that failed, and
+	 * hoping the userspace driver can replay things from there. This all
+	 * sounds very complicated for little gain.
+	 *
+	 * Instead, we should just flag the VM as unusable, and fail any
+	 * further request targeting this VM.
+	 *
+	 * As an analogy, this would be mapped to a VK_ERROR_DEVICE_LOST
+	 * situation, where the logical device needs to be re-created.
+	 */
+	bool unusable;
 };
 #define to_msm_vm(x) container_of(x, struct msm_gem_vm, base)
 
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 5e37e1dad5bb..79bbe552f23e 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -668,6 +668,9 @@  int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	if (args->pad)
 		return -EINVAL;
 
+	if (to_msm_vm(ctx->vm)->unusable)
+		return UERR(EPIPE, dev, "context is unusable");
+
 	/* for now, we just have 3d pipe.. eventually this would need to
 	 * be more clever to dispatch to appropriate gpu module:
 	 */
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 5def12abac6c..72e5ad69a08c 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -377,8 +377,20 @@  static void recover_worker(struct kthread_work *work)
 
 	/* Increment the fault counts */
 	submit->queue->faults++;
-	if (submit->vm)
-		to_msm_vm(submit->vm)->faults++;
+	if (submit->vm) {
+		struct msm_gem_vm *vm = to_msm_vm(submit->vm);
+
+		vm->faults++;
+
+		/*
+		 * If userspace has opted-in to VM_BIND (and therefore userspace
+		 * management of the VM), faults mark the VM as unusuable.  This
+		 * matches vulkan expectations (vulkan is the main target for
+		 * VM_BIND)
+		 */
+		if (!vm->managed)
+			vm->unusable = true;
+	}
 
 	get_comm_cmdline(submit, &comm, &cmd);