On 4/11/2025 7:54 PM, Alex Deucher wrote:
On Fri, Apr 11, 2025 at 9:05 AM Sunil Khatri <sunil.kha...@amd.com> wrote:
add process and pid information in the userqueue error
logging to make it more useful in resolving the error
by logs.

Sample log:
[   42.444297] [drm:amdgpu_userqueue_wait_for_signal [amdgpu]] *ERROR* Timed 
out waiting for fence f=000000001c74d978 for comm:Xwayland pid:3427
[   42.444669] [drm:amdgpu_userqueue_suspend [amdgpu]] *ERROR* Not suspending 
userqueue, timeout waiting for comm:Xwayland pid:3427
[   42.824729] [drm:amdgpu_userqueue_wait_for_signal [amdgpu]] *ERROR* Timed 
out waiting for fence f=0000000074407d3e for comm:systemd-logind pid:1058
[   42.825082] [drm:amdgpu_userqueue_suspend [amdgpu]] *ERROR* Not suspending 
userqueue, timeout waiting for comm:systemd-logind pid:1058

Signed-off-by: Sunil Khatri <sunil.kha...@amd.com>
---
  drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c | 45 +++++++++++++++----
  1 file changed, 37 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c
index ecd49cf15b2a..5b58c41618ee 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c
@@ -62,12 +62,17 @@ amdgpu_userqueue_cleanup(struct amdgpu_userq_mgr *uq_mgr,
         struct amdgpu_device *adev = uq_mgr->adev;
         const struct amdgpu_userq_funcs *uq_funcs = 
adev->userq_funcs[queue->queue_type];
         struct dma_fence *f = queue->last_fence;
+       struct drm_file *file;
+       char proc_log[50];
         int ret;

         if (f && !dma_fence_is_signaled(f)) {
                 ret = dma_fence_wait_timeout(f, true, msecs_to_jiffies(100));
                 if (ret <= 0) {
-                       DRM_ERROR("Timed out waiting for fence f=%p\n", f);
+                       file = uq_mgr->file;
+                       drm_process_info(file, proc_log, sizeof(proc_log));
+                       DRM_ERROR("Timed out waiting for fence f=%p for %s\n",
+                                 f, proc_log);
user drm_err() here and below so we get proper handling of multiple devices.

Alex
Sure Alex. Once i have the main drm patch reviewed would update these too.
Sunil

                         return;
                 }
         }
@@ -427,6 +432,8 @@ amdgpu_userqueue_resume_all(struct amdgpu_userq_mgr *uq_mgr)
         const struct amdgpu_userq_funcs *userq_funcs;
         struct amdgpu_usermode_queue *queue;
         int queue_id;
+       struct drm_file *file;
+       char proc_log[50];
         int ret = 0;

         /* Resume all the queues for this process */
@@ -435,8 +442,12 @@ amdgpu_userqueue_resume_all(struct amdgpu_userq_mgr 
*uq_mgr)
                 ret = userq_funcs->resume(uq_mgr, queue);
         }

-       if (ret)
-               DRM_ERROR("Failed to resume all the queue\n");
+       if (ret) {
+               file = uq_mgr->file;
+               drm_process_info(file, proc_log, sizeof(proc_log));
+               DRM_ERROR("Failed to resume all the queue for %s\n",
+                         proc_log);
+               }
         return ret;
  }

@@ -585,6 +596,8 @@ amdgpu_userqueue_suspend_all(struct amdgpu_userq_mgr 
*uq_mgr)
         const struct amdgpu_userq_funcs *userq_funcs;
         struct amdgpu_usermode_queue *queue;
         int queue_id;
+       struct drm_file *file;
+       char proc_log[50];
         int ret = 0;

         /* Try to suspend all the queues in this process ctx */
@@ -593,8 +606,12 @@ amdgpu_userqueue_suspend_all(struct amdgpu_userq_mgr 
*uq_mgr)
                 ret += userq_funcs->suspend(uq_mgr, queue);
         }

-       if (ret)
-               DRM_ERROR("Couldn't suspend all the queues\n");
+       if (ret) {
+               file = uq_mgr->file;
+               drm_process_info(file, proc_log, sizeof(proc_log));
+               DRM_ERROR("Couldn't suspend all the queues for %s\n",
+                         proc_log);
+               }
         return ret;
  }

@@ -602,6 +619,8 @@ static int
  amdgpu_userqueue_wait_for_signal(struct amdgpu_userq_mgr *uq_mgr)
  {
         struct amdgpu_usermode_queue *queue;
+       struct drm_file *file;
+       char proc_log[50];
         int queue_id, ret;

         idr_for_each_entry(&uq_mgr->userq_idr, queue, queue_id) {
@@ -611,7 +630,10 @@ amdgpu_userqueue_wait_for_signal(struct amdgpu_userq_mgr 
*uq_mgr)
                         continue;
                 ret = dma_fence_wait_timeout(f, true, msecs_to_jiffies(100));
                 if (ret <= 0) {
-                       DRM_ERROR("Timed out waiting for fence f=%p\n", f);
+                       file = uq_mgr->file;
+                       drm_process_info(file, proc_log, sizeof(proc_log));
+                       DRM_ERROR("Timed out waiting for fence f=%p for %s\n",
+                                 f, proc_log);
                         return -ETIMEDOUT;
                 }
         }
@@ -624,19 +646,26 @@ amdgpu_userqueue_suspend(struct amdgpu_userq_mgr *uq_mgr,
                          struct amdgpu_eviction_fence *ev_fence)
  {
         int ret;
+       struct drm_file *file;
+       char proc_log[50];
         struct amdgpu_fpriv *fpriv = uq_mgr_to_fpriv(uq_mgr);
         struct amdgpu_eviction_fence_mgr *evf_mgr = &fpriv->evf_mgr;

         /* Wait for any pending userqueue fence work to finish */
         ret = amdgpu_userqueue_wait_for_signal(uq_mgr);
         if (ret) {
-               DRM_ERROR("Not suspending userqueue, timeout waiting for 
work\n");
+               file = uq_mgr->file;
+               drm_process_info(file, proc_log, sizeof(proc_log));
+               DRM_ERROR("Not suspending userqueue, timeout waiting for %s\n",
+                         proc_log);
                 return;
         }

         ret = amdgpu_userqueue_suspend_all(uq_mgr);
         if (ret) {
-               DRM_ERROR("Failed to evict userqueue\n");
+               file = uq_mgr->file;
+               drm_process_info(file, proc_log, sizeof(proc_log));
+               DRM_ERROR("Failed to evict userqueue for %s\n", proc_log);
                 return;
         }

--
2.34.1

Reply via email to