drm/amdgpu: Fix SMU error failure
authorAndrey Grodzovsky <andrey.grodzovsky@amd.com>
Mon, 24 Aug 2020 15:02:18 +0000 (11:02 -0400)
committerAlex Deucher <alexander.deucher@amd.com>
Tue, 15 Sep 2020 21:24:55 +0000 (17:24 -0400)
Wait for HW/PSP initiated ASIC reset to complete before
starting the recovery operations.

v2: Remove typo

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

index 2d15535..36a8f12 100644 (file)
@@ -4843,14 +4843,32 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev)
 {
        struct drm_device *dev = pci_get_drvdata(pdev);
        struct amdgpu_device *adev = drm_to_adev(dev);
-       int r;
+       int r, i;
        bool vram_lost;
+       u32 memsize;
 
        DRM_INFO("PCI error: slot reset callback!!\n");
 
+       /* wait for asic to come out of reset */
+       msleep(500);
+
        pci_restore_state(pdev);
 
-       adev->in_pci_err_recovery = true;
+       /* confirm  ASIC came out of reset */
+       for (i = 0; i < adev->usec_timeout; i++) {
+               memsize = amdgpu_asic_get_config_memsize(adev);
+
+               if (memsize != 0xffffffff)
+                       break;
+               udelay(1);
+       }
+       if (memsize == 0xffffffff) {
+               r = -ETIME;
+               goto out;
+       }
+
+       /* TODO Call amdgpu_pre_asic_reset instead */
+       adev->in_pci_err_recovery = true;       
        r = amdgpu_device_ip_suspend(adev);
        adev->in_pci_err_recovery = false;
        if (r)