[PATCH v2 3/7] drm/admgpu: Serialize RAS recovery work directly into reset domain queue.

[Andrey Grodzovsky]

Save the extra usless work schedule. Also swith to delayed work.

Signed-off-by: Andrey Grodzovsky <andrey.grodzov...@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 12 +++++++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h |  2 +-
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index a653cf3b3d13..7e8c7bcc7303 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -35,6 +35,8 @@
 #include "amdgpu_xgmi.h"
 #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h"
 #include "atom.h"
+#include "amdgpu_reset.h"
+
 #ifdef CONFIG_X86_MCE_AMD
 #include <asm/mce.h>
 
@@ -1889,7 +1891,7 @@ static int amdgpu_ras_badpages_read(struct amdgpu_device 
*adev,
 static void amdgpu_ras_do_recovery(struct work_struct *work)
 {
        struct amdgpu_ras *ras =
-               container_of(work, struct amdgpu_ras, recovery_work);
+               container_of(work, struct amdgpu_ras, recovery_work.work);
        struct amdgpu_device *remote_adev = NULL;
        struct amdgpu_device *adev = ras->adev;
        struct list_head device_list, *device_list_handle =  NULL;
@@ -1916,7 +1918,7 @@ static void amdgpu_ras_do_recovery(struct work_struct 
*work)
        }
 
        if (amdgpu_device_should_recover_gpu(ras->adev))
-               amdgpu_device_gpu_recover(ras->adev, NULL);
+               amdgpu_device_gpu_recover_imp(ras->adev, NULL);
        atomic_set(&ras->in_recovery, 0);
 }
 
@@ -2148,7 +2150,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev)
        }
 
        mutex_init(&con->recovery_lock);
-       INIT_WORK(&con->recovery_work, amdgpu_ras_do_recovery);
+       INIT_DELAYED_WORK(&con->recovery_work, amdgpu_ras_do_recovery);
        atomic_set(&con->in_recovery, 0);
        con->eeprom_control.bad_channel_bitmap = 0;
 
@@ -2217,7 +2219,7 @@ static int amdgpu_ras_recovery_fini(struct amdgpu_device 
*adev)
        if (!data)
                return 0;
 
-       cancel_work_sync(&con->recovery_work);
+       cancel_delayed_work_sync(&con->recovery_work);
 
        mutex_lock(&con->recovery_lock);
        con->eh_data = NULL;
@@ -2910,7 +2912,7 @@ int amdgpu_ras_reset_gpu(struct amdgpu_device *adev)
        struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
        if (atomic_cmpxchg(&ras->in_recovery, 0, 1) == 0)
-               schedule_work(&ras->recovery_work);
+               amdgpu_reset_domain_schedule(ras->adev->reset_domain, 
&ras->recovery_work);
        return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index b9a6fac2b8b2..f7e21c2abc61 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -347,7 +347,7 @@ struct amdgpu_ras {
        struct ras_manager *objs;
 
        /* gpu recovery */
-       struct work_struct recovery_work;
+       struct delayed_work recovery_work;
        atomic_t in_recovery;
        struct amdgpu_device *adev;
        /* error handler data */
-- 
2.25.1