drm/lima: Expose job_hang_limit module parameter
Subsystems (3)
PATCH 1/1 drm/lima: Expose job_hang_limit module parameter
From: Andrey Lebedev <[email protected]>
Some pp or gp jobs can be successfully repeated even after they time outs. Introduce lima module parameter to specify number of times a job can hang before being dropped.
---
Hello,
This patch allows to work around a freezing problem as discussed in
https://gitlab.freedesktop.org/lima/linux/-/issues/33
drivers/gpu/drm/lima/lima_drv.c | 4 ++++ drivers/gpu/drm/lima/lima_drv.h | 1 + drivers/gpu/drm/lima/lima_sched.c | 5 +++-- 3 files changed, 8 insertions(+), 2 deletions(-)
drivers/gpu/drm/lima/lima_drv.c +4 -0
@@ -19,6 +19,7 @@int lima_sched_timeout_ms;uint lima_heap_init_nr_pages =8 ;uint lima_max_error_tasks; +int lima_job_hang_limit;MODULE_PARM_DESC (sched_timeout_ms,"task run timeout in ms" );module_param_named (sched_timeout_ms, lima_sched_timeout_ms, int,0444 );
@@ -29,6 +30,9 @@ module_param_named(heap_init_nr_pages, lima_heap_init_nr_pages, uint, 0444);MODULE_PARM_DESC (max_error_tasks,"max number of error tasks to save" );module_param_named (max_error_tasks, lima_max_error_tasks, uint,0644 ); +MODULE_PARM_DESC (job_hang_limit,"number of times to allow a job to hang before dropping it (default 0)" ); +module_param_named (job_hang_limit, lima_job_hang_limit, int,0444 ); +static int lima_ioctl_get_param (struct drm_device *dev,void *data,struct drm_file *file) {struct drm_lima_get_param *args = data;
drivers/gpu/drm/lima/lima_drv.h +1 -0
@@ -11,6 +11,7 @@extern int lima_sched_timeout_ms;extern uint lima_heap_init_nr_pages;extern uint lima_max_error_tasks; +extern int lima_job_hang_limit;struct lima_vm ;struct lima_bo ;
drivers/gpu/drm/lima/lima_sched.c +3 -2
@@ -503,8 +503,9 @@ int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)INIT_WORK (&pipe->recover_work, lima_sched_recover_work); -return drm_sched_init (&pipe->base, &lima_sched_ops,1 ,0 , -msecs_to_jiffies (timeout), name); +return drm_sched_init (&pipe->base, &lima_sched_ops,1 , + lima_job_hang_limit,msecs_to_jiffies (timeout), + name); }void lima_sched_pipe_fini (struct lima_sched_pipe *pipe)
Signature · 1 line
2.25.1
Quoted 25 lines · Andrey Lebedev
From: Andrey Lebedev <[email protected]>
Some pp or gp jobs can be successfully repeated even after they time outs. Introduce lima module parameter to specify number of times a job can hang before being dropped.
Signed-off-by Andrey Lebedev <[email protected]>---
Hello,
This patch allows to work around a freezing problem as discussed in
https://gitlab.freedesktop.org/lima/linux/-/issues/33drivers/gpu/drm/lima/lima_drv.c | 4 ++++ drivers/gpu/drm/lima/lima_drv.h | 1 + drivers/gpu/drm/lima/lima_sched.c | 5 +++-- 3 files changed, 8 insertions(+), 2 deletions(-)drivers/gpu/drm/lima/lima_drv.c +0
-0
drivers/gpu/drm/lima/lima_drv.c line 22
Quoted 4 lines
drivers/gpu/drm/lima/lima_drv.c +1
-0@@ -19,6 +19,7 @@int lima_sched_timeout_ms;uint lima_heap_init_nr_pages =8 ;uint lima_max_error_tasks; +int lima_job_hang_limit;
Better be an "uint" to avoid negative check. With this fixed, patch is:
Regards,
Qiang
Quoted 44 lines
drivers/gpu/drm/lima/lima_drv.c +3
-0MODULE_PARM_DESC (sched_timeout_ms,"task run timeout in ms" );module_param_named (sched_timeout_ms, lima_sched_timeout_ms, int,0444 );@@ -29,6 +30,9 @@ module_param_named(heap_init_nr_pages, lima_heap_init_nr_pages, uint, 0444);MODULE_PARM_DESC (max_error_tasks,"max number of error tasks to save" );module_param_named (max_error_tasks, lima_max_error_tasks, uint,0644 ); +MODULE_PARM_DESC (job_hang_limit,"number of times to allow a job to hang before dropping it (default 0)" ); +module_param_named (job_hang_limit, lima_job_hang_limit, int,0444 ); +static int lima_ioctl_get_param (struct drm_device *dev,void *data,struct drm_file *file) {struct drm_lima_get_param *args = data;drivers/gpu/drm/lima/lima_drv.h +1
-0@@ -11,6 +11,7 @@extern int lima_sched_timeout_ms;extern uint lima_heap_init_nr_pages;extern uint lima_max_error_tasks; +extern int lima_job_hang_limit;struct lima_vm ;struct lima_bo ;drivers/gpu/drm/lima/lima_sched.c +3
-3@@ -503,8 +503,9 @@ int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)INIT_WORK (&pipe->recover_work, lima_sched_recover_work); -return drm_sched_init (&pipe->base, &lima_sched_ops,1 ,0 , -msecs_to_jiffies (timeout), name); +return drm_sched_init (&pipe->base, &lima_sched_ops,1 , + lima_job_hang_limit,msecs_to_jiffies (timeout), + name); }void lima_sched_pipe_fini (struct lima_sched_pipe *pipe) --2.25.1
PATCH 1/1 drm/lima: Expose job_hang_limit module parameter
From: Andrey Lebedev <[email protected]>
Some pp or gp jobs can be successfully repeated even after they time outs. Introduce lima module parameter to specify number of times a job can hang before being dropped.
---
Changed type of lima_job_hang_limit to unsigned int.
drivers/gpu/drm/lima/lima_drv.c | 4 ++++ drivers/gpu/drm/lima/lima_drv.h | 1 + drivers/gpu/drm/lima/lima_sched.c | 5 +++-- 3 files changed, 8 insertions(+), 2 deletions(-)
drivers/gpu/drm/lima/lima_drv.c +4 -0
@@ -19,6 +19,7 @@int lima_sched_timeout_ms;uint lima_heap_init_nr_pages =8 ;uint lima_max_error_tasks; +uint lima_job_hang_limit;MODULE_PARM_DESC (sched_timeout_ms,"task run timeout in ms" );module_param_named (sched_timeout_ms, lima_sched_timeout_ms, int,0444 );
@@ -29,6 +30,9 @@ module_param_named(heap_init_nr_pages, lima_heap_init_nr_pages, uint, 0444);MODULE_PARM_DESC (max_error_tasks,"max number of error tasks to save" );module_param_named (max_error_tasks, lima_max_error_tasks, uint,0644 ); +MODULE_PARM_DESC (job_hang_limit,"number of times to allow a job to hang before dropping it (default 0)" ); +module_param_named (job_hang_limit, lima_job_hang_limit, int,0444 ); +static int lima_ioctl_get_param (struct drm_device *dev,void *data,struct drm_file *file) {struct drm_lima_get_param *args = data;
drivers/gpu/drm/lima/lima_drv.h +1 -0
@@ -11,6 +11,7 @@extern int lima_sched_timeout_ms;extern uint lima_heap_init_nr_pages;extern uint lima_max_error_tasks; +extern int lima_job_hang_limit;struct lima_vm ;struct lima_bo ;
drivers/gpu/drm/lima/lima_sched.c +3 -2
@@ -503,8 +503,9 @@ int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)INIT_WORK (&pipe->recover_work, lima_sched_recover_work); -return drm_sched_init (&pipe->base, &lima_sched_ops,1 ,0 , -msecs_to_jiffies (timeout), name); +return drm_sched_init (&pipe->base, &lima_sched_ops,1 , + lima_job_hang_limit,msecs_to_jiffies (timeout), + name); }void lima_sched_pipe_fini (struct lima_sched_pipe *pipe)
Signature · 1 line
2.25.1
PATCH 1/1 drm/lima: Expose job_hang_limit module parameter
In reply to Qiang YuFrom: Andrey Lebedev <[email protected]>
Some pp or gp jobs can be successfully repeated even after they time outs. Introduce lima module parameter to specify number of times a job can hang before being dropped.
---
Fixes for the embarrassing build error
drivers/gpu/drm/lima/lima_drv.c | 4 ++++ drivers/gpu/drm/lima/lima_drv.h | 1 + drivers/gpu/drm/lima/lima_sched.c | 5 +++-- 3 files changed, 8 insertions(+), 2 deletions(-)
drivers/gpu/drm/lima/lima_drv.c +4 -0
@@ -19,6 +19,7 @@int lima_sched_timeout_ms;uint lima_heap_init_nr_pages =8 ;uint lima_max_error_tasks; +uint lima_job_hang_limit;MODULE_PARM_DESC (sched_timeout_ms,"task run timeout in ms" );module_param_named (sched_timeout_ms, lima_sched_timeout_ms, int,0444 );
@@ -29,6 +30,9 @@ module_param_named(heap_init_nr_pages, lima_heap_init_nr_pages, uint, 0444);MODULE_PARM_DESC (max_error_tasks,"max number of error tasks to save" );module_param_named (max_error_tasks, lima_max_error_tasks, uint,0644 ); +MODULE_PARM_DESC (job_hang_limit,"number of times to allow a job to hang before dropping it (default 0)" ); +module_param_named (job_hang_limit, lima_job_hang_limit, int,0444 ); +static int lima_ioctl_get_param (struct drm_device *dev,void *data,struct drm_file *file) {struct drm_lima_get_param *args = data;
drivers/gpu/drm/lima/lima_drv.h +1 -0
@@ -11,6 +11,7 @@extern int lima_sched_timeout_ms;extern uint lima_heap_init_nr_pages;extern uint lima_max_error_tasks; +extern uint lima_job_hang_limit;struct lima_vm ;struct lima_bo ;
drivers/gpu/drm/lima/lima_sched.c +3 -2
@@ -503,8 +503,9 @@ int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)INIT_WORK (&pipe->recover_work, lima_sched_recover_work); -return drm_sched_init (&pipe->base, &lima_sched_ops,1 ,0 , -msecs_to_jiffies (timeout), name); +return drm_sched_init (&pipe->base, &lima_sched_ops,1 , + lima_job_hang_limit,msecs_to_jiffies (timeout), + name); }void lima_sched_pipe_fini (struct lima_sched_pipe *pipe)
Signature · 1 line
2.25.1
Qiang Yu
maintainerQuoted 31 lines · Andrey Lebedev
From: Andrey Lebedev <[email protected]>
Some pp or gp jobs can be successfully repeated even after they time outs. Introduce lima module parameter to specify number of times a job can hang before being dropped.
Signed-off-by Andrey Lebedev <[email protected]>---
Fixes for the embarrassing build error
Reported-by kernel test robot <[email protected]>drivers/gpu/drm/lima/lima_drv.c | 4 ++++ drivers/gpu/drm/lima/lima_drv.h | 1 + drivers/gpu/drm/lima/lima_sched.c | 5 +++-- 3 files changed, 8 insertions(+), 2 deletions(-)drivers/gpu/drm/lima/lima_drv.c +1
-0@@ -19,6 +19,7 @@int lima_sched_timeout_ms;uint lima_heap_init_nr_pages =8 ;uint lima_max_error_tasks; +uint lima_job_hang_limit;MODULE_PARM_DESC (sched_timeout_ms,"task run timeout in ms" );module_param_named (sched_timeout_ms, lima_sched_timeout_ms, int,0444 );
drivers/gpu/drm/lima/lima_drv.c module_param_named(heap_init_nr_pages, lima_heap_init_nr_pages, uint, 0444); · line 35
Quoted 6 lines
drivers/gpu/drm/lima/lima_drv.c +3
-0@@ -29,6 +30,9 @@ module_param_named(heap_init_nr_pages, lima_heap_init_nr_pages, uint, 0444);MODULE_PARM_DESC (max_error_tasks,"max number of error tasks to save" );module_param_named (max_error_tasks, lima_max_error_tasks, uint,0644 ); +MODULE_PARM_DESC (job_hang_limit,"number of times to allow a job to hang before dropping it (default 0)" ); +module_param_named (job_hang_limit, lima_job_hang_limit, int,0444 ); +
Still miss this "int" to "uint".
Regards,
Qiang
Quoted 32 lines
drivers/gpu/drm/lima/lima_drv.c +0
-0static int lima_ioctl_get_param (struct drm_device *dev,void *data,struct drm_file *file) {struct drm_lima_get_param *args = data;drivers/gpu/drm/lima/lima_drv.h +1
-0@@ -11,6 +11,7 @@extern int lima_sched_timeout_ms;extern uint lima_heap_init_nr_pages;extern uint lima_max_error_tasks; +extern uint lima_job_hang_limit;struct lima_vm ;struct lima_bo ;drivers/gpu/drm/lima/lima_sched.c +3
-3@@ -503,8 +503,9 @@ int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)INIT_WORK (&pipe->recover_work, lima_sched_recover_work); -return drm_sched_init (&pipe->base, &lima_sched_ops,1 ,0 , -msecs_to_jiffies (timeout), name); +return drm_sched_init (&pipe->base, &lima_sched_ops,1 , + lima_job_hang_limit,msecs_to_jiffies (timeout), + name); }void lima_sched_pipe_fini (struct lima_sched_pipe *pipe) --
Quoted 2 lines
2.25.1
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
PATCH 1/1 drm/lima: Expose job_hang_limit module parameter
From: Andrey Lebedev <[email protected]>
Some pp or gp jobs can be successfully repeated even after they time outs. Introduce lima module parameter to specify number of times a job can hang before being dropped.
---
Now all types are correct (uint).
drivers/gpu/drm/lima/lima_drv.c | 4 ++++ drivers/gpu/drm/lima/lima_drv.h | 1 + drivers/gpu/drm/lima/lima_sched.c | 5 +++-- 3 files changed, 8 insertions(+), 2 deletions(-)
drivers/gpu/drm/lima/lima_drv.c +4 -0
@@ -19,6 +19,7 @@int lima_sched_timeout_ms;uint lima_heap_init_nr_pages =8 ;uint lima_max_error_tasks; +uint lima_job_hang_limit;MODULE_PARM_DESC (sched_timeout_ms,"task run timeout in ms" );module_param_named (sched_timeout_ms, lima_sched_timeout_ms, int,0444 );
@@ -29,6 +30,9 @@ module_param_named(heap_init_nr_pages, lima_heap_init_nr_pages, uint, 0444);MODULE_PARM_DESC (max_error_tasks,"max number of error tasks to save" );module_param_named (max_error_tasks, lima_max_error_tasks, uint,0644 ); +MODULE_PARM_DESC (job_hang_limit,"number of times to allow a job to hang before dropping it (default 0)" ); +module_param_named (job_hang_limit, lima_job_hang_limit, uint,0444 ); +static int lima_ioctl_get_param (struct drm_device *dev,void *data,struct drm_file *file) {struct drm_lima_get_param *args = data;
drivers/gpu/drm/lima/lima_drv.h +1 -0
@@ -11,6 +11,7 @@extern int lima_sched_timeout_ms;extern uint lima_heap_init_nr_pages;extern uint lima_max_error_tasks; +extern uint lima_job_hang_limit;struct lima_vm ;struct lima_bo ;
drivers/gpu/drm/lima/lima_sched.c +3 -2
@@ -503,8 +503,9 @@ int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)INIT_WORK (&pipe->recover_work, lima_sched_recover_work); -return drm_sched_init (&pipe->base, &lima_sched_ops,1 ,0 , -msecs_to_jiffies (timeout), name); +return drm_sched_init (&pipe->base, &lima_sched_ops,1 , + lima_job_hang_limit,msecs_to_jiffies (timeout), + name); }void lima_sched_pipe_fini (struct lima_sched_pipe *pipe)
Signature · 1 line
2.25.1
Hello guys,
What is the status of this patch? Was this committed to any branch? Is it pending for merge to the mainline? Do I have to do anything in order to make it mergeable?
Quoted 67 lines · Andrey Lebedev
From: Andrey Lebedev <[email protected]>
Some pp or gp jobs can be successfully repeated even after they time outs. Introduce lima module parameter to specify number of times a job can hang before being dropped.
Signed-off-by Andrey Lebedev <[email protected]>---
Now all types are correct (uint).
drivers/gpu/drm/lima/lima_drv.c | 4 ++++ drivers/gpu/drm/lima/lima_drv.h | 1 + drivers/gpu/drm/lima/lima_sched.c | 5 +++-- 3 files changed, 8 insertions(+), 2 deletions(-)drivers/gpu/drm/lima/lima_drv.c +4
-0@@ -19,6 +19,7 @@int lima_sched_timeout_ms;uint lima_heap_init_nr_pages =8 ;uint lima_max_error_tasks; +uint lima_job_hang_limit;MODULE_PARM_DESC (sched_timeout_ms,"task run timeout in ms" );module_param_named (sched_timeout_ms, lima_sched_timeout_ms, int,0444 );@@ -29,6 +30,9 @@ module_param_named(heap_init_nr_pages, lima_heap_init_nr_pages, uint, 0444);MODULE_PARM_DESC (max_error_tasks,"max number of error tasks to save" );module_param_named (max_error_tasks, lima_max_error_tasks, uint,0644 ); +MODULE_PARM_DESC (job_hang_limit,"number of times to allow a job to hang before dropping it (default 0)" ); +module_param_named (job_hang_limit, lima_job_hang_limit, uint,0444 ); +static int lima_ioctl_get_param (struct drm_device *dev,void *data,struct drm_file *file) {struct drm_lima_get_param *args = data;drivers/gpu/drm/lima/lima_drv.h +1
-0@@ -11,6 +11,7 @@extern int lima_sched_timeout_ms;extern uint lima_heap_init_nr_pages;extern uint lima_max_error_tasks; +extern uint lima_job_hang_limit;struct lima_vm ;struct lima_bo ;drivers/gpu/drm/lima/lima_sched.c +3
-2@@ -503,8 +503,9 @@ int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)INIT_WORK (&pipe->recover_work, lima_sched_recover_work); -return drm_sched_init (&pipe->base, &lima_sched_ops,1 ,0 , -msecs_to_jiffies (timeout), name); +return drm_sched_init (&pipe->base, &lima_sched_ops,1 , + lima_job_hang_limit,msecs_to_jiffies (timeout), + name); }void lima_sched_pipe_fini (struct lima_sched_pipe *pipe)
Signature · 3 lines
Andrey Lebedev aka -.- . -.. -.. . .-. Software engineer Homepage: http://lebedev.lt/
Applied to drm-misc-next:
https://cgit.freedesktop.org/drm/drm-misc/
Sorry for the late response.
Regards,
Qiang
Quoted 80 lines · Andrey Lebedev
Hello guys,
What is the status of this patch? Was this committed to any branch? Is it pending for merge to the mainline? Do I have to do anything in order to make it mergeable?
Quoted 57 lines · Andrey Lebedev
From: Andrey Lebedev <[email protected]>
Some pp or gp jobs can be successfully repeated even after they time outs. Introduce lima module parameter to specify number of times a job can hang before being dropped.
Signed-off-by Andrey Lebedev <[email protected]>---
Now all types are correct (uint).
drivers/gpu/drm/lima/lima_drv.c | 4 ++++ drivers/gpu/drm/lima/lima_drv.h | 1 + drivers/gpu/drm/lima/lima_sched.c | 5 +++-- 3 files changed, 8 insertions(+), 2 deletions(-)drivers/gpu/drm/lima/lima_drv.c +4
-0@@ -19,6 +19,7 @@int lima_sched_timeout_ms;uint lima_heap_init_nr_pages =8 ;uint lima_max_error_tasks; +uint lima_job_hang_limit;MODULE_PARM_DESC (sched_timeout_ms,"task run timeout in ms" );module_param_named (sched_timeout_ms, lima_sched_timeout_ms, int,0444 );@@ -29,6 +30,9 @@ module_param_named(heap_init_nr_pages, lima_heap_init_nr_pages, uint, 0444);MODULE_PARM_DESC (max_error_tasks,"max number of error tasks to save" );module_param_named (max_error_tasks, lima_max_error_tasks, uint,0644 ); +MODULE_PARM_DESC (job_hang_limit,"number of times to allow a job to hang before dropping it (default 0)" ); +module_param_named (job_hang_limit, lima_job_hang_limit, uint,0444 ); +static int lima_ioctl_get_param (struct drm_device *dev,void *data,struct drm_file *file) {struct drm_lima_get_param *args = data;drivers/gpu/drm/lima/lima_drv.h +1
-0@@ -11,6 +11,7 @@extern int lima_sched_timeout_ms;extern uint lima_heap_init_nr_pages;extern uint lima_max_error_tasks; +extern uint lima_job_hang_limit;struct lima_vm ;struct lima_bo ;drivers/gpu/drm/lima/lima_sched.c +0
-0@@ -503,8 +503,9 @@ int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)drivers/gpu/drm/lima/lima_sched.c int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name) · line 511
Quoted 10 lines
drivers/gpu/drm/lima/lima_sched.c +3
-2INIT_WORK (&pipe->recover_work, lima_sched_recover_work); -return drm_sched_init (&pipe->base, &lima_sched_ops,1 ,0 , -msecs_to_jiffies (timeout), name); +return drm_sched_init (&pipe->base, &lima_sched_ops,1 , + lima_job_hang_limit,msecs_to_jiffies (timeout), + name); }void lima_sched_pipe_fini (struct lima_sched_pipe *pipe)--
Andrey Lebedev aka -.- . -.. -.. . .-.
Software engineer
Homepage: http://lebedev.lt/