gpu: nvgpu: hold power ref for deterministic channels

To support deterministic channels even with platforms where railgating
is supported, have each deterministic-marked channel hold a power
reference during their lifetime, and skip taking power refs for jobs in
submit path for those.

Previously, railgating blocked deterministic submits in general because
of gk20a_busy()/gk20a_idle() calls in submit path possibly taking time
and more significantly because the gpu may need turning on which takes a
nondeterministic and long amount of time.

As an exception, gk20a_do_idle() can still block deterministic submits
until gk20a_do_unidle() is called. Add a rwsem to guard this. VPR resize
needs do_idle, which conflicts with deterministic channels' requirement
to keep the GPU on. This is documented in the ioctl header now.

Make NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_NO_JOBTRACKING always
set in the gpu characteristics now that it's supported. The only thing
left now blocking NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_FULL is
the sync framework.

Make the channel debug dump show which channels are deterministic.

Bug 200291300
Jira NVGPU-70

Change-Id: I47b6f3a8517cd6e4255f6ca2855e3dd912e4f5f3
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1483038
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

2 files changed, 10 insertions, 0 deletions

diff --git a/drivers/gpu/nvgpu/common/linux/driver_common.c b/drivers/gpu/nvgpu/common/linux/driver_common.c
index bd9a4e77..a00880ed 100644
--- a/drivers/gpu/nvgpu/common/linux/driver_common.c
+++ b/drivers/gpu/nvgpu/common/linux/driver_common.c
@@ -39,6 +39,7 @@ static void nvgpu_init_vars(struct gk20a *g)
         gk20a_init_gr(g);
 
         init_rwsem(&g->busy_lock);
+        init_rwsem(&g->deterministic_busy);
 
         nvgpu_spinlock_init(&g->mc_enable_lock);
 
diff --git a/drivers/gpu/nvgpu/common/linux/module.c b/drivers/gpu/nvgpu/common/linux/module.c
index 34a0ded6..cbad3993 100644
--- a/drivers/gpu/nvgpu/common/linux/module.c
+++ b/drivers/gpu/nvgpu/common/linux/module.c
@@ -298,6 +298,12 @@ int __gk20a_do_idle(struct device *dev, bool force_reset)
         bool is_railgated;
         int err = 0;
 
+        /*
+         * Hold back deterministic submits and changes to deterministic
+         * channels - this must be outside the power busy locks.
+         */
+        gk20a_channel_deterministic_idle(g);
+
         /* acquire busy lock to block other busy() calls */
         down_write(&g->busy_lock);
 
@@ -403,6 +409,7 @@ fail_drop_usage_count:
 fail_timeout:
         nvgpu_mutex_release(&platform->railgate_lock);
         up_write(&g->busy_lock);
+        gk20a_channel_deterministic_unidle(g);
         return -EBUSY;
 }
 
@@ -456,6 +463,8 @@ int __gk20a_do_unidle(struct device *dev)
         nvgpu_mutex_release(&platform->railgate_lock);
         up_write(&g->busy_lock);
 
+        gk20a_channel_deterministic_unidle(g);
+
         return 0;
 }