gpu: nvgpu: Convert VIDMEM work_struct to thread

Convert the work_struct used by the vidmem background clearing to
a thread to make it more cross platform. The thread waits on a
condition variable to determine when work needs to be done. The
signal comes from the DMA API when it enqueues a new nvgpu_mem that
needs clearing.

Add logic for handling suspend: the CE cannot be accessed while
the GPU is suspended. As such the background thread must be paused
while the GPU is suspended and the CE is not available.

Several other changes were also made:

  o  Move the code that enqueues a nvgpu_mem from the DMA API
     code to a function in the VIDMEM code.
  o  Move nvgpu_vidmem_get_pending_alloc() to the Linux specific
     code as this function is only used there. It's a trivial
     function that QNX can easily implement as well.
  o  Remove the was_empty logic from the enqueue. Now just always
     signal the condition variable when anew nvgpu_mem comes in.
  o  Move CE suspend to after MM suspend.

JIRA NVGPU-30
JIRA NVGPU-138

Change-Id: Ie9286ae5a127c3fced86dfb9794e7d81eab0491c
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1574498
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>

3 files changed, 13 insertions, 6 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/gk20a.c b/drivers/gpu/nvgpu/gk20a/gk20a.c
index e1bf2b4b..02baf683 100644
--- a/drivers/gpu/nvgpu/gk20a/gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/gk20a.c
@@ -33,6 +33,7 @@
 #include <nvgpu/pmu.h>
 #include <nvgpu/gmmu.h>
 #include <nvgpu/ltc.h>
+#include <nvgpu/vidmem.h>
 
 #include <trace/events/gk20a.h>
 
@@ -97,8 +98,6 @@ int gk20a_prepare_poweroff(struct gk20a *g)
         if (gk20a_fifo_is_engine_busy(g))
                 return -EBUSY;
 
-        gk20a_ce_suspend(g);
-
         ret = gk20a_channel_suspend(g);
         if (ret)
                 return ret;
@@ -111,6 +110,8 @@ int gk20a_prepare_poweroff(struct gk20a *g)
         ret |= gk20a_mm_suspend(g);
         ret |= gk20a_fifo_suspend(g);
 
+        gk20a_ce_suspend(g);
+
         /* Disable GPCPLL */
         if (g->ops.clk.suspend_clk_support)
                 ret |= g->ops.clk.suspend_clk_support(g);
@@ -323,6 +324,8 @@ int gk20a_finalize_poweron(struct gk20a *g)
                 }
         }
 
+        nvgpu_vidmem_thread_unpause(&g->mm);
+
 #if defined(CONFIG_TEGRA_GK20A_NVHOST) && defined(CONFIG_TEGRA_19x_GPU)
         if (gk20a_platform_has_syncpoints(g) && g->syncpt_unit_size) {
                 if (!nvgpu_mem_is_valid(&g->syncpt_mem)) {
diff --git a/drivers/gpu/nvgpu/gk20a/mm_gk20a.c b/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
index 687951a9..67ab307f 100644
--- a/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
@@ -978,9 +978,7 @@ int gk20a_mm_suspend(struct gk20a *g)
 {
         gk20a_dbg_fn("");
 
-#if defined(CONFIG_GK20A_VIDMEM)
-        cancel_work_sync(&g->mm.vidmem.clear_mem_worker);
-#endif
+        nvgpu_vidmem_thread_pause_sync(&g->mm);
 
         g->ops.mm.cbc_clean(g);
         g->ops.mm.l2_flush(g, false);
diff --git a/drivers/gpu/nvgpu/gk20a/mm_gk20a.h b/drivers/gpu/nvgpu/gk20a/mm_gk20a.h
index 556cb234..13698cd7 100644
--- a/drivers/gpu/nvgpu/gk20a/mm_gk20a.h
+++ b/drivers/gpu/nvgpu/gk20a/mm_gk20a.h
@@ -36,6 +36,8 @@
 #include <nvgpu/rbtree.h>
 #include <nvgpu/kref.h>
 #include <nvgpu/atomic.h>
+#include <nvgpu/cond.h>
+#include <nvgpu/thread.h>
 
 struct nvgpu_pd_cache;
 
@@ -272,7 +274,11 @@ struct mm_gk20a {
                 struct nvgpu_list_node clear_list_head;
                 struct nvgpu_mutex clear_list_mutex;
 
-                struct work_struct clear_mem_worker;
+                struct nvgpu_cond clearing_thread_cond;
+                struct nvgpu_thread clearing_thread;
+                struct nvgpu_mutex clearing_thread_lock;
+                nvgpu_atomic_t pause_count;
+
                 nvgpu_atomic64_t bytes_pending;
         } vidmem;
 };