gpu: nvgpu: Wait for idle via FIFO registers

Wait for engine idle via FIFO's engine status instead of submitting
WFI to channel. Submitting WFI and waiting is not robust, and wait
might invoke debug dump which cannot be done while powering down.

Bug 1499214

Change-Id: I4d52e8558e1a862ad4292036594d81ebfbd5f36b
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/432151
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Tested-by: Sachin Nikam <snikam@nvidia.com>

1 files changed, 4 insertions, 19 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/channel_gk20a.c b/drivers/gpu/nvgpu/gk20a/channel_gk20a.c
index 33681c2a..b4d9c785 100644
--- a/drivers/gpu/nvgpu/gk20a/channel_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/channel_gk20a.c
@@ -1948,29 +1948,14 @@ int gk20a_channel_suspend(struct gk20a *g)
         struct fifo_gk20a *f = &g->fifo;
         u32 chid;
         bool channels_in_use = false;
-        struct device *d = dev_from_gk20a(g);
         int err;
 
         gk20a_dbg_fn("");
 
-        /* idle the engine by submitting WFI on non-KEPLER_C channel */
-        for (chid = 0; chid < f->num_channels; chid++) {
-                struct channel_gk20a *c = &f->channel[chid];
-                if (c->in_use && c->obj_class != KEPLER_C && !c->has_timedout) {
-                        err = gk20a_channel_submit_wfi(c);
-                        if (err) {
-                                gk20a_err(d, "cannot idle channel %d\n",
-                                                chid);
-                                return err;
-                        }
-
-                        if (c->sync)
-                                c->sync->wait_cpu(c->sync,
-                                                  &c->last_submit.post_fence,
-                                                  500000);
-                        break;
-                }
-        }
+        /* wait for engine idle */
+        err = gk20a_fifo_wait_engine_idle(g);
+        if (err)
+                return err;
 
         for (chid = 0; chid < f->num_channels; chid++) {
                 if (f->channel[chid].in_use) {