gpu: nvgpu: do not suspend/resume recovered channel

Already torn down channels should not be suspended or
resumed. A channel reference could still be available
even if it is recovered but not closed. Use ch_timedout
status to check if channel is already recovered/aborted.

Bug 2404865

Change-Id: I718eab6032ee94a9322da7a239a978b388de2b01
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1929338
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
(cherry picked from commit 88cff206ae31999240945a870683e8a10be584d7
in dev-kernel)
Reviewed-on: https://git-master.nvidia.com/r/2016994
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

1 files changed, 25 insertions, 8 deletions

diff --git a/drivers/gpu/nvgpu/common/fifo/channel.c b/drivers/gpu/nvgpu/common/fifo/channel.c
index b5ae42d4..2ada3911 100644
--- a/drivers/gpu/nvgpu/common/fifo/channel.c
+++ b/drivers/gpu/nvgpu/common/fifo/channel.c
@@ -2291,7 +2291,13 @@ int gk20a_channel_suspend(struct gk20a *g)
         for (chid = 0; chid < f->num_channels; chid++) {
                 struct channel_gk20a *ch = gk20a_channel_from_id(g, chid);
 
-                if (ch != NULL) {
+                if (ch == NULL) {
+                        continue;
+                }
+                if (gk20a_channel_check_timedout(ch)) {
+                        nvgpu_log_info(g, "do not suspend recovered "
+                                                "channel %d", chid);
+                } else {
                         nvgpu_log_info(g, "suspend channel %d", chid);
                         /* disable channel */
                         gk20a_disable_channel_tsg(g, ch);
@@ -2304,10 +2310,9 @@ int gk20a_channel_suspend(struct gk20a *g)
 
                         channels_in_use = true;
 
-                        active_runlist_ids |= BIT(ch->runlist_id);
-
-                        gk20a_channel_put(ch);
+                        active_runlist_ids |= (u32) BIT64(ch->runlist_id);
                 }
+                gk20a_channel_put(ch);
         }
 
         if (channels_in_use) {
@@ -2317,7 +2322,13 @@ int gk20a_channel_suspend(struct gk20a *g)
                         struct channel_gk20a *ch = gk20a_channel_from_id(g, chid);
 
                         if (ch != NULL) {
-                                g->ops.fifo.unbind_channel(ch);
+                                if (gk20a_channel_check_timedout(ch)) {
+                                        nvgpu_log_info(g, "do not unbind "
+                                                        "recovered channel %d",
+                                                        chid);
+                                } else {
+                                        g->ops.fifo.unbind_channel(ch);
+                                }
                                 gk20a_channel_put(ch);
                         }
                 }
@@ -2339,13 +2350,19 @@ int gk20a_channel_resume(struct gk20a *g)
         for (chid = 0; chid < f->num_channels; chid++) {
                 struct channel_gk20a *ch = gk20a_channel_from_id(g, chid);
 
-                if (ch != NULL) {
+                if (ch == NULL) {
+                        continue;
+                }
+                if (gk20a_channel_check_timedout(ch)) {
+                        nvgpu_log_info(g, "do not resume recovered "
+                                                "channel %d", chid);
+                } else {
                         nvgpu_log_info(g, "resume channel %d", chid);
                         g->ops.fifo.bind_channel(ch);
                         channels_in_use = true;
-                        active_runlist_ids |= BIT(f->channel[chid].runlist_id);
-                        gk20a_channel_put(ch);
+                        active_runlist_ids |= (u32) BIT64(ch->runlist_id);
                 }
+                gk20a_channel_put(ch);
         }
 
         if (channels_in_use) {