nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-24 10:34:43 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: use spinlock for ch timeout lock

Browse Source 
The channel timeout lock guards a very small critical section. Use a
spinlock instead of a mutex for performance.

Bug 1795076

Change-Id: I94940f3fbe84ed539bcf1bc76ca6ae7a0ef2fe13
Signed-off-by: Aingara Paramakuru <aparamakuru@nvidia.com>
Reviewed-on: http://git-master/r/1200803
(cherry picked from commit 4fa9e973da141067be145d9eba2ea74e96869dcd)
Reviewed-on: http://git-master/r/1203799
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
This commit is contained in:
Aingara Paramakuru
2016-08-10 20:03:49 -04:00
committed by mobile promotions
parent 7ff4a760a8
commit c0dd9ea9c8
2 changed files with 16 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
drivers/gpu/nvgpu/gk20a/channel_gk20a.c
View File

@@ -1660,35 +1660,35 @@ static void gk20a_channel_timeout_start(struct channel_gk20a *ch,
if (!ch->wdt_enabled)
return;
mutex_lock(&ch->timeout.lock);
raw_spin_lock(&ch->timeout.lock);
if (ch->timeout.initialized) {
mutex_unlock(&ch->timeout.lock);
raw_spin_unlock(&ch->timeout.lock);
return;
}
ch->timeout.job = job;
ch->timeout.initialized = true;
raw_spin_unlock(&ch->timeout.lock);
schedule_delayed_work(&ch->timeout.wq,
msecs_to_jiffies(gk20a_get_channel_watchdog_timeout(ch)));
mutex_unlock(&ch->timeout.lock);
}
static void gk20a_channel_timeout_stop(struct channel_gk20a *ch)
{
mutex_lock(&ch->timeout.lock);
raw_spin_lock(&ch->timeout.lock);
if (!ch->timeout.initialized) {
mutex_unlock(&ch->timeout.lock);
raw_spin_unlock(&ch->timeout.lock);
return;
}
mutex_unlock(&ch->timeout.lock);
raw_spin_unlock(&ch->timeout.lock);
cancel_delayed_work_sync(&ch->timeout.wq);
mutex_lock(&ch->timeout.lock);
raw_spin_lock(&ch->timeout.lock);
ch->timeout.initialized = false;
mutex_unlock(&ch->timeout.lock);
raw_spin_unlock(&ch->timeout.lock);
}
void gk20a_channel_timeout_restart_all_channels(struct gk20a *g)
@@ -1700,13 +1700,13 @@ void gk20a_channel_timeout_restart_all_channels(struct gk20a *g)
struct channel_gk20a *ch = &f->channel[chid];
if (gk20a_channel_get(ch)) {
mutex_lock(&ch->timeout.lock);
raw_spin_lock(&ch->timeout.lock);
if (!ch->timeout.initialized) {
mutex_unlock(&ch->timeout.lock);
raw_spin_unlock(&ch->timeout.lock);
gk20a_channel_put(ch);
continue;
}
mutex_unlock(&ch->timeout.lock);
raw_spin_unlock(&ch->timeout.lock);
cancel_delayed_work_sync(&ch->timeout.wq);
if (!ch->has_timedout)
@@ -1740,10 +1740,10 @@ static void gk20a_channel_timeout_handler(struct work_struct *work)
ch->hw_chid);
/* Get timed out job and reset the timer */
mutex_lock(&ch->timeout.lock);
raw_spin_lock(&ch->timeout.lock);
job = ch->timeout.job;
ch->timeout.initialized = false;
mutex_unlock(&ch->timeout.lock);
raw_spin_unlock(&ch->timeout.lock);
if (gk20a_fence_is_expired(job->post_fence)) {
gk20a_err(dev_from_gk20a(g),
@@ -2333,7 +2333,7 @@ int gk20a_init_channel_support(struct gk20a *g, u32 chid)
mutex_init(&c->ioctl_lock);
spin_lock_init(&c->jobs_lock);
mutex_init(&c->last_submit.fence_lock);
mutex_init(&c->timeout.lock);
raw_spin_lock_init(&c->timeout.lock);
mutex_init(&c->sync_lock);
INIT_DELAYED_WORK(&c->timeout.wq, gk20a_channel_timeout_handler);
INIT_DELAYED_WORK(&c->clean_up.wq, gk20a_channel_clean_up_jobs);

2
drivers/gpu/nvgpu/gk20a/channel_gk20a.h
View File

@@ -72,7 +72,7 @@ struct channel_gk20a_job {
struct channel_gk20a_timeout {
struct delayed_work wq;
struct mutex lock;
raw_spinlock_t lock;
bool initialized;
struct channel_gk20a_job *job;
};