video_core: Block in WaitFence.
This function is called rarely and blocks quite often for a long time. So don't waste power and let the CPU sleep. This might also increase the performance as the other cores might be allowed to clock higher.
This commit is contained in:
parent
f846e3d6d0
commit
cb9dd01ffd
3 changed files with 9 additions and 5 deletions
|
@ -66,19 +66,20 @@ const DmaPusher& GPU::DmaPusher() const {
|
|||
return *dma_pusher;
|
||||
}
|
||||
|
||||
void GPU::WaitFence(u32 syncpoint_id, u32 value) const {
|
||||
void GPU::WaitFence(u32 syncpoint_id, u32 value) {
|
||||
// Synced GPU, is always in sync
|
||||
if (!is_async) {
|
||||
return;
|
||||
}
|
||||
MICROPROFILE_SCOPE(GPU_wait);
|
||||
while (syncpoints[syncpoint_id].load(std::memory_order_relaxed) < value) {
|
||||
}
|
||||
std::unique_lock lock{sync_mutex};
|
||||
sync_cv.wait(lock, [=]() { return syncpoints[syncpoint_id].load() >= value; });
|
||||
}
|
||||
|
||||
void GPU::IncrementSyncPoint(const u32 syncpoint_id) {
|
||||
syncpoints[syncpoint_id]++;
|
||||
std::lock_guard lock{sync_mutex};
|
||||
sync_cv.notify_all();
|
||||
if (!syncpt_interrupts[syncpoint_id].empty()) {
|
||||
u32 value = syncpoints[syncpoint_id].load();
|
||||
auto it = syncpt_interrupts[syncpoint_id].begin();
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue