CPU/Timing: Use an approximated amortized amount of ticks when advancing timing.

We divide the number of ticks to add by the number of cores (4) to obtain a more or less rough estimate of the actual number of ticks added. This assumes that all 4 cores are doing similar work. Previously we were adding ~4 times the number of ticks, thus making the games think that time was going way too fast.

This lets us bypass certain hangs in some games like Breath of the Wild.

We should modify our CoreTiming to support multiple cores (both running in a single thread, and in multiple host threads).
This commit is contained in:
Subv 2018-08-12 20:41:28 -05:00
parent a9877c8f65
commit d923766042
2 changed files with 11 additions and 1 deletions

View File

@ -86,7 +86,16 @@ public:
}
void AddTicks(u64 ticks) override {
CoreTiming::AddTicks(ticks - num_interpreted_instructions);
// Divide the number of ticks by the amount of CPU cores. TODO(Subv): This yields only a
// rough approximation of the amount of executed ticks in the system, it may be thrown off
// if not all cores are doing a similar amount of work. Instead of doing this, we should
// device a way so that timing is consistent across all cores without increasing the ticks 4
// times.
u64 amortized_ticks = (ticks - num_interpreted_instructions) / Core::NUM_CPU_CORES;
// Always execute at least one tick.
amortized_ticks = std::max<u64>(amortized_ticks, 1);
CoreTiming::AddTicks(amortized_ticks);
num_interpreted_instructions = 0;
}
u64 GetTicksRemaining() override {

View File

@ -90,6 +90,7 @@ void Cpu::RunLoop(bool tight_loop) {
LOG_TRACE(Core, "Core-{} idling", core_index);
if (IsMainCore()) {
// TODO(Subv): Only let CoreTiming idle if all 4 cores are idling.
CoreTiming::Idle();
CoreTiming::Advance();
}