Software backend: Delete forked PixelEngine. #224
Conversation
|
IIRC, performance counters are disabled by default (woo, .ini options hiding things from developers) in every game except Super Mario Sunshine, where they're used in that one "Scrubbing Sirena Beach" level to determine how much goop you've washed away. If you can finish that level normally (e.g. not instantly, which was the result way back when), you can probably assume you didn't break them. |
|
@pauldacheez It may also be the case that the level lasts indefinitely and cannot be completed in the given time frame. |
| @@ -378,7 +381,7 @@ void VideoSoftware::RegisterCPMMIO(MMIO::Mapping* mmio, u32 base) | |||
|
|
|||
| void VideoSoftware::RegisterPEMMIO(MMIO::Mapping* mmio, u32 base) | |||
| { | |||
| SWPixelEngine::RegisterMMIO(mmio, base); | |||
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
Mostly just zapping a bunch of duplicated code; the only interesting thing going on here is the changes to the performance counter implementation.
|
I think I've addressed all the review comments here. |
|
@neobrain I'll let you merge this one. |
| @@ -51,4 +51,17 @@ | |||
| void BypassXFB(u8* texture, u32 fbWidth, u32 fbHeight, const EFBRectangle& sourceRc, float Gamma); | |||
|
|
|||
| void DoState(PointerWrap &p); | |||
|
|
|||
| extern u32 perf_values[PQ_NUM_MEMBERS]; | |||
| inline void IncPerfCounterQuadCount(PerfQueryType type) { | |||
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
|
Any other review comments? |
|
LGTM |
Software backend: Delete forked PixelEngine.
|
I'd have preferred Super Mario Sunshine to be tested with the software renderer and/or a perf counter hardware test to be written, but alas - the chances of this breaking something look low enough to nod this through. If you're bored one day I highly welcome hardware tests for the feature, though ;) |
Mostly just zapping a bunch of duplicated code; the only interesting thing
going on here is the changes to the performance counter implementation.
My rough testing shows I didn't break anything, but I don't know how to test the performance counters.