In recording benches, record to and from the same format.
The recording bench must record some source material into some sort of
display list, and fundamentally cannot separate the timing of the two.
This CL makes it so the source material and display list are of the same type.
So instead of previous:
--nolite: SkRecord-based picture -> SkRecord-based picture
--lite: SkRecord-based picture -> threadsafe SkLiteDL
Now this times
--nolite: SkRecord-based picture -> SkRecord-based picture
--lite: SkLiteDL -> threadsafe SkLiteDL
This makes it easier to profile SkLiteDL and explore both recording and playback overhead hot spots.
The threadsafety is incidental for the source (and doesn't affect playback speed),
but I think it's handy to keep around on the destination to make a more fair comparison.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2230323002
Review-Url: https://codereview.chromium.org/2230323002
diff --git a/bench/RecordingBench.h b/bench/RecordingBench.h
index a5793b3..8e0e12e 100644
--- a/bench/RecordingBench.h
+++ b/bench/RecordingBench.h
@@ -10,6 +10,7 @@
#include "Benchmark.h"
#include "SkPicture.h"
+#include "SkLiteDL.h"
class RecordingBench : public Benchmark {
public:
@@ -24,8 +25,8 @@
private:
sk_sp<const SkPicture> fSrc;
SkString fName;
+ sk_sp<SkLiteDL> fDL;
bool fUseBBH;
- bool fLite;
typedef Benchmark INHERITED;
};