mtklein
04d53a5cd5
Revert of nanobench: --veryVerbose for more Win7 debugging ( https://codereview.chromium.org/401663002/ )
...
Reason for revert:
Shouldn't be needed anymore.
Original issue's description:
> nanobench: --veryVerbose for more Win7 debugging
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/e57452d
R=mtklein@chromium.org
TBR=mtklein@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Author: mtklein@google.com
Review URL: https://codereview.chromium.org/404543004
2014-07-17 11:23:07 -07:00
Mike Klein
e57452debd
nanobench: --veryVerbose for more Win7 debugging
...
BUG=skia:
Review URL: https://codereview.chromium.org/401663002
2014-07-17 13:19:22 -04:00
mtklein
55b0ffc486
nanobench: Protect more against infinite loops.
...
BUG=skia:
R=mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/404543002
2014-07-17 08:38:23 -07:00
Mike Klein
912947737a
Use __rdtsc on Windows.
...
This seems to be ~100x higher resolution than QueryPerformanceCounter. AFAIK, all our Windows perf bots have constant_tsc, so we can be a bit more direct about using rdtsc directly: it'll always tick at the max CPU frequency.
Now, the question remains, what is the max CPU frequency to divide through by? It looks like QueryPerformanceFrequency actually gives the CPU frequency in kHz, suspiciously exactly what we need to divide through to get elapsed milliseconds. That was a freebie.
I did some before/after comparison on slow benchmarks. Timings look the same. Going to land this without review tonight to see what happens on the bots; happy to review carefully tomorrow.
R=mtklein@google.com
TBR=bungeman
BUG=skia:
Review URL: https://codereview.chromium.org/394363003
2014-07-16 19:59:32 -04:00
Mike Klein
e3631364e9
Try to debug Windows infinite loops in nanobench.
...
NOTREECHECKS=true
BUG=skia:
Review URL: https://codereview.chromium.org/396753004
2014-07-15 17:56:37 -04:00
Mike Klein
3944a1d237
destroyContexts after each bench instead of before
...
BUG=skia:
Review URL: https://codereview.chromium.org/394893002
2014-07-15 13:40:19 -04:00
mtklein
6238688af0
Give windows boring bars and use 'us' for microseconds.
...
NOTREECHECKS=true
BUG=skia:
R=bsalomon@google.com , mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/393673006
2014-07-15 10:30:31 -07:00
mtklein
1e319f7342
Add --resetGpuContext to both DM and nanobench.
...
Defaulting to true to be conservative for now.
BUG=skia:
NOTREECHECKS=true
R=mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/398483005
2014-07-15 08:27:06 -07:00
mtklein
a189ccdb4d
nanobench: add --runOnce.
...
BUG=skia:
R=egdaniel@google.com , mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/392583005
2014-07-14 12:28:48 -07:00
mtklein
60317d0ffb
nanobench: support --outResultsFile
...
BUG=skia:
R=kelvinly@google.com , mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/390933002
2014-07-14 11:30:37 -07:00
mtklein
5d9d10e821
nanobench: add a cute bar chart
...
Give this a try? Helpful, or gets in the way?
BUG=skia:
R=krajcevski@google.com , mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/390483002
2014-07-11 11:57:07 -07:00
mtklein
40b32be371
nanobench: add median and --cpu/--gpu
...
BUG=skia:
R=krajcevski@google.com , mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/377283002
2014-07-09 08:46:50 -07:00
mtklein
9bc86ed052
Hey, maybe we should _call_ glFinish.
...
BUG=skia:
CQ_EXTRA_TRYBOTS=tryserver.skia:Build-Mac10.7-Clang-x86-Debug-Trybot
R=mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/345063004
2014-07-01 10:02:43 -07:00
mtklein
bb6a028239
draft gpu support in nanobench
...
BUG=skia:
R=bsalomon@google.com , mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/359473004
2014-07-01 08:43:42 -07:00
mtklein
f372321de3
Add nanobench, a fresh look at our microbenchmark harness.
...
out/Release/nanobench runs 4-5x faster on my Z620 and ~2x faster on my N5 than out/Release/bench with the same configs.
(Debug isn't really comparable. I haven't added the Debug -> single loop check.)
BUG=skia:
R=djsollen@google.com , mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/347443002
2014-06-25 14:08:00 -07:00
