skia2/dm
mtklein 31f8867571 Turn on NVPR 4x MSAA by default when supported in DM and nanobench.
This brings DM and nanobench's default configs in line with GM's.

BUG=skia:

Review URL: https://codereview.chromium.org/704563003
2014-11-05 12:28:26 -08:00
..
DM.cpp Separate JSON functions from DMWriteTask. 2014-11-04 07:21:10 -08:00
DMCpuGMTask.cpp Don't test the old picture backend anymore. 2014-09-29 08:44:46 -07:00
DMCpuGMTask.h Remove expectations / -r from DM entirely. 2014-09-09 07:36:57 -07:00
DMGpuGMTask.cpp Turn on NVPR 4x MSAA by default when supported in DM and nanobench. 2014-11-05 12:28:26 -08:00
DMGpuGMTask.h Remove expectations / -r from DM entirely. 2014-09-09 07:36:57 -07:00
DMGpuSupport.h Turn on NVPR 4x MSAA by default when supported in DM and nanobench. 2014-11-05 12:28:26 -08:00
DMJsonWriter.cpp Separate JSON functions from DMWriteTask. 2014-11-04 07:21:10 -08:00
DMJsonWriter.h Separate JSON functions from DMWriteTask. 2014-11-04 07:21:10 -08:00
DMPDFRasterizeTask.cpp Update DM JSON format. 2014-09-09 07:59:46 -07:00
DMPDFRasterizeTask.h SkData to SkStreamAsset to avoid unneeded copying 2014-08-26 10:38:07 -07:00
DMPDFTask.cpp Add --matrix to DM, to play around with and maybe use on a bot. 2014-09-11 12:30:12 -07:00
DMPDFTask.h Remove expectations / -r from DM entirely. 2014-09-09 07:36:57 -07:00
DMPipeTask.cpp Remove underscores from mode identifiers. 2014-09-23 09:20:14 -07:00
DMPipeTask.h DM: Add --skps. 2014-05-14 17:55:32 +00:00
DMQuiltTask.cpp SkTaskGroup::batch(fn, args, N) 2014-10-29 14:17:13 -07:00
DMQuiltTask.h Don't test the old picture backend anymore. 2014-09-29 08:44:46 -07:00
DMReporter.cpp Print max RSS in GM and nanobench too. 2014-08-19 15:55:55 -07:00
DMReporter.h DM tweaks 2014-05-29 20:14:48 +00:00
DMSerializeTask.cpp Don't test the old picture backend anymore. 2014-09-29 08:44:46 -07:00
DMSerializeTask.h Don't test the old picture backend anymore. 2014-09-29 08:44:46 -07:00
DMSKPTask.cpp Update DM JSON format. 2014-09-09 07:59:46 -07:00
DMSKPTask.h Remove expectations / -r from DM entirely. 2014-09-09 07:36:57 -07:00
DMTask.cpp DM: --gpu and --cpu should only control top-level tasks. 2014-09-17 12:26:18 -07:00
DMTask.h SkThreadPool ~~> SkTaskGroup 2014-09-03 15:34:37 -07:00
DMTaskRunner.cpp SkThreadPool ~~> SkTaskGroup 2014-09-03 15:34:37 -07:00
DMTaskRunner.h SkThreadPool ~~> SkTaskGroup 2014-09-03 15:34:37 -07:00
DMTestTask.cpp SkThreadPool ~~> SkTaskGroup 2014-09-03 15:34:37 -07:00
DMTestTask.h SkThreadPool ~~> SkTaskGroup 2014-09-03 15:34:37 -07:00
DMUtil.cpp Don't test the old picture backend anymore. 2014-09-29 08:44:46 -07:00
DMUtil.h Don't test the old picture backend anymore. 2014-09-29 08:44:46 -07:00
DMWriteTask.cpp Separate JSON functions from DMWriteTask. 2014-11-04 07:21:10 -08:00
DMWriteTask.h Separate JSON functions from DMWriteTask. 2014-11-04 07:21:10 -08:00
README Clarify what DM means in the README file. 2014-10-09 09:00:30 -07:00

DM (Diamond Master, a.k.a Dungeon master, a.k.a GM 2).

DM is like GM, but multithreaded.  It doesn't do everything GM does.

DM's design is based around Tasks and a TaskRunner.

A Task represents an independent unit of work that might fail.  We make a task
for each GM/configuration pair we want to run.  Tasks can kick off new tasks
themselves.  For example, a CpuTask can kick off a ReplayTask to make sure
recording and playing back an SkPicture gives the same result as direct
rendering.

The TaskRunner runs all tasks on one of two threadpools, whose sizes are
configurable by --cpuThreads and --gpuThreads.  Ideally we'd run these on a
single threadpool but it can swamp the GPU if we shove too much work into it at
once.  --cpuThreads defaults to the number of cores on the machine.
--gpuThreads defaults to 1, but you may find 2 or 4 runs a little faster.

So the main flow of DM is:

    for each GM:
        for each configuration:
            kick off a new task
    < tasks run, maybe fail, and maybe kick off new tasks >
    wait for all tasks to finish
    report failures