aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/core/SkThreadedBMPDevice.cpp
Commit message (Collapse)AuthorAge
* handle large rects, rename helperGravatar Mike Reed2018-01-17
| | | | | | | | | | | To fix gm/bigrect, needed to do adjust "largest" rect so it doesn't become empty when round-tripping with SkRect/SkIRect. I renamed it after this. Bug: skia: Change-Id: I747782c8456da603cf298275d2300ea1996e7629 Reviewed-on: https://skia-review.googlesource.com/95563 Commit-Queue: Mike Reed <reed@google.com> Reviewed-by: Yuqian Li <liyuqian@google.com>
* Compare drawBounds with MakeLargest instead of isFiniteGravatar Yuqian Li2018-01-09
| | | | | | | | | | | | In some cases, transformedBounds is finite but its rounded out IRect is empty so we can't do that. TBR: reed@google.com Bug: skia: Change-Id: Ife78e91a494964042fa09b8cbe86cabbd2572fb9 Reviewed-on: https://skia-review.googlesource.com/92625 Reviewed-by: Yuqian Li <liyuqian@google.com> Commit-Queue: Yuqian Li <liyuqian@google.com>
* Remove macro in the threaded backendGravatar Yuqian Li2018-01-09
| | | | | | | | | | | | This is a smaller CL distilled from our init-once CL (87784). Note that the thread alloc is unused here but it will be used by init-once. Bug: skia: Change-Id: If4bc2febfb3fbcd85bf973a01e60b9b64c873a19 Reviewed-on: https://skia-review.googlesource.com/92621 Commit-Queue: Yuqian Li <liyuqian@google.com> Reviewed-by: Herb Derby <herb@google.com>
* move largest apis into privateGravatar Mike Reed2018-01-08
| | | | | | | | | Related to https://skia-review.googlesource.com/c/skia/+/91860 Bug: skia: Change-Id: Ia8fd981b422bbab75541b078277d2e09e1fc9d41 Reviewed-on: https://skia-review.googlesource.com/91940 Reviewed-by: Brian Salomon <bsalomon@google.com>
* Try to bypass the false-positive clang thread analysis warningGravatar Yuqian Li2017-12-01
| | | | | | | | | | | | | | | | This is identical to 70221 except for 3 lines inside SkFlexibleTaskGroup2D::work to bypass the false-positive warning. We cannot reproduce the error the android roller generated so we'll have to try and see. The detailed warning can be found in: https://skia-review.googlesource.com/c/skia/+/78720 TBR: mtklein@google.com Bug: skia: Change-Id: I3e2f414581dbd1398482cf45cf4f43eaf0535651 Reviewed-on: https://skia-review.googlesource.com/79321 Commit-Queue: Yuqian Li <liyuqian@google.com> Reviewed-by: Yuqian Li <liyuqian@google.com>
* Revert "Renaming and refactoring to prepare for init-once threaded backend"Gravatar Leon Scroggins2017-11-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 07a42411f8d4a00d992bafdd18f2183f58c4547a. Reason for revert: This looks to be breaking the Android roll. See https://googleplex-android-review.googlesource.com/c/platform/external/skia/+/3295488 external/skia/src/core/SkTaskGroup2D.cpp:73:28: error: releasing mutex 'rowData.fMutex' that was not held [-Werror,-Wthread-safety-analysis] rowData.fMutex.unlock(); Original change's description: > Renaming and refactoring to prepare for init-once threaded backend > > Bug: skia: > Change-Id: I39b1d73b612a9c133847dd7361d634a0351659f1 > Reviewed-on: https://skia-review.googlesource.com/70221 > Reviewed-by: Herb Derby <herb@google.com> > Commit-Queue: Yuqian Li <liyuqian@google.com> TBR=mtklein@google.com,herb@google.com,liyuqian@google.com Change-Id: I701e6c62f6f437a6a285953b45b2e58cf7f0b6e5 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia: Reviewed-on: https://skia-review.googlesource.com/78720 Reviewed-by: Leon Scroggins <scroggo@google.com> Commit-Queue: Leon Scroggins <scroggo@google.com>
* Renaming and refactoring to prepare for init-once threaded backendGravatar Yuqian Li2017-11-30
| | | | | | | | Bug: skia: Change-Id: I39b1d73b612a9c133847dd7361d634a0351659f1 Reviewed-on: https://skia-review.googlesource.com/70221 Reviewed-by: Herb Derby <herb@google.com> Commit-Queue: Yuqian Li <liyuqian@google.com>
* Add FIFO thread pool, use it for most things.Gravatar Mike Klein2017-09-01
| | | | | | | | | | | | | | | | | | | We like a LIFO default thread pool in tools like DM for better memory/time locality... the bots use less memory this way, and generally run faster. But most use cases want a FIFO queue, so that they can get going on the next parts of early work while later work is still running. This splits the implementation into one using SkTArray and pop_back for LIFO, and a new one using std::deque and pop_front for FIFO. Change-Id: Ief203b6869a00f1f8084019431a781d15fc63750 Reviewed-on: https://skia-review.googlesource.com/41849 Commit-Queue: Mike Klein <mtklein@chromium.org> Commit-Queue: Brian Osman <brianosman@google.com> Reviewed-by: Yuqian Li <liyuqian@google.com> Reviewed-by: Brian Osman <brianosman@google.com>
* Allow providing SkExecutor to SkThreadedBMPDeviceGravatar Yuqian Li2017-08-30
| | | | | | | | | | | So that callers could manage the thread pool themselves instead of relying on SkThreadBMPDevice. Bug: skia: Change-Id: Idbeafc9ac5d111370fd1b74f6eab9ac8df179940 Reviewed-on: https://skia-review.googlesource.com/39760 Reviewed-by: Mike Klein <mtklein@google.com> Commit-Queue: Yuqian Li <liyuqian@google.com>
* Flush threaded device when reaching max queue sizeGravatar Yuqian Li2017-08-07
| | | | | | | | | | | Our SkThreadedBMPDevice is very experimental so I didn't handle this edge case earlier. Maybe it's now a good time to fix it. Bug: skia: Change-Id: Ie3938475449c1341d34200ff3afe4589836950fc Reviewed-on: https://skia-review.googlesource.com/31203 Reviewed-by: Florin Malita <fmalita@chromium.org> Commit-Queue: Yuqian Li <liyuqian@google.com>
* SkDevice::drawBitmap takes x,y, not matrixGravatar Hal Canary2017-06-27
| | | | | | | | | Motivation: a simpler call should make it easier for SkPDF to learn how to do drawBitmap with A8 bitmap and a maskfilter. Change-Id: I1a5d190b40b0e9e08fa8876d265a9835a3e9987d Reviewed-on: https://skia-review.googlesource.com/20961 Reviewed-by: Mike Reed <reed@google.com> Commit-Queue: Hal Canary <halcanary@google.com>
* we can std::move work to avoid copying itGravatar Mike Klein2017-06-07
| | | | | | | | | clang-tidy pointed out this opportunity in the Google3 build. Change-Id: Id3c7b291fec12cbf860e7fc2583c1a34ad23f697 Reviewed-on: https://skia-review.googlesource.com/18927 Reviewed-by: Yuqian Li <liyuqian@google.com> Commit-Queue: Mike Klein <mtklein@chromium.org>
* Add TiledDrawScheduler so we can concurrently draw and enqueGravatar Yuqian Li2017-06-05
| | | | | | | | | | | | | | | | | (instead of finishing enque before draw). The highlight is that we can now achieve 9x speedup compared to 5x in all our previous approaches (including multi-picture draw). The schedulers here are experimental. I'd like to move on to try initializing once for each draw before further polishing and optimizing the schedule mechanism. Bug: skia: Change-Id: Idc3d030d475af9645c24c5372ff62b9a402206cc Reviewed-on: https://skia-review.googlesource.com/17826 Reviewed-by: Mike Reed <reed@google.com> Reviewed-by: Herb Derby <herb@google.com> Commit-Queue: Yuqian Li <liyuqian@google.com>
* Add SkThreadedBMPDevice for Threaded Raster BackendGravatar Yuqian Li2017-04-03
BUG=skia: Change-Id: I882b6563c735796f3a4dcd19f6c79e7efd8306ae Reviewed-on: https://skia-review.googlesource.com/10505 Commit-Queue: Yuqian Li <liyuqian@google.com> Reviewed-by: Mike Klein <mtklein@chromium.org>
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-03 13:49:24 +0000