aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/gpu/GrOpFlushState.cpp
Commit message (Collapse)AuthorAge
* Handle too many (or too large) paths in GrDefaultPathRendererGravatar Brian Osman2017-06-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PathGeoBuilder constructs the geometry with the same basic technique as before, but allows interrupting the process to emit multiple draws. Original test case was 2000 non-AA stroked circles, which created ~66000 vertices. That now renders, as do various tests with a single large path (as well as filled paths). Added a new set of 'AtLeast' allocators for vertex and index data. These take a minimum size and a fallback size. If the minimum size can be satisfied by an existing block, then the caller gets *all* memory in that block, otherwise they get a new block sized for the fallback amount. The previous allocation scheme wasn't a good fit for the new use-case, and because we don't usually need many verts, the flexible approach seems appropriate. TODO: I think that this could be extracted and re-used for MSAA path renderer without too much work? I need to read that code more carefully to make sure it lines up. Re-land of: https://skia-review.googlesource.com/18360 Re-land of: https://skia-review.googlesource.com/18983 Bug: skia:6695 Change-Id: I09ac1273e5af67ed0e3e886de90e2970c3d0b239 Reviewed-on: https://skia-review.googlesource.com/19480 Commit-Queue: Brian Osman <brianosman@google.com> Reviewed-by: Brian Salomon <bsalomon@google.com>
* Retract GrTexture*.h & GrRenderTarget*.h from other headersGravatar Robert Phillips2017-06-13
| | | | | | | | | This does push some additional work (& includes) into the .cpp files. Change-Id: I27c847e371802270d13594dcc22aae44039990bb Reviewed-on: https://skia-review.googlesource.com/19660 Commit-Queue: Robert Phillips <robertphillips@google.com> Reviewed-by: Brian Salomon <bsalomon@google.com>
* Revert "Revert "Expand partial render target write pixels workaround.""Gravatar Brian Salomon2017-06-12
| | | | | | | | | | | New version limits the workaround to unorm configs. This reverts commit fdd117025fb846771e9e47cbf825d9071745fa98. Change-Id: I47a08a0ea4cf1acd88ca3c1bf9922cf0a8d215cc Reviewed-on: https://skia-review.googlesource.com/19490 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Revert "Expand partial render target write pixels workaround."Gravatar Brian Salomon2017-05-31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 0cc507d22566bb7e28e5fe21e4b3fc7b682d86a2. Reason for revert: Causing IntTextureText to fail on bots Original change's description: > Expand partial render target write pixels workaround. > > The workaround is extended in the following ways: > > 1) It now applies to any texture whose base level has *ever* been attached to a FBO. > 2) It applies to Adreno 5xx in addition to Adreno 4xx > 3) It applies in the atlas upload code path. > > This workaround (and a similar one) are narrowed to GLCaps rather than Caps. > > Bug: skia: > Change-Id: Id600e9739bb97bf6766075ea2a987fd2039e53e5 > Reviewed-on: https://skia-review.googlesource.com/18150 > Reviewed-by: Robert Phillips <robertphillips@google.com> > Commit-Queue: Brian Salomon <bsalomon@google.com> > TBR=bsalomon@google.com,robertphillips@google.com No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia: Change-Id: I53c8dd7ea7f614da57331470fcc24c6d84aba354 Reviewed-on: https://skia-review.googlesource.com/18229 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Expand partial render target write pixels workaround.Gravatar Brian Salomon2017-05-31
| | | | | | | | | | | | | | | | The workaround is extended in the following ways: 1) It now applies to any texture whose base level has *ever* been attached to a FBO. 2) It applies to Adreno 5xx in addition to Adreno 4xx 3) It applies in the atlas upload code path. This workaround (and a similar one) are narrowed to GLCaps rather than Caps. Bug: skia: Change-Id: Id600e9739bb97bf6766075ea2a987fd2039e53e5 Reviewed-on: https://skia-review.googlesource.com/18150 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Revert "Revert "Remove GrPipeline from GrDrawOp.""Gravatar Brian Salomon2017-03-21
| | | | | | | | | This reverts commit c48af934608bbb65650641f66adb51f2102d4274. Change-Id: I4ba78fd7e5a7d406b88223ca6f7245c029b60f76 Reviewed-on: https://skia-review.googlesource.com/9981 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Revert "Remove GrPipeline from GrDrawOp."Gravatar Brian Salomon2017-03-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 2bf4b3a97b770811d9e0558dbbfbdb57cfafbdb7. Reason for revert: nanobench assertion Original change's description: > Remove GrPipeline from GrDrawOp. > > GrDrawOp subclasses are now free to construct their pipelines at flush time and now in theory could use multiple GrPipelines for multipass rendering. > > GrProcessorSet may be used to retain the processors from a GrPaint with "pending execution" style refs. > > NVPR and Instanced rendering are updated to create their pipelines at flush time without a GrPipelineBuilder. > > The monolithic pipeline creation/management that was on GrDrawOp is moved to GrMeshDrawOp. However, this is temporary and will be removed in coming changes. > > Change-Id: I124282e3cea5d070970b5460c8a679fcaf7a8eff > Reviewed-on: https://skia-review.googlesource.com/7279 > Commit-Queue: Brian Salomon <bsalomon@google.com> > Reviewed-by: Robert Phillips <robertphillips@google.com> > TBR=bsalomon@google.com,robertphillips@google.com,csmartdalton@google.com,reviews@skia.org NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true Change-Id: I1bc64f6cbbd5f482417637a034342c2b5371dc5c Reviewed-on: https://skia-review.googlesource.com/9817 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Remove GrPipeline from GrDrawOp.Gravatar Brian Salomon2017-03-16
| | | | | | | | | | | | | | | GrDrawOp subclasses are now free to construct their pipelines at flush time and now in theory could use multiple GrPipelines for multipass rendering. GrProcessorSet may be used to retain the processors from a GrPaint with "pending execution" style refs. NVPR and Instanced rendering are updated to create their pipelines at flush time without a GrPipelineBuilder. The monolithic pipeline creation/management that was on GrDrawOp is moved to GrMeshDrawOp. However, this is temporary and will be removed in coming changes. Change-Id: I124282e3cea5d070970b5460c8a679fcaf7a8eff Reviewed-on: https://skia-review.googlesource.com/7279 Commit-Queue: Brian Salomon <bsalomon@google.com> Reviewed-by: Robert Phillips <robertphillips@google.com>
* Rename GrBatchAtlas files to GrDrawOpAtlasGravatar Brian Salomon2016-12-16
| | | | | | | | | This intermediary change only exists to make the actual class rename change readable on gerrit due to gerrit not recognizing file renames correctly. Change-Id: I919f84837fb17191ca49f00f82e56330f84766da Reviewed-on: https://skia-review.googlesource.com/6190 Reviewed-by: Brian Osman <brianosman@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Rename GrBatchFlushState to GrOpFlushStateGravatar Brian Salomon2016-12-07
Change-Id: I1a8052c61d7624929caf45ba44e2a465cd0dc1c2 Reviewed-on: https://skia-review.googlesource.com/5649 Commit-Queue: Brian Salomon <bsalomon@google.com> Reviewed-by: Brian Osman <brianosman@google.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-03 07:22:50 +0000