| Summary: | [CoordinatedGraphics] Remove unnecessary two virtual methods from TiledBackingStoreClient | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Ryuan Choi <ryuan.choi> | ||||||
| Component: | WebKit EFL | Assignee: | Ryuan Choi <ryuan.choi> | ||||||
| Status: | RESOLVED FIXED | ||||||||
| Severity: | Normal | CC: | cmarcelo, commit-queue, gyuyoung.kim, kondapallykalyan, lucas.de.marchi, luiz, noam, yoon | ||||||
| Priority: | P2 | ||||||||
| Version: | 528+ (Nightly build) | ||||||||
| Hardware: | Unspecified | ||||||||
| OS: | Unspecified | ||||||||
| Attachments: |
|
||||||||
|
Description
Ryuan Choi
2015-07-20 19:30:22 PDT
Created attachment 257156 [details]
Patch
Comment on attachment 257156 [details]
Patch
Any reason you didn't submit to EWS ?
(In reply to comment #2) > Comment on attachment 257156 [details] > Patch > > Any reason you didn't submit to EWS ? Sorry for the late answer. I didn't know it. I submit to EWS Comment on attachment 257156 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=257156&action=review > Source/WebCore/platform/graphics/texmap/coordinated/TiledBackingStore.cpp:-132 > - IntRect dirtyRect = mapFromContents(contentsRect); When we don't adjust mapFromContents(), is there any problem ? mapFromContents() applies a content scale factor. IntRect TiledBackingStore::mapFromContents(const IntRect& rect) const { return enclosingIntRect(FloatRect(rect.x() * m_contentsScale, rect.y() * m_contentsScale, rect.width() * m_contentsScale, rect.height() * m_contentsScale)); } Comment on attachment 257156 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=257156&action=review >> Source/WebCore/platform/graphics/texmap/coordinated/TiledBackingStore.cpp:-132 >> - IntRect dirtyRect = mapFromContents(contentsRect); > > When we don't adjust mapFromContents(), is there any problem ? > > mapFromContents() applies a content scale factor. > > IntRect TiledBackingStore::mapFromContents(const IntRect& rect) const > { > return enclosingIntRect(FloatRect(rect.x() * m_contentsScale, > rect.y() * m_contentsScale, > rect.width() * m_contentsScale, > rect.height() * m_contentsScale)); > } With this patch, coverageRatio will get dirtyRect which is already scaled, while previous coverageRatio get contentsRect. > Source/WebCore/platform/graphics/texmap/coordinated/TiledBackingStore.cpp:151 > +void TiledBackingStore::createTiles(const IntRect& visibleRect, const IntRect& contentsRect) I will rename these as scaledXXX not to confuse. Created attachment 260026 [details]
Patch
Comment on attachment 260026 [details]
Patch
I can't find behavior change here. So rs=me that there is no behavior change on the assumption. However someone(yoon?) might want to take a final look before landing.
Comment on attachment 260026 [details] Patch Clearing flags on attachment: 260026 Committed r189161: <http://trac.webkit.org/changeset/189161> All reviewed patches have been landed. Closing bug. |