Disable GPU clip masks

Bug: skia:
Change-Id: I3abacfd17ea6247d0bdb2ab8a2584a88bf468dad
Reviewed-on: https://skia-review.googlesource.com/140776
Reviewed-by: Stephen White <senorblanco@chromium.org>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>

1 files changed, 4 insertions, 0 deletions

diff --git a/src/gpu/GrClipStackClip.cpp b/src/gpu/GrClipStackClip.cpp
index 7704d929f9..92af020e78 100644
--- a/src/gpu/GrClipStackClip.cpp
+++ b/src/gpu/GrClipStackClip.cpp
@@ -148,6 +148,10 @@ bool GrClipStackClip::UseSWOnlyPath(GrContext* context,
                                     bool hasUserStencilSettings,
                                     const GrRenderTargetContext* renderTargetContext,
                                     const GrReducedClip& reducedClip) {
+    // TODO: right now it appears that GPU clip masks are strictly slower than software. We may
+    // want to revisit this assumption once we can test with render target sorting.
+    return true;
+
     // TODO: generalize this function so that when
     // a clip gets complex enough it can just be done in SW regardless
     // of whether it would invoke the GrSoftwarePathRenderer.