Bug 207160 – [ Mac ] css2.1/20110323/replaced-intrinsic-ratio-001.htm is flaky failing

• Component: CSS
• Status: NEW

[Jacob Uphoff]

css2.1/20110323/replaced-intrinsic-ratio-001.htm

This test has been flaky failing on debug for all visible history

I was able to reproduce the failure with running in iterations x5000 (it failed 8 times)

History:

https://results.webkit.org/?suite=layout-tests&test=css2.1%2F20110323%2Freplaced-intrinsic-ratio-001.htm&limit=50000

Diff:

--- /Volumes/Data/slave/catalina-debug-tests-wk1/build/layout-test-results/css2.1/20110323/replaced-intrinsic-ratio-001-expected.txt
+++ /Volumes/Data/slave/catalina-debug-tests-wk1/build/layout-test-results/css2.1/20110323/replaced-intrinsic-ratio-001-actual.txt
@@ -54,9 +54,9 @@
                     RenderSVGPath {path} at (26,8) size 124x28 [fill={[type=SOLID] [color=#008000]}] [data="M 500 50 L 150 200 L 850 200 Z"]
       RenderBlock (floating) {P} at (1,799) size 240x16
         RenderEmbeddedObject {OBJECT} at (16,0) size 208x76 [bgcolor=#FF0000] [border: (16px solid #0000FF)]
-          layer at (0,0) size 176x44
-            RenderView at (0,0) size 176x44
-          layer at (0,0) size 176x44
-            RenderSVGRoot {svg} at (0,0) size 176x44
-              RenderSVGRect {rect} at (0,0) size 176x44 [stroke={[type=SOLID] [color=#008000] [stroke width=12.00]}] [fill={[type=SOLID] [color=#00FF00]}] [x=0.00] [y=0.00] [width=1000.00] [height=250.00]
-              RenderSVGPath {path} at (26,8) size 124x28 [fill={[type=SOLID] [color=#008000]}] [data="M 500 50 L 150 200 L 850 200 Z"]
+          layer at (0,0) size 300x150
+            RenderView at (0,0) size 300x150
+          layer at (0,0) size 300x150
+            RenderSVGRoot {svg} at (0,0) size 300x77
+              RenderSVGRect {rect} at (0,0) size 300x77 [stroke={[type=SOLID] [color=#008000] [stroke width=12.00]}] [fill={[type=SOLID] [color=#00FF00]}] [x=0.00] [y=0.00] [width=1000.00] [height=250.00]
+              RenderSVGPath {path} at (45,15) size 210x45 [fill={[type=SOLID] [color=#008000]}] [data="M 500 50 L 150 200 L 850 200 Z"]

[Radar WebKit Bug Importer]

<rdar://problem/59130476>

[Jacob Uphoff]

Created attachment 389589 [details]
Update Test Expectations

[Ryan Haddad]

Comment on attachment 389589 [details]
Update Test Expectations

Clearing flags on attachment: 389589

Committed r255613: <https://trac.webkit.org/changeset/255613>

[Alexey Proskuryakov]

This test doesn't have any timers, animations and such. I wondered if it was similar to bug 206914, but there is no canvas either. What gives?

[alan]

(In reply to Alexey Proskuryakov from comment #4)
> This test doesn't have any timers, animations and such. I wondered if it was
> similar to bug 206914, but there is no canvas either. What gives?
The value of "300x150" indicates that it might be similar in the sense that this is also racy (layout vs. paint) and we end up using the default inline replaced dimension instead of a computed/used one.

[Simon Fraser (smfr)]

It does seem similar.

[Jason Lawrence]

This test is flaky failing on Mac wk2 now also. The diff is the same.

The flaky failures appear in the visual history for Mac wk2 on 03/05/2020.

History:
https://results.webkit.org/?suite=layout-tests&test=css2.1%2F20110323%2Freplaced-intrinsic-ratio-001.htm&limit=50000&platform=mac

diff: 
--- /Volumes/Data/slave/catalina-release-tests-wk2/build/layout-test-results/css2.1/20110323/replaced-intrinsic-ratio-001-expected.txt
+++ /Volumes/Data/slave/catalina-release-tests-wk2/build/layout-test-results/css2.1/20110323/replaced-intrinsic-ratio-001-actual.txt
@@ -54,9 +54,9 @@
                     RenderSVGPath {path} at (26,8) size 124x28 [fill={[type=SOLID] [color=#008000]}] [data="M 500 50 L 150 200 L 850 200 Z"]
       RenderBlock (floating) {P} at (1,799) size 240x16
         RenderEmbeddedObject {OBJECT} at (16,0) size 208x76 [bgcolor=#FF0000] [border: (16px solid #0000FF)]
-          layer at (0,0) size 176x44
-            RenderView at (0,0) size 176x44
-          layer at (0,0) size 176x44
-            RenderSVGRoot {svg} at (0,0) size 176x44
-              RenderSVGRect {rect} at (0,0) size 176x44 [stroke={[type=SOLID] [color=#008000] [stroke width=12.00]}] [fill={[type=SOLID] [color=#00FF00]}] [x=0.00] [y=0.00] [width=1000.00] [height=250.00]
-              RenderSVGPath {path} at (26,8) size 124x28 [fill={[type=SOLID] [color=#008000]}] [data="M 500 50 L 150 200 L 850 200 Z"]
+          layer at (0,0) size 300x150
+            RenderView at (0,0) size 300x150
+          layer at (0,0) size 300x150
+            RenderSVGRoot {svg} at (0,0) size 300x77
+              RenderSVGRect {rect} at (0,0) size 300x77 [stroke={[type=SOLID] [color=#008000] [stroke width=12.00]}] [fill={[type=SOLID] [color=#00FF00]}] [x=0.00] [y=0.00] [width=1000.00] [height=250.00]
+              RenderSVGPath {path} at (45,15) size 210x45 [fill={[type=SOLID] [color=#008000]}] [data="M 500 50 L 150 200 L 850 200 Z"]

[Jason Lawrence]

I have marked this test as flaky failing in wk2 also here: https://trac.webkit.org/changeset/260026/webkit

[Said Abou-Hallawa]

This is the current Mac failure diff

--- /Volumes/Data/worker/bigsur-debug-tests-wk2/build/layout-test-results/css2.1/20110323/replaced-intrinsic-ratio-001-expected.txt
+++ /Volumes/Data/worker/bigsur-debug-tests-wk2/build/layout-test-results/css2.1/20110323/replaced-intrinsic-ratio-001-actual.txt
@@ -1,4 +1,4 @@
-layer at (0,0) size 785x887
+layer at (0,0) size 785x886
   RenderView at (0,0) size 785x600
 layer at (0,0) size 785x823
   RenderBlock {HTML} at (0,0) size 785x823