Bug 223223 - [ macOS Debug ARM64 ] 2X w3c/svg tests are constantly text failing
Summary: [ macOS Debug ARM64 ] 2X w3c/svg tests are constantly text failing
Status: NEW
Alias: None
Product: WebKit
Classification: Unclassified
Component: SVG (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2021-03-15 17:10 PDT by Robert Jenner
Modified: 2021-03-17 11:24 PDT (History)
7 users (show)

See Also:


Attachments
diffs from text failing tests (3.54 KB, application/zip)
2021-03-15 17:11 PDT, Robert Jenner
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Robert Jenner 2021-03-15 17:10:08 PDT
imported/w3c/web-platform-tests/svg/import/paths-data-03-f-manual.svg
svg/W3C-SVG-1.1/paths-data-03-f.svg 

Are constantly text failing in macOS Debug on Apple Silicon only. 

HISTORY URL:
https://results.webkit.org/?suite=layout-tests&suite=layout-tests&test=svg%2FW3C-SVG-1.1%2Fpaths-data-03-f.svg&test=imported%2Fw3c%2Fweb-platform-tests%2Fsvg%2Fimport%2Fpaths-data-03-f-manual.svg

While only one test notes that it is imported, both tests appear to be imported tests from MIT.

Here is the text diff for the first test. The diff for the second test is very very similar. I have also attached both diff files to the bug:

--- /Volumes/Data/worker/bigsur-debug-applesilicon-tests-wk1/build/layout-test-results/svg/W3C-SVG-1.1/paths-data-03-f-expected.txt
+++ /Volumes/Data/worker/bigsur-debug-applesilicon-tests-wk1/build/layout-test-results/svg/W3C-SVG-1.1/paths-data-03-f-actual.txt
@@ -24,7 +24,7 @@
           chunk 1 text run 1 at (390.00,300.00) startOffset 0 endOffset 4 width 23.34: "M, a"
       RenderSVGRect {rect} at (348,243) size 4x4 [fill={[type=SOLID] [color=#0000FF]}] [x=348.00] [y=243.00] [width=4.00] [height=4.00]
       RenderSVGRect {rect} at (428,303) size 4x4 [fill={[type=SOLID] [color=#0000FF]}] [x=428.00] [y=303.00] [width=4.00] [height=4.00]
-      RenderSVGPath {path} at (257,29) size 201x100 [fill={[type=SOLID] [color=#C0C000] [fill rule=EVEN-ODD]}] [data="M 270 30 C 252.272 50.1015 253.547 80.6061 272.891 99.158 C 292.234 117.71 322.766 117.71 342.109 99.158 C 361.453 80.6061 362.728 50.1015 345 30 C 325.402 41.3151 315.847 64.383 321.704 86.2422 C 327.561 108.101 347.37 123.301 370 123.301 C 392.63 123.301 412.439 108.101 418.296 86.2422 C 424.153 64.383 414.598 41.3151 395 30 C 370.621 36.2946 354.726 59.7348 357.898 84.7125 C 361.071 109.69 382.322 128.412 407.5 128.412 C 432.678 128.412 453.929 109.69 457.101 84.7125 C 460.274 59.7348 444.379 36.2946 420 30 Z"]
+      RenderSVGPath {path} at (257,29) size 201x100 [fill={[type=SOLID] [color=#C0C000] [fill rule=EVEN-ODD]}] [data="M 270 30 C 252.272 50.1015 253.547 80.6061 272.891 99.158 C 292.234 117.71 322.766 117.71 342.109 99.158 C 361.453 80.6061 362.728 50.1015 345 30 C 325.402 41.3151 315.847 64.383 321.704 86.2422 C 327.561 108.101 347.37 123.301 370 123.301 C 392.63 123.301 412.439 108.101 418.296 86.2422 C 424.153 64.383 414.598 41.3151 395 30 C 370.621 36.2946 354.726 59.7348 357.898 84.7125 C 361.071 109.69 382.322 128.412 407.5 128.412 C 432.678 128.412 453.929 109.69 457.102 84.7125 C 460.274 59.7348 444.379 36.2946 420 30 Z"]
       RenderSVGText {text} at (280,124) size 65x14 contains 1 chunk(s)
         RenderSVGInlineText {#text} at (0,0) size 65x14
           chunk 1 text run 1 at (280.00,135.00) startOffset 0 endOffset 13 width 64.02: "M, A, a, a, z"
Comment 1 Robert Jenner 2021-03-15 17:11:21 PDT
Created attachment 423265 [details]
diffs from text failing tests

Attaching two diffs from text failing tests.
Comment 2 Radar WebKit Bug Importer 2021-03-15 17:11:59 PDT
<rdar://problem/75455148>
Comment 3 Robert Jenner 2021-03-15 17:12:55 PDT
These tests only fail on Apple Silicon Macs, and as such I cannot reproduce the failures because I do not have access to an AS Mac.
Comment 4 Robert Jenner 2021-03-15 17:27:06 PDT
Updated test expectations to [ Failure ] while test is being looked at. 
https://trac.webkit.org/changeset/274454/webkit
Comment 5 Alexey Proskuryakov 2021-03-17 11:24:52 PDT
Looks like a rounding issue, but a pretty large one: 457.102 vs. 457.101.