Title: [261026] trunk/JSTests
Revision
261026
Author
[email protected]
Date
2020-05-01 15:03:15 -0700 (Fri, 01 May 2020)

Log Message

Fix stress/big-int-negate-jit.js
https://bugs.webkit.org/show_bug.cgi?id=211321

Reviewed by Saam Barati.

This test is assuming that 100000 iteration of 2 negateBigInt calls makes negateBigInt DFG-compiled.
But this is wrong if BigInt negation is fast enough. We made BigInt faster and now this test starts
failing because running this iteration finishes earlier than DFG compilation finishes. We should use `useConcurrentJIT`
to ensure this status.

* stress/big-int-negate-jit.js:

Modified Paths

Diff

Modified: trunk/JSTests/ChangeLog (261025 => 261026)


--- trunk/JSTests/ChangeLog	2020-05-01 22:00:00 UTC (rev 261025)
+++ trunk/JSTests/ChangeLog	2020-05-01 22:03:15 UTC (rev 261026)
@@ -1,5 +1,19 @@
 2020-05-01  Yusuke Suzuki  <[email protected]>
 
+        Fix stress/big-int-negate-jit.js
+        https://bugs.webkit.org/show_bug.cgi?id=211321
+
+        Reviewed by Saam Barati.
+
+        This test is assuming that 100000 iteration of 2 negateBigInt calls makes negateBigInt DFG-compiled.
+        But this is wrong if BigInt negation is fast enough. We made BigInt faster and now this test starts
+        failing because running this iteration finishes earlier than DFG compilation finishes. We should use `useConcurrentJIT`
+        to ensure this status.
+
+        * stress/big-int-negate-jit.js:
+
+2020-05-01  Yusuke Suzuki  <[email protected]>
+
         REGRESSION (r260990): 8 test262 unicode tests failing
         https://bugs.webkit.org/show_bug.cgi?id=211298
 

Modified: trunk/JSTests/stress/big-int-negate-jit.js (261025 => 261026)


--- trunk/JSTests/stress/big-int-negate-jit.js	2020-05-01 22:00:00 UTC (rev 261025)
+++ trunk/JSTests/stress/big-int-negate-jit.js	2020-05-01 22:03:15 UTC (rev 261026)
@@ -1,5 +1,5 @@
 //@ skip if not $jitTests
-//@ runBigIntEnabled
+//@ runDefault("--useConcurrentJIT=0")
 
 function assert(a, b) {
     if (a !== b)
_______________________________________________
webkit-changes mailing list
[email protected]
https://lists.webkit.org/mailman/listinfo/webkit-changes

Reply via email to