-
Bug
-
Resolution: Duplicate
-
P4
-
23
-
x86_64
This is the same failure as reported in JDK-8323641 and JDK-8327978.
I suspect that it was a regression ofJDK-8324517: CastII nodes took up 90%+ of compile time, and compile time was extremely high 100sec+.
AVX512 seems required to reach the required unrolling factor.
Now this change was backed-out, and may be re-introduced in JDK-8332827.
Test: compiler/loopopts/superword/TestAlignVectorFuzzer.java#VerifyAlignVector-NoTieredCompilation-Xbatch
ITERATION 2 of 5. Test testDUBCFH, time allowance: 110626, until timeout: 418626
62649 403 b compiler.loopopts.superword.TestAlignVectorFuzzer::testDUBCFH (1237 bytes)
Timeout signalled after 480 seconds
TEST PASSED: hit maximal time allownance during iteration 2, time allowance: -383253, until timeout: -75253
I suspect that it was a regression of
AVX512 seems required to reach the required unrolling factor.
Now this change was backed-out, and may be re-introduced in JDK-8332827.
Test: compiler/loopopts/superword/TestAlignVectorFuzzer.java#VerifyAlignVector-NoTieredCompilation-Xbatch
ITERATION 2 of 5. Test testDUBCFH, time allowance: 110626, until timeout: 418626
62649 403 b compiler.loopopts.superword.TestAlignVectorFuzzer::testDUBCFH (1237 bytes)
Timeout signalled after 480 seconds
TEST PASSED: hit maximal time allownance during iteration 2, time allowance: -383253, until timeout: -75253
- duplicates
-
JDK-8332829 [BACKOUT] C2: crash in compiled code because of dependency on removed range check CastIIs
- Resolved
- relates to
-
JDK-8327978 C2 SuperWord: Fix compilation time regression in dependency graph traversal after JDK-8325651
- Resolved
-
JDK-8332537 C2: High memory usage reported for compiler/loopopts/superword/TestAlignVectorFuzzer.java
- Resolved
-
JDK-8324517 C2: crash in compiled code because of dependency on removed range check CastIIs
- Closed
-
JDK-8323641 Test compiler/loopopts/superword/TestAlignVectorFuzzer.java timed out
- Resolved