RFR: 8261671: X86 I2L conversion can be skipped for certain masked positive values [v3]

Vladimir Kozlov kvn at openjdk.java.net
Thu Mar 4 21:18:54 UTC 2021


On Thu, 4 Mar 2021 19:38:30 GMT, Vladimir Kozlov <kvn at openjdk.org> wrote:

>> Hi @vnkozlov. I've added a BMI2.java and a compiler/intrinsics/bmi/verifycode/BzhiTestI2L.java test as you suggested. I'd appreciate any further review or comments.
>
> Thank you! I will run our testing and let you know results.

compiler/intrinsics/bmi/TestBzhiI2L.java test failed:
 stdout: [For random generator using seed: -6986231732124296613
To re-run test with same seed value please add "-Djdk.test.lib.random.seed=-6986231732124296613" to command line.
];
 stderr: [Exception in thread "main" java.lang.UnsupportedOperationException
	at compiler.intrinsics.bmi.Expr.longExpr(Expr.java:97)
	at compiler.intrinsics.bmi.BMITestRunner$Executor.runUnaryLongRegTest(BMITestRunner.java:280)
	at compiler.intrinsics.bmi.BMITestRunner$Executor.runTests(BMITestRunner.java:219)
	at compiler.intrinsics.bmi.BMITestRunner$Executor.main(BMITestRunner.java:190)
]
 exitValue = 1

java.lang.RuntimeException: Expected to get exit value of [0]

	at jdk.test.lib.process.OutputAnalyzer.shouldHaveExitValue(OutputAnalyzer.java:474)
	at compiler.intrinsics.bmi.BMITestRunner.runTest(BMITestRunner.java:151)
	at compiler.intrinsics.bmi.BMITestRunner.runTests(BMITestRunner.java:87)
	at compiler.intrinsics.bmi.TestBzhiI2L.main(TestBzhiI2L.java:49)

-------------

PR: https://git.openjdk.java.net/jdk/pull/2590


More information about the hotspot-compiler-dev mailing list