Re: RFR: 8255616: Disable AOT and Graal in Oracle OpenJDK

On Fri, 30 Oct 2020 17:52:09 GMT, Igor Ignatyev wrote: >> We shipped Ahead-of-Time compilation (the jaotc tool) in JDK 9, as an >> experimental feature. We shipped Graal as an experimental JIT compiler in >> JDK 10. We haven't seen much use of these features, and the effort required >> to

Re: RFR: 8223347: Integration of Vector API (Incubator) [v4]

On Wed, 14 Oct 2020 00:34:04 GMT, Sandhya Viswanathan wrote: >> There are several gc tests crashed in panama-vector tier3 testing which >> seems are not observed in openjdk repo. >> The crashes look like: >> # assert(oopDesc::is_oop(obj)) failed: not an oop: 0xfff1 >> # >> # JRE

Re: RFR: 8223347: Integration of Vector API (Incubator) [v4]

On Mon, 12 Oct 2020 12:56:10 GMT, Erik Joelsson wrote: >> Paul Sandoz has updated the pull request with a new target base due to a >> merge or a rebase. The pull request now >> contains ten commits: >> - Merge master >> - Fix related to merge >> - HotspotIntrinsicCandidate to

Re: RFR(T/XS) 8231145: [Graal] org.graalvm.compiler.debug.test.DebugContextTest fails because DebugContextTest.testLogging.input is not available

to BUILD_VM_COMPILER_TESTS and keep the SUFFIXES you already added, and skip the addition to SRCS. /Erik On 2019-09-17 15:45, Ekaterina Pavlova wrote: Hi, please review the following change which fixes org.graalvm.compiler.debug.test.DebugContextTest. The test fails because it tries to read

RFR(T/XS) 8231145: [Graal] org.graalvm.compiler.debug.test.DebugContextTest fails because DebugContextTest.testLogging.input is not available

Hi, please review the following change which fixes org.graalvm.compiler.debug.test.DebugContextTest. The test fails because it tries to read DebugContextTest.testLogging.input file which is not available at runtime. The fix copies testLogging.input file into jdk.vm.compiler.tests.jar.

Re: RFR (XXS) 8214557: Filter out VM flags which don't affect AOT code generation

Thanks Erik for prompt review. -katya On 11/30/18 2:54 PM, Erik Joelsson wrote: Looks ok. /Erik On 2018-11-30 14:47, Ekaterina Pavlova wrote: Hi, AOT requires that the same java runtime configuration should be used during AOT compilation and execution. This implies the same VM flags used

RFR (XXS) 8214557: Filter out VM flags which don't affect AOT code generation

Hi, AOT requires that the same java runtime configuration should be used during AOT compilation and execution. This implies the same VM flags used by java and jaotc. However some flags could dramatically slow down jaotc while don't affecting AOT code generation at all. This fix filters out

Re: RFR(M) 8152988: [AOT] Update test batch definitions to include aot-ed java.base module mode into hs-comp testing

yes, it is at the same location: http://cr.openjdk.java.net/~epavlova//8152988/webrev.01/ -katya On 11/1/18 3:15 PM, Magnus Ihse Bursie wrote: Do you have an updated webrev? /Magnus 1 nov. 2018 kl. 22:28 skrev Ekaterina Pavlova : Magnus, thanks for the review. I fixed everything you

Re: RFR(M) 8152988: [AOT] Update test batch definitions to include aot-ed java.base module mode into hs-comp testing

Magnus, thanks for the review. I fixed everything you pointed to. thanks, -katya On 11/1/18 6:46 AM, Magnus Ihse Bursie wrote: Hi Katya, Sorry for the late response. :-( On 2018-10-29 21:24, Ekaterina Pavlova wrote: Vladimir, I added "-XX:+UseAOTStrictLoading" to "java -ver

Re: RFR(M) 8152988: [AOT] Update test batch definitions to include aot-ed java.base module mode into hs-comp testing

+$1_JTREG_BASIC_OPTIONS += -vmoptions:"$$($1_AOT_OPTIONS)" + endif Please review updated webrev: http://cr.openjdk.java.net/~epavlova//8152988/webrev.01/ thanks, -katya On 10/23/18 4:40 PM, Vladimir Kozlov wrote: On 10/22/18 9:36 AM, Ekaterina Pavlova wrote: Vladimir, thanks a

Re: RFR(XXS) 8212877: Restore JTREG_VERBOSE value for mach5 testing

Thanks David. -katya On 10/23/18 3:04 PM, David Holmes wrote: LGTM. Thanks, David On 24/10/2018 6:33 AM, Ekaterina Pavlova wrote: Hi All, Please review this extra small change which restores JTREG_VERBOSE value for mach5 testing. Default JTREG_VERBOSE value was changed from "fail,

RFR(XXS) 8212877: Restore JTREG_VERBOSE value for mach5 testing

Hi All, Please review this extra small change which restores JTREG_VERBOSE value for mach5 testing. Default JTREG_VERBOSE value was changed from "fail,error,time" to "fail,error,summary" after switching to run-test framework (JDK-8212028). Having compact tests output is good for local users

Re: RFR(M) 8152988: [AOT] Update test batch definitions to include aot-ed java.base module mode into hs-comp testing

atya On 10/19/18 12:21 AM, Magnus Ihse Bursie wrote: On 2018-10-18 08:29, Ekaterina Pavlova wrote: Hi All, Please review the changes which add support to build AOTed jdk modules and then use them during the testing. Note, building of AOTed libraries needs to be done at the same machine t

Re: RFR(M) 8152988: [AOT] Update test batch definitions to include aot-ed java.base module mode into hs-comp testing

Vladimir, thanks a lot for the review. "-XX:+PrintAOT" was only passed to "java -version" which is done before the testing to verify that AOTed libraries work. It also perhaps could be useful debug info in case some tests starts fail. But I can change -XX:+PrintAOT to -XX:+UseAOTStrictLoading

RFR(M) 8152988: [AOT] Update test batch definitions to include aot-ed java.base module mode into hs-comp testing

Hi All, Please review the changes which add support to build AOTed jdk modules and then use them during the testing. Note, building of AOTed libraries needs to be done at the same machine the testing is going to be started. So, this could not be done at jdk build time. Updated files: -

Re: RFR(S) 8205078: [Graal] org.graalvm.compiler.core.test.VerifyDebugUsageTest fails with "Expected exception: VerificationError"

Doug, Vladimir, thanks for your reviews. -katya On 8/7/18 9:10 AM, Vladimir Kozlov wrote: Okay. Changes are good then. Vladimir On 8/6/18 11:40 PM, Doug Simon wrote: On 7 Aug 2018, at 08:07, Ekaterina Pavlova wrote: Vladimir, thanks for prompt review. I can revert the changes and add

Re: RFR(S) 8205078: [Graal] org.graalvm.compiler.core.test.VerifyDebugUsageTest fails with "Expected exception: VerificationError"

Based on what Doug said: "We don't support the indy string pattern in these tests". Note, Graal itself (and JVMCI) are compiled without Indy strings concat: http://hg.openjdk.java.net/jdk/jdk/file/c23f1e4910f3/make/CompileJavaModules.gmk#l425 Thanks, Vladimir On 8/6/18 1:42 PM, Ekaterina Pav

RFR(S) 8205078: [Graal] org.graalvm.compiler.core.test.VerifyDebugUsageTest fails with "Expected exception: VerificationError"

Hi All, please review the change which introduces additional make target in JtregGraalUnit.gmk to compile org/graalvm/compiler/core/test/VerifyDebugUsageTest.java with -XDstringConcat=inline. Otherwise the test fails. Also moved building of jdk.vm.compiler.replacements.verifier.jar into

Re: RFR(XXS): 8206113: Troubles configuring graal tests

On 7/2/18 2:17 PM, David Holmes wrote: Hi Katya, On 3/07/2018 5:10 AM, Ekaterina Pavlova wrote: Hi All, please review this extra small change which makes make more silent by removing info message in case graal unit tests are not built. Happy to see the build-time warning go, but I thought

RFR(XXS): 8206113: Troubles configuring graal tests

Hi All, please review this extra small change which makes make more silent by removing info message in case graal unit tests are not built. JBS: https://bugs.openjdk.java.net/browse/JDK-8206113 webrev: http://cr.openjdk.java.net/~epavlova//8206113/webrev.00/index.html thanks, -katya p.s.

Re: RFR(XXS) : 8206088 : 8205207 broke builds

Good, thanks for quick response and webrev! -katya On 6/28/18 9:42 PM, Igor Ignatyev wrote: http://cr.openjdk.java.net/~iignatyev//8206088/webrev.00/index.html 1 line changed: 0 ins; 0 del; 1 mod; Hi all, could you please review this one liner fix? webrev:

Re: RFR(M): 8205207: Port Graal unit tests under jtreg

Thanks Erik! -katya On 6/27/18 8:36 AM, Erik Joelsson wrote: On 2018-06-27 00:29, Ekaterina Pavlova wrote: well, INCLUDE_GRAAL is not defined at the time we run tests. I can try to guard by something like  ifeq ($(OPENJDK_TARGET_OS)-$(OPENJDK_TARGET_CPU), $(filter $(OPENJDK_TARGET_OS

Re: RFR(M): 8205207: Port Graal unit tests under jtreg

On 6/26/18 9:08 AM, Ekaterina Pavlova wrote: Hello Magnus, On 6/26/18 12:50 AM, Magnus Ihse Bursie wrote: 23 juni 2018 kl. 00:22 skrev Ekaterina Pavlova mailto:ekaterina.pavl...@oracle.com>>: Fixed and regenerated webrev at the same location: http://cr.openjdk.java.net/~epavlova//8

Re: RFR(M): 8205207: Port Graal unit tests under jtreg

Hello Magnus, On 6/26/18 12:50 AM, Magnus Ihse Bursie wrote: 23 juni 2018 kl. 00:22 skrev Ekaterina Pavlova mailto:ekaterina.pavl...@oracle.com>>: Fixed and regenerated webrev at the same location: http://cr.openjdk.java.net/~epavlova//8205207/webrev.01/index.html Better, but not do

Re: RFR(M): 8205207: Port Graal unit tests under jtreg

Thanks Erik! -katya On 6/25/18 8:52 AM, Erik Joelsson wrote: Looks ok to me. /Erik On 2018-06-22 15:22, Ekaterina Pavlova wrote: Fixed and regenerated webrev at the same location:  http://cr.openjdk.java.net/~epavlova//8205207/webrev.01/index.html Erik, thanks again for your detailed

Re: RFR(M): 8205207: Port Graal unit tests under jtreg

. In JtregGraalUnit.gmk: Some minor style nits: 54: align )) with first eval line as you have done with the other eval calls 118: Since 117 is a one line parameter, please move comma to 117 133: Same as for 118 130-132: Please indent 4 spaces for continuation /Erik On 2018-06-22 14:16, Ekaterina Pavlova wrote

Re: RFR(M): 8205207: Port Graal unit tests under jtreg

On 6/22/18 2:29 PM, Doug Simon wrote: On 22 Jun 2018, at 23:16, Ekaterina Pavlova wrote: Erik, Doug, thank you a lot for your reviews and advises. I fixed everything what Erik has pointed out, please see my answers inlined. As about moving more staff in 'updategraalinopenjdk' can we

Re: RFR(M): 8205207: Port Graal unit tests under jtreg

procedures?) On 2018-06-18 22:26, Ekaterina Pavlova wrote: Hi All, please review porting of Graal unit tests under jtreg. The main idea of this porting is to keep Graal unit tests as is (located in src/jdk.internal.vm.compiler/share/classes/*.test) and run them similar way they are run in Graal

Re: RFR(S) 8197453 : Add support of extra problem list

as excluded on all platforms, the convention is to use "generic-all". -- Jon On 2/26/18 11:47 AM, Igor Ignatyev wrote: adding build-dev alias -- Igor On Feb 8, 2018, at 3:08 PM, Ekaterina Pavlova <ekaterina.pavl...@oracle.com> wrote: Hi all, ProblemList.txt files used by mak