From dean.long at oracle.com Thu Jul 2 00:08:35 2020 From: dean.long at oracle.com (Dean Long) Date: Wed, 1 Jul 2020 17:08:35 -0700 Subject: RFR(XL) 8247922: Update Graal Message-ID: https://bugs.openjdk.java.net/browse/JDK-8247922 http://cr.openjdk.java.net/~dlong/8247922/webrev/ This is a Graal update.? Changes since the last update (JDK-8243380) are listed in the bug description. dl From vladimir.kozlov at oracle.com Thu Jul 2 17:12:00 2020 From: vladimir.kozlov at oracle.com (Vladimir Kozlov) Date: Thu, 2 Jul 2020 10:12:00 -0700 Subject: RFR(XL) 8247922: Update Graal In-Reply-To: References: Message-ID: Looks good. Thank you for linking failures to bugs. Vladimir K. On 7/1/20 5:08 PM, Dean Long wrote: > https://bugs.openjdk.java.net/browse/JDK-8247922 > http://cr.openjdk.java.net/~dlong/8247922/webrev/ > > This is a Graal update.? Changes since the last update (JDK-8243380) are listed in the bug description. > > dl > > > From dean.long at oracle.com Thu Jul 2 20:01:30 2020 From: dean.long at oracle.com (Dean Long) Date: Thu, 2 Jul 2020 13:01:30 -0700 Subject: RFR(XL) 8247922: Update Graal In-Reply-To: References: Message-ID: <8afc61d8-a542-0952-7854-13bfca718d2f@oracle.com> Thanks Vladimir. dl On 7/2/20 10:12 AM, Vladimir Kozlov wrote: > Looks good. Thank you for linking failures to bugs. > > Vladimir K. > > On 7/1/20 5:08 PM, Dean Long wrote: >> https://bugs.openjdk.java.net/browse/JDK-8247922 >> http://cr.openjdk.java.net/~dlong/8247922/webrev/ >> >> This is a Graal update.? Changes since the last update (JDK-8243380) >> are listed in the bug description. >> >> dl >> >> >> From VOJIN.JOVANOVIC at ORACLE.COM Mon Jul 6 09:56:05 2020 From: VOJIN.JOVANOVIC at ORACLE.COM (Vojin Jovanovic) Date: Mon, 6 Jul 2020 11:56:05 +0200 Subject: Reduced set of classes that are initialized at build time by Native Image Message-ID: <3C612F04-AA6C-4BDC-B82C-46AE8DC2909F@ORACLE.COM> Hi all, We plan to reduce the set of classes that are initialized at build time by default in GraalVM Native Image. We plan to do this because for many JDK packages (e.g., `java.awt?) the cost of maintaining initialization at build time outweighs the performance benefits. The issue that covers this change is https://github.com/oracle/graal/issues/2650 and the corresponding PR is https://github.com/oracle/graal/pull/2656 The plan is to merge this feature at the end of the week (before the cut-off for the 20.2 release). I will try to shrink the list of initialized packages even further (https://github.com/oracle/graal/pull/2656/files#diff-fe1c43eaa0fc9417273054562118a946). However, future changes of this list should be small compared to what is already done. This might be a breaking change for any project that uses Native Image and initializes classes at build time: JDK classes that got accidentally initialized will report an error, and JDK objects that end up in the image heap will report an error. The fix should be straightforward: either don?t initialize your project classes at build time, or simply add the initialized (or snapshotted) classes to the list of initializations with `?initialize-at-build-time`. Please comment on the PR if you have ideas, concerns, or unexpected issues with your project. Thanks, - Vojin From shelajev at gmail.com Mon Jul 6 12:22:32 2020 From: shelajev at gmail.com (=?UTF-8?Q?Oleg_=C5=A0elajev?=) Date: Mon, 6 Jul 2020 15:22:32 +0300 Subject: Reduced set of classes that are initialized at build time by Native Image In-Reply-To: <3C612F04-AA6C-4BDC-B82C-46AE8DC2909F@ORACLE.COM> References: <3C612F04-AA6C-4BDC-B82C-46AE8DC2909F@ORACLE.COM> Message-ID: Adding GraalVM mailing lists to which this email about reducing the set of classes that are initialized at build time by default in GraalVM Native Image could be interesting. This might be a breaking change. Oleg On Mon, 6 Jul 2020 at 13:15, Vojin Jovanovic wrote: > Hi all, > > We plan to reduce the set of classes that are initialized at build time by > default in GraalVM Native Image. We plan to do this because for many JDK > packages (e.g., `java.awt?) the cost of maintaining initialization at build > time outweighs the performance benefits. The issue that covers this change > is > > https://github.com/oracle/graal/issues/2650 > > and the corresponding PR is > > https://github.com/oracle/graal/pull/2656 > > The plan is to merge this feature at the end of the week (before the > cut-off for the 20.2 release). I will try to shrink the list of initialized > packages even further ( > https://github.com/oracle/graal/pull/2656/files#diff-fe1c43eaa0fc9417273054562118a946). > However, future changes of this list should be small compared to what is > already done. > > This might be a breaking change for any project that uses Native Image and > initializes classes at build time: JDK classes that got accidentally > initialized will report an error, and JDK objects that end up in the image > heap will report an error. > > The fix should be straightforward: either don?t initialize your project > classes at build time, or simply add the initialized (or snapshotted) > classes to the list of initializations with `?initialize-at-build-time`. > > Please comment on the PR if you have ideas, concerns, or unexpected issues > with your project. > > Thanks, > - Vojin From dean.long at oracle.com Wed Jul 29 19:48:01 2020 From: dean.long at oracle.com (Dean Long) Date: Wed, 29 Jul 2020 12:48:01 -0700 Subject: [15] RFR(XS) 8248597: [Graal] api/java_security/SignatureSpi/DelegationTests.html fails with Method "javasoft.sqe.tests.api.java.security.SignatureSpi.JCKSignatureSpi.clear" doesn't exist. Message-ID: <688dfe7e-c8ad-3c37-d2f7-432609f45e2f@oracle.com> https://bugs.openjdk.java.net/browse/JDK-8248597 http://cr.openjdk.java.net/~dlong/8248597/webrev/ This change fixes an issue with frame states in Graal that was causing a JCK test to fail.? The fix is from Tom Rodriguez.? This change has already been reviewed and merged into Graal.? A new unit test was added to detect the problem.? After this is reviewed I'll be requesting permission to push this into 15. dl From vladimir.kozlov at oracle.com Wed Jul 29 20:49:20 2020 From: vladimir.kozlov at oracle.com (Vladimir Kozlov) Date: Wed, 29 Jul 2020 13:49:20 -0700 Subject: [15] RFR(XS) 8248597: [Graal] api/java_security/SignatureSpi/DelegationTests.html fails with Method "javasoft.sqe.tests.api.java.security.SignatureSpi.JCKSignatureSpi.clear" doesn't exist. In-Reply-To: <688dfe7e-c8ad-3c37-d2f7-432609f45e2f@oracle.com> References: <688dfe7e-c8ad-3c37-d2f7-432609f45e2f@oracle.com> Message-ID: Looks good. Thanks, Vladimir On 7/29/20 12:48 PM, Dean Long wrote: > https://bugs.openjdk.java.net/browse/JDK-8248597 > http://cr.openjdk.java.net/~dlong/8248597/webrev/ > > This change fixes an issue with frame states in Graal that was causing a JCK test to fail.? The fix is from Tom > Rodriguez.? This change has already been reviewed and merged into Graal.? A new unit test was added to detect the > problem.? After this is reviewed I'll be requesting permission to push this into 15. > > dl From dean.long at oracle.com Wed Jul 29 20:52:43 2020 From: dean.long at oracle.com (Dean Long) Date: Wed, 29 Jul 2020 13:52:43 -0700 Subject: [15] RFR(XS) 8248597: [Graal] api/java_security/SignatureSpi/DelegationTests.html fails with Method "javasoft.sqe.tests.api.java.security.SignatureSpi.JCKSignatureSpi.clear" doesn't exist. In-Reply-To: References: <688dfe7e-c8ad-3c37-d2f7-432609f45e2f@oracle.com> Message-ID: Thanks Vladimir. dl On 7/29/20 1:49 PM, Vladimir Kozlov wrote: > Looks good. > > Thanks, > Vladimir > > On 7/29/20 12:48 PM, Dean Long wrote: >> https://bugs.openjdk.java.net/browse/JDK-8248597 >> http://cr.openjdk.java.net/~dlong/8248597/webrev/ >> >> This change fixes an issue with frame states in Graal that was >> causing a JCK test to fail.? The fix is from Tom Rodriguez. This >> change has already been reviewed and merged into Graal.? A new unit >> test was added to detect the problem.? After this is reviewed I'll be >> requesting permission to push this into 15. >> >> dl