From edward.nevill at gmail.com Wed Oct 14 13:50:48 2015 From: edward.nevill at gmail.com (Edward Nevill) Date: Wed, 14 Oct 2015 14:50:48 +0100 Subject: Project proposal: AArch32 port Message-ID: <1444830648.7802.34.camel@mylittlepony.linaroharston> Hi, I would like to propose the creation of a new project to port OpenJDK to ARM's aarch32 architecture. The aarch32 architecture is the 32 bit side of the ARMv8 architecture (the other side being aarch64). There is already a port for aarch64 which has been upstreamed into JDK9, thanks to the great work of Andrew Haley, Andrew Dinn and others however 32 bit support is missing. The aarch64 architecture is broadly compatible with ARMv7 with only a few minor differences most of which are not applicable to application level code. It is proposed that the aarch32 port will also support ARMv7. ARMv6 could also be supported, but this would depend on the level of community interest (ARMv6 is the architecture in the Raspberry Pi so there may be significant community interest and involvement if we were also to target ARMv6). We already have a complete template interpreter thanks to the excellent work of Joseph Joyce who presented his work at the recent Linaro Connect conference. A copy of his slides is available at http://openjdk.linaro.org/sfo15-openjdk-arm32.pdf I have created a JIRA issue and uploaded webrevs of this port based on JDK 9 to cr.openjdk.java.net. JIRA Issue: https://bugs.openjdk.java.net/browse/JDK-8139303 Webrevs: http://cr.openjdk.java.net/~enevill/8139303/webrev.hotspot/ http://cr.openjdk.java.net/~enevill/8139303/webrev.hs-comp/ http://cr.openjdk.java.net/~enevill/8139303/webrev.jdk/ The template interpreter is complete and runs a large variety of applications (eclipse, jedit, galaxians, javac, although sadly it cannot build itself as yet:-). It is intended that the port will be a complete port of OpenJDK to include Template Interpreter / C1 and possibly C2. Can I please request the support of the Porters Group as the Sponsoring Group with myself as project lead. Many thanks, Edward Nevill From dalibor.topic at oracle.com Thu Oct 22 09:01:06 2015 From: dalibor.topic at oracle.com (dalibor topic) Date: Thu, 22 Oct 2015 11:01:06 +0200 Subject: Project proposal: AArch32 port In-Reply-To: <1444830648.7802.34.camel@mylittlepony.linaroharston> References: <1444830648.7802.34.camel@mylittlepony.linaroharston> Message-ID: <5628A5D2.40909@oracle.com> Hi Edward, As the OpenJDK Porters Group Lead, I agree that the Porters Group would be a good fit for a sponsoring Group for this Project. Can you elaborate a bit on the initial set of Committers/Reviewers you would see participating in the Project - would it be just you & Joseph from Linaro (or somewhere else?) for the start? cheers, dalibor topic On 14.10.2015 15:50, Edward Nevill wrote: > Hi, > > I would like to propose the creation of a new project to port OpenJDK to > ARM's aarch32 architecture. The aarch32 architecture is the 32 bit side > of the ARMv8 architecture (the other side being aarch64). > > There is already a port for aarch64 which has been upstreamed into JDK9, > thanks to the great work of Andrew Haley, Andrew Dinn and others however > 32 bit support is missing. > > The aarch64 architecture is broadly compatible with ARMv7 with only a > few minor differences most of which are not applicable to application > level code. It is proposed that the aarch32 port will also support > ARMv7. > > ARMv6 could also be supported, but this would depend on the level of > community interest (ARMv6 is the architecture in the Raspberry Pi so > there may be significant community interest and involvement if we were > also to target ARMv6). > > We already have a complete template interpreter thanks to the excellent > work of Joseph Joyce who presented his work at the recent Linaro Connect > conference. A copy of his slides is available at > > http://openjdk.linaro.org/sfo15-openjdk-arm32.pdf > > I have created a JIRA issue and uploaded webrevs of this port based on > JDK 9 to cr.openjdk.java.net. > > JIRA Issue: https://bugs.openjdk.java.net/browse/JDK-8139303 > > Webrevs: > > http://cr.openjdk.java.net/~enevill/8139303/webrev.hotspot/ > http://cr.openjdk.java.net/~enevill/8139303/webrev.hs-comp/ > http://cr.openjdk.java.net/~enevill/8139303/webrev.jdk/ > > The template interpreter is complete and runs a large variety of > applications (eclipse, jedit, galaxians, javac, although sadly it cannot > build itself as yet:-). > > It is intended that the port will be a complete port of OpenJDK to > include Template Interpreter / C1 and possibly C2. > > Can I please request the support of the Porters Group as the Sponsoring > Group with myself as project lead. > > Many thanks, > Edward Nevill > > -- Dalibor Topic | Principal Product Manager Phone: +494089091214 | Mobile: +491737185961 ORACLE Deutschland B.V. & Co. KG | K?hneh?fe 5 | 22761 Hamburg ORACLE Deutschland B.V. & Co. KG Hauptverwaltung: Riesstr. 25, D-80992 M?nchen Registergericht: Amtsgericht M?nchen, HRA 95603 Komplement?rin: ORACLE Deutschland Verwaltung B.V. Hertogswetering 163/167, 3543 AS Utrecht, Niederlande Handelsregister der Handelskammer Midden-Niederlande, Nr. 30143697 Gesch?ftsf?hrer: Alexander van der Ven, Astrid Kepper, Val Maher Oracle is committed to developing practices and products that help protect the environment From edward.nevill at gmail.com Thu Oct 22 12:46:54 2015 From: edward.nevill at gmail.com (Edward Nevill) Date: Thu, 22 Oct 2015 13:46:54 +0100 Subject: Project proposal: AArch32 port In-Reply-To: <5628A5D2.40909@oracle.com> References: <1444830648.7802.34.camel@mylittlepony.linaroharston> <5628A5D2.40909@oracle.com> Message-ID: <1445518014.29998.4.camel@mylittlepony.linaroharston> On Thu, 2015-10-22 at 11:01 +0200, dalibor topic wrote: > Hi Edward, > > As the OpenJDK Porters Group Lead, I agree that the Porters Group would > be a good fit for a sponsoring Group for this Project. > > Can you elaborate a bit on the initial set of Committers/Reviewers you > would see participating in the Project - would it be just you & Joseph > from Linaro (or somewhere else?) for the start? > > cheers, > dalibor topic Hi Dalibor, The initial set of Committers I propose are Andrew Dinn ('adinn'), Andrew Haley ('aph') and myself to start with. I do not propose having any formal review on this project so there is no need for formal Reviewers. All the best, Ed. From dalibor.topic at oracle.com Fri Oct 23 12:53:15 2015 From: dalibor.topic at oracle.com (dalibor topic) Date: Fri, 23 Oct 2015 14:53:15 +0200 Subject: Project proposal: AArch32 port In-Reply-To: <1445518014.29998.4.camel@mylittlepony.linaroharston> References: <1444830648.7802.34.camel@mylittlepony.linaroharston> <5628A5D2.40909@oracle.com> <1445518014.29998.4.camel@mylittlepony.linaroharston> Message-ID: <562A2DBB.9000404@oracle.com> On 22.10.2015 14:46, Edward Nevill wrote: > Hi Dalibor, > > The initial set of Committers I propose are > > Andrew Dinn ('adinn'), Andrew Haley ('aph') and myself to start with. Thanks, Edward. That's a very accomplished group! Since you're sending this e-mail from a non-Linaro account, while CC:ing a number of people at Linaro, I am wondering which OCAs would you and other Committers plan to make contributions under. Can you elaborate, please? > I do not propose having any formal review on this project so there is no > need for formal Reviewers. That sounds fine to me. I noticed that in your initial set of Committers Joseph Joyce is missing, while at the same time you seem to suggest using his externally developed code to jump-start this Project. That seems a bit unusual - typically the original author of a new feature would want to continue to be actively involved during that feature's further development in an OpenJDK Project, so I'm curious for the rationale behind doing things differently in this case. Can you and Joseph explain if you intend to use the template interpreter code you referenced in the new Project, and if so, how (i.e. under which OCA) it would be contributed to it? I'm asking since I don't see Joseph Joyce mentioned at [0], and slides you referenced unfortunately provide little information on that. Last but not least, is there anyone else who has contributed to the code beside yourself, Joseph? Is there any code (or other materials) taken from other third party or open source projects in it? cheers, dalibor topic [0] http://www.oracle.com/technetwork/community/oca-486395.html -- Dalibor Topic | Principal Product Manager Phone: +494089091214 | Mobile: +491737185961 ORACLE Deutschland B.V. & Co. KG | K?hneh?fe 5 | 22761 Hamburg ORACLE Deutschland B.V. & Co. KG Hauptverwaltung: Riesstr. 25, D-80992 M?nchen Registergericht: Amtsgericht M?nchen, HRA 95603 Komplement?rin: ORACLE Deutschland Verwaltung B.V. Hertogswetering 163/167, 3543 AS Utrecht, Niederlande Handelsregister der Handelskammer Midden-Niederlande, Nr. 30143697 Gesch?ftsf?hrer: Alexander van der Ven, Astrid Kepper, Val Maher Oracle is committed to developing practices and products that help protect the environment From edward.nevill at gmail.com Fri Oct 23 13:41:42 2015 From: edward.nevill at gmail.com (Edward Nevill) Date: Fri, 23 Oct 2015 14:41:42 +0100 Subject: Project proposal: AArch32 port In-Reply-To: <562A2DBB.9000404@oracle.com> References: <1444830648.7802.34.camel@mylittlepony.linaroharston> <5628A5D2.40909@oracle.com> <1445518014.29998.4.camel@mylittlepony.linaroharston> <562A2DBB.9000404@oracle.com> Message-ID: <1445607702.28722.19.camel@mint> On Fri, 2015-10-23 at 14:53 +0200, dalibor topic wrote: > On 22.10.2015 14:46, Edward Nevill wrote: > > Hi Dalibor, > > > > Andrew Dinn ('adinn'), Andrew Haley ('aph') and myself to start with. > > Thanks, Edward. That's a very accomplished group! Thanks:) > Since you're sending this e-mail from a non-Linaro account, while CC:ing > a number of people at Linaro, I am wondering which OCAs would you and > other Committers plan to make contributions under. Can you elaborate, > please? Andrew Dinn and Andrew Haley will make contributions under the RedHat OCA. I will make contributions under the Linaro OCA. I currently use my gmail address for contributing to JDK 9. > > I noticed that in your initial set of Committers Joseph Joyce is > missing, while at the same time you seem to suggest using his externally > developed code to jump-start this Project. Yes. Joseph worked as an intern at Linaro for 14 weeks over this summer and did a great job porting the template interpreter in that time. Unfortunately Joseph is no longer working for us. However Linaro owns the copyright of all the work he did while he was working for us and we can contribute it under the Linaro OCA. > > That seems a bit unusual - typically the original author of a new > feature would want to continue to be actively involved during that > feature's further development in an OpenJDK Project, so I'm curious for > the rationale behind doing things differently in this case. Joseph has indeed expressed an interest in continuing to contribute on a voluntary basis. However, to the best of my knowledge, Joseph has not signed the OCA in a personal capacity, therefore I did not include him in the list of initial committers. If Joseph does sign the OCA in the future then we would be happy to have him as a committer. > > Can you and Joseph explain if you intend to use the template interpreter > code you referenced in the new Project, and if so, how (i.e. under which > OCA) it would be contributed to it? I'm asking since I don't see Joseph > Joyce mentioned at [0], and slides you referenced unfortunately provide > little information on that. Yes, under the Linaro OCA. > > Last but not least, is there anyone else who has contributed to the code > beside yourself, Joseph? Is there any code (or other materials) taken > from other third party or open source projects in it? To the best of my knowledge no, but I didn't write it. Joseph: Would you be able to confirm this? Many thanks for your help Dalibor, All the best, Ed. From david.holmes at oracle.com Thu Oct 29 02:22:31 2015 From: david.holmes at oracle.com (David Holmes) Date: Thu, 29 Oct 2015 12:22:31 +1000 Subject: RFR: 8129526: Solaris: clean up another remnant of interruptible I/O Message-ID: <563182E7.4040701@oracle.com> bug: https://bugs.openjdk.java.net/browse/JDK-8129526 webrev: http://cr.openjdk.java.net/~dholmes/8129526/webrev/ Nothing but code deletion here :) Interruptible I/O was finally removed some time ago but we found a remnant left over in the Solaris code in the use of SIGinterrupt. But without interruptible I/O we don't need SIGinterrupt, nor its cross-platform alter-ego INTERRUPT_SIGNAL. Some code related to INTERRUPT_SIGNAL was copied from the Solaris code to the Linux code (even though interruptible I/O never existed on linux) and it was later copied from the Linux code to BSD and AIX. All of that has now been cleaned up too - hence the cc to porters-dev to make sure the AIX folk see this. Testing: JPRT Thanks, David From david.holmes at oracle.com Thu Oct 29 10:34:52 2015 From: david.holmes at oracle.com (David Holmes) Date: Thu, 29 Oct 2015 20:34:52 +1000 Subject: RFR: 8129526: Solaris: clean up another remnant of interruptible I/O In-Reply-To: <5631ED90.6030503@oracle.com> References: <563182E7.4040701@oracle.com> <5631ED90.6030503@oracle.com> Message-ID: <5631F64C.407@oracle.com> Thanks Fred! David On 29/10/2015 7:57 PM, Frederic Parain wrote: > David, > > Thanks for continuing the interruptible I/O clean up effort. > Changes look good to me. > > Fred > > On 10/29/2015 03:22 AM, David Holmes wrote: >> bug: https://bugs.openjdk.java.net/browse/JDK-8129526 >> webrev: http://cr.openjdk.java.net/~dholmes/8129526/webrev/ >> >> Nothing but code deletion here :) >> >> Interruptible I/O was finally removed some time ago but we found a >> remnant left over in the Solaris code in the use of SIGinterrupt. But >> without interruptible I/O we don't need SIGinterrupt, nor its >> cross-platform alter-ego INTERRUPT_SIGNAL. Some code related to >> INTERRUPT_SIGNAL was copied from the Solaris code to the Linux code >> (even though interruptible I/O never existed on linux) and it was later >> copied from the Linux code to BSD and AIX. All of that has now been >> cleaned up too - hence the cc to porters-dev to make sure the AIX folk >> see this. >> >> Testing: JPRT >> >> Thanks, >> David > From daniel.daugherty at oracle.com Thu Oct 29 13:50:07 2015 From: daniel.daugherty at oracle.com (Daniel D. Daugherty) Date: Thu, 29 Oct 2015 07:50:07 -0600 Subject: RFR: 8129526: Solaris: clean up another remnant of interruptible I/O In-Reply-To: <563182E7.4040701@oracle.com> References: <563182E7.4040701@oracle.com> Message-ID: <5632240F.7020108@oracle.com> On 10/28/15 8:22 PM, David Holmes wrote: > bug: https://bugs.openjdk.java.net/browse/JDK-8129526 > webrev: http://cr.openjdk.java.net/~dholmes/8129526/webrev/ src/os/aix/vm/jvm_aix.cpp No comments. src/os/aix/vm/jvm_aix.h No comments. src/os/aix/vm/os_aix.cpp No comments. src/os/bsd/vm/jvm_bsd.cpp No comments. src/os/bsd/vm/jvm_bsd.h No comments. src/os/bsd/vm/os_bsd.cpp No comments. src/os/linux/vm/jvm_linux.cpp No comments. src/os/linux/vm/jvm_linux.h No comments. src/os/linux/vm/os_linux.cpp No comments. src/os/solaris/vm/jvm_solaris.cpp No comments. src/os/solaris/vm/jvm_solaris.h No comments. src/os/solaris/vm/os_solaris.cpp No comments. src/os/solaris/vm/os_solaris.hpp No comments. src/os_cpu/solaris_sparc/vm/os_solaris_sparc.cpp No comments. src/os_cpu/solaris_x86/vm/os_solaris_x86.cpp No comments. Nice cleanup. Thumbs up. Dan > > Nothing but code deletion here :) > > Interruptible I/O was finally removed some time ago but we found a > remnant left over in the Solaris code in the use of SIGinterrupt. But > without interruptible I/O we don't need SIGinterrupt, nor its > cross-platform alter-ego INTERRUPT_SIGNAL. Some code related to > INTERRUPT_SIGNAL was copied from the Solaris code to the Linux code > (even though interruptible I/O never existed on linux) and it was > later copied from the Linux code to BSD and AIX. All of that has now > been cleaned up too - hence the cc to porters-dev to make sure the AIX > folk see this. > > Testing: JPRT > > Thanks, > David From david.holmes at oracle.com Thu Oct 29 22:20:43 2015 From: david.holmes at oracle.com (David Holmes) Date: Fri, 30 Oct 2015 08:20:43 +1000 Subject: RFR: 8129526: Solaris: clean up another remnant of interruptible I/O In-Reply-To: <5632240F.7020108@oracle.com> References: <563182E7.4040701@oracle.com> <5632240F.7020108@oracle.com> Message-ID: <56329BBB.5080400@oracle.com> Thanks Dan! David On 29/10/2015 11:50 PM, Daniel D. Daugherty wrote: > On 10/28/15 8:22 PM, David Holmes wrote: >> bug: https://bugs.openjdk.java.net/browse/JDK-8129526 >> webrev: http://cr.openjdk.java.net/~dholmes/8129526/webrev/ > > src/os/aix/vm/jvm_aix.cpp > No comments. > > src/os/aix/vm/jvm_aix.h > No comments. > > src/os/aix/vm/os_aix.cpp > No comments. > > src/os/bsd/vm/jvm_bsd.cpp > No comments. > > src/os/bsd/vm/jvm_bsd.h > No comments. > > src/os/bsd/vm/os_bsd.cpp > No comments. > > src/os/linux/vm/jvm_linux.cpp > No comments. > > src/os/linux/vm/jvm_linux.h > No comments. > > src/os/linux/vm/os_linux.cpp > No comments. > > src/os/solaris/vm/jvm_solaris.cpp > No comments. > > src/os/solaris/vm/jvm_solaris.h > No comments. > > src/os/solaris/vm/os_solaris.cpp > No comments. > > src/os/solaris/vm/os_solaris.hpp > No comments. > > src/os_cpu/solaris_sparc/vm/os_solaris_sparc.cpp > No comments. > > src/os_cpu/solaris_x86/vm/os_solaris_x86.cpp > No comments. > > Nice cleanup. Thumbs up. > > Dan > > >> >> Nothing but code deletion here :) >> >> Interruptible I/O was finally removed some time ago but we found a >> remnant left over in the Solaris code in the use of SIGinterrupt. But >> without interruptible I/O we don't need SIGinterrupt, nor its >> cross-platform alter-ego INTERRUPT_SIGNAL. Some code related to >> INTERRUPT_SIGNAL was copied from the Solaris code to the Linux code >> (even though interruptible I/O never existed on linux) and it was >> later copied from the Linux code to BSD and AIX. All of that has now >> been cleaned up too - hence the cc to porters-dev to make sure the AIX >> folk see this. >> >> Testing: JPRT >> >> Thanks, >> David > From volker.simonis at gmail.com Fri Oct 30 10:26:03 2015 From: volker.simonis at gmail.com (Volker Simonis) Date: Fri, 30 Oct 2015 11:26:03 +0100 Subject: RFR: 8129526: Solaris: clean up another remnant of interruptible I/O In-Reply-To: <563182E7.4040701@oracle.com> References: <563182E7.4040701@oracle.com> Message-ID: Hi David, I'm a little bit late to the game but I just wanted to mention that your changes work perfectly on AIX as well. Thanks, Volker On Thu, Oct 29, 2015 at 3:22 AM, David Holmes wrote: > bug: https://bugs.openjdk.java.net/browse/JDK-8129526 > webrev: http://cr.openjdk.java.net/~dholmes/8129526/webrev/ > > Nothing but code deletion here :) > > Interruptible I/O was finally removed some time ago but we found a remnant > left over in the Solaris code in the use of SIGinterrupt. But without > interruptible I/O we don't need SIGinterrupt, nor its cross-platform > alter-ego INTERRUPT_SIGNAL. Some code related to INTERRUPT_SIGNAL was copied > from the Solaris code to the Linux code (even though interruptible I/O never > existed on linux) and it was later copied from the Linux code to BSD and > AIX. All of that has now been cleaned up too - hence the cc to porters-dev > to make sure the AIX folk see this. > > Testing: JPRT > > Thanks, > David From david.holmes at oracle.com Fri Oct 30 12:14:11 2015 From: david.holmes at oracle.com (David Holmes) Date: Fri, 30 Oct 2015 22:14:11 +1000 Subject: RFR: 8129526: Solaris: clean up another remnant of interruptible I/O In-Reply-To: References: <563182E7.4040701@oracle.com> Message-ID: <56335F13.3070002@oracle.com> Thanks Volker! I was confident this would not be an issue. David On 30/10/2015 8:26 PM, Volker Simonis wrote: > Hi David, > > I'm a little bit late to the game but I just wanted to mention that > your changes work perfectly on AIX as well. > > Thanks, > Volker > > > On Thu, Oct 29, 2015 at 3:22 AM, David Holmes wrote: >> bug: https://bugs.openjdk.java.net/browse/JDK-8129526 >> webrev: http://cr.openjdk.java.net/~dholmes/8129526/webrev/ >> >> Nothing but code deletion here :) >> >> Interruptible I/O was finally removed some time ago but we found a remnant >> left over in the Solaris code in the use of SIGinterrupt. But without >> interruptible I/O we don't need SIGinterrupt, nor its cross-platform >> alter-ego INTERRUPT_SIGNAL. Some code related to INTERRUPT_SIGNAL was copied >> from the Solaris code to the Linux code (even though interruptible I/O never >> existed on linux) and it was later copied from the Linux code to BSD and >> AIX. All of that has now been cleaned up too - hence the cc to porters-dev >> to make sure the AIX folk see this. >> >> Testing: JPRT >> >> Thanks, >> David