On Fri, 21 Apr 2023 20:26:12 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, AARCH64, RISCV64, S390, PPC64`
> Note that `amd64` and
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, AARCH64, RISCV64, S390, PPC64`
> Note that `amd64` and
On Mon, 17 Apr 2023 20:59:06 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Wed, 19 Apr 2023 13:22:54 GMT, Martin Doerr wrote:
>> Roger Riggs has updated the pull request with a new target base due to a
>> merge or a rebase. The pull request now contains 17 commits:
>>
>> - Merge branch 'master' into 8304915-arch-enum
>> - ArchTest on Debian RISC-V 64 confirmed by
On Mon, 17 Apr 2023 20:59:06 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Mon, 17 Apr 2023 20:59:06 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Mon, 17 Apr 2023 20:59:06 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Mon, 17 Apr 2023 20:59:06 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, AARCH64, RISCV64, S390, PPC64`
> Note that `amd64` and
On Sat, 15 Apr 2023 17:17:13 GMT, Glavo wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Fixed isPPC64().
>> Consolidated switch cases in ArchTest.
>> Moved mapping of build TARGET_OS and TARGET_CPU to the build
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, AARCH64, RISCV64, S390, PPC64`
> Note that `amd64` and
On Wed, 12 Apr 2023 17:31:49 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Fri, 14 Apr 2023 17:08:10 GMT, Erik Joelsson wrote:
>> make/modules/java.base/gensrc/GensrcMisc.gmk line 72:
>>
>>> 70: endif
>>> 71:
>>> 72: $(eval $(call SetupTextFileProcessing, BUILD_PLATFORMPROPERTIES_JAVA, \
>>
>> @erikj79 Is there a better/good way to do these mappings, or select "lo
On Fri, 14 Apr 2023 14:31:31 GMT, Roger Riggs wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Fixed isPPC64().
>> Consolidated switch cases in ArchTest.
>> Moved mapping of build TARGET_OS and TARGET_CPU to the
On Wed, 12 Apr 2023 17:31:49 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Wed, 12 Apr 2023 17:31:49 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Wed, 12 Apr 2023 09:07:34 GMT, Martin Doerr wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct mapping and test of ppc64
>
> Works on PPC64 Big Endian, now. However, little Endian fails:
> STARTEDArchTes
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, AARCH64, RISCV64, S390, PPC64`
> Note that `amd64` and
On Wed, 12 Apr 2023 03:44:45 GMT, Amit Kumar wrote:
> Another remark: Old JDK on s390 used "os.arch = zArch_64", current one
> "os.arch = s390x". @offamitkumar: You probably want to take a look.
zArch_64 is not relevant/not used in the OpenJDK port to IBM System z. As noted
elsewhere in the PR
On Tue, 11 Apr 2023 21:09:43 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Tue, 11 Apr 2023 18:07:41 GMT, Martin Doerr wrote:
> Another remark: Old JDK on s390 used "os.arch = zArch_64", current one
> "os.arch = s390x". @offamitkumar: You probably want to take a look.
Martin, only concern was that I didn't have a good experience with `s390x`
string in
[past](http
On Tue, 11 Apr 2023 21:09:43 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, AARCH64, RISCV64, S390, PPC64`
> Note that `amd64` and
On Tue, 11 Apr 2023 18:35:56 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Tue, 11 Apr 2023 18:35:56 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Tue, 11 Apr 2023 19:33:28 GMT, Roger Riggs wrote:
>>> > Would be great if you could support "os.arch = ppc64" for AIX and legacy
>>> > linux, too.
>>>
>>> Changing os.arch is out of scope for this PR. The best way for that would
>>> someone supporting ppc to develop and propose a PR.
>>
>>
On Tue, 11 Apr 2023 18:37:11 GMT, Glavo wrote:
>>> Would be great if you could support "os.arch = ppc64" for AIX and legacy
>>> linux, too.
>>
>> Changing os.arch is out of scope for this PR.
>> The best way for that would someone supporting ppc to develop and propose a
>> PR.
>
>> > Would be
On Tue, 11 Apr 2023 16:58:13 GMT, Roger Riggs wrote:
> > Would be great if you could support "os.arch = ppc64" for AIX and legacy
> > linux, too.
>
> Changing os.arch is out of scope for this PR. The best way for that would
> someone supporting ppc to develop and propose a PR.
It seems that y
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, AARCH64, RISCV64, S390, PPC64`
> Note that `amd64` and
On Tue, 11 Apr 2023 18:02:14 GMT, Martin Doerr wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Modified test to check Architecture is64bits() and isLittleEndian()
>> against Unsafe respective values.
>> Relocate
On Tue, 11 Apr 2023 17:58:54 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
On Tue, 11 Apr 2023 17:58:54 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, AARCH64, RISCV64, S390, PPC64`
> Note that `amd64` and
On Tue, 11 Apr 2023 10:39:39 GMT, Martin Doerr wrote:
>> This should (probably) be correct:
>> Suggestion:
>>
>> case PPC64 -> !OperatingSystem.isAix() &&
>> Architecture.isLittleEndian();
>
> Looks correct, but makes the test pointless for any linux on PPC64.
Will change to compar
On Tue, 11 Apr 2023 11:41:24 GMT, Glavo wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Remove unused static and import of Stabile
>
> src/java.base/share/classes/jdk/internal/util/PlatformProps.java.template
> lin
On Sat, 8 Apr 2023 18:00:53 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Tue, 11 Apr 2023 10:26:02 GMT, ExE Boss wrote:
>> test/jdk/jdk/internal/util/ArchTest.java line 128:
>>
>>> 126: case RISCV64 -> true;
>>> 127: case S390 -> false;
>>> 128: case PPC64 -> true;
>>
>> This is not always true. The PPC64 architecture supports
On Tue, 11 Apr 2023 10:15:27 GMT, Martin Doerr wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Remove unused static and import of Stabile
>
> test/jdk/jdk/internal/util/ArchTest.java line 128:
>
>> 126:
On Sat, 8 Apr 2023 18:00:53 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, IA64, ARM, AARCH64, RISCV64, S390X,
> PPC64LE`
> Note
On Fri, 7 Apr 2023 21:13:03 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Fri, 7 Apr 2023 21:15:11 GMT, Roger Riggs wrote:
> Refactored the way that build time architecture values are mapped to the
> Architecture enum.
Thank you for this, Roger.
-
PR Review Comment: https://git.openjdk.org/jdk/pull/13357#discussion_r1161075537
On Fri, 7 Apr 2023 21:13:03 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Fri, 7 Apr 2023 06:03:11 GMT, Thomas Stuefe wrote:
>>> What did you use as the example that would not compile on the other
>>> architecture?
>>
>> https://github.com/openjdk/jdk/blob/52ca4a70fc3de9e285964f9545ea8cd54e2d9924/src/java.base/share/classes/jdk/internal/util/OperatingSystemProps.
On Fri, 7 Apr 2023 12:05:28 GMT, Lutz Schmidt wrote:
>> Okay, Lutz is the expert here. Sorry for the noise.
>
> Just to let my voice be heard directly after being cited several times: s390
> is used to designate the CPU architecture. The arch-specific files are stored
> in src/hotspot/cpu/s390
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, IA64, ARM, AARCH64, RISCV64, S390X,
> PPC64LE`
> Note
On Thu, 6 Apr 2023 22:44:52 GMT, Phil Race wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Remove unneeded qualified export from java.base to jdk.attach
>
> src/jdk.accessibility/windows/classes/com/sun/java/accessi
On Fri, 7 Apr 2023 11:21:33 GMT, Thomas Stuefe wrote:
>> Hello Thomas, that change was based on the review comment here
>> https://github.com/openjdk/jdk/pull/13357#discussion_r1159810942
>
> Okay, Lutz is the expert here. Sorry for the noise.
Just to let my voice be heard directly after being
On Fri, 7 Apr 2023 10:52:47 GMT, Jaikiran Pai wrote:
>> src/java.base/share/classes/jdk/internal/util/Architecture.java line 41:
>>
>>> 39: AARCH64(),
>>> 40: RISCV64(),
>>> 41: S390(),
>>
>> Why getting rid of the X in s390x? There has not been an s390 linux kernel
>> in almost te
On Fri, 7 Apr 2023 10:24:20 GMT, Thomas Stuefe wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Remove unneeded qualified export from java.base to jdk.attach
>
> src/java.base/share/classes/jdk/internal/util/Architec
On Thu, 6 Apr 2023 19:25:19 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Fri, 7 Apr 2023 09:28:18 GMT, Thomas Stuefe wrote:
> > > What I meant was: You define PPCLE. PPCLE specifies ppc, little endian.
> > > We also have PPC big-endian, it is used by AIX (and you can also run
> > > Linux with PPC big-endian). Why omit that?
> > > os.arch for AIX is "ppc64".
> >
On Fri, 7 Apr 2023 07:42:51 GMT, Alan Bateman wrote:
> > What I meant was: You define PPCLE. PPCLE specifies ppc, little endian. We
> > also have PPC big-endian, it is used by AIX (and you can also run Linux
> > with PPC big-endian). Why omit that?
> > os.arch for AIX is "ppc64".
>
> So I thin
On Thu, 6 Apr 2023 19:25:19 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Fri, 7 Apr 2023 06:33:08 GMT, Thomas Stuefe wrote:
> What I meant was: You define PPCLE. PPCLE specifies ppc, little endian. We
> also have PPC big-endian, it is used by AIX (and you can also run Linux with
> PPC big-endian). Why omit that?
>
> os.arch for AIX is "ppc64".
So I think you ar
On Thu, 6 Apr 2023 20:39:41 GMT, Roger Riggs wrote:
> > What about PPC (big endian)? Used on AIX?
>
> I'm not aware of any code (in OpenJDK) related to big/little endian that is
> derived from `os.arch`.
>
> > On Arm, it may be useful to know whether we built for thumb mode (We
> > recently h
On Thu, 6 Apr 2023 21:17:25 GMT, Glavo wrote:
>> The point of Architecture is reflect the architecture as supported by the
>> build and the runtime mutually and to reflect dependencies on the target
>> hardware.
>>
>> What did you use as the example that would not compile on the other
>> arc
On Thu, 6 Apr 2023 19:25:19 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Thu, 6 Apr 2023 20:56:42 GMT, Roger Riggs wrote:
> What did you use as the example that would not compile on the other
> architecture?
https://github.com/openjdk/jdk/blob/52ca4a70fc3de9e285964f9545ea8cd54e2d9924/src/java.base/share/classes/jdk/internal/util/OperatingSystemProps.java.templat
On Thu, 6 Apr 2023 20:42:26 GMT, Glavo wrote:
>> There is no benefit to preemptively defining a full set of architectures; we
>> only need those that are used in the OpenJDK runtime selection of options or
>> parameters.
>> The other and unknown cases can be handled in code using switch as `def
On Thu, 6 Apr 2023 20:38:02 GMT, Thomas Stuefe wrote:
> Another question, how does this work with Zero?
Zero is orthogonal to architecture; the interpreter is compiled for a specific
target architecture.
-
PR Comment: https://git.openjdk.org/jdk/pull/13357#issuecomment-1499606550
On Thu, 6 Apr 2023 20:24:27 GMT, Roger Riggs wrote:
>> src/java.base/share/classes/jdk/internal/util/Architecture.java line 100:
>>
>>> 98: */
>>> 99: public static Architecture current() {
>>> 100: return archValues[OperatingSystemProps.CURRENT_ARCH_ORDINAL];
>>
>> I think the
On Thu, 6 Apr 2023 20:28:29 GMT, Thomas Stuefe wrote:
> What about PPC (big endian)? Used on AIX?
I'm not aware of any code (in OpenJDK) related to big/little endian that is
derived from `os.arch`.
>
> On Arm, it may be useful to know whether we built for thumb mode (We recently
> had this p
On Thu, 6 Apr 2023 19:25:19 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Thu, 6 Apr 2023 19:25:19 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Thu, 6 Apr 2023 19:55:07 GMT, Glavo wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Remove unneeded qualified export from java.base to jdk.attach
>
> src/java.base/share/classes/jdk/internal/util/Architecture.jav
On Thu, 6 Apr 2023 19:25:19 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, IA64, ARM, AARCH64, RISCV64, S390X,
> PPC64LE`
> Note
On Thu, 6 Apr 2023 19:01:39 GMT, Roger Riggs wrote:
> Removed.
Good, that makes it simpler and it means the qualified export can be removed
too.
-
PR Review Comment: https://git.openjdk.org/jdk/pull/13357#discussion_r1160162760
On Thu, 6 Apr 2023 18:44:02 GMT, Alan Bateman wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Revert changes to Foreign API to avoid class with another PR
>> Rename S390X to S390, representing just the s390 archit
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, IA64, ARM, AARCH64, RISCV64, S390X,
> PPC64LE`
> Note
On Thu, 6 Apr 2023 17:17:35 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, IA64, ARM, AARCH64, RISCV64, S390X,
> PPC64LE`
> Note
On Thu, 6 Apr 2023 15:27:49 GMT, David M. Lloyd wrote:
>> src/java.base/share/classes/jdk/internal/util/Architecture.java line 47:
>>
>>> 45:
>>> 46: // Cache a copy of the array for lightweight indexing
>>> 47: private static final Architecture[] archValues =
>>> Architecture.values()
On Thu, 6 Apr 2023 08:00:43 GMT, ExE Boss wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct spelling of isAARCH64 in WIndows AttachProviderImpl
>
> src/java.base/share/classes/jdk/internal/util/Architecture.ja
On Thu, 6 Apr 2023 03:17:57 GMT, ExE Boss wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct spelling of isAARCH64 in WIndows AttachProviderImpl
>
> src/java.base/share/classes/jdk/internal/util/Architecture.ja
On Thu, 6 Apr 2023 08:05:14 GMT, ExE Boss wrote:
>> src/java.base/share/classes/jdk/internal/util/Architecture.java line 85:
>>
>>> 83: */
>>> 84: @ForceInline
>>> 85: public static boolean isRISCV64() {
>>
>> Are all the "isers" necessary to provide constant code folding or is the
On Thu, 6 Apr 2023 07:41:48 GMT, Per Minborg wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct spelling of isAARCH64 in WIndows AttachProviderImpl
>
> src/java.base/share/classes/jdk/internal/util/Architecture
On Wed, 5 Apr 2023 19:20:08 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Wed, 5 Apr 2023 23:39:00 GMT, Jorn Vernee wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct spelling of isAARCH64 in WIndows AttachProviderImpl
>
> test/jdk/java/foreign/TestUnsupportedLinker.java line 26:
On Thu, 6 Apr 2023 03:17:37 GMT, ExE Boss wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct spelling of isAARCH64 in WIndows AttachProviderImpl
>
> src/java.base/share/classes/jdk/internal/util/Architecture.ja
On Thu, 6 Apr 2023 08:11:45 GMT, Glavo wrote:
> os.arch is always the target architecture for JVM, even using 32-bit JVM on
> 64-bit system, its value remains x86.
Okay, I could have sworn we had to add the `ADDRESS_SIZE` check because of this
issue where a 32-bit VM running on x64 would erron
On Wed, 5 Apr 2023 23:46:33 GMT, Jorn Vernee wrote:
> Right, it is still possible to run a 32-bit VM build on a 64-bit platform.
> `os.arch` can be `amd64`, the runtime platform, in this case, but the VM
> variant can still be 32 bits.
`os.arch` is always the target architecture for JVM, even
On Thu, 6 Apr 2023 07:40:50 GMT, Per Minborg wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct spelling of isAARCH64 in WIndows AttachProviderImpl
>
> src/java.base/share/classes/jdk/internal/util/Architecture
On Wed, 5 Apr 2023 19:20:08 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Thu, 6 Apr 2023 07:36:36 GMT, Per Minborg wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct spelling of isAARCH64 in WIndows AttachProviderImpl
>
> src/java.base/share/classes/jdk/internal/util/Architecture
On Wed, 5 Apr 2023 19:20:08 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Wed, 5 Apr 2023 19:20:08 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Wed, 5 Apr 2023 21:36:57 GMT, Glavo wrote:
>> src/java.base/share/classes/jdk/internal/util/Architecture.java line 77:
>>
>>> 75: */
>>> 76: @ForceInline
>>> 77: public static boolean isARM() {
>>
>> It should define what’s the difference to aarch64 for example will aarch64
>>
On Wed, 5 Apr 2023 19:20:08 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Wed, 5 Apr 2023 21:18:43 GMT, Glavo wrote:
>> There should be no difference; I was hesitant to drop the ADDRESS_SIZE check
>> without knowing more about the foreign api dependencies. ADDRESS_SIZE is
>> computed (I think) from `UNSAFE.ADDRESS_SIZE * 8`.
>> But I can't think of how it can be
On Wed, 5 Apr 2023 19:20:08 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
On Wed, 5 Apr 2023 20:28:17 GMT, Bernd wrote:
> It should define what’s the difference to aarch64 for example will aarch64
> also be arm, but arm32 wont? (Or remove)
I think x86 and ARM are a bit confusing in this regard, as they can refer to
32-bit architectures in a narrow sense and 32-bit o
On Wed, 5 Apr 2023 20:40:32 GMT, Roger Riggs wrote:
>> src/java.base/share/classes/jdk/internal/foreign/CABI.java line 48:
>>
>>> 46: // might be running in a 32-bit VM on a 64-bit platform.
>>> 47: // addressSize will be correctly 32
>>> 48: if (Architecture.isX64() && A
On Wed, 5 Apr 2023 20:31:43 GMT, Bernd wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct spelling of isAARCH64 in WIndows AttachProviderImpl
>
> test/jdk/jdk/internal/util/ArchTest.java line 58:
>
>> 56:
On Wed, 5 Apr 2023 20:25:43 GMT, Bernd wrote:
>> Roger Riggs has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Correct spelling of isAARCH64 in WIndows AttachProviderImpl
>
> src/java.base/share/classes/jdk/internal/foreign/CABI.java line
On Wed, 5 Apr 2023 19:20:08 GMT, Roger Riggs wrote:
>> Define an internal jdk.internal.util.Architecture enumeration and static
>> methods to replace uses of the system property `os.arch`.
>> The enumeration values are defined to match those used in the build.
>> The initial values are: `X64, X8
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, IA64, ARM, AARCH64, RISCV64, S390X,
> PPC64LE`
> Note
> Define an internal jdk.internal.util.Architecture enumeration and static
> methods to replace uses of the system property `os.arch`.
> The enumeration values are defined to match those used in the build.
> The initial values are: `X64, X86, IA64, ARM, AARCH64, RISCV64, S390X,
> PPC64LE`
> Note
1 - 100 of 107 matches
Mail list logo
