Hi Volker,
thanks for the hint. GNU grep works, so I close this as wontfix.
..Thomas
On Wed, Jan 11, 2017 at 6:51 PM, Volker Simonis
wrote:
> Hi Thomas,
>
> why not simply using GNU grep? We already insist for so many other tools
> on the GNU version that this wouldn't be a big thing.
>
> And
In JDK-8066474, all references to OPENJDK_TARGET_CPU_JLI_CFLAGS except
one was removed.
Bug: https://bugs.openjdk.java.net/browse/JDK-8172702
Patch inline:
diff --git a/common/autoconf/buildjdk-spec.gmk.in
b/common/autoconf/buildjdk-spec.gmk.in
--- a/common/autoconf/buildjdk-spec.gmk.in
+++ b/
Looks good.
/Erik
On 2017-01-12 10:07, Magnus Ihse Bursie wrote:
In JDK-8066474, all references to OPENJDK_TARGET_CPU_JLI_CFLAGS except
one was removed.
Bug: https://bugs.openjdk.java.net/browse/JDK-8172702
Patch inline:
diff --git a/common/autoconf/buildjdk-spec.gmk.in
b/common/autoconf/bu
On 2017-01-12 09:56, Thomas Stüfe wrote:
Hi Volker,
thanks for the hint. GNU grep works, so I close this as wontfix.
You might want to add a test in configure that it is picking up a
correct grep.
/Magnus
..Thomas
On Wed, Jan 11, 2017 at 6:51 PM, Volker Simonis
wrote:
Hi Thomas,
why n
jtreg 4.2 b05 was recently promoted with some important fixes. Please review
the change below to upgrade jdk9-dev to the new version. I have run jdk-tier1,
jdk-tier2 and jdk-tier3 on linux and os x with this change and did not see any
new test failures.
Thanks,
/Staffan
diff --git a/common/co
> On 12 Jan 2017, at 10:43, Staffan Larsen wrote:
>
> jtreg 4.2 b05 was recently promoted with some important fixes. Please review
> the change below to upgrade jdk9-dev to the new version. I have run
> jdk-tier1, jdk-tier2 and jdk-tier3 on linux and os x with this change and did
> not see an
Looks good.
/Erik
On 2017-01-12 11:43, Staffan Larsen wrote:
jtreg 4.2 b05 was recently promoted with some important fixes. Please review
the change below to upgrade jdk9-dev to the new version. I have run jdk-tier1,
jdk-tier2 and jdk-tier3 on linux and os x with this change and did not see
On 2017-01-11 17:40, Erik Joelsson wrote:
Hello,
This patch fixes bug in the bundle logic, specifically the part that
tries to work around files with spaces in them. In one of my recent
changes in Bundles.gmk, I managed to break this completely.
Bug: https://bugs.openjdk.java.net/browse/JDK-
On 12/01/2017 10:43, Staffan Larsen wrote:
jtreg 4.2 b05 was recently promoted with some important fixes. Please review
the change below to upgrade jdk9-dev to the new version. I have run jdk-tier1,
jdk-tier2 and jdk-tier3 on linux and os x with this change and did not see any
new test failu
> On 12 Jan 2017, at 12:51, Alan Bateman wrote:
>
>
>
> On 12/01/2017 10:43, Staffan Larsen wrote:
>> jtreg 4.2 b05 was recently promoted with some important fixes. Please review
>> the change below to upgrade jdk9-dev to the new version. I have run
>> jdk-tier1, jdk-tier2 and jdk-tier3 on l
We do not expand the value of @BUILD_HOTSPOT@ and @THEPWDCMD@ anymore in
the generated spec.gmk.
No-one seems to be suffering from this (and I've double-checked with
a grep) so we should just remove the declarations.
Bug: https://bugs.openjdk.java.net/browse/JDK-8172714
Patch inline:
diff --g
Dear all,
please take a look at this fix. This should prevent configure from using a
grep which is not able to handle pattern lists with empty patterns.
Bug:
https://bugs.openjdk.java.net/browse/JDK-8172712
webrevs:
http://cr.openjdk.java.net/~stuefe/webrevs/8172712-configure-should-check-that-g
On Thu, Jan 12, 2017 at 10:32 AM, Magnus Ihse Bursie <
magnus.ihse.bur...@oracle.com> wrote:
> On 2017-01-12 09:56, Thomas Stüfe wrote:
>
> Hi Volker,
>
> thanks for the hint. GNU grep works, so I close this as wontfix.
>
> You might want to add a test in configure that it is picking up a correct
On 12.01.2017 11:43, Staffan Larsen wrote:
> jtreg 4.2 b05 was recently promoted with some important fixes. Please review
> the change below to upgrade jdk9-dev to the new version. I have run
> jdk-tier1, jdk-tier2 and jdk-tier3 on linux and os x with this change and did
> not see any new test f
Looks good to me.
Since the closed generated-configure needs to be regenerated, I'll
sponsor and push this patch.
/Magnus
On 2017-01-12 14:04, Thomas Stüfe wrote:
Dear all,
please take a look at this fix. This should prevent configure from using a
grep which is not able to handle pattern li
/common/autoconf/flags.m4 contains duplicated code to add
-D__solaris__ to the JDK compiler flags when the target os is solaris.
Probably a result of a merge error during development of the new hotspot
build system.
Bug: https://bugs.openjdk.java.net/browse/JDK-8162750
Patch inline:
diff --gi
In JDK-8170284, the hotspot build logic was better integrated with the
top make level. However, the calls to make there was not provided with
MAKE_ARGS.
Bug: https://bugs.openjdk.java.net/browse/JDK-8170863
WebRev:
http://cr.openjdk.java.net/~ihse/JDK-8170863-always-pass-MAKE_ARGS/webrev.01
Thank you Magnus!
Patch is tested on AIX, Solaris, Linux and Windows, no issues.
Kind Regards, Thomas
On Thu, Jan 12, 2017 at 2:36 PM, Magnus Ihse Bursie <
magnus.ihse.bur...@oracle.com> wrote:
> Looks good to me.
>
> Since the closed generated-configure needs to be regenerated, I'll sponsor
>
Looks good.
/Erik
On 2017-01-12 13:33, Magnus Ihse Bursie wrote:
We do not expand the value of @BUILD_HOTSPOT@ and @THEPWDCMD@ anymore
in the generated spec.gmk.
No-one seems to be suffering from this (and I've double-checked
with a grep) so we should just remove the declarations.
Bug: ht
Looks good.
/Erik
On 2017-01-12 14:42, Magnus Ihse Bursie wrote:
/common/autoconf/flags.m4 contains duplicated code to add
-D__solaris__ to the JDK compiler flags when the target os is solaris.
Probably a result of a merge error during development of the new
hotspot build system.
Bug: htt
Looks good. Are you sure MAKE_ARGS should be added to bootcycle-images?
I assume you have tested this.
/Erik
On 2017-01-12 14:47, Magnus Ihse Bursie wrote:
In JDK-8170284, the hotspot build logic was better integrated with the
top make level. However, the calls to make there was not provided
Hi,
the tag in http://hg.openjdk.java.net/code-tools/jtreg still says jtreg4.2-b04.
See: http://hg.openjdk.java.net/code-tools/jtreg/file/4b0cd55e7741/.hgtags
Wouldn't that need some adoption first?
Best regards
Christoph
> -Original Message-
> From: jdk9-dev [mailto:jdk9-dev-boun...@
b05 is not yet officially announced or tagged, although I'll get to that
today.
I'm not aware that any tests require any new functionality (yet), but as
Staffan mentioned, there are fixes in the handling for timeout handling
that are important for some systems using jtreg.
-- Jon
On 1/12/17
Magnus:
Looks good to me as well.
Tim
On 01/12/17 06:08, Erik Joelsson wrote:
Looks good.
/Erik
On 2017-01-12 13:33, Magnus Ihse Bursie wrote:
We do not expand the value of @BUILD_HOTSPOT@ and @THEPWDCMD@ anymore
in the generated spec.gmk.
No-one seems to be suffering from this (and I've
I also ran into this problem a couple of weeks ago with some other
"misbehaving” grep. Thanks for adding this check!
Cheers,
Mikael
> On Jan 12, 2017, at 5:04 AM, Thomas Stüfe wrote:
>
> Dear all,
>
> please take a look at this fix. This should prevent configure from using a
> grep which is
Thank you!
On Thu, 12 Jan 2017 at 18:30, Mikael Vidstedt
wrote:
>
> I also ran into this problem a couple of weeks ago with some other
> "misbehaving” grep. Thanks for adding this check!
>
> Cheers,
> Mikael
>
>
> On Jan 12, 2017, at 5:04 AM, Thomas Stüfe wrote:
>
> Dear all,
>
> please take a
Hi.
I'm trying to build OpenJDK 8 on macOS 10.12, and having some trouble
getting it to accept the path to Xcode.
$ ./configure "--with-xcode-path=$HOME/DevEnv/Applications/Xcode4/Xcode.app"
checking Determining if we need to set DEVELOPER_DIR... yes
(/Users/tester/DevEnv/Applications/Xc
27 matches
Mail list logo