On Fri, Sep 12, 2014 at 2:03 PM, Joe Slater wrote:
> -+#ifdef __arm__
> ++#if defined(__arm__) || defined(__aarch64__)
> +/*
This code defined memory barriers for 32bit, Are you sure it should be
same for 64bit as well ?
how was this tested ?
--
___
O
From: Roy Li
DEPENDS does not support the override, fix it by checking the DISTRO_FEATURE
with ptest
Signed-off-by: Roy Li
---
meta/recipes-devtools/flex/flex.inc |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/recipes-devtools/flex/flex.inc
b/meta/recipes-devtools
From: Wenzong Fan
There's not bash specific syntax in the xtests scripts:
$ cd Linux-PAM-1.1.6/xtests
# replace /bin/bash to /bin/sh and check the bashisms:
$ checkbashisms *.sh
No output
So the runtime dependency to bash could be removed.
Signed-off-by: Wenzong Fan
---
.../libpam-xt
From: Wenzong Fan
There's not bash specific syntax in the xtests scripts:
$ cd Linux-PAM-1.1.6/xtests
# replace /bin/bash to /bin/sh and check the bashisms:
$ checkbashisms *.sh
No output
So the runtime dependency to bash could be removed.
The following changes since commit 429802fe669
From: "Roy.Li"
Signed-off-by: Roy.Li
---
.../ltp/ltp/add-knob-for-numa.patch| 39 ++
meta/recipes-extended/ltp/ltp_20140422.bb | 3 ++
2 files changed, 42 insertions(+)
create mode 100644 meta/recipes-extended/ltp/ltp/add-knob-for-numa.patch
diff
We have libxml2-python for native and nativesdk, but don't have it for
target, and can't find the reason from the git log, libxml2-python is
widely used, after looked into it's configure.in, we can add it
by PACKAGECONFIG.
The previous --with-python=${STAGING_BINDIR}/python is incorrect, it
acted
Changes of V2:
* Use PACKAGECONFIG as Ross suggested.
// Robert
The following changes since commit 10df0718d6a626d99beb68cde8d914ee0820d7eb:
classes/populate_sdk_base: enable adding custom commands to SDK install
script (2014-09-11 17:44:40 +0100)
are available in the git repository at:
g
> -Original Message-
> From: Burton, Ross [mailto:ross.bur...@intel.com]
> Sent: Monday, September 15, 2014 10:23 PM
> To: Huang, Jie (Jackie)
> Cc: OE-core
> Subject: Re: [OE-core] [PATCH] libgcrypt: Fix ARM assembly when building
> __PIC__
>
> On 15 September 2014 11:16, wrote:
> >
Signed-off-by: Nobuhiro Iwamatsu
---
V2: Update commit message.
.../packagegroup-core-standalone-sdk-target.bb |1 +
1 file changed, 1 insertion(+)
diff --git
a/meta/recipes-core/packagegroups/packagegroup-core-standalone-sdk-target.bb
b/meta/recipes-core/packagegroups/packagegroup-
On Mon, Sep 15, 2014 at 4:54 PM, Nobuhiro Iwamatsu
wrote:
> When a user takes coverage in gcc of SDK, it becomes the link error in SDK
> because there is not libgcov.
Looks good. The commit message could be improved so to simply say ..
"Add libgcov to on device SDK"
>
>
> ld: cannot find -l
When a user takes coverage in gcc of SDK, it becomes the link error in SDK
because there is not libgcov.
ld: cannot find -lgcov
collect2: error: ld returned 1 exit status
This adds libgcov-dev to target sysroot.
Signed-off-by: Nobuhiro Iwamatsu
---
.../packagegroup-core-standalone-sd
Add aarch64 architecture to liburcu and lttng-ust.
Internally it is treated the same as "arm".
Signed-off-by: Joe Slater
---
.../lttng/lttng-ust/add-aarch64.patch | 19 +++
meta/recipes-kernel/lttng/lttng-ust_2.5.0.bb |1 +
meta/recipes-support/liburcu/fi
On 15 September 2014 21:20, Dan McGregor wrote:
>> git-am is not happy with this patch, claiming that it's malformed.
>> Can you try re-sending or pushing to a branch somewhere?
>
> Sure: http://bitbucket.org/dankm/openembedded-core.git, branch "sanitizers".
>
> It has unrelated changes in it, but
On 15 September 2014 09:00, Burton, Ross wrote:
> On 15 September 2014 05:00, Dan McGregor wrote:
>> From: Dan McGregor
>>
>> AddressSanitizer is a fast memory error detector.
>> ThreadSanitizer detects data races.
>> UBSanitizer detectes undefined behaviour.
>>
>> All consist of compiler instru
On 15 September 2014 18:20, Slater, Joseph W (Wind River)
wrote:
>> On 12 September 2014 22:03, Joe Slater wrote:
>> > +++ b/meta/recipes-kernel/sysprof/files/rmb-arm.patch
>>
>> This patch is missing a signed-off-by and upstream-status.
>
> I am not creating this patch, just modifying it. The e
> -Original Message-
> From: Burton, Ross [mailto:ross.bur...@intel.com]
> Sent: Sunday, September 14, 2014 1:44 PM
> To: Slater, Joseph
> Cc: OE-core
> Subject: Re: [OE-core] [oe-core][PATCH 1/1] sysprof: add aarch64 support
>
> On 12 September 2014 22:03, Joe Slater wrote:
> > +++ b/m
On Mon, Sep 15, 2014 at 4:23 PM, Burton, Ross wrote:
> On 15 September 2014 11:16, wrote:
>> +++
>> b/meta/recipes-support/libgcrypt/files/libgcrypt-1.6.1-make-arm-asm-fPIC-friendly.patch
>
> This is missing upstream-status/signed-off-by tags.
>
> Also, Andrea said:
>> Instead of patching 1.6.1
On 09/15/2014 06:11 AM, Bruce Ashfield wrote:
On 14-09-12 10:49 PM, Armin Kuster wrote:
Not sure if this this is the correct process.
I'll grab the patch and carry it here, and update the SRCREVs at the
same time as the other BSPs.
Great, thanks.
As for adding qemuppc64 fully to the sup
On 15 September 2014 15:47, Mike Looijmans wrote:
> It's just that the try/except version will always work on any system,
> whereas the device comparison may fail on some setups. And given the habit
> of programmers to copy these things from other locations, chances are that a
> device comparison
On 15 September 2014 05:00, Dan McGregor wrote:
> From: Dan McGregor
>
> AddressSanitizer is a fast memory error detector.
> ThreadSanitizer detects data races.
> UBSanitizer detectes undefined behaviour.
>
> All consist of compiler instrumentation and a run-time library.
> The compiler instrumen
On 09/15/2014 10:43 PM, Burton, Ross wrote:
On 15 September 2014 15:38, Robert Yang wrote:
Ah, yes, then we should set PACKAGECONFIG ??= "python" by default ?
Otherwise
I'm afraid that there would be errors.
No need to set it by default - although my argument here is that we
don't currently
On 09/15/2014 04:12 PM, Richard Purdie wrote:
On Mon, 2014-09-15 at 15:58 +0200, Mike Looijmans wrote:
I can probably come up with scenarios where "os.stat(src).st_dev ==
os.stat(destdir).st_dev" but os.link will still fail.
Less code and does not assume that when src and destdir are on the s
On 15 September 2014 15:38, Robert Yang wrote:
> Ah, yes, then we should set PACKAGECONFIG ??= "python" by default ?
> Otherwise
> I'm afraid that there would be errors.
No need to set it by default - although my argument here is that we
don't currently. This should work:
PACKAGECONFIG ??= ""
PA
On Mon, Sep 15, 2014 at 11:36 AM, Burton, Ross wrote:
> On 15 September 2014 15:33, Otavio Salvador wrote:
>> I think it is a way better base for work but I really don't care as I
>> use my own recipe in fsl-arm and so does TI.
>
> And that's the problem really isn't it. :) Why do you both use y
On 09/15/2014 10:01 PM, Burton, Ross wrote:
On 15 September 2014 14:05, Robert Yang wrote:
+PACKAGECONFIG ??= ""
+PACKAGECONFIG[python] =
+EXTRA_OECONF_class-native = "--with-python=${PYTHON} --without-legacy
--without-docbook --with-c14n --without-lzma --with-zlib"
You should remove the py
On 15 September 2014 06:10, Ed Swarthout wrote:
> It looks like udev ptest is the culprit:
>
> /f/deb/home/yocto/yocto/build2014/tmp/sysroots/qemuppc/usr/lib/udev/ptest/test/sys/block/loop0/subsystem/loop0/subsystem:
> total used in directory 8 available 55515512
> drwxr-xr-x 2 ed ed0 4096 Au
On 15 September 2014 15:33, Otavio Salvador wrote:
> I think it is a way better base for work but I really don't care as I
> use my own recipe in fsl-arm and so does TI.
And that's the problem really isn't it. :) Why do you both use your
own recipe, instead of keeping the u-boot in oe-core up to
On 15 September 2014 06:48, Rongqing Li wrote:
> Could we move irda-utils out of oe-core?
> I think nearly no one is using it
I tend to agree, irda is rarely used these days. Feel free to propose
a patch to remove it from oe-core (and move it to some other layer in
meta-oe) once 1.7 is released.
On Mon, Sep 15, 2014 at 11:21 AM, Burton, Ross wrote:
> On 15 September 2014 14:33, Otavio Salvador wrote:
>> I am using 2014.07 in meta-fsl-arm for a while and I support the merge
>> of this for 1.8.
>
> But as we're post-freeze, endorsement doesn't qualify as a reason to
> break the freeze...
>
On 15 September 2014 09:27, wrote:
> From: Roy Li
>
> Setting INHIBIT_PACKAGE_STRIP is added by 61cb45869c[kmod: ptest fixes], but I
> can not reproduce the described issue, and seems it make wrong logical [
> when enable ptest, INHIBIT_PACKAGE_STRIP is always 1; and when disable ptest,
> INHIBI
On 15 September 2014 10:01, wrote:
> +PACKAGECONFIG ?= "acl attr"
Note that this should be ??=. I've fixed this when merging to my
branch so no need to re-send.
Ross
--
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http
On 15 September 2014 11:16, wrote:
> +++
> b/meta/recipes-support/libgcrypt/files/libgcrypt-1.6.1-make-arm-asm-fPIC-friendly.patch
This is missing upstream-status/signed-off-by tags.
Also, Andrea said:
> Instead of patching 1.6.1, wouldn't it be better to update to libgcrypt_1.6.2
> ?
No, we
On 15 September 2014 14:33, Otavio Salvador wrote:
> I am using 2014.07 in meta-fsl-arm for a while and I support the merge
> of this for 1.8.
But as we're post-freeze, endorsement doesn't qualify as a reason to
break the freeze...
(queued in my ross/unfreeze branch for when we've thawed)
Ross
On Mon, 2014-09-15 at 15:58 +0200, Mike Looijmans wrote:
> I can probably come up with scenarios where "os.stat(src).st_dev ==
> os.stat(destdir).st_dev" but os.link will still fail.
>
> Less code and does not assume that when src and destdir are on the same
> device, they can be linked:
>
> t
On 15 September 2014 14:05, Robert Yang wrote:
> +PACKAGECONFIG ??= ""
> +PACKAGECONFIG[python] =
> +EXTRA_OECONF_class-native = "--with-python=${PYTHON} --without-legacy
> --without-docbook --with-c14n --without-lzma --with-zlib"
You should remove the python arguments from all overrides of
EXTRA
I can probably come up with scenarios where "os.stat(src).st_dev ==
os.stat(destdir).st_dev" but os.link will still fail.
Less code and does not assume that when src and destdir are on the same
device, they can be linked:
try:
os.link(src, dst)
except:
shutil.copyfile(src, dst)
Mike.
On Sun, Sep 14, 2014 at 6:40 PM, Burton, Ross wrote:
> On 14 September 2014 04:34, Denys Dmytriyenko wrote:
>> From: Denys Dmytriyenko
>>
>> * Update U-boot and corresponding tools to version 2014.07
>> * Small cleanups with indentation and consolidation of u-boot.inc
>
> This missed the feature
On Mon, Sep 15, 2014 at 10:33 AM, Otavio Salvador
wrote:
> On Sun, Sep 14, 2014 at 6:40 PM, Burton, Ross wrote:
>> On 14 September 2014 04:34, Denys Dmytriyenko wrote:
>>> From: Denys Dmytriyenko
>>>
>>> * Update U-boot and corresponding tools to version 2014.07
>>> * Small cleanups with indent
On 14-09-12 10:49 PM, Armin Kuster wrote:
Not sure if this this is the correct process.
I'll grab the patch and carry it here, and update the SRCREVs at the
same time as the other BSPs.
As for adding qemuppc64 fully to the supported set of emulated BSPs
for 1.7, that's Richard's call (and I ha
On 09/15/2014 06:43 AM, Burton, Ross wrote:
On 9 September 2014 08:34, Robert Yang wrote:
We have libxml2-python for native and nativesdk, but don't have it for
target, and can't find the reason from the git log, libxml2-python is
widely used, after looked into it's configure.in, we can enable
On 09/14/2014 06:30 PM, Burton, Ross wrote:
> On 11 September 2014 21:33, Soren Brinkmann
> wrote:
>> Add a distro variable to set the hostname. This allows easily assigning
>> a hostname in local.conf.
>
> It's already as easy as assigning a hostname in local.conf:
>
> hostname_pn-base-files =
On Mon, Sep 15, 2014 at 12:16 PM, wrote:
> From: Jackie Huang
>
> libgcrypt.so.20 contains .text relocations, backport a patch
> to fix it.
>
> Signed-off-by: Jackie Huang
> ---
> ...ibgcrypt-1.6.1-make-arm-asm-fPIC-friendly.patch | 169
>
> meta/recipes-support/libgcryp
From: Jackie Huang
libgcrypt.so.20 contains .text relocations, backport a patch
to fix it.
Signed-off-by: Jackie Huang
---
...ibgcrypt-1.6.1-make-arm-asm-fPIC-friendly.patch | 169
meta/recipes-support/libgcrypt/libgcrypt_1.6.1.bb |2 +
2 files changed, 171 insertion
From: Roy Li
Add PACKAGECONFIG for acl/attr
and setting the default value to enable them
Signed-off-by: Roy Li
---
meta/recipes-devtools/rsync/rsync_3.1.0.bb |4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/meta/recipes-devtools/rsync/rsync_3.1.0.bb
b/meta/recipes-dev
From: Roy Li
Setting INHIBIT_PACKAGE_STRIP is added by 61cb45869c[kmod: ptest fixes], but I
can not reproduce the described issue, and seems it make wrong logical [
when enable ptest, INHIBIT_PACKAGE_STRIP is always 1; and when disable ptest,
INHIBIT_PACKAGE_STRIP is always 0]
Cc: Tudor Florea
45 matches
Mail list logo
