State of Sumo

State of Sumo

[Burton, Ross]

Hi,

We'll hopefully be firing the 2.5 RC build on Monday, so I did a build
today to see where we stand. As is typical we got two new failures on
the autobuilder...  My test build included the pseudo bump which
caused an almost complete rebuild, which is a great way to shake out
intermittent failures.

https://bugzilla.yoctoproject.org/show_bug.cgi?id=12704
Go appears to be using the target CFLAGS when using the host compiler,
with predictable failures when the host is old and doesn't support new
flags.

https://bugzilla.yoctoproject.org/show_bug.cgi?id=12705
u-boot build failure

CC'ing the usual maintainers of those two recipes, if you get a moment
to look at these then that would be much appreciated.

I'll also say that if you have any outstanding patches that *need* to
be in 2.5 and can't wait for 2.5.1 then say so now, or forever hold
your peace (and submit them for master/backport to 2.5.1).

Regards,
Ross

Re: State of Sumo

[Marek Vasut]

On 04/20/2018 11:57 PM, Burton, Ross wrote:
> Hi,
> 
> We'll hopefully be firing the 2.5 RC build on Monday, so I did a build
> today to see where we stand. As is typical we got two new failures on
> the autobuilder...  My test build included the pseudo bump which
> caused an almost complete rebuild, which is a great way to shake out
> intermittent failures.
> 
> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12704
> Go appears to be using the target CFLAGS when using the host compiler,
> with predictable failures when the host is old and doesn't support new
> flags.
> 
> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12705
> u-boot build failure

Is there any chance you can tar-up the whole U-Boot build dir and make
it available, so it could be debugged ?

-- 
Best regards,
Marek Vasut

Re: State of Sumo

[Andre McCurdy]

On Fri, Apr 20, 2018 at 2:57 PM, Burton, Ross <ross.burton@intel.com> wrote:
> Hi,
>
> We'll hopefully be firing the 2.5 RC build on Monday, so I did a build
> today to see where we stand. As is typical we got two new failures on
> the autobuilder...  My test build included the pseudo bump which
> caused an almost complete rebuild, which is a great way to shake out
> intermittent failures.
>
> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12704
> Go appears to be using the target CFLAGS when using the host compiler,
> with predictable failures when the host is old and doesn't support new
> flags.

Maybe worth resurrecting something like:

  http://lists.openembedded.org/pipermail/openembedded-core/2016-February/118212.html

to flush out similar issues?

> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12705
> u-boot build failure
>
> CC'ing the usual maintainers of those two recipes, if you get a moment
> to look at these then that would be much appreciated.
>
> I'll also say that if you have any outstanding patches that *need* to
> be in 2.5 and can't wait for 2.5.1 then say so now, or forever hold
> your peace (and submit them for master/backport to 2.5.1).
>
> Regards,
> Ross
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: State of Sumo

[Burton, Ross]

On 20 April 2018 at 23:09, Andre McCurdy <armccurdy@gmail.com> wrote:
> On Fri, Apr 20, 2018 at 2:57 PM, Burton, Ross <ross.burton@intel.com> wrote:
>> Hi,
>>
>> We'll hopefully be firing the 2.5 RC build on Monday, so I did a build
>> today to see where we stand. As is typical we got two new failures on
>> the autobuilder...  My test build included the pseudo bump which
>> caused an almost complete rebuild, which is a great way to shake out
>> intermittent failures.
>>
>> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12704
>> Go appears to be using the target CFLAGS when using the host compiler,
>> with predictable failures when the host is old and doesn't support new
>> flags.
>
> Maybe worth resurrecting something like:
>
>   http://lists.openembedded.org/pipermail/openembedded-core/2016-February/118212.html
>
> to flush out similar issues?

Thank you for finding that, I remembered there was a patch but
couldn't find it. :)

Ross

Re: State of Sumo

[Burton, Ross]

Okay, fired at https://autobuilder.yocto.io/builders/nightly-x86-64-lsb/builds/969.
That won't be catching people using CC_FOR_BUILD in target recipes to
build pieces using the host compiler, but it will be interesting to
see if it catches anything.

Ross

On 21 April 2018 at 09:08, Burton, Ross <ross.burton@intel.com> wrote:
> On 20 April 2018 at 23:09, Andre McCurdy <armccurdy@gmail.com> wrote:
>> On Fri, Apr 20, 2018 at 2:57 PM, Burton, Ross <ross.burton@intel.com> wrote:
>>> Hi,
>>>
>>> We'll hopefully be firing the 2.5 RC build on Monday, so I did a build
>>> today to see where we stand. As is typical we got two new failures on
>>> the autobuilder...  My test build included the pseudo bump which
>>> caused an almost complete rebuild, which is a great way to shake out
>>> intermittent failures.
>>>
>>> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12704
>>> Go appears to be using the target CFLAGS when using the host compiler,
>>> with predictable failures when the host is old and doesn't support new
>>> flags.
>>
>> Maybe worth resurrecting something like:
>>
>>   http://lists.openembedded.org/pipermail/openembedded-core/2016-February/118212.html
>>
>> to flush out similar issues?
>
> Thank you for finding that, I remembered there was a patch but
> couldn't find it. :)
>
> Ross

Re: State of Sumo

[Burton, Ross]

Good news is that went green, so the problem is restricted to
CC_FOR_BUILD usage I guess.

Ross

On 21 April 2018 at 09:09, Burton, Ross <ross.burton@intel.com> wrote:
> Okay, fired at https://autobuilder.yocto.io/builders/nightly-x86-64-lsb/builds/969.
> That won't be catching people using CC_FOR_BUILD in target recipes to
> build pieces using the host compiler, but it will be interesting to
> see if it catches anything.
>
> Ross
>
> On 21 April 2018 at 09:08, Burton, Ross <ross.burton@intel.com> wrote:
>> On 20 April 2018 at 23:09, Andre McCurdy <armccurdy@gmail.com> wrote:
>>> On Fri, Apr 20, 2018 at 2:57 PM, Burton, Ross <ross.burton@intel.com> wrote:
>>>> Hi,
>>>>
>>>> We'll hopefully be firing the 2.5 RC build on Monday, so I did a build
>>>> today to see where we stand. As is typical we got two new failures on
>>>> the autobuilder...  My test build included the pseudo bump which
>>>> caused an almost complete rebuild, which is a great way to shake out
>>>> intermittent failures.
>>>>
>>>> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12704
>>>> Go appears to be using the target CFLAGS when using the host compiler,
>>>> with predictable failures when the host is old and doesn't support new
>>>> flags.
>>>
>>> Maybe worth resurrecting something like:
>>>
>>>   http://lists.openembedded.org/pipermail/openembedded-core/2016-February/118212.html
>>>
>>> to flush out similar issues?
>>
>> Thank you for finding that, I remembered there was a patch but
>> couldn't find it. :)
>>
>> Ross

Re: State of Sumo

[Matt Madison]

I've got a handle on the go-runtime issue, will send a patch shortly.

-Matt

On Sat, Apr 21, 2018 at 3:02 AM, Burton, Ross <ross.burton@intel.com> wrote:
> Good news is that went green, so the problem is restricted to
> CC_FOR_BUILD usage I guess.
>
> Ross
>
> On 21 April 2018 at 09:09, Burton, Ross <ross.burton@intel.com> wrote:
>> Okay, fired at https://autobuilder.yocto.io/builders/nightly-x86-64-lsb/builds/969.
>> That won't be catching people using CC_FOR_BUILD in target recipes to
>> build pieces using the host compiler, but it will be interesting to
>> see if it catches anything.
>>
>> Ross
>>
>> On 21 April 2018 at 09:08, Burton, Ross <ross.burton@intel.com> wrote:
>>> On 20 April 2018 at 23:09, Andre McCurdy <armccurdy@gmail.com> wrote:
>>>> On Fri, Apr 20, 2018 at 2:57 PM, Burton, Ross <ross.burton@intel.com> wrote:
>>>>> Hi,
>>>>>
>>>>> We'll hopefully be firing the 2.5 RC build on Monday, so I did a build
>>>>> today to see where we stand. As is typical we got two new failures on
>>>>> the autobuilder...  My test build included the pseudo bump which
>>>>> caused an almost complete rebuild, which is a great way to shake out
>>>>> intermittent failures.
>>>>>
>>>>> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12704
>>>>> Go appears to be using the target CFLAGS when using the host compiler,
>>>>> with predictable failures when the host is old and doesn't support new
>>>>> flags.
>>>>
>>>> Maybe worth resurrecting something like:
>>>>
>>>>   http://lists.openembedded.org/pipermail/openembedded-core/2016-February/118212.html
>>>>
>>>> to flush out similar issues?
>>>
>>> Thank you for finding that, I remembered there was a patch but
>>> couldn't find it. :)
>>>
>>> Ross
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: State of Sumo

[Khem Raj]

[-- Attachment #1: Type: text/plain, Size: 2087 bytes --]

On Fri, Apr 20, 2018 at 3:09 PM Andre McCurdy <armccurdy@gmail.com> wrote:

> On Fri, Apr 20, 2018 at 2:57 PM, Burton, Ross <ross.burton@intel.com>
> wrote:
> > Hi,
> >
> > We'll hopefully be firing the 2.5 RC build on Monday, so I did a build
> > today to see where we stand. As is typical we got two new failures on
> > the autobuilder...  My test build included the pseudo bump which
> > caused an almost complete rebuild, which is a great way to shake out
> > intermittent failures.
> >
> > https://bugzilla.yoctoproject.org/show_bug.cgi?id=12704
> > Go appears to be using the target CFLAGS when using the host compiler,
> > with predictable failures when the host is old and doesn't support new
> > flags.
>
> Maybe worth resurrecting something like:
>
>
> http://lists.openembedded.org/pipermail/openembedded-core/2016-February/118212.html
>
> to flush out similar issues?


The issue at hand is a bit different where portions
Are built using native compiler in target recipe itself simple fix would be
to pass right c flags

This patch in itself does have merits and we could consider this for 2.6 it
would be intrusive but would result in cleanup

>
>
> > https://bugzilla.yoctoproject.org/show_bug.cgi?id=12705
> > u-boot build failure
> >
> > CC'ing the usual maintainers of those two recipes, if you get a moment
> > to look at these then that would be much appreciated.
> >
> > I'll also say that if you have any outstanding patches that *need* to
> > be in 2.5 and can't wait for 2.5.1 then say so now, or forever hold
> > your peace (and submit them for master/backport to 2.5.1).
> >
> > Regards,
> > Ross
> > --
> > _______________________________________________
> > Openembedded-core mailing list
> > Openembedded-core@lists.openembedded.org
> > http://lists.openembedded.org/mailman/listinfo/openembedded-core
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core
>

[-- Attachment #2: Type: text/html, Size: 3508 bytes --]

Re: State of Sumo

[Peter Kjellerstedt]

> -----Original Message-----
> From: openembedded-core-bounces@lists.openembedded.org
> [mailto:openembedded-core-bounces@lists.openembedded.org] On Behalf Of
> Burton, Ross
> Sent: den 20 april 2018 23:58
> To: OE-core <openembedded-core@lists.openembedded.org>
> Cc: Marek Vasut <marex@denx.de>; Otavio Salvador
> <otavio@ossystems.com.br>
> Subject: [OE-core] State of Sumo
> 
> Hi,
> 
> We'll hopefully be firing the 2.5 RC build on Monday, so I did a build
> today to see where we stand. As is typical we got two new failures on
> the autobuilder...  My test build included the pseudo bump which
> caused an almost complete rebuild, which is a great way to shake out
> intermittent failures.
> 
> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12704
> Go appears to be using the target CFLAGS when using the host compiler,
> with predictable failures when the host is old and doesn't support new
> flags.
> 
> https://bugzilla.yoctoproject.org/show_bug.cgi?id=12705
> u-boot build failure
> 
> CC'ing the usual maintainers of those two recipes, if you get a moment
> to look at these then that would be much appreciated.
> 
> I'll also say that if you have any outstanding patches that *need* to
> be in 2.5 and can't wait for 2.5.1 then say so now, or forever hold
> your peace (and submit them for master/backport to 2.5.1).

Not sure what the status is for patches for bitbake itself, but I have 
one pending ("lib/bb/utils: prevent movefile from changing dest dir 
permissions" sent to bitbake-devel on April 10), which I was hoping 
would make it into Sumo. The effect when the bug strikes is pretty 
weird, in our case it causes /etc to end up with mode 644 rather than 
755 when building an SDK in a Docker container...

> Regards,
> Ross

//Peter

Re: State of Sumo

[Burton, Ross]

On 20 April 2018 at 23:42, Peter Kjellerstedt
<peter.kjellerstedt@axis.com> wrote:
> Not sure what the status is for patches for bitbake itself, but I have
> one pending ("lib/bb/utils: prevent movefile from changing dest dir
> permissions" sent to bitbake-devel on April 10), which I was hoping
> would make it into Sumo. The effect when the bug strikes is pretty
> weird, in our case it causes /etc to end up with mode 644 rather than
> 755 when building an SDK in a Docker container...

I wouldn't call that a release blocker, so if we're good to go on
Monday that can go to master instead and be backported to the Sumo
branch asap.

Ross

Re: State of Sumo

[Peter Kjellerstedt]

> -----Original Message-----
> From: Burton, Ross [mailto:ross.burton@intel.com]
> Sent: den 21 april 2018 10:08
> To: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
> Cc: OE-core <openembedded-core@lists.openembedded.org>
> Subject: Re: [OE-core] State of Sumo
> 
> On 20 April 2018 at 23:42, Peter Kjellerstedt
> <peter.kjellerstedt@axis.com> wrote:
> > Not sure what the status is for patches for bitbake itself, but I
> have
> > one pending ("lib/bb/utils: prevent movefile from changing dest dir
> > permissions" sent to bitbake-devel on April 10), which I was hoping
> > would make it into Sumo. The effect when the bug strikes is pretty
> > weird, in our case it causes /etc to end up with mode 644 rather than
> > 755 when building an SDK in a Docker container...
> 
> I wouldn't call that a release blocker, so if we're good to go on
> Monday that can go to master instead and be backported to the Sumo
> branch asap.
> 
> Ross

Well, I cannot disagree with you on the status of the patch vs the 
release state of OE-Core. However, it has been a little bit 
disappointing to see more invasive patches being posted to the mailing 
list and subsequently being integrated, while this quite trivial patch 
has been neglected.

//Peter