On 12/29/2016 10:36 PM, Andreas Bombe wrote:
> If I upload now, it won't be part of the stretch release.
Ok, I actually overlooked the fact that the package hasn't entered
testing yet at all. Then it's indeed a good idea to wait first
and then upload the fixed package.
> but who guarantees that
On Thu, Dec 29, 2016 at 10:18:07PM +0100, John Paul Adrian Glaubitz wrote:
> That's correct. But I usually rather prefer fixing a package everywhere
> than just saying "Works on amd64, I don't care about the rest because
> I don't use anything else.".
>
> My point is, if you upload it now, you
On Wed, Dec 28, 2016 at 04:29:52PM +0100, John Paul Adrian Glaubitz wrote:
> On 12/28/2016 04:18 PM, Radovan Birdic wrote:
> > Here is new generic patch that should work for all architectures.
> > I added "--as-needed" flag that ensures linking with latomic library only
> > if it is necessary.
>
On 12/29/2016 10:15 PM, Andreas Bombe wrote:
> 1) Right now would still be too late with the 10 day transition in
>effect.
Well, it would reset the transition counter back to 10 days.
> 2) Unless I'm mistaken, a build failure on release architectures is only
>a blocker if it previously
On Thu, Dec 29, 2016 at 10:02:01PM +0100, John Paul Adrian Glaubitz wrote:
> On 12/29/2016 09:57 PM, Andreas Bombe wrote:
> > Thanks for the work. I was aware of the build failure but I will wait
> > until the package has made it to testing before uploading a fix.
>
> Not sure why. It affects two
On 12/29/2016 09:57 PM, Andreas Bombe wrote:
> Thanks for the work. I was aware of the build failure but I will wait
> until the package has made it to testing before uploading a fix.
Not sure why. It affects two release architectures, so the package would
be missing in testing for release
On 12/28/2016 04:18 PM, Radovan Birdic wrote:
> Here is new generic patch that should work for all architectures.
> I added "--as-needed" flag that ensures linking with latomic library only if
> it is necessary.
>
> With this patch package builds successfully on mips, mipsel, mips64el and
>
Hi,
Here is new generic patch that should work for all architectures.
I added "--as-needed" flag that ensures linking with latomic library only if it
is necessary.
With this patch package builds successfully on mips, mipsel, mips64el and i386.
Regards,
Radovan---
On 12/28/2016 12:00 PM, John Paul Adrian Glaubitz wrote:
>> I have created and attached a patch that resolves this issue.
>> With this patch package builds successfully on mips, mipsel, mips64el and
>> i386.
>
> Sounds good. Although it would probably a better idea not to make the patch
>
Hi Radovan!
On 12/28/2016 11:04 AM, Radovan Birdic wrote:
> The problem occurs because mips 32-bit architectures does not support use of
> 8-byte atomics functions.
> To solve this problem, it is necessary to provide linking with libatomic
> library.
Thanks for the patch. I'm not the usual guy
Package: limesuite
Version: 16.8.23.819+git20161221+dfsg-1
Severity: important
Tags: sid + patch
Justification: FTBFS
User: debian-m...@lists.debian.org
Usertags: mips-patch
Package limesuite_16.8.23.819+git20161221+dfsg-1 FTBFS on mips and mipsel with
following error:
>
11 matches
Mail list logo
