[SCM] audacity/master: Build without sbsms to fix FTBFS on non-x86 archs. (Closes: #882173)
jcowgill at debian.org
Sat Nov 25 10:16:34 UTC 2017
On 24/11/17 17:39, Jaromír Mikeš wrote:
> 2017-11-24 14:37 GMT+01:00 James Cowgill <jcowgill at debian.org>:
>> On 24/11/17 11:08, Jaromír Mikeš wrote:
>>> 2017-11-20 21:08 GMT+01:00 <mira-guest at users.alioth.debian.org>:
>>>> The following commit has been merged in the master branch:
>>>> commit fa79381b2f34c2af651c05a2b433ae20e4ec9007
>>>> Author: Jaromír Mikeš <mira.mikes at seznam.cz>
>>>> Date: Mon Nov 20 21:08:12 2017 +0100
>>>> Build without sbsms to fix FTBFS on non-x86 archs. (Closes: #882173)
>>>> diff --git a/debian/rules b/debian/rules
>>>> index 34d3389..6970493 100755
>>>> --- a/debian/rules
>>>> +++ b/debian/rules
>>>> @@ -21,10 +21,9 @@ configure_flags := \
>>>> --with-midi=system \
>>>> --with-portaudio=system \
>>>> --with-portsmf=system \
>>>> -# --with-sbsms=system \
>>>> --with-soundtouch=system \
>>>> + --without-sbsms \
>>>> dh $@
>>> maybe even better would be build without sbsms conditionally on all archs
>>> except amd64?
>>> Can someone implement such a condition in d/rules?
>> Your patch doesn't actually fix the bug (use of sse on non-amd64). The
>> attached patch allows sbsms to work on all architectures while disabling
>> sse on all non-amd64 architectures.
>> Also note that this line of your original code was wrong:
>> # --with-sbsms=system \
>> Because there is a line continuation, this will also comment out the
>> "--with-soundtouch=system" on the next line. I've fixed that by using
>> "--with-sbsms=local" instead.
> Thank you James for fixing this.
> Can you provide DM flag for me to I can upload audacity do unstable now?
Should be done now.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 833 bytes
Desc: OpenPGP digital signature
More information about the pkg-multimedia-maintainers