2012/11/22 Alberto Luaces :
>> I have tried with the other machine, and it built fine. Diff attached
>> (I commited to VCS anyway). Since release managers approved with
>> those changes enabled, I will go ahead later, but if you have the
>> opportunity confirm that it's OK in the meantime, please
"Manuel A. Fernandez Montecelo" writes:
> 2012/11/22 Alberto Luaces :
>>> Failed to build again, it looks like ~7GB isn't enough... will have to
>>> make more room to compile or try to grab the beefier machine soon.
>>
>> Yes, looking at the build tree it says
>>
>> $ du -sm
>> 8079
>
> I have tri
2012/11/22 Alberto Luaces :
>> Failed to build again, it looks like ~7GB isn't enough... will have to
>> make more room to compile or try to grab the beefier machine soon.
>
> Yes, looking at the build tree it says
>
> $ du -sm
> 8079
I have tried with the other machine, and it built fine. Diff a
"Manuel A. Fernandez Montecelo" writes:
> 2012/11/21 Manuel A. Fernandez Montecelo :
>> Lastly, I couldn't build it in a beefier machine today at work, but
>> I'm trying again now, will tell the result later.
>
> Failed to build again, it looks like ~7GB isn't enough... will have to
> make more ro
"Manuel A. Fernandez Montecelo" writes:
> Question: in the following paragraphs, ${LINKER_FLAGS} should actually
> be ${LDFLAGS}, or am I missing something?
Good catch! That was the remaining of a special fix for an older bug
with arm's gcc that it's not used currently. I have removed those
occu
2012/11/21 Manuel A. Fernandez Montecelo :
> Lastly, I couldn't build it in a beefier machine today at work, but
> I'm trying again now, will tell the result later.
Failed to build again, it looks like ~7GB isn't enough... will have to
make more room to compile or try to grab the beefier machine s
2012/11/21 Alberto Luaces :
> "Manuel A. Fernandez Montecelo" writes:
>
>> I asked release managers in the other e-mail, so if they give the
>> go-ahead it's better for us. You can defer/avoid investigating
>> until/unless they don't want us to use dh level 9.
>
> Ok. Anyway, I did the test of ch
"Manuel A. Fernandez Montecelo" writes:
> I asked release managers in the other e-mail, so if they give the
> go-ahead it's better for us. You can defer/avoid investigating
> until/unless they don't want us to use dh level 9.
Ok. Anyway, I did the test of changing debian/compat to "7", and it
w
2012/11/21 Alberto Luaces :
>> Any clue? Maybe because of the changes of DESTDIR and so on? There
>> is no "No space left on device" kind of error, but maybe it's because
>> of that (/tmp as tmpfs, and no previous errors for similar
>> files/paths).
>
> I will investigate; in my build, osgdb_seri
"Manuel A. Fernandez Montecelo" writes:
> 2012/11/20 Alberto Luaces :
>> "Manuel A. Fernandez Montecelo" writes:
>>
>>> 2012/11/19 Alberto Luaces :
>>
>> [...]
>>
>>>
>>> So, indeed, lintian also shows these warnings for some of my packages ;)
>>
>> You were right again, in fact I now looked close
2012/11/20 Alberto Luaces :
> "Manuel A. Fernandez Montecelo" writes:
>
>> 2012/11/19 Alberto Luaces :
>
> [...]
>
>>
>> So, indeed, lintian also shows these warnings for some of my packages ;)
>
> You were right again, in fact I now looked closely to the list of
> warning messages and noticed that
"Manuel A. Fernandez Montecelo" writes:
> 2012/11/19 Alberto Luaces :
[...]
>
> So, indeed, lintian also shows these warnings for some of my packages ;)
You were right again, in fact I now looked closely to the list of
warning messages and noticed that, for example, libosgDB.so was absent
from
2012/11/19 Alberto Luaces :
> Ok, I will follow your examples. I had already taken into account that
> CMake problem, and in fact I am already adding CPPFLAGS into CFLAGS.
> However, if lintian throws those warnings for me and not for you,
> something has to be different between those packages. I
"Manuel A. Fernandez Montecelo" writes:
[...]
Thanks a lot for the pointers. Indeed, I tested my patch and it worked,
but as I said, I felt it was too much hacky. I knew that a fresh look
at the rules would reveal a much cleaner approach. I had completely
forgotten about the target installatio
Brr, too many things at this time of the night, hope to get things
right this time :-(
2012/11/15 Manuel A. Fernandez Montecelo :
> That is, the "root" of the target system is set to "debian/tmp", the
> application is then installed in prefix "/usr" in the target system,
> thus "debian/tmp/usr", a
2012/11/15 Alberto Luaces :
>> if I remember correctly, there were some problems if one just do as
>> suggested in the bug report. Let me check this; I will have an answer
>> before the weekend.
>
> Indeed, there was more than setting the prefix. Doing so places all the
> installation products un
Alberto Luaces Fernández writes:
>> Hello,
>>
>> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=690878
>>
>> Should we do something about this before the release? I think that
>> the prefix should be /usr, I am not sure if it would qualify as an RC
>> bug and a new version would be accepted at
> Hello,
>
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=690878
>
> Should we do something about this before the release? I think that
> the prefix should be /usr, I am not sure if it would qualify as an RC
> bug and a new version would be accepted at this point, but I think
> that it's feas
Hello,
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=690878
Should we do something about this before the release? I think that
the prefix should be /usr, I am not sure if it would qualify as an RC
bug and a new version would be accepted at this point, but I think
that it's feasible.
Cheers.
19 matches
Mail list logo
