Either Bero or pcpa over last years showed 0% interest in uclibc, which imo
is good :)
uClibc idea is yours, and only you is trying to maintain it.

i586 build is still broken, please fix it ASAP.


2015-06-23 19:18 GMT+02:00 Per Øyvind Karlsen <[email protected]>:

> Well, as I told you earlier, only component of mine from systemd that I'm
> actually currently  using is udevd, so if in a rush without me avaliable to
> help, just disable the particular feature causing issues and blocking your
> work.
>
> I can always have a look at later, consider whether missing functionality
> in uClibc should/needs to be implemented, be it feasible and worthwhile..
> Often it's just a matter of adding new definitions like. ie. new flags for
> ie. mounting, for which the functionality in question has already been
> implemented in the kernel, with the flag definitions there just to make
> them available for user space tools to use and similar..
> If I do make sure the functioality to be implemented and available i
> uClibc, I'll reenable whatever feature you had to disable to make it build
> (just be sure to disable only what you need to build, not just the uClibc
> build entirely.
> Judging by the functio name, I guess I most likely would find it of
> relevace and sufficiently worthwhile to take care oif it though. :)
>
> So in general, anything missing can most often be easily added by simply
>  looking at glibc or other libcs, and sync their headers etc., and some
> occations you end up just "borrowing" it as-is from other libc projects,
> ie. something most people should actively contributing should be equipped
> with the competence & reasoning skills to idependently & figure out how to
> solve >75% of these things themself.
>
> For this matter however, most wouldn't, it was a false regression in gcc
> internals with recent gcc it relied on the behaviour of, as early as during
> _start() in the dynamic linker, so not really possible to make anything
> dynamically linked working then... ;)
>
> But keep in mind that we have two (at least one) greater toolchains
> talents with better competence and expertize than mine, both bero & pcpa,
> which you probably should be able turning to for assistane. And if that
> still fails, and without me available, there's always the quite common and
> heavily relied on resort we've always been dependent on and would make it
> impossible for us to carry around all our software packages without perfect
> knowledge and understanding of all the huge amount of packages we maintain
> (together), namely upstream..  ;) *
>
> I got home from the office around 13:00 after having been there working
> all night, and I don't just currently have any linux systems that's not in
> parts and already ready and running for me to attend to this right away,
> nor do I really feel like assembling any equipment to do, I was quite happy
> leaving the office today.. ;)
>
> * By using list more, I'll be better able to help (not just) yourself
> (same goes for and will benefit contributors in general, not just on the
> technical issues at stake, but at properly communicate through the right
> channels again) with such matters not just myself personally and directly,
> but help give advice and on where to point to for help when I'm not able to
> or unavailable for much more than communication on list.
> It'll be easier for me to help you getting help and communicate with
> upstream, assist you in diagnosing issues and better equipped with insight
> and understanding built and provided while we try do initial diagnosing
> together, to make it as efficient and less of a hazzle to make upstream
> both fully understand and be able to help you out fixing the issue faster.
>
> I intend andd expect to be more actively avilable and involved again now
> as my prescriptions has finally been sorted out again and I have an office
> to work at, so for the immediate future, I'm basically available and
> willing to work full time on cooker again, and if it were to be beginning
> to bear fruits and going a more succesful and motivation direction again,
> I'm likely too dumb and tenacious to avoid the not most healthy at times
> level of participation and involvedment as before. :)
>
> I hope to continue from where I left of with announcing last weeks
> strategy of mine, serving hopefully as advice/suggestion for others to join
> behind..
> And I hope to get better as such again on a more frequent basis as well,
> incidents when I was actively involved/leading the project back then, where
> online politics and what not in public were found to be wastly more tiring
> & harder to deal with than all other tasks combined, and made me extremely
> reluctant from writing much nor or say much in open on publi lists etc.
> since, hence explaining my perceived absence by most.
>
> Only from list though, still have been doing  a lot of work on distro for
> periods that I intend to prepare for inclusion direclty into cooker over
> the next couple of weeks or so..:)
>
>
>
> 2015-06-23 12:46 GMT+02:00 Tomasz Gajc <[email protected]>:
>
>> Per your new uclibc still does not work on i586 error is the same as in
>> previous message.
>> https://abf.io/build_lists/2512965
>>
>> Second uclibc is missing syncfs from glibc:
>>
>> /builddir/build/BUILD/systemd-221/src/shared/machine-pool.c:281:16: error: 
>> implicit declaration of function 'syncfs' 
>> [-Werror=implicit-function-declaration]
>>          (void) syncfs(fd);
>>
>> https://abf.io/build_lists/2512967
>>
>>
>>
>> 2015-05-31 2:29 GMT+02:00 Tomasz Paweł Gajc <[email protected]>:
>>
>>> Dnia wtorek, 26 maja 2015 09:03:41 Per Øyvind Karlsen pisze:
>>> > I'll try get around to attend to this later today (no promises though,
>>> > which many of you might already guess from my rather sporadic activity
>>> and
>>> > all over the last months).
>>> >
>>>
>>> Per can you please take a look on attached config.log for i586 build on
>>> ABF?
>>>
>>> configure:3541: uclibc-gcc -V >&5
>>> gcc: error: unrecognized command line option '-V'
>>> gcc: fatal error: no input files
>>> compilation terminated.
>>> configure:3552: $? = 1
>>> configure:3541: uclibc-gcc -qversion >&5
>>> gcc: error: unrecognized command line option '-qversion'
>>> gcc: fatal error: no input files
>>> compilation terminated.
>>> configure:3552: $? = 1
>>> configure:3572: checking whether the C compiler works
>>> configure:3594: uclibc-gcc -Oz -Wa,--compress-debug-sections -gdwarf-4 -
>>> Wstrict-aliasing=2 -pipe -Wformat -Werror=format-security
>>> -D_FORTIFY_SOURCE=2
>>> -fstack-protector --param=ssp-buffer-size=4  -fomit-frame-pointer
>>> -mtune=atom
>>> -march=i586 -fasynchronous-unwind-tables -fno-stack-protector -Oz
>>>  -Wl,--no-
>>> undefined  conftest.c  >&5
>>> /usr/bin/ld: warning: /usr/uclibc/usr/lib/crt1.o: missing .note.GNU-stack
>>> section implies executable stack
>>> /usr/bin/ld: warning: /usr/uclibc/usr/lib/crti.o: missing .note.GNU-stack
>>> section implies executable stack
>>> /usr/bin/ld: warning: /usr/uclibc/usr/lib/crtn.o: missing .note.GNU-stack
>>> section implies executable stack
>>> configure:3598: $? = 0
>>> configure:3646: result: yes
>>> configure:3649: checking for C compiler default output file name
>>> configure:3651: result: a.out
>>> configure:3657: checking for suffix of executables
>>> configure:3664: uclibc-gcc -o conftest -Oz -Wa,--compress-debug-sections
>>> -
>>> gdwarf-4 -Wstrict-aliasing=2 -pipe -Wformat -Werror=format-security -
>>> D_FORTIFY_SOURCE=2 -fstack-protector --param=ssp-buffer-size=4
>>> -fomit-frame-
>>> pointer -mtune=atom -march=i586 -fasynchronous-unwind-tables -fno-stack-
>>> protector -Oz   -Wl,--no-undefined  conftest.c  >&5
>>> /usr/bin/ld: warning: /usr/uclibc/usr/lib/crt1.o: missing .note.GNU-stack
>>> section implies executable stack
>>> /usr/bin/ld: warning: /usr/uclibc/usr/lib/crti.o: missing .note.GNU-stack
>>> section implies executable stack
>>> /usr/bin/ld: warning: /usr/uclibc/usr/lib/crtn.o: missing .note.GNU-stack
>>> section implies executable stack
>>> configure:3668: $? = 0
>>> configure:3690: result:
>>> configure:3712: checking whether we are cross compiling
>>> configure:3720: uclibc-gcc -o conftest -Oz -Wa,--compress-debug-sections
>>> -
>>> gdwarf-4 -Wstrict-aliasing=2 -pipe -Wformat -Werror=format-security -
>>> D_FORTIFY_SOURCE=2 -fstack-protector --param=ssp-buffer-size=4
>>> -fomit-frame-
>>> pointer -mtune=atom -march=i586 -fasynchronous-unwind-tables -fno-stack-
>>> protector -Oz   -Wl,--no-undefined  conftest.c  >&5
>>> /usr/bin/ld: warning: /usr/uclibc/usr/lib/crt1.o: missing .note.GNU-stack
>>> section implies executable stack
>>> /usr/bin/ld: warning: /usr/uclibc/usr/lib/crti.o: missing .note.GNU-stack
>>> section implies executable stack
>>> /usr/bin/ld: warning: /usr/uclibc/usr/lib/crtn.o: missing .note.GNU-stack
>>> section implies executable stack
>>> configure:3724: $? = 0
>>> configure:3731: ./conftest
>>> /builddir/build/BUILD/systemd-220/configure: line 3733: 30759
>>> Segmentation
>>> fault      ./conftest$ac_cv_exeext
>>> configure:3735: $? = 139
>>> configure:3742: error: in `/builddir/build/BUILD/systemd-220/uclibc':
>>> configure:3744: error: cannot run C compiled programs.
>>> If you meant to cross compile, use `--host'.
>>>
>>>
>>> --
>>> Cheers
>>> TPG
>>
>>
>>
>
_______________________________________________
OM-Cooker mailing list
[email protected]
http://ml.openmandriva.org/listinfo.cgi/om-cooker-openmandriva.org

Reply via email to