[arch-general] Build pacman statically
To exchange information I want to let know this list that I have filled a feature request form to ask for a statically builded pacman. https://bugs.archlinux.org/task/30993 Comments welcome in the bug manager. جاك الفضة
On Fri, 3 Aug 2012 10:31:06 -0400 Jack Silver <jacksilver045@gmail.com> wrote:
To exchange information I want to let know this list that I have filled a feature request form to ask for a statically builded pacman.
https://bugs.archlinux.org/task/30993
Comments welcome in the bug manager.
جاك الفضة
Well, bugtracker is not a place for comments, it's for solutions. Anyway... statically compiling things is not a way of avoiding trouble, at least not in a self-sustained fashion. So, if you propose to have pacman in [core] statically compiled against all needed libraries, I would be against that as the package will be an unmaintainable mess. That might work for a rescue (read embedded-like system with ulibc) system, but probably booting from a live media is much easier and more robust. -- Leonid Isaev GnuPG key: 0x164B5A6D Fingerprint: C0DF 20D0 C075 C3F1 E1BE 775A A7AE F6CB 164B 5A6D
On 3 August 2012 12:29, Leonid Isaev <lisaev@umail.iu.edu> wrote:
On Fri, 3 Aug 2012 10:31:06 -0400 Jack Silver <jacksilver045@gmail.com> wrote:
To exchange information I want to let know this list that I have filled a feature request form to ask for a statically builded pacman.
https://bugs.archlinux.org/task/30993
Comments welcome in the bug manager.
جاك الفضة
Well, bugtracker is not a place for comments, it's for solutions.
Anyway... statically compiling things is not a way of avoiding trouble, at least not in a self-sustained fashion. So, if you propose to have pacman in [core] statically compiled against all needed libraries, I would be against that as the package will be an unmaintainable mess. That might work for a rescue (read embedded-like system with ulibc) system, but probably booting from a live media is much easier and more robust.
-- Leonid Isaev GnuPG key: 0x164B5A6D Fingerprint: C0DF 20D0 C075 C3F1 E1BE 775A A7AE F6CB 164B 5A6D
Meh it does have it's advantages. I've screwed up my system to a point where I can't use pacman (because the libraries it depends on have gone wild). Having a staticly compiled pacman would allow for me to use it no matter what. LiveCD is not always easier.... Calvin
On Fri, Aug 3, 2012 at 11:29 AM, Leonid Isaev <lisaev@umail.iu.edu> wrote:
On Fri, 3 Aug 2012 10:31:06 -0400 Jack Silver <jacksilver045@gmail.com> wrote:
To exchange information I want to let know this list that I have filled a feature request form to ask for a statically builded pacman.
https://bugs.archlinux.org/task/30993
Comments welcome in the bug manager.
جاك الفضة
Well, bugtracker is not a place for comments, it's for solutions.
Anyway... statically compiling things is not a way of avoiding trouble, at least not in a self-sustained fashion. So, if you propose to have pacman in [core] statically compiled against all needed libraries, I would be against that as the package will be an unmaintainable mess.
Why would it be a unmaintainable mess?
On Fri, 3 Aug 2012 11:35:10 -0500 Sander Jansen <s.jansen@gmail.com> wrote:
On Fri, Aug 3, 2012 at 11:29 AM, Leonid Isaev <lisaev@umail.iu.edu> wrote:
On Fri, 3 Aug 2012 10:31:06 -0400 Jack Silver <jacksilver045@gmail.com> wrote:
To exchange information I want to let know this list that I have filled a feature request form to ask for a statically builded pacman.
https://bugs.archlinux.org/task/30993
Comments welcome in the bug manager.
جاك الفضة
Well, bugtracker is not a place for comments, it's for solutions.
Anyway... statically compiling things is not a way of avoiding trouble, at least not in a self-sustained fashion. So, if you propose to have pacman in [core] statically compiled against all needed libraries, I would be against that as the package will be an unmaintainable mess.
Why would it be a unmaintainable mess?
Because it is _statically_ compiled so the whole binary has to be rebuilt even after a minor update of one of the libraries. This is assuming that you can actually make such binary with gcc... -- Leonid Isaev GnuPG key: 0x164B5A6D Fingerprint: C0DF 20D0 C075 C3F1 E1BE 775A A7AE F6CB 164B 5A6D
On 3 August 2012 13:03, Leonid Isaev <lisaev@umail.iu.edu> wrote:
On Fri, 3 Aug 2012 11:35:10 -0500 Sander Jansen <s.jansen@gmail.com> wrote:
On Fri, Aug 3, 2012 at 11:29 AM, Leonid Isaev <lisaev@umail.iu.edu> wrote:
On Fri, 3 Aug 2012 10:31:06 -0400 Jack Silver <jacksilver045@gmail.com> wrote:
To exchange information I want to let know this list that I have filled a feature request form to ask for a statically builded pacman.
https://bugs.archlinux.org/task/30993
Comments welcome in the bug manager.
جاك الفضة
Well, bugtracker is not a place for comments, it's for solutions.
Anyway... statically compiling things is not a way of avoiding trouble, at least not in a self-sustained fashion. So, if you propose to have pacman in [core] statically compiled against all needed libraries, I would be against that as the package will be an unmaintainable mess.
Why would it be a unmaintainable mess?
Because it is _statically_ compiled so the whole binary has to be rebuilt even after a minor update of one of the libraries. This is assuming that you can actually make such binary with gcc...
No. It only needs to be recompile when the compiler feels like it. If the perceived benefit of the newer library to link against is greater than the time and energy it takes to recompile and package a product, then the compiler won't do it. If curl does a minor bump fixing a function that pacman doesn't even use for example, we then we probably wouldn't bother. Now if it was a critical update then yes, we would obviously do it. there is a whole discussion on why static linking is good on http://sta.li Calvin
On Fri, Aug 3, 2012 at 8:40 PM, Calvin Morrison <mutantturkey@gmail.com> wrote:
there is a whole discussion on why static linking is good on http://sta.li
Including such gems as: "Of course Ulrich Drepper thinks that dynamic linking is great, but clearly that’s because of his lack of experience and his delusions of grandeur." I'm fine with the minimalism of plan9, but I'd much rather have my system free from arguments ad hominem, even if I'm not sure if the former quote was serious or not. -- Mantas Mikulėnas
On 3 August 2012 14:22, Mantas Mikulėnas <grawity@gmail.com> wrote:
On Fri, Aug 3, 2012 at 8:40 PM, Calvin Morrison <mutantturkey@gmail.com> wrote:
there is a whole discussion on why static linking is good on http://sta.li
Including such gems as: "Of course Ulrich Drepper thinks that dynamic linking is great, but clearly that’s because of his lack of experience and his delusions of grandeur."
I'm fine with the minimalism of plan9, but I'd much rather have my system free from arguments ad hominem, even if I'm not sure if the former quote was serious or not.
-- Mantas Mikulėnas
Yes the motley crew of plan9/suckless/cat-v.org drip with irony and bad taste Calvin
On Fri, 3 Aug 2012 13:40:15 -0400 Calvin Morrison <mutantturkey@gmail.com> wrote:
On 3 August 2012 13:03, Leonid Isaev <lisaev@umail.iu.edu> wrote:
On Fri, 3 Aug 2012 11:35:10 -0500 Sander Jansen <s.jansen@gmail.com> wrote:
On Fri, Aug 3, 2012 at 11:29 AM, Leonid Isaev <lisaev@umail.iu.edu> wrote:
On Fri, 3 Aug 2012 10:31:06 -0400 Jack Silver <jacksilver045@gmail.com> wrote:
To exchange information I want to let know this list that I have filled a feature request form to ask for a statically builded pacman.
https://bugs.archlinux.org/task/30993
Comments welcome in the bug manager.
جاك الفضة
Well, bugtracker is not a place for comments, it's for solutions.
Anyway... statically compiling things is not a way of avoiding trouble, at least not in a self-sustained fashion. So, if you propose to have pacman in [core] statically compiled against all needed libraries, I would be against that as the package will be an unmaintainable mess.
Why would it be a unmaintainable mess?
Because it is _statically_ compiled so the whole binary has to be rebuilt even after a minor update of one of the libraries. This is assuming that you can actually make such binary with gcc...
No.
It only needs to be recompile when the compiler feels like it. If the perceived benefit of the newer library to link against is greater than the time and energy it takes to recompile and package a product, then the compiler won't do it.
If curl does a minor bump fixing a function that pacman doesn't even use for example, we then we probably wouldn't bother. Now if it was a critical update then yes, we would obviously do it.
Making this decision IS a maintainance effort...
there is a whole discussion on why static linking is good on http://sta.li
Calvin
-- Leonid Isaev GnuPG key: 0x164B5A6D Fingerprint: C0DF 20D0 C075 C3F1 E1BE 775A A7AE F6CB 164B 5A6D
there is a whole discussion on why static linking is good on http://sta.li
There is also a whole distribution that statically compiles everything and gives various arguments for doing so. Pacman depends on shell scripts too. I guess the location of those would also need to be considered. -- _______________________________________________________________________ 'Write programs that do one thing and do it well. Write programs to work together. Write programs to handle text streams, because that is a universal interface' (Doug McIlroy) _______________________________________________________________________
participants (6)
-
Calvin Morrison
-
Jack Silver
-
Kevin Chadwick
-
Leonid Isaev
-
Mantas Mikulėnas
-
Sander Jansen