[arch-dev-public] [signoff] new toolchain
I've updated both architectures to: gcc-libs-4.3.1-1 gcc-4.3.1-1 - minor bugfixes kernel-headers-2.6.25.6-1 glibc-2.8-1 - major bump binutils-2.18-7 - minor fixes tzdata-2008c-1 glibc doesn't ship tarballs anymore. we use cvs checkouts from the stable branch. so we get all latest bugfixes that went upstream. we do it the same way now for binutils. please test these packages a bit more intensive. some gcc related updates for packages in extra will follow soon. please signoff after testing. (my own "will OpenOffice build and run" test is still pending) When the toolchain has moved to core I plan to update "db" with another nice .so bump... -Andy glibc changes: Version 2.8 * New locales: bo_CN, bo_IN. * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8. * Sorting rules for some Indian languages (Devanagari and Gujarati). Implemented by Pravin Satpute. * IPV6 addresses in /etc/resolv.conf can now have a scope ID * nscd caches now all timeouts for DNS entries Implemented by Ulrich Drepper. * nscd is more efficient and wakes up less often. Implemented by Ulrich Drepper. * More checking functions: asprintf, dprintf, obstack_printf, vasprintf, vdprintf, and obstack_vprintf. Implemented by Jakub Jelinek. * Faster memset for x86-64. Implemented by Harsha Jagasia and H.J. Lu. * Faster memcpy on x86. Implemented by Ulrich Drepper. * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX). Implemented by Ulrich Drepper. * Faster sqrt and sqrtf implemention for some PPC variants. Implemented by Stephen Munroe.
On Wed, 11 Jun 2008, Andreas Radke wrote:
I've updated both architectures to:
gcc-libs-4.3.1-1 gcc-4.3.1-1 - minor bugfixes kernel-headers-2.6.25.6-1 glibc-2.8-1 - major bump binutils-2.18-7 - minor fixes tzdata-2008c-1
glibc doesn't ship tarballs anymore. we use cvs checkouts from the stable branch. so we get all latest bugfixes that went upstream. we do it the same way now for binutils. please test these packages a bit more intensive.
some gcc related updates for packages in extra will follow soon.
please signoff after testing.
(my own "will OpenOffice build and run" test is still pending)
When the toolchain has moved to core I plan to update "db" with another nice .so bump...
-Andy
I've just noticed that findutils (which is out-of-date, BTW) will need to be rebuilt against the glibc in testing. I could bug report it if you prefer or even fix it myself as tpowa is inactive. Eric -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
On Tue, 10 Jun 2008, Eric Belanger wrote:
On Wed, 11 Jun 2008, Andreas Radke wrote:
I've updated both architectures to:
gcc-libs-4.3.1-1 gcc-4.3.1-1 - minor bugfixes kernel-headers-2.6.25.6-1 glibc-2.8-1 - major bump binutils-2.18-7 - minor fixes tzdata-2008c-1
glibc doesn't ship tarballs anymore. we use cvs checkouts from the stable branch. so we get all latest bugfixes that went upstream. we do it the same way now for binutils. please test these packages a bit more intensive.
some gcc related updates for packages in extra will follow soon.
please signoff after testing.
(my own "will OpenOffice build and run" test is still pending)
When the toolchain has moved to core I plan to update "db" with another nice .so bump...
-Andy
I've just noticed that findutils (which is out-of-date, BTW) will need to be rebuilt against the glibc in testing. I could bug report it if you prefer or even fix it myself as tpowa is inactive.
Eric
On second thought, I'll just go ahead and add a rebuilt findutils in testing. I'll update it at the same time. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
(my own "will OpenOffice build and run" test is still pending)
done. got OOo 2.4.1 and den OOo 3.0 devel snapshot compiled and running with gcc-gcj. i had to add one fix again to glibc-2.8-2 to solve an old issue flooding the memory in certain situations. so from my view the toolchain is ready to move. i will start to build some more updates with it. i'm waiting for signoffs now. -Andy
On Wed, Jun 11, 2008 at 4:10 PM, Andreas Radke
(my own "will OpenOffice build and run" test is still pending)
done. got OOo 2.4.1 and den OOo 3.0 devel snapshot compiled and running with gcc-gcj.
i had to add one fix again to glibc-2.8-2 to solve an old issue flooding the memory in certain situations.
so from my view the toolchain is ready to move. i will start to build some more updates with it. i'm waiting for signoffs now.
Thanks Andy, I will test and signoff by Friday, but I have a quick question first (I trust your opinion here) - do you think these changes warrant a rebuild of [core] entirely? I got a personal email wondering the same - if you think it is a good idea, I'd be more than willing to do some of the work, but if you think it's useless, then we won't do it. (Anyway, off for the day. I will see your response tomorrow when I get back to work). Cheers, Aaron
Am Wed, 11 Jun 2008 17:50:57 -0500
schrieb "Aaron Griffin"
Thanks Andy, I will test and signoff by Friday, but I have a quick question first (I trust your opinion here) - do you think these changes warrant a rebuild of [core] entirely? I got a personal email wondering the same - if you think it is a good idea, I'd be more than willing to do some of the work, but if you think it's useless, then we won't do it.
We don't need to rebuild whole repositories with every major glibc/gcc update. It's recommended to gain some speed improvements but it's not a must. So far we did it only once with the core repo. This glibc release this time has some improvements for x86_64 and gcc4.3.x had some speedups at all. But nothing worth to spend much time on it. Once we have too much manpower or good scripts for using something like automated makeworld or even a dedicated build machine we can recompile even [extra] where some packages are still from 2004. Right now we should focus on other stuff. -Andy
On Wednesday 11 June 2008 00:38:28 Andreas Radke wrote:
please signoff after testing.
I just built KDE 4.0.82 without any problems. So here is my sign-off for x86_64. -- archlinux.de
On Fri, 13 Jun 2008, Pierre Schmitz wrote:
On Wednesday 11 June 2008 00:38:28 Andreas Radke wrote:
please signoff after testing.
I just built KDE 4.0.82 without any problems. So here is my sign-off for x86_64.
gimp-devel built fine on i686. So signing off for i686. Plus, I haven't noticed any problems with daily usage on both i686 and x86_64. BTW, I've also built findutils on both arches with the testing toolchain. Also, findutils will need to be moved to core with the new toolchain but it still needs a x86_64 signoff. Please test it and move it to core whenever you'll move the toolchain. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
participants (4)
-
Aaron Griffin
-
Andreas Radke
-
Eric Belanger
-
Pierre Schmitz