Re: [arch-general] [arch-dev-public] [signoff] glibc-2.12 toolchain build
On 28/05/10 00:55, Jan de Groot wrote:
On Sat, 2010-05-22 at 20:45 +1000, Allan McRae wrote:
Hi,
Glibc-2.12 toolchain rebuild. Fedora has been running something like the glibc-2.12 release for the last few weeks so as always it should be nice and safe.
linux-api-headers-2.6.34-1 - update - fix license
glibc-2.12-1 - update - clean up old stuff from install file
binutils-2.20.1-3 - new cvs snapshot
gcc-4.5.0-3 - just a rebuild
valgrind - rebuild for glibc-2.12 - patched for glibc-2.12 and future 2.12.x versions - build fix patch
Signoff both, Allan
Signoff both. Didn't test valgrind, but building a full xorg-server stack on top of testing with this new toolchain should give proof that this toolchain works fine. Does gdb still work with this version of glibc btw, or does it need a rebuild just like valgrind?
I think gdb is fine... but I have not used it extensively. I'd appreciate reports from anyone who has extensively used gdb and valgrind with the new glibc. Allan
On Thu, May 27, 2010 at 11:03 AM, Allan McRae <allan@archlinux.org> wrote:
I think gdb is fine... but I have not used it extensively. I'd appreciate reports from anyone who has extensively used gdb and valgrind with the new glibc.
Allan
I've been using valgrind on the new glibc. It reports a whole new slew of warnings for previously "quiet" code, but otherwise performs as expected. Curl example: http://codepad.org/htorjWF9 dave
On Thu, May 27, 2010 at 12:22 PM, dave reisner <d@falconindy.com> wrote:
On Thu, May 27, 2010 at 11:03 AM, Allan McRae <allan@archlinux.org> wrote:
I think gdb is fine... but I have not used it extensively. I'd appreciate reports from anyone who has extensively used gdb and valgrind with the new glibc.
Allan
I've been using valgrind on the new glibc. It reports a whole new slew of warnings for previously "quiet" code, but otherwise performs as expected.
Curl example: http://codepad.org/htorjWF9
dave
Followup: I'm not able to find a whole lot of hard evidence from other people about this behavior aside from a Cygwin related post [1] which I can't personally reproduce. I've opened FS#19634 with a possible resolution. [1] http://www.cygwin.com/ml/libc-alpha/2010-05/msg00024.html d
participants (2)
-
Allan McRae
-
dave reisner