Bug#554574: libstdc++6: apt segfaults on hppa
On 05.11.2009 14:30, Domenico Andreoli wrote:
Package: libstdc++6
Version: 4.4.2-2
Severity: grave
Hi,
after this morning upgrade apt started to systematically segfault
on my hppa box, not even an update was allowed. i reported this on the
hppa porters list [0].
of course suspect no. 1 is the NPTL switch but downgrading eglibc to
2.10.1-3 or even -2 did not improve the situation. i also tried with
standard kernel 2.6.30-2-parisc-smp (at that moment i was running
2.6.32-rc5) but still nothing.
then i tried to rebuild apt with the latest stuff, just in case NPTL
switch introduced any compatibility issue but g++ died at least one
time during usual autoconf g++ usability tests.
finally i tried downgrading libstdc++6 to 4.4.2-1 and it looks like
the problem is gone. indeed this package was upgraded this morning.
in the meanwhile 2.6.32-rc6 had been built and booted nicely. i've
then been able to use apt again and re-upgraded eglibc to 2.10.1-5 and
everything still works.
upgrading g++-4.4& related stuff back to 4.4.2-2 not only reintroduces
the problem but it does not even get past the whole process dying before
the end. downgrading to 4.4.2-1 settles everything again.
frankly i do not know what to do next, besides trying to rebuild gcc-4.4
4.4.2-1 with latest eglibc to see if it is the culprit
or rebuild against eglibc-2.9. could you do this as a test?
or it is rather
the svn merge between gcc 4.4.2-1 and -2.
test results for 4.4.2-1:
http://gcc.gnu.org/ml/gcc-testresults/2009-10/msg01919.html
for 4.4.2-2:
http://gcc.gnu.org/ml/gcc-testresults/2009-11/msg00351.html
there are some differences, which are not seen in Dave's build:
http://gcc.gnu.org/ml/gcc-testresults/2009-11/msg00047.html
there are some parisc scpecific changes:
2009-10-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
Backport from mainline:
2009-08-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
* pa.md (reload_inhi, reload_outhi, reload_inqi, reload_outqi): New
patterns.
* pa.c (emit_move_sequence): Check if address of operand1 is valid
for mode mode of operand0 when doing secondary reload for SAR.
2009-10-20 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
Backport from mainline:
2009-10-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
PR target/41702
* pa.md (casesi): Use sign extended index in call to
gen_casesi64p.
(casesi64p): Update pattern to reflect above.
thanks,
Domenico
[0] http://lists.debian.org/debian-hppa/2009/11/msg00005.html
-- System Information:
Debian Release: squeeze/sid
APT prefers unstable
APT policy: (990, 'unstable'), (1, 'experimental')
Architecture: hppa (parisc)
Kernel: Linux 2.6.32-rc6 (SMP w/2 CPU cores)
Locale: LANG=C, LC_CTYPE=it_IT (charmap=ISO-8859-1)
Shell: /bin/sh linked to /bin/bash
Versions of packages libstdc++6 depends on:
hi gcc-4.4-base 4.4.2-1 The GNU Compiler Collection (base
ii libc6 2.10.1-5 GNU C Library: Shared libraries
hi libgcc4 4.4.2-1 GCC support library
Reply to: