On Tue, 2003-10-28 at 07:45, Matt Chorman wrote: > On Monday 27 October 2003 08:35 pm, you wrote: > > On Tue, 2003-10-28 at 06:21, Matt Chorman wrote: > > > On Monday 27 October 2003 08:09 pm, Martin Schlemmer wrote: > > > > On Tue, 2003-10-28 at 01:46, Matt Chorman wrote: > > > > > My experience with ntpl has been exaclty like this - even downgrading > > > > > did not help. Try recompiling glibc WITHOUT use=ntpl. Your binaries > > > > > will (should?) start to work again. Mine did.... > > > > > > > > What CPU is in this box ? > > > > > > This is running an AthlonXP 2000+.. Ye Olde AMD chips... ;-) > > > > What gcc/binutils ? > > I initially tried with gcc 3.3.1-r5, binutils 2.14.90.0.6-r6, glibc 2.3.2-r6. > After seeing the thread on the stability of nptl this AM I decided to try > re-emerging glibc with USE=nptl (famous last words). I sync'd and emerge'd > glibc-2.3.2-r8 (with nptl). I dropped out of X and went to get back in when I > started getting segfaults. I decided to bump gcc to 3.3.2-r1 and binutils to > 2.14.90.0.6-r7 (with the new glibc). I was still getting segfaults, so I > emerged glibc with the new gcc/binutils and the segfaults were still there. > Tried again by going back to glibc-2.3.2-r7 (with nptl) - no joy. I removed > nptl from the use flags, re-emerged glibc-2.3.2-r8 and voila, everything is > back to normal. In the process, I did try to recompile xfree but there was no > change. I also use prelink - so somewhere during this process I ran a prelink > -ua but still had segfaults. >
What kernel ? > Is there a bug open for this that I should be posting at? Need any more info? > Not that I know of. Thanks, -- Martin Schlemmer Gentoo Linux Developer, Desktop/System Team Developer Cape Town, South Africa
signature.asc
Description: This is a digitally signed message part
