Re: nss problems

2004-03-10 Thread Alfred M. Szmidt
- ftpfs segfaults when calling gethostbyname_r. According to Alfred that does not happen in his test program. I guess that test program did not use the stack that much... - Emacs hangs if it cannot find /etc/hosts to lookup `localhost'. (I found that out when crosshurd did n

ext2fs locking [PATCH]

2004-03-10 Thread Marco Gerards
Hi, This patch implements what Marcus and Roland proposed to fix the fatfs locking issue. This patch only deals with the reading problem of fatfs, the writing part will be a lot more work. I will do that later. Thanks, Marco 2004-03-08 Marco Gerards <[EMAIL PROTECTED]> * diskfs.h (d

Re: Core dump (semi) update

2004-03-10 Thread Alfred M. Szmidt
[sorry for the delay, busy with univ.] I feel really utterly stupid, I was using a old objdump that didn't have the needed tweak you posted (bfd groking lwpstatus_t.pr_fpreg). I still do not know what is wrong with gdb not being able to read the core dump, but here is the output from the objdump i

Re: nss problems

2004-03-10 Thread Roland McGrath
The stack should be huge in a single-threaded program, so I don't know why Emacs has trouble. For multithreaded programs, linuxthreads and NPTL define __libc_alloca_cutoff to grok their threads' stack limits. What we need to do is define that in the Hurd libc as well, so that cthreads-using progr

[Fwd: ext2fs patch for large stores, RC1+20040304]

2004-03-10 Thread Ognyan Kulev
bug-hurd ignored this mail. I hope that this time it won't. Original Message Subject: ext2fs patch for large stores, RC1+20040304 Date: Thu, 04 Mar 2004 17:14:01 +0200 From: Ognyan Kulev <[EMAIL PROTECTED]> Organization: Faculty of Mathematics and Informatics, University of Sofi