userspace drivers

2003-01-19 Thread Robert Millan
Hi! I'd like to play a bit with userspace drivers, but i have some doubts on the interfaces: - how can i wait for a hardware interruption in a userspace process? - despite it's not necessary due to a well-known bug, how can i ask Mach for permission to access io ports? also, is there any ongoi

Bug#177486: POSIX semaphore

2003-01-19 Thread Robert Millan
Package: hurd Version: unavailable; reported 2003-01-20 Severity: normal Tags: patch This is just a reminder that a patch to provide posix semaphores, made by Michal 'hramrach' Suchanek, is available here: http://mail.gnu.org/archive/html/bug-hurd/2002-11/msg00316.html Atm it still applies clean

Re: Glibc sunrpc failure?

2003-01-19 Thread Jeff Bailey
On Sun, Jan 19, 2003 at 12:37:30PM -0800, Jeff Bailey wrote: > (gdb) print *stat_info > $5 = {st_fstype = 23, st_fsid = 63, st_ino = 130590, st_gen = 331987, > st_rdev = 142541374619648, st_mode = 1, st_nlink = 1000, st_uid = 1000, > st_gid = 2823, st_size = 4479340028585050112, st_atime = 99

Glibc sunrpc failure?

2003-01-19 Thread Jeff Bailey
I'm tracking down the glibc sunrpc failure, and here where I've gotten to (Mostly an update in case someone says "Hey that's so obvious" and can save me some hacking time): The actualy EOVERFLOW message is from xstat64_conv, which is seeing this for the st_size: st_size = 4479340028585050112 An

Hang while tracking sunrpc bug

2003-01-19 Thread Jeff Bailey
When stepping through glibc's sunrpc/rpcgen to figure out why stat is broken, I got this far: (gdb) step __io_stat (stat_object=4294967295, stat_info=0x10179e0) at /disks/7/cvstree/build.glibc/hurd/RPC_io_stat.c:71 71 register Request *InP = &Mess.In; (gdb) 72 regist