Tom Spindler --> current-users (2006-06-29 18:24:19 -0700):
> And although it's bad form to reply to my own mail...

That's why I reply to yours again instead of mine ;-)


> This definitely seems like a regression.

...which must have taken place in the last few days. I just compared
with running two different kernel - from June 25 and June 19
respectively - and the select(2) problem did not show up with the older
one. Here's a diff of the ident output from both kernels:

$ diff working_kernel non-working_kernel
<      $NetBSD: GENERIC,v 1.757 2006/06/16 22:13:35 jmcneill Exp $
>      $NetBSD: GENERIC,v 1.758 2006/06/19 15:44:44 gdamore Exp $
<      $NetBSD: cgd.c,v 1.35 2006/05/14 21:42:26 elad Exp $
>      $NetBSD: cgd.c,v 1.36 2006/06/20 03:20:44 christos Exp $
<      $NetBSD: com.c,v 1.244 2006/06/07 22:33:34 kardel Exp $
>      $NetBSD: com.c,v 1.245 2006/06/25 00:39:21 perry Exp $
<      $NetBSD: ext2fs_subr.c,v 1.23 2006/06/07 22:34:18 kardel Exp $
>      $NetBSD: ext2fs_subr.c,v 1.24 2006/06/23 14:13:02 yamt Exp $
<      $NetBSD: ffs_alloc.c,v 1.92 2006/06/07 22:34:19 kardel Exp $
>      $NetBSD: ffs_alloc.c,v 1.93 2006/06/23 14:13:02 yamt Exp $
<      $NetBSD: ffs_inode.c,v 1.82 2006/06/07 22:34:19 kardel Exp $
>      $NetBSD: ffs_inode.c,v 1.83 2006/06/23 14:13:02 yamt Exp $
<      $NetBSD: identcpu.c,v 1.32 2006/05/29 17:35:41 rpaulo Exp $
>      $NetBSD: identcpu.c,v 1.33 2006/06/20 03:23:09 christos Exp $
<      $NetBSD: init_main.c,v 1.268 2006/06/09 22:47:56 kardel Exp $
<      $NetBSD: init_sysctl.c,v 1.72 2006/06/17 06:54:58 yamt Exp $
>      $NetBSD: init_main.c,v 1.269 2006/06/25 08:00:01 yamt Exp $
>      $NetBSD: init_sysctl.c,v 1.74 2006/06/21 13:46:17 christos Exp $
<      $NetBSD: kern_proc.c,v 1.90 2006/06/11 07:32:18 rjs Exp $
>      $NetBSD: kern_proc.c,v 1.91 2006/06/25 08:12:54 yamt Exp $
<      $NetBSD: kern_sa.c,v 1.74 2006/06/11 07:32:18 rjs Exp $
>      $NetBSD: kern_sa.c,v 1.81 2006/06/25 08:13:28 yamt Exp $
<      $NetBSD: kern_synch.c,v 1.161 2006/05/14 21:15:11 elad Exp $
>      $NetBSD: kern_synch.c,v 1.162 2006/06/24 05:23:06 mrg Exp $
<      $NetBSD: kernfs_vnops.c,v 1.121 2006/06/07 22:33:41 kardel Exp $
>      $NetBSD: kernfs_vnops.c,v 1.125 2006/06/23 20:54:21 christos Exp $
<      $NetBSD: lfs_itimes.c,v 1.9 2006/06/07 22:34:43 kardel Exp $
<      $NetBSD: lfs_segment.c,v 1.182 2006/06/07 22:34:43 kardel Exp $
>      $NetBSD: lfs_itimes.c,v 1.10 2006/06/23 14:13:02 yamt Exp $
>      $NetBSD: lfs_segment.c,v 1.184 2006/06/24 05:28:54 perseant Exp $
<      $NetBSD: lfs_vnops.c,v 1.178 2006/05/18 23:15:09 perseant Exp $
>      $NetBSD: lfs_vnops.c,v 1.179 2006/06/24 05:28:54 perseant Exp $
<      $NetBSD: pci_machdep.c,v 1.14 2006/02/07 20:38:43 bouyer Exp $
>      $NetBSD: pci_machdep.c,v 1.15 2006/06/25 03:46:56 soren Exp $
<      $NetBSD: procfs_linux.c,v 1.24 2005/12/11 12:24:51 christos Exp $
>      $NetBSD: procfs_linux.c,v 1.25 2006/06/24 16:34:02 christos Exp $
<      $NetBSD: ptyfs_vfsops.c,v 1.15 2006/05/14 21:31:52 elad Exp $
>      $NetBSD: ptyfs_vfsops.c,v 1.17 2006/06/20 09:17:14 tron Exp $
<      $NetBSD: subr_callback.c,v 1.2 2006/05/27 07:42:42 yamt Exp $
>      $NetBSD: subr_callback.c,v 1.4 2006/06/23 16:17:23 drochner Exp $
>      $NetBSD: subr_kmem.c,v 1.2 2006/06/25 08:10:04 yamt Exp $
>      $NetBSD: subr_vmem.c,v 1.1 2006/06/25 08:00:01 yamt Exp $
<      $NetBSD: ufs_lookup.c,v 1.77 2006/06/07 22:34:43 kardel Exp $
>      $NetBSD: ufs_lookup.c,v 1.78 2006/06/23 14:13:02 yamt Exp $
<      $NetBSD: uhci.c,v 1.196 2006/05/12 01:25:00 mrg Exp $
>      $NetBSD: uhci.c,v 1.197 2006/06/19 15:44:45 gdamore Exp $
<      $NetBSD: uipc_socket.c,v 1.120 2006/06/13 21:19:56 ginsbach Exp $
<      $NetBSD: uipc_socket2.c,v 1.71 2006/05/14 21:15:12 elad Exp $
>      $NetBSD: uipc_socket.c,v 1.121 2006/06/21 12:55:12 yamt Exp $
>      $NetBSD: uipc_socket2.c,v 1.72 2006/06/21 12:55:12 yamt Exp $
<      $NetBSD: uvm_amap.c,v 1.73 2006/04/21 14:04:45 yamt Exp $
>      $NetBSD: uvm_amap.c,v 1.74 2006/06/25 08:03:46 yamt Exp $
<      $NetBSD: uvm_map.c,v 1.226 2006/05/25 14:27:28 yamt Exp $
>      $NetBSD: uvm_map.c,v 1.227 2006/06/25 08:03:46 yamt Exp $
<      $NetBSD: vfs_subr.c,v 1.266 2006/05/14 21:15:12 elad Exp $
>      $NetBSD: vfs_subr.c,v 1.267 2006/06/23 14:13:02 yamt Exp $
<      $NetBSD: wd.c,v 1.325 2006/06/05 18:22:23 bouyer Exp $
>      $NetBSD: wd.c,v 1.326 2006/06/24 04:49:40 gendalia Exp $

So, which one is the culprit?


Regards, Jukka

-- 
bashian roulette:
$ ((RANDOM%6)) || rm -rf ~