Date: Wed, 13 Oct 2004 14:10:36 +0300 From: Andriy Gapon <avg@icyb.net.ua> To: freebsd-emulation@freebsd.org Subject: Re: linux threaded application hang Message-ID: <416D0D2C.6020001@icyb.net.ua> In-Reply-To: <416CF40A.3070008@icyb.net.ua> References: <416CF40A.3070008@icyb.net.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
on 13.10.2004 12:23 Andriy Gapon said the following:
> (gdb) p set
> $1 = (sigset_t *) 0x482c87f0
> (gdb) p *set
> $2 = {__val = {1250168, 1208405312, 1208363760, 1209738430, 1209738446,
> 1210126092, 1209738478, 1209738494, 1210131344, 1209738526, 1209738542,
> 1209738558,
> 1209738574, 1209738590, 1209554352, 1209738622, 1209738638,
> 1209738654, 1209738670, 1209738686, 1209738702, 1209738718, 1209738734,
> 1209738750,
> 1209738766, 1209738782, 1209738798, 1210090948, 1209738830,
> 1209738846, 1209738862, 1209738878}}
>
> if I understand correctly linux sigset_t then mask for signal 32 should
> be bit 31 (highest) of __val[0]:
>
> 1250168 == 100110001001101111000
well it seems that frame 0 was actually reported wrong by gdb, or
something like that
> (gdb) p mask
> $3 = {__val = {0, 0, 0, 0, 0, 1208347125, 0, 3217021232, 0, 0, 0, 0, 0,
> 0, 3, 0, 0, 0, 0, 0, 16777216, 0, 1209525676, 102, 1209536638, 1209529244,
> 1209524044, 1208404144, 7, 1210900512, 3, 1210877936}}
and this is correct:
`ps -O sigmask` reported empty mask for initial thread, so it looks that
it was linux_kill that actually failed.
--
Andriy Gapon
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?416D0D2C.6020001>