Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 25 Jul 2006 08:22:47 +0200
From:      des@des.no (Dag-Erling =?iso-8859-1?Q?Sm=F8rgrav?=)
To:        John Baldwin <jhb@freebsd.org>
Cc:        Tom Rhodes <trhodes@freebsd.org>, src-committers@freebsd.org, cvs-src@freebsd.org, cvs-all@freebsd.org, Robert Watson <rwatson@freebsd.org>, Greg 'groggy' Lehey <grog@freebsd.org>, Xin LI <delphij@freebsd.org>
Subject:   Re: truss status
Message-ID:  <86hd16gq4o.fsf@xps.des.no>
In-Reply-To: <200607241342.38412.jhb@freebsd.org> (John Baldwin's message of "Mon, 24 Jul 2006 13:42:37 -0400")
References:  <200607210845.k6L8j0Ep026008@repoman.freebsd.org> <20060724000259.GE16230@wantadilla.lemis.com> <20060724130731.D44945@fledge.watson.org> <200607241342.38412.jhb@freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
John Baldwin <jhb@freebsd.org> writes:
> Well, the biggest thing I found when doing the PHOLD/P_WEXIT stuff that w=
as=20
> missing was PIOCWAIT and PIOCCONT (IIRC).  Specifically, we don't current=
ly=20
> have any ptrace() equivalent to that.

I'm not sure that's a problem.  I had at one point a version of truss,
rewritten from scratch, that worked with ptrace().  I never committed
it because there were issues with attaching to children of the traced
process; I tried to get rid of the reparenting stuff by having a
debugging hierarchy parallel to the process hierarchy (i.e. each
process has a real parent and a debugging parent, instead of the
current situation where a debugged process is reparented to the
debugging process, which makes it impossible to debug e.g. a shell),
this required major changes to exit1() and kern_wait() which I never
completed.  At some point I accidentally blew away the tree; this was
before we had Perforce, and I had no backup.

DES
--=20
Dag-Erling Sm=F8rgrav - des@des.no



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?86hd16gq4o.fsf>