Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 21 May 2010 20:31:17 -0700
From:      Marcus Reid <marcus@blazingdot.com>
To:        Kevin Oberman <oberman@es.net>
Cc:        Nikolay Denev <ndenev@gmail.com>, freebsd-stable@freebsd.org
Subject:   Re: dtrace problem?
Message-ID:  <20100522033117.GB73888@blazingdot.com>
In-Reply-To: <20100522025444.8FE2F1CC3A@ptavv.es.net>
References:  <20100522014817.GA73888@blazingdot.com> <20100522025444.8FE2F1CC3A@ptavv.es.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, May 21, 2010 at 07:54:44PM -0700, Kevin Oberman wrote:
> > Date: Fri, 21 May 2010 18:48:17 -0700
> > From: Marcus Reid <marcus@blazingdot.com>
> > Sender: owner-freebsd-stable@freebsd.org
> > 
> > On Fri, May 21, 2010 at 09:22:17AM +0300, Nikolay Denev wrote:
> > > On May 21, 2010, at 9:03 AM, Marcus Reid wrote:
> > > 
> > > > Hi,
> > > > 
> > > > Running a recent RELENG_8 (FreeBSD 8.1-PRERELEASE #0: Tue May 18 23:37:37
> > > > PDT 2010), I'm having a problem using dtrace.  For every .d file I attempt
> > > > to compile, I get:
> > > > 
> > > >  dtrace: failed to compile script test.d: "/usr/lib/dtrace/psinfo.d", line 37: syntax error near "uid_t"
> > > > 
> > > > This is my first attempt to use dtrace, so I can't be sure if it worked
> > > > before.  It happens with some scripts that can be assumed to be valid,
> > > > so it's not just me.  Is it broken for anyone else?
> > > > 
> > > > Thanks,
> > > > 
> > > > Marcus
> > > 
> > > Hi,
> > > 
> > > Have you rebuilt your kernel as described here : http://wiki.freebsd.org/DTrace
> > > 
> > > I was once getting "uid_t" errors when my kernel was not compiled with "WITH_CTF" option.
> > 
> > Yes, that would probably be it.  The handbook is explicit about
> > building with WITH_CTF=1, so I put it in make.conf.
> > 
> >   http://www.freebsd.org/doc/en/books/handbook/dtrace-enable.html
> > 
> > Thanks, I'll rebuild without it.
> 
> I believe WITH_CTF=1 would probably be placed in /etc/src.conf.

Ah, right you are.  That is, if world could be built with 'WITH_CTF=1'.
That appears to be where my breakage was; you have to build kernel with
it set but world without it.

Marcus



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100522033117.GB73888>