Re: kern/43611: kernel-trace (via ktruss) reports wrong parameter/result values

[christos%zoulas.com@localhost (Christos Zoulas)]

The following reply was made to PR kern/43611; it has been noted by GNATS.

From: christos%zoulas.com@localhost (Christos Zoulas)
To: gnats-bugs%NetBSD.org@localhost, kern-bug-people%netbsd.org@localhost, 
        gnats-admin%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost, 
        Wolfgang.Stukenbrock%nagler-company.com@localhost
Cc: 
Subject: Re: kern/43611: kernel-trace (via ktruss) reports wrong 
parameter/result values
Date: Tue, 13 Jul 2010 13:42:06 -0400

 On Jul 13,  5:40pm, david%l8s.co.uk@localhost (David Laight) wrote:
 -- Subject: Re: kern/43611: kernel-trace (via ktruss) reports wrong parameter
 
 | The following reply was made to PR kern/43611; it has been noted by GNATS.
 | 
 | From: David Laight <david%l8s.co.uk@localhost>
 | To: gnats-bugs%NetBSD.org@localhost
 | Cc: 
 | Subject: Re: kern/43611: kernel-trace (via ktruss) reports wrong 
parameter/result values
 | Date: Tue, 13 Jul 2010 18:40:14 +0100
 | 
 |  On Tue, Jul 13, 2010 at 12:35:00PM +0000, 
Wolfgang.Stukenbrock%nagler-company.com@localhost wrote:
 |  > >Synopsis:       kernel-trace (via ktruss) reports wrong parameter/result 
values
 |  ...
 |  >   While debugging a tape-device problem we have traced the dd(1) command 
in order to validate the correct system calls.
 |  
 |  ktrace+kdump tends to work much better than ktruss.
 |  
 |  ktruss tries to convert the ktrace data to system call parameters and
 |  fails when the data is unexpected (due to extra traces).
 |  
 |  ktruss probably should be rewritten to use system call entry/exit
 |  breakpoints and then userspace controlled reads of the traced programs
 |  memory (which is how truss works on SYSV).
 
 Well, this is PT_SYSCALL and /usr/pkg/bin/strace :-)
 
 christos