The following reply was made to PR port-mips/29395; it has been noted by GNATS.

From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: kilbi@rad.rwth-aachen.de
Cc: gnats-bugs@NetBSD.org, port-mips-maintainer@NetBSD.org,
	gnats-admin@NetBSD.org, tsutsui@ceres.dti.ne.jp
Subject: Re: port-mips/29395 (r5k (specific?) cache problem / data corruption)
Date: Wed, 22 Mar 2006 18:34:28 +0900

 > I've repeated my former stress test (mentioned within this PR) last
 > night:
 > 
 > I was no longer able to (re-)produce any (file) data corruption. So,
 > this PR could be closed according to your comment.
 
 Ok, thanks.
 
 > my Qube2 runs quite actual -current:
 > 
 >   NetBSD 3.99.17 (COCHISE) #2: Tue Mar 21 21:29:39 MET 2006
 > 
 > But: I've got several (silent) system hangups during this stress
 > testing (from my impression accompanied/correlated with the additional
 > network traffic load)!: No further message, no reaction on serial
 > console, could not get into ddb with break signal.
 
 Could you please mention more details of your environments
 and your "stress test"?
 i.e. it could also happen with GENERIC or not,
 any special options (config or mk.conf) on building your kernel,
 what devices your qube has (dmesg would help),
 what commands you tried on the test (whole log would be useful)
 etc.
 
 > Weren't there any complaints/PR about that kind of problem, yet?
 
 With a quick grance at gnats page:
 http://www.NetBSD.org/Gnats/
 no similar PR around mips ports.
 ---