Subject: Re: kern/35351 (Re: CVS commit: src/sys/kern)
To: None <kern-bug-people@netbsd.org, gnats-admin@netbsd.org,>
From: YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp>
List: netbsd-bugs
Date: 04/07/2007 16:00:06
The following reply was made to PR kern/35351; it has been noted by GNATS.
From: yamt@mwd.biglobe.ne.jp (YAMAMOTO Takashi)
To: e@murder.org
Cc: rmind@NetBSD.org, tech-kern@netbsd.org, gnats-bugs@netbsd.org
Subject: Re: kern/35351 (Re: CVS commit: src/sys/kern)
Date: Sun, 8 Apr 2007 00:58:32 +0900 (JST)
[ remove core@ and source-changes@, add tech-kern@ ]
> YAMAMOTO Takashi wrote:
> >> YAMAMOTO Takashi wrote:
> >>>>> ok, then you mean not to fix it for netbsd-4?
> >>>>>
> >>>>> YAMAMOTO Takashi
> >>>> no, sorry. it's not a problem for netbsd-4.
> >>>>
> >>>> -e.
> >>> i don't understand why it isn't a problem. can you explain?
> >>>
> >>> YAMAMOTO Takashi
> >> I think I already did.
> >
> > sorry, where/when?
> >
> >> can you tell me why it *is* a problem?
> >
> > because, afaik, there is no difference wrt this problem between
> > netbsd-4 and current.
> >
> > YAMAMOTO Takashi
>
> are we talking about the pr itself or the null dereference commit?
i'm talking about the PR.
> whichever one it is, I'd like to see a scenario of where it can be
> a problem.
an example which i told you some time ago:
1. thread A calls fileassoc_file_add and sleeps in malloc.
2. thread B calls fileassoc_file_add and successfully allocates and inserts
a new entry.
3. thread A wakes up and inserts a duplicated entry.
do you mean it isn't a problem?
YAMAMOTO Takashi