Subject: Re: nfs/ubc panic with bad pointers
To: None <>
From: YAMAMOTO Takashi <>
List: tech-kern
Date: 10/28/2002 10:37:07
i see. thanks.
so, there's inconsistency between ffs and nfs.
ffs truncates the file to original size when error.  but nfs doesn't.
maybe we should define what VOP_WRITE should do on error.


> YAMAMOTO Takashi <> writes:
> > btw, what's the meaning of do-while around uiomove in nfs_write?
> > it never loops, i think.  to make it similar with ffs version?
> It loops, since the size of data moved in a iteration may be limited
> by the size of ubc window.
> enami.