> > > > doesn't it make write(2) and mmap'ed accesses update atime
> > > > in somewhat unpredictable manner?
> > > 
> > > For mmap'ed accesses, probably. But I don't think this is a problem,
> > > because if we're faulting in a page via mmap(), we're reading the file,
> > > isn't it ?
> > 
> > i agree that it shouldn't be a problem, because it's
> > the pre-ubc behaviour, iirc.
> 
> Ha yes, so this should be restored too.

i'm not sure if it should.
i have not heard about any actual problem due to it.
besides, netbsd is not only os which doesn't update atime for mmap.

> > > Hum, I just tested. Yes the atime is updated on a write, and testing
> > > PGO_OVERWRITE doesn't fix the problem :(
> > 
> > i think testing access_type works for -current.
> > it doesn't work for netbsd-2, though.
> 
> Hum, I tried this (see attached patch), and atime is still updated when writing
> to an existing file (my test is: echo >> some_file).

due to VOP_GETPAGES called by ufs_balloc_range?

YAMAMOTO Takashi