Yeah, this is basically just centered on fdesc doing the wrong thing.

Anything else should get it's own thread :-)

James

>
>On Thu, 22 Feb 2001, Jaromír Dolecek wrote:
>
>> Jaromír Dolecek wrote:
>> > I think the code in fdesc_open() for Fdesc case should mask off
>> > all flags other than O_RDONLY, O_WRONLY and O_RDWR. Note that currently,
>> > it does not change them at all.
>> 
>> Also one cosmetic - sys_fhopen() should call VOP_OPEN() _before_
>> it checks O_TRUNC.
>
>Why? sys_fhopen follows the exact same procedure as vn_open(), just w/o
>the namei() calls & setup.
>
>The desire was that fhopen(2) acted EXACTLY like open(2) except that (1)
>the file had to exist (O_CREAT makes no sense), and 2) you passed in a
>filehandle rather than a path. While I came into this discussion late, I
>haven't seen a reason why we should change sys_fhopen().
>
>If we change vn_open(), then yes we should also change sys_fhopen(). But I
>thought the discussion was talking about the fdesc filesystem instead. :-)
>
>Take care,
>
>Bill
>
>
>
>
>