xfs
[Top] [All Lists]

Re: [DMAPI] code error in dm_ip_to_handle()

To: Dean Roehrich <roehrich@xxxxxxx>
Subject: Re: [DMAPI] code error in dm_ip_to_handle()
From: Aurelien Degremont - Stagiaire <degremont@xxxxxxxxxxx>
Date: Tue, 17 May 2005 10:52:02 +0200
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <20050516195240.86BD04FE8A@xxxxxxxxxxxxxxxxxxxxxxxx>
References: <20050516195240.86BD04FE8A@xxxxxxxxxxxxxxxxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; fr-FR; rv:1.2.1) Gecko/20030225
Dean Roehrich a écrit:
From:  Aurelien Degremont - Stagiaire <degremont@xxxxxxxxxxx>


Here is a very small patch that change the parameter of fh_to_inode() and inode_to_fh().
I just replace the dm_fsfid_t by a dm_fid_t.


I've been using your patch today.  So far it's good.

Here's the rest of it, to completely remove dm_fsfid.  Do these
dm_copyin_handle() changes look right to you?


It was really a good idea to fix this since I found, in the dmapi user space library, in the parse_handle() function, that you copy the opaque buffer ha_fid in a xfs_fid2_t structure (argh ! an xfs reference in a code which is supposed to be independent :)). And from this xfs_fid2, the igen and ino are extracted even if the ha_fid is supposed to be opaque ! The igen and ino number should have been extracted using FS specific call (from fsys_vector).

But, now this is solved, fortunately :) since we used dm_fid_t.

(As dmapi.h is included in this file, maybe the XFS struct could be replaced by their DMAPI equivalent ?)

Aurélien


<Prev in Thread] Current Thread [Next in Thread>