xfs
[Top] [All Lists]

Re: [REVIEW #2] Update kernel code to compile in userspace (libxfs)

To: "Christoph Hellwig" <hch@xxxxxxxxxxxxx>
Subject: Re: [REVIEW #2] Update kernel code to compile in userspace (libxfs)
From: "Barry Naujok" <bnaujok@xxxxxxx>
Date: Mon, 01 Sep 2008 12:06:54 +1000
Cc: "xfs@xxxxxxxxxxx" <xfs@xxxxxxxxxxx>
In-reply-to: <20080901015950.GA16825@infradead.org>
Organization: SGI
References: <op.ugmy3u2o3jf8g2@pc-bnaujok.melbourne.sgi.com> <op.ugm1ttzb3jf8g2@pc-bnaujok.melbourne.sgi.com> <20080829181120.GA9455@infradead.org> <op.ugr4zpyv3jf8g2@pc-bnaujok.melbourne.sgi.com> <20080901015950.GA16825@infradead.org>
Sender: xfs-bounce@xxxxxxxxxxx
User-agent: Opera Mail/9.51 (Win32)
On Mon, 01 Sep 2008 11:59:50 +1000, Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote:

On Mon, Sep 01, 2008 at 11:56:03AM +1000, Barry Naujok wrote:
On Sat, 30 Aug 2008 04:11:20 +1000, Christoph Hellwig <hch@xxxxxxxxxxxxx>
wrote:


On Fri, Aug 29, 2008 at 05:59:43PM +1000, Barry Naujok wrote:
On Fri, 29 Aug 2008 17:00:56 +1000, Barry Naujok <bnaujok@xxxxxxx>
wrote:

Mostly header changes associated with #ifdef __KERNEL__ stuff.

Some function prototypes had to be reordered to separate
shared and kernel only functions.

fs/xfs/xfs_inode.h patch looked pretty horrid (and there was a missing declaration), here's a better version:

I still don't like it very much :) What about just moving the xfs_ictimestamp and xfs_icdinode defintions next to the ifork so that we can save one ifdef __KERNEL__? Also might be worth to move the xfs_iflock & co inlines before the end of that __KERNEL__ block to save another ifdef.

Ok, this now only has one blob of #ifdef __KERNEL__ in it now:

Looks good.


What about the rest ;)


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