xfs
[Top] [All Lists]

Re: [PATCH 2/9] xfs: convert internal buffer functions to pass maps

To: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Subject: Re: [PATCH 2/9] xfs: convert internal buffer functions to pass maps
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Fri, 22 Jun 2012 16:48:54 +1000
Cc: xfs@xxxxxxxxxxx
In-reply-to: <20120620064824.GC14760@xxxxxxxxxxxxx>
References: <1339133914-11148-1-git-send-email-david@xxxxxxxxxxxxx> <1339133914-11148-3-git-send-email-david@xxxxxxxxxxxxx> <20120620064824.GC14760@xxxxxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Wed, Jun 20, 2012 at 02:48:24AM -0400, Christoph Hellwig wrote:
> > +STATIC int
> > +_xfs_buf_get_maps(
> 
> Should this really be STATIC which prevents inlining even if the
> compiler thinks that's a good idea?  (Same for the free side)

Not really, just habit. I'll change it.

> Also any reason for the underscore in the naming?

Just that other internal functions used it. I'll remove it.

> > +   bp->b_bn = map[0].bm_bn;
> > +   bp->b_length = 0;
> > +   for (i = 0; i < nmaps; i++) {
> > +           bp->b_maps[i].bm_bn = map[i].bm_bn;
> > +           bp->b_maps[i].bm_len = map[i].bm_len;
> > +           bp->b_length += map[i].bm_len;
> > +   }
> > +   bp->b_io_length = bp->b_length;
> > +
> 
> > +   for (i = 0; i < nmaps; i++) {
> > +           numblks += map[i].bm_len;
> > +   }
> 
> no need for the curly braces here.

 ok.

> 
> > +   struct xfs_buf  *bp,
> > +   int             map,
> > +   int             *buf_offset,
> > +   int             *count,
> > +   int             rw)
> >  {
> > -   int                     rw, map_i, total_nr_pages, nr_pages;
> > +   int                     page_index;
> 
> The indentation for the arguments and local variables doesn't seem to
> line up here.

Oops, my bad. Will fix.

-- 
Dave Chinner
david@xxxxxxxxxxxxx

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