xfs
[Top] [All Lists]

Re: [PATCH v2 01/12] xfs: add a helper to figure out the space log reser

To: Jeff Liu <jeff.liu@xxxxxxxxxx>
Subject: Re: [PATCH v2 01/12] xfs: add a helper to figure out the space log reservation per item
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Sun, 20 Jan 2013 09:43:54 +1100
Cc: Mark Tinguely <tinguely@xxxxxxx>, xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <50FA12D2.7090209@xxxxxxxxxx>
References: <50EEC68A.1030306@xxxxxxxxxx> <50F9D4A5.6050800@xxxxxxx> <50FA12D2.7090209@xxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Sat, Jan 19, 2013 at 11:28:18AM +0800, Jeff Liu wrote:
> On 01/19/2013 07:03 AM, Mark Tinguely wrote:
> > On 01/10/13 07:47, Jeff Liu wrote:
> >> Add a helper xfs_calc_buf_res() to calcuate out the space log reservations 
> >> per item which were got
> >> involved in a transaction.  It will call the xfs_buf_log_overhead() to 
> >> figure out the extra space
> >> for the struct xfs_buf_log_format that gets written into the log for every 
> >> buffer as well as a little
> >> bit of extra space because there also needs a log opheader, i.e. struct 
> >> xlog_op_header.
> >>
> > The commit message is too wide.
> > 
> >> Signed-off-by: Jie Liu<jeff.liu@xxxxxxxxxx>
> >> ---
> >>   fs/xfs/xfs_trans.c |   31 +++++++++++++++++++++++++++++++
> >>   1 file changed, 31 insertions(+)
> >>
> >> diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
> >> index 06ed520..6c601ea 100644
> >> --- a/fs/xfs/xfs_trans.c
> >> +++ b/fs/xfs/xfs_trans.c
> >> @@ -37,14 +37,45 @@
> >>   #include "xfs_extent_busy.h"
> >>   #include "xfs_bmap.h"
> >>   #include "xfs_quota.h"
> >> +#include "xfs_qm.h"
> >>   #include "xfs_trans_priv.h"
> >>   #include "xfs_trans_space.h"
> >>   #include "xfs_inode_item.h"
> >> +#include "xfs_log_priv.h"
> >> +#include "xfs_buf_item.h"
> >>   #include "xfs_trace.h"
> >>
> >>   kmem_zone_t      *xfs_trans_zone;
> >>   kmem_zone_t      *xfs_log_item_desc_zone;
> >>
> >> +/*
> >> + * A buffer has a format structure overhead in the log in addition
> >> + * to the data, so we need to take this into account when reserving
> >> + * space in a transaction for a buffer.  Round the space required up
> >> + * to a multiple of 128 bytes so that we don't change the historical
> >> + * reservation that has been used for this overhead.
> >> + */
> >> +STATIC uint
> >> +xfs_buf_log_overhead(void)
> >> +{
> >> +  return round_up(sizeof(struct xlog_op_header) +
> >> +                  sizeof(struct xfs_buf_log_format), 128);
> >> +}
> > 
> > Not used directly except below. inline or fold below?
> I should be inline.  Maybe it's proper to make xfs_calc_buf_res() inline
> as well?

It's static. The compiler will inline it for us if appropriate.

Cheers,

Dave.
-- 
Dave Chinner
david@xxxxxxxxxxxxx

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