Re: 2.6.5-mm1 [PATCH]

Re: 2.6.5-mm1 [PATCH]

[Ray Lee]

Hello,

Probably the least important email you'll receive all day, but...

> no-quota-inode-shrinkage.patch 
>  shrink inode when quota is disabled

Could I suggest an alternate version, below? It limits the knowledge of
the CONFIG_QUOTA option to the quota header file, and still shrinks the
inode by two pointers. The only functional difference between this and
Matt Mackall's version is the below will still leave in a call to
memset, but with a zero length. On the plus side, it keeps fs/inode.c
free of preprocessor noise, which seems worth the trade-off.

 quota.h |    4 ++++
 1 files changed, 4 insertions(+)

diff -NurX ../dontdiff linus-2.6/include/linux/quota.h linus-2.6-inode-shrinkage/include/linux/quota.h
--- linus-2.6/include/linux/quota.h	2004-04-03 08:46:35.000000000 -0800
+++ linus-2.6-inode-shrinkage/include/linux/quota.h	2004-04-03 08:45:19.000000000 -0800
@@ -57,7 +57,11 @@
 #define kb2qb(x) ((x) >> (QUOTABLOCK_BITS-10))
 #define toqb(x) (((x) + QUOTABLOCK_SIZE - 1) >> QUOTABLOCK_BITS)
 
+#ifdef CONFIG_QUOTA
 #define MAXQUOTAS 2
+#else
+#define MAXQUOTAS 0
+#endif
 
 #define USRQUOTA  0		/* element used for user quotas */
 #define GRPQUOTA  1		/* element used for group quotas */


--
Ray Lee

Re: 2.6.5-mm1 [PATCH]

[Andrew Morton]

Ray Lee <ray-lk@madrabbit.org> wrote:
>
> Could I suggest an alternate version, below? It limits the knowledge of
>  the CONFIG_QUOTA option to the quota header file, and still shrinks the
>  inode by two pointers. The only functional difference between this and
>  Matt Mackall's version is the below will still leave in a call to
>  memset, but with a zero length. On the plus side, it keeps fs/inode.c
>  free of preprocessor noise, which seems worth the trade-off.
> 
>   quota.h |    4 ++++
>   1 files changed, 4 insertions(+)
> 
>  diff -NurX ../dontdiff linus-2.6/include/linux/quota.h linus-2.6-inode-shrinkage/include/linux/quota.h
>  --- linus-2.6/include/linux/quota.h	2004-04-03 08:46:35.000000000 -0800
>  +++ linus-2.6-inode-shrinkage/include/linux/quota.h	2004-04-03 08:45:19.000000000 -0800
>  @@ -57,7 +57,11 @@
>   #define kb2qb(x) ((x) >> (QUOTABLOCK_BITS-10))
>   #define toqb(x) (((x) + QUOTABLOCK_SIZE - 1) >> QUOTABLOCK_BITS)
>   
>  +#ifdef CONFIG_QUOTA
>   #define MAXQUOTAS 2
>  +#else
>  +#define MAXQUOTAS 0
>  +#endif

The advantage of the ifdeffy one is that if someone accesses i_dquot
outside CONFIG_QUOTA, they get a compile failure rather than runtime inode
corruption.

Re: 2.6.5-mm1 [PATCH]

[Ray Lee]

On Sun, 2004-04-04 at 23:55, Andrew Morton wrote:
> The advantage of the ifdeffy one is that if someone accesses i_dquot
> outside CONFIG_QUOTA, they get a compile failure rather than runtime inode
> corruption.

Ah, that's useful. Point taken.