Skip to contentSkip to conversation

: ext3 reservations: Renumber the ext3 reservations ioctls

PATCH In review Stephen Tweedie 2 participants 4 messages Sep 7, 2004 – Sep 9, 2004

lore
Share & tools
lore
#0 lore raw

PATCH 2/6 : ext3 reservations: Renumber the ext3 reservations ioctls

All ext2/3 ioctls apart from these ones use 'f' as the root char for their macro-generated ioctl numbers; make reservations consistent with this.

include/linux/ext3_fs.h renamed from include/linux/ext3_fs.h.=K0001=.orig +2 -2
@@ -235,8 +235,8 @@ struct ext3_new_group_data {
 #ifdef CONFIG_JBD_DEBUG
 #define EXT3_IOC_WAIT_FOR_READONLY	_IOR('f', 99, long)
 #endif
-#define EXT3_IOC_GETRSVSZ		_IOR('r', 1, long)
-#define EXT3_IOC_SETRSVSZ		_IOW('r', 2, long)
+#define EXT3_IOC_GETRSVSZ		_IOR('f', 5, long)
+#define EXT3_IOC_SETRSVSZ		_IOW('f', 6, long)
 
 /*
  * Structure of an inode on the disk
#1 lore raw
Quoted 3 lines · Stephen Tweedie

All ext2/3 ioctls apart from these ones use 'f' as the root char for their macro-generated ioctl numbers; make reservations consistent with this.

Maybe you could reuse the XFS reservation ioctls instead of adding another set? Having incompatible APIs for the same thing on different filesystems sounds like the wrong way to go.

#2 lore raw

Hi,

Quoted 3 lines · Christoph Hellwig

Maybe you could reuse the XFS reservation ioctls instead of adding another set? Having incompatible APIs for the same thing on different filesystems sounds like the wrong way to go.

I don't mind either way. But I just looked, and I think they are doing different things. If I'm reading the XFS bits right, the XFS ioctls actively reserve/free disk space; but the ext3 ones do nothing except set/query the size of the per-inode sliding reservation window.

So sounds like they are best kept separate for now.

Cheers,
Stephen

#3 lore raw
Quoted 8 lines · Stephen C. Tweedie

Hi,

Quoted 3 lines · Christoph Hellwig

Maybe you could reuse the XFS reservation ioctls instead of adding another set? Having incompatible APIs for the same thing on different filesystems sounds like the wrong way to go.

Quoted 6 lines

I don't mind either way. But I just looked, and I think they are doing different things. If I'm reading the XFS bits right, the XFS ioctls actively reserve/free disk space; but the ext3 ones do nothing except set/query the size of the per-inode sliding reservation window.

So sounds like they are best kept separate for now.

makes sense.