Skip to contentSkip to conversation

fs: inode_set_flags() replace opencoded set_mask_bits()

PATCH In review Vineet Gupta 2 participants 3 messages Aug 7, 2015 VFS

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 fs: inode_set_flags() replace opencoded set_mask_bits()

It seems that 5f16f3225b0624 and 00a1a053ebe5, both with same commitlog ("ext4: atomically set inode->i_flags in ext4_set_inode_flags()") introduced the set_mask_bits API, but somehow missed not using it in ext4 in the end

Also, set_mask_bits is used in fs quite a bit and we can possibly come up with a generic llsc based implementation (w/o the cmpxchg loop)

Cc Alexander Viro <[email protected]>
Cc Theodore Ts'o <[email protected]>
Cc Peter Zijlstra (Intel) <[email protected]>
---
 fs/inode.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)
fs/inode.c +1 -7
index d30640f7a193..d892ff711615 100644
@@ -2005,13 +2005,7 @@ EXPORT_SYMBOL(inode_dio_wait);
 void inode_set_flags(struct inode *inode, unsigned int flags,
 		     unsigned int mask)
 {
-	unsigned int old_flags, new_flags;
-
 	WARN_ON_ONCE(flags & ~mask);
-	do {
-		old_flags = ACCESS_ONCE(inode->i_flags);
-		new_flags = (old_flags & ~mask) | flags;
-	} while (unlikely(cmpxchg(&inode->i_flags, old_flags,
-				  new_flags) != old_flags));
+	set_mask_bits(&inode->i_flags, mask, flags);
 }
 EXPORT_SYMBOL(inode_set_flags);
Signature · 1 line
1.9.1
#1 lore raw
Quoted 7 lines · Vineet Gupta

It seems that 5f16f3225b0624 and 00a1a053ebe5, both with same commitlog ("ext4: atomically set inode->i_flags in ext4_set_inode_flags()") introduced the set_mask_bits API, but somehow missed not using it in ext4 in the end

Also, set_mask_bits is used in fs quite a bit and we can possibly come up with a generic llsc based implementation (w/o the cmpxchg loop)

May I also suggest changing the return value of set_mask_bits() to old.

You can compute the new value given old, but you cannot compute the old value given new, therefore old is the better return value. Also, no current user seems to use the return value, so changing it is without risk.

#2 lore raw
Quoted 6 lines · Peter Zijlstra
Quoted 5 lines · Vineet Gupta
Quoted 5 lines

It seems that 5f16f3225b0624 and 00a1a053ebe5, both with same commitlog ("ext4: atomically set inode->i_flags in ext4_set_inode_flags()") introduced the set_mask_bits API, but somehow missed not using it in ext4 in the end

Quoted 8 lines
Quoted 2 lines
Quoted 2 lines

Also, set_mask_bits is used in fs quite a bit and we can possibly come up with a generic llsc based implementation (w/o the cmpxchg loop)

May I also suggest changing the return value of set_mask_bits() to old.

You can compute the new value given old, but you cannot compute the old value given new, therefore old is the better return value. Also, no current user seems to use the return value, so changing it is without risk.

Makes sense - will do that early next week !