Skip to contentSkip to conversation

ext3 IO latency measurements

DISCUSSION Bodo Eggert 1 participant 4 messages Mar 26, 2009 – Mar 27, 2009 20 messages not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

#12 lore raw
Quoted 3 lines · Linus Torvalds

brains, and we'd better change the default - and if some distro really _thinks_ about it, and decides that they really want old-fashioned atime, let them do that".

That's something a 2.7 kernel series might do - collect all sane defaults you wished to set all the time.

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

#16 lore raw

Re: [PATCH 1/2] Add a strictatime mount option

Quoted 3 lines · Matthew Garrett

Add support for explicitly requesting full atime updates. This makes it possible for kernels to default to relatime but still allow userspace to override it.

Maybe the *atime should be consolidated to atime=foo.

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

#22 lore raw

Re: [PATCH] Allow relatime to update atime once a day

Review

Quoted 6 lines · Matthew Garrett

--- a/fs/inode.c

(quoted diff) +4 -0
+
+     if (!relatime_need_update(mnt, inode, now))
+             goto out;
+

-> if (timespec_equal(&inode->i_atime, &now))
-> goto out;

timespec_equal is now duplicate, because:

Quoted 1 line

+static int relatime_need_update(struct vfsmount *mnt, struct inode *inode,

..

Quoted 2 lines
(quoted diff) +2 -0
+     if (timespec_compare(&inode->i_mtime, &inode->i_atime) >= 0)
+             return 1;
#23 lore raw

Re: [PATCH] Allow relatime to update atime once a day

In reply to a message not in the archive

fs/inode.c

Quoted 11 lines · Matthew Garrett
fs/inode.c +4 -1
index 0487ddb..057c92b 100644
--- a/fs/inode.c
 
 now = current_fs_time(inode->i_sb);
+
+     if (!relatime_need_update(mnt, inode, now))
+             goto out;
+
 if (timespec_equal(&inode->i_atime, &now))
 goto out;

Forget what I just said, I should rather read than assume.

But I'm wondering if inlining this once-used function would be a good thing, since relatime is supposed to be a common option? Otherwise, I'd pull the flags check out and avoid the function call.