...@openvz.org
Acked-by: Ian Kent ra...@themaw.net
I was working on this too.
I prefer this patch over my own.
Haven't tested it yet but it is very similar to the patch I was working
with which had seen some basic testing.
---
fs/namei.c |3 ++-
1 files changed, 2 insertions(+), 1 deletions
On Thu, 2014-07-17 at 18:04 +1000, NeilBrown wrote:
> On Thu, 17 Jul 2014 13:00:56 +0800 Ian Kent wrote:
>
> > On Wed, 2014-07-16 at 14:56 +0800, Ian Kent wrote:
> > > > That looks a bit messy ... I wonder if we could have a new "ino" flag
> > > &
On Thu, 2014-07-17 at 18:04 +1000, NeilBrown wrote:
On Thu, 17 Jul 2014 13:00:56 +0800 Ian Kent ra...@themaw.net wrote:
On Wed, 2014-07-16 at 14:56 +0800, Ian Kent wrote:
That looks a bit messy ... I wonder if we could have a new ino flag
which
says This dentry is mounted
On Wed, 2014-07-16 at 14:56 +0800, Ian Kent wrote:
> > That looks a bit messy ... I wonder if we could have a new "ino" flag which
> > says "This dentry is mounted-on if it needs to be. Gets set by ->lookup
> > and cleared by ->d_automount or when ->d
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
> ->fs_lock protects AUTOFS_INF_EXPIRING. We need to be sure
> that once the flag is set, no new references beneath the dentry
> are taken. So rcu-walk currently needs to take fs_lock before
> checking the flag. This hurts performance.
>
>
On Mon, 2014-07-14 at 10:53 +1000, NeilBrown wrote:
> Here is a revised version of this one patch.
> This one fixes a problem with refcounts on dentry and adds a comment to
> clarify the behaviour of should_expire().
I had some problems with this patch.
Looked like it got munged by the email
On Wed, 2014-07-16 at 16:00 +1000, NeilBrown wrote:
> On Wed, 16 Jul 2014 11:24:58 +0800 Ian Kent wrote:
>
> > On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
> > > autofs4 currently doesn't support RCU-walk - it immediately
> > > aborts any attempt at RCU-w
On Wed, 2014-07-16 at 15:51 +1000, NeilBrown wrote:
> On Wed, 16 Jul 2014 12:44:17 +0800 Ian Kent wrote:
>
> > On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
> > > Any attempt to look up a pathname that passes though an
> > > autofs4 mount is currently forced
On Wed, 2014-07-16 at 15:51 +1000, NeilBrown wrote:
On Wed, 16 Jul 2014 12:44:17 +0800 Ian Kent ra...@themaw.net wrote:
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
Any attempt to look up a pathname that passes though an
autofs4 mount is currently forced out of RCU-walk
On Wed, 2014-07-16 at 16:00 +1000, NeilBrown wrote:
On Wed, 16 Jul 2014 11:24:58 +0800 Ian Kent ra...@themaw.net wrote:
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
autofs4 currently doesn't support RCU-walk - it immediately
aborts any attempt at RCU-walk to force REF-walk
On Mon, 2014-07-14 at 10:53 +1000, NeilBrown wrote:
Here is a revised version of this one patch.
This one fixes a problem with refcounts on dentry and adds a comment to
clarify the behaviour of should_expire().
I had some problems with this patch.
Looked like it got munged by the email client.
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
-fs_lock protects AUTOFS_INF_EXPIRING. We need to be sure
that once the flag is set, no new references beneath the dentry
are taken. So rcu-walk currently needs to take fs_lock before
checking the flag. This hurts performance.
Change
On Wed, 2014-07-16 at 14:56 +0800, Ian Kent wrote:
That looks a bit messy ... I wonder if we could have a new ino flag which
says This dentry is mounted-on if it needs to be. Gets set by -lookup
and cleared by -d_automount or when -d_manage returns -EISDIR.
At one point
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
> Any attempt to look up a pathname that passes though an
> autofs4 mount is currently forced out of RCU-walk into
> REF-walk.
>
> This can significantly hurt performance of many-thread work
> loads on many-core systems, especially if the
tive() might benefit from a
similar addition. Multiple calls to ->lookup() made while the dentry is
still unhashed should have enforced ordering due to the directory
i_mutex so there shouldn't be a problem adding this. But perhaps you
haven't seen delays in that function.
Acked-by: Ian Ken
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
> Signed-off-by: NeilBrown
Obviously straight forward.
Acked-by: Ian Kent
> ---
> fs/autofs4/autofs_i.h | 14 --
> 1 file changed, 14 deletions(-)
>
> diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/
oduced in
> commit b5b801779d59165c4ecf1009009109545bd1f642
> autofs4: Add d_manage() dentry operation
>
> Signed-off-by: NeilBrown
Again, an obvious cleanup, thanks for that.
Acked-by: Ian Kent
> ---
> fs/autofs4/expire.c |1 -
> 1 file changed, 1 deletion(-)
>
>
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
> autofs4 currently doesn't support RCU-walk - it immediately
> aborts any attempt at RCU-walk to force REF-walk for path name
> lookup.
As discussed I don't have time to properly test these just now but I'll
do my best to review the patches and
On Tue, 2014-07-15 at 14:05 +1000, NeilBrown wrote:
> On Tue, 15 Jul 2014 11:48:33 +0800 Ian Kent wrote:
> ...
> > So far I've tracked this to something that was introduced between 3.11
> > and 3.12. One change that went into 3.12 was Jeff Laytons' umount
> > specific pa
On Tue, 2014-07-15 at 14:05 +1000, NeilBrown wrote:
On Tue, 15 Jul 2014 11:48:33 +0800 Ian Kent ra...@themaw.net wrote:
...
So far I've tracked this to something that was introduced between 3.11
and 3.12. One change that went into 3.12 was Jeff Laytons' umount
specific path resolution
b5b801779d59165c4ecf1009009109545bd1f642
autofs4: Add d_manage() dentry operation
Signed-off-by: NeilBrown ne...@suse.de
Again, an obvious cleanup, thanks for that.
Acked-by: Ian Kent ra...@themaw.net
---
fs/autofs4/expire.c |1 -
1 file changed, 1 deletion(-)
diff --git a/fs/autofs4/expire.c b
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
autofs4 currently doesn't support RCU-walk - it immediately
aborts any attempt at RCU-walk to force REF-walk for path name
lookup.
As discussed I don't have time to properly test these just now but I'll
do my best to review the patches and
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
Signed-off-by: NeilBrown ne...@suse.de
Obviously straight forward.
Acked-by: Ian Kent ra...@themaw.net
---
fs/autofs4/autofs_i.h | 14 --
1 file changed, 14 deletions(-)
diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4
that autofs4_lookup_active() might benefit from a
similar addition. Multiple calls to -lookup() made while the dentry is
still unhashed should have enforced ordering due to the directory
i_mutex so there shouldn't be a problem adding this. But perhaps you
haven't seen delays in that function.
Acked-by: Ian Kent ra
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
Any attempt to look up a pathname that passes though an
autofs4 mount is currently forced out of RCU-walk into
REF-walk.
This can significantly hurt performance of many-thread work
loads on many-core systems, especially if the automounted
On Mon, 2014-07-14 at 10:53 +1000, NeilBrown wrote:
> Here is a revised version of this one patch.
> This one fixes a problem with refcounts on dentry and adds a comment to
> clarify the behaviour of should_expire().
I have some bad news I'm afraid, not about the patches.
As I mentioned I'm well
On Mon, 2014-07-14 at 10:53 +1000, NeilBrown wrote:
Here is a revised version of this one patch.
This one fixes a problem with refcounts on dentry and adds a comment to
clarify the behaviour of should_expire().
I have some bad news I'm afraid, not about the patches.
As I mentioned I'm well
On Thu, 2014-07-10 at 18:25 +1000, NeilBrown wrote:
> On Thu, 10 Jul 2014 15:43:40 +0800 Ian Kent wrote:
>
> > On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
> > > autofs4 currently doesn't support RCU-walk - it immediately
> > > aborts any attempt at RCU-w
On Thu, 2014-07-10 at 15:43 +0800, Ian Kent wrote:
>
> Thanks for your effort Bruce,
Oh boy, where did that come from, sorry Neil, ;)
> Ian
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.o
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
> autofs4 currently doesn't support RCU-walk - it immediately
> aborts any attempt at RCU-walk to force REF-walk for path name
> lookup.
>
> This can cause a significant performance impact on multi-core
> systems.
> I have a client with a test
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
autofs4 currently doesn't support RCU-walk - it immediately
aborts any attempt at RCU-walk to force REF-walk for path name
lookup.
This can cause a significant performance impact on multi-core
systems.
I have a client with a test case
On Thu, 2014-07-10 at 15:43 +0800, Ian Kent wrote:
Thanks for your effort Bruce,
Oh boy, where did that come from, sorry Neil, ;)
Ian
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info
On Thu, 2014-07-10 at 18:25 +1000, NeilBrown wrote:
On Thu, 10 Jul 2014 15:43:40 +0800 Ian Kent ra...@themaw.net wrote:
On Thu, 2014-07-10 at 09:41 +1000, NeilBrown wrote:
autofs4 currently doesn't support RCU-walk - it immediately
aborts any attempt at RCU-walk to force REF-walk
On Wed, 2014-06-25 at 13:06 -0700, Andrew Morton wrote:
> On Wed, 25 Jun 2014 20:49:39 +0800 Ian Kent wrote:
>
> > On strict build environments we can see:
> >
> > fs/autofs4/inode.c: In function 'autofs4_fill_super':
> > fs/autofs4/inode.c:312: error: 'p
On Wed, 2014-06-25 at 08:15 +0200, Heiko Carstens wrote:
> On Tue, Jun 24, 2014 at 04:52:22PM -0700, David Rientjes wrote:
> > On Mon, 23 Jun 2014, Andrew Morton wrote:
> > > On Sat, 21 Jun 2014 11:10:58 +0200 Heiko Carstens
> > > wrote:
> > > > On Wed, Jun 18, 2014 at 02:29:31PM -0700, Andrew
for unfinished jobs
This is due to the use of pgrp_set being used to indicate pgrp has
has been set rather than initializing pgrp itself.
Signed-off-by: Ian Kent
---
fs/autofs4/inode.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
for unfinished jobs
This is due to the use of pgrp_set being used to indicate pgrp has
has been set rather than initializing pgrp itself.
Signed-off-by: Ian Kent ra...@themaw.net
---
fs/autofs4/inode.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/autofs4/inode.c b/fs
On Wed, 2014-06-25 at 08:15 +0200, Heiko Carstens wrote:
On Tue, Jun 24, 2014 at 04:52:22PM -0700, David Rientjes wrote:
On Mon, 23 Jun 2014, Andrew Morton wrote:
On Sat, 21 Jun 2014 11:10:58 +0200 Heiko Carstens
heiko.carst...@de.ibm.com wrote:
On Wed, Jun 18, 2014 at 02:29:31PM
On Wed, 2014-06-25 at 13:06 -0700, Andrew Morton wrote:
On Wed, 25 Jun 2014 20:49:39 +0800 Ian Kent ra...@themaw.net wrote:
On strict build environments we can see:
fs/autofs4/inode.c: In function 'autofs4_fill_super':
fs/autofs4/inode.c:312: error: 'pgrp' may be used uninitialized
On Wed, 2014-06-11 at 23:52 -0700, David Rientjes wrote:
> On Thu, 12 Jun 2014, Ian Kent wrote:
>
> > > > diff --git a/fs/seq_file.c b/fs/seq_file.c
> > > > index 1d641bb108d2..fca78a04c0d1 100644
> > > > --- a/fs/seq_file.c
> > > > +++ b/fs
On Wed, 2014-06-11 at 15:29 -0700, David Rientjes wrote:
> On Wed, 11 Jun 2014, Heiko Carstens wrote:
>
> > Alternatively we could also change the seqfile code to fall back to
> > vmalloc allocations. That would probably "fix" all single_open usages
> > where large contiguous memory areas are
On Wed, 2014-06-11 at 15:29 -0700, David Rientjes wrote:
On Wed, 11 Jun 2014, Heiko Carstens wrote:
Alternatively we could also change the seqfile code to fall back to
vmalloc allocations. That would probably fix all single_open usages
where large contiguous memory areas are needed and
On Wed, 2014-06-11 at 23:52 -0700, David Rientjes wrote:
On Thu, 12 Jun 2014, Ian Kent wrote:
diff --git a/fs/seq_file.c b/fs/seq_file.c
index 1d641bb108d2..fca78a04c0d1 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -8,8 +8,10 @@
#include linux/fs.h
#include
On Wed, 2014-05-28 at 15:37 -0700, Andrew Morton wrote:
> On Wed, 28 May 2014 11:01:53 +0200 Heiko Carstens
> wrote:
>
> > Now, /proc/stat uses single_open() for showing information. This means
> > the all data will be gathered and buffered once to a (big) buf.
> >
> > Now, /proc/stat shows
On Wed, 2014-05-28 at 15:37 -0700, Andrew Morton wrote:
On Wed, 28 May 2014 11:01:53 +0200 Heiko Carstens heiko.carst...@de.ibm.com
wrote:
Now, /proc/stat uses single_open() for showing information. This means
the all data will be gathered and buffered once to a (big) buf.
Now,
Hi all,
A lot of testing has been done to check that the amd format map
implementation doesn't break existing function. There's more to
do but it's time to release 5.1.0.
autofs
==
The package can be found at:
ftp://ftp.kernel.org/pub/linux/daemons/autofs/v5
It is autofs-5.1.0.tar.[gz|xz]
Hi all,
A lot of testing has been done to check that the amd format map
implementation doesn't break existing function. There's more to
do but it's time to release 5.1.0.
autofs
==
The package can be found at:
ftp://ftp.kernel.org/pub/linux/daemons/autofs/v5
It is autofs-5.1.0.tar.[gz|xz]
On Wed, 2014-05-28 at 19:06 +0800, Ian Kent wrote:
> On Wed, 2014-05-28 at 10:59 +0200, Heiko Carstens wrote:
> > The number of bytes contained 'within' /proc/stat depends on the number
> > of online cpus and not of the numbe of possible cpus.
> >
> > This reduces th
On Wed, 2014-05-28 at 10:59 +0200, Heiko Carstens wrote:
> The number of bytes contained 'within' /proc/stat depends on the number
> of online cpus and not of the numbe of possible cpus.
>
> This reduces the number of bytes requested for the initial buffer allocation
> within stat_open(). Which
On Wed, 2014-05-28 at 10:59 +0200, Heiko Carstens wrote:
The number of bytes contained 'within' /proc/stat depends on the number
of online cpus and not of the numbe of possible cpus.
This reduces the number of bytes requested for the initial buffer allocation
within stat_open(). Which is
On Wed, 2014-05-28 at 19:06 +0800, Ian Kent wrote:
On Wed, 2014-05-28 at 10:59 +0200, Heiko Carstens wrote:
The number of bytes contained 'within' /proc/stat depends on the number
of online cpus and not of the numbe of possible cpus.
This reduces the number of bytes requested
On Wed, May 21, 2014 at 02:25:21PM +0200, Heiko Carstens wrote:
> Hi all,
>
> I'm just wondering why /proc/stat is a single_open() seq_file and not a
> regular seq_file with an iterator (say 48 online cpus for each iteration
> or something similar).
>
> Of course, in theory, the "intr" line may
On Wed, May 21, 2014 at 02:25:21PM +0200, Heiko Carstens wrote:
Hi all,
I'm just wondering why /proc/stat is a single_open() seq_file and not a
regular seq_file with an iterator (say 48 online cpus for each iteration
or something similar).
Of course, in theory, the intr line may be very
On Wed, 2014-04-23 at 14:46 -0700, Andrew Morton wrote:
> On Wed, 23 Apr 2014 13:20:36 +0800 Ian Kent wrote:
>
> > autofs needs to be able to see private data dentry flags for
> > its dentrys that are being created but not yet hashed and for
> > its dentys that have be
On Wed, 2014-04-23 at 14:46 -0700, Andrew Morton wrote:
On Wed, 23 Apr 2014 13:20:36 +0800 Ian Kent ra...@themaw.net wrote:
autofs needs to be able to see private data dentry flags for
its dentrys that are being created but not yet hashed and for
its dentys that have been rmdir()ed
ion to a reference count of 0 before
being freed so autofs can occassionally use a dentry that
is invalid which can lead to a panic.
Signed-off-by: Ian Kent
---
fs/autofs4/root.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
in
to a reference count of 0 before
being freed so autofs can occassionally use a dentry that
is invalid which can lead to a panic.
Signed-off-by: Ian Kent ra...@themaw.net
---
fs/autofs4/root.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/autofs4/root.c b/fs/autofs4
From: Sasha Levin
There wasn't any check of the size passed from userspace before
trying to allocate the memory required.
This meant that userspace might request more space than allowed,
triggering an OOM.
Signed-off-by: Sasha Levin
Signed-off-by: Ian Kent
---
fs/autofs4/dev-ioctl.c |3
-by: Ian Kent ra...@themaw.net
---
fs/autofs4/dev-ioctl.c |3 +++
1 file changed, 3 insertions(+)
diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c
index 3182c0e..232e03d 100644
--- a/fs/autofs4/dev-ioctl.c
+++ b/fs/autofs4/dev-ioctl.c
@@ -103,6 +103,9 @@ static struct autofs_dev_ioctl
On Sat, 2014-04-05 at 23:11 -0400, Sasha Levin wrote:
> On 04/05/2014 11:03 PM, Ian Kent wrote:
> > On Sat, 2014-04-05 at 11:06 -0400, Sasha Levin wrote:
> >> Ping? Anyone wants to take this?
> >
> > Is this causing a problem for users?
> > If it is
On Sat, 2014-04-05 at 11:06 -0400, Sasha Levin wrote:
> Ping? Anyone wants to take this?
Is this causing a problem for users?
If it is I'll send it to Andrew straight away.
I do have this on my queue but don't have any other patches to send so
it's just sitting there.
If it worries you I'll
On Sat, 2014-04-05 at 11:06 -0400, Sasha Levin wrote:
Ping? Anyone wants to take this?
Is this causing a problem for users?
If it is I'll send it to Andrew straight away.
I do have this on my queue but don't have any other patches to send so
it's just sitting there.
If it worries you I'll send
On Sat, 2014-04-05 at 23:11 -0400, Sasha Levin wrote:
On 04/05/2014 11:03 PM, Ian Kent wrote:
On Sat, 2014-04-05 at 11:06 -0400, Sasha Levin wrote:
Ping? Anyone wants to take this?
Is this causing a problem for users?
If it is I'll send it to Andrew straight away.
I do have
Hi all,
This is an initial beta release which adds the ability of autofs
to understand the amd map format.
Making autofs understand the amd map format is a significant change
and, due to the way it's implemented, forms the basis for further
changes for autofs 5.1. However, most of these changes
Hi all,
This is an initial beta release which adds the ability of autofs
to understand the amd map format.
Making autofs understand the amd map format is a significant change
and, due to the way it's implemented, forms the basis for further
changes for autofs 5.1. However, most of these changes
Hi all,
Here is a bug fix release for autofs.
For those that have clones of the git repository the master branch
will progress to autofs-5.1.0 and autofs-5.0 will continue to be
updated with bug fixes (and periodic releases) on the new branch
autofs-5.0-stable.
For those that have git clones
Hi all,
Here is a bug fix release for autofs.
For those that have clones of the git repository the master branch
will progress to autofs-5.1.0 and autofs-5.0 will continue to be
updated with bug fixes (and periodic releases) on the new branch
autofs-5.0-stable.
For those that have git clones
On Fri, 2013-12-27 at 11:53 +0100, Sedat Dilek wrote:
> On Fri, Dec 27, 2013 at 5:32 AM, Ian Kent wrote:
>
> Hi,
>
> saw some typos...
Right, I'm pretty tired, and seem to be so a lot these days.
Andrew, let me fix the typos and re-submit the patch.
>
> > From: Ian K
On Fri, 2013-12-27 at 11:53 +0100, Sedat Dilek wrote:
On Fri, Dec 27, 2013 at 5:32 AM, Ian Kent ra...@themaw.net wrote:
Hi,
saw some typos...
Right, I'm pretty tired, and seem to be so a lot these days.
Andrew, let me fix the typos and re-submit the patch.
From: Ian Kent ik
This patch obsoletes both the current patches:
autofs-fix-symlinks-arent-checked-for-expiry.patch
autofs-fix-fix-symlinks-arent-checked-for-expiry.patch
It's a merge of both of these patches with an updated description.
---
Ian Kent (1):
autofs - fix symlinks aren't checked for expiry
From: Ian Kent
The autofs4 module doesn't consider symlinks for expire as it did
in the older autofs v3 module (so it's actually a long stnding
regression).
The user space daemon has focused on the use of bind mounts instead
of symlinks for a long time now and that's why this has not been
On Thu, 2013-12-26 at 17:19 -0800, Andrew Morton wrote:
> On Fri, 27 Dec 2013 09:09:52 +0800 Ian Kent wrote:
>
> > On Thu, 2013-12-26 at 13:42 -0800, Andrew Morton wrote:
> > > On Tue, 24 Dec 2013 17:44:59 +0800 Ian Kent wrote:
> > >
> > > > When
On Thu, 2013-12-26 at 13:42 -0800, Andrew Morton wrote:
> On Tue, 24 Dec 2013 17:44:59 +0800 Ian Kent wrote:
>
> > When following a symlink the last_used counter is unconditionally
> > updated causing the expire checks from user space to prevent
> > expiry. Opps!
>
On Thu, 2013-12-26 at 13:42 -0800, Andrew Morton wrote:
On Tue, 24 Dec 2013 17:44:59 +0800 Ian Kent ra...@themaw.net wrote:
When following a symlink the last_used counter is unconditionally
updated causing the expire checks from user space to prevent
expiry. Opps!
A bit unclear
On Thu, 2013-12-26 at 17:19 -0800, Andrew Morton wrote:
On Fri, 27 Dec 2013 09:09:52 +0800 Ian Kent ra...@themaw.net wrote:
On Thu, 2013-12-26 at 13:42 -0800, Andrew Morton wrote:
On Tue, 24 Dec 2013 17:44:59 +0800 Ian Kent ra...@themaw.net wrote:
When following a symlink
This patch obsoletes both the current patches:
autofs-fix-symlinks-arent-checked-for-expiry.patch
autofs-fix-fix-symlinks-arent-checked-for-expiry.patch
It's a merge of both of these patches with an updated description.
---
Ian Kent (1):
autofs - fix symlinks aren't checked for expiry
From: Ian Kent ik...@redhat.com
The autofs4 module doesn't consider symlinks for expire as it did
in the older autofs v3 module (so it's actually a long stnding
regression).
The user space daemon has focused on the use of bind mounts instead
of symlinks for a long time now and that's why
When following a symlink the last_used counter is unconditionally
updated causing the expire checks from user space to prevent
expiry. Opps!
Signed-off-by: Ian Kent
---
fs/autofs4/symlink.c |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/fs/autofs4/symlink.c b/fs
When following a symlink the last_used counter is unconditionally
updated causing the expire checks from user space to prevent
expiry. Opps!
Signed-off-by: Ian Kent ra...@themaw.net
---
fs/autofs4/symlink.c |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/fs/autofs4
From: Rui Xiang
Use the helper macro !IS_ROOT to replace parent != dentry->d_parent.
Just clean up.
Signed-off-by: Rui Xiang
Acked-by: Ian Kent
---
fs/autofs4/root.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
in
From: Rui Xiang
While kzallocing sbi/ino fails, it should return -ENOMEM.
And it should return the err value from autofs_prepare_pipe.
Signed-off-by: Rui Xiang
Acked-by: Ian Kent
---
fs/autofs4/inode.c | 13 -
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git a/fs
Hi Andrew,
Here are a couple of fixes for autofs.
The first patch fixes a couple of potential incorrect error returns
from autofs4_fill_super() so it's probably worthwhile to mege this
sooner rather than later.
The later is purely an improvement to using the IS_ROOT() macro.
---
Rui Xiang
From: Rui Xiang rui.xi...@huawei.com
While kzallocing sbi/ino fails, it should return -ENOMEM.
And it should return the err value from autofs_prepare_pipe.
Signed-off-by: Rui Xiang rui.xi...@huawei.com
Acked-by: Ian Kent ra...@themaw.net
---
fs/autofs4/inode.c | 13 -
1 file
Hi Andrew,
Here are a couple of fixes for autofs.
The first patch fixes a couple of potential incorrect error returns
from autofs4_fill_super() so it's probably worthwhile to mege this
sooner rather than later.
The later is purely an improvement to using the IS_ROOT() macro.
---
Rui Xiang
From: Rui Xiang rui.xi...@huawei.com
Use the helper macro !IS_ROOT to replace parent != dentry-d_parent.
Just clean up.
Signed-off-by: Rui Xiang rui.xi...@huawei.com
Acked-by: Ian Kent ra...@themaw.net
---
fs/autofs4/root.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff
From: Ian Kent
The autofs4 module doesn't consider symlinks for expire.
The user space daemon has focused on the use of bind mounts
instead of symlinks for a long time now. But, with the
future addition of amd map parsing in automount(8), not
to mention amd iteslf, symlink expiry is needed
From: Ian Kent ik...@redhat.com
The autofs4 module doesn't consider symlinks for expire.
The user space daemon has focused on the use of bind mounts
instead of symlinks for a long time now. But, with the
future addition of amd map parsing in automount(8), not
to mention amd iteslf, symlink
On Sat, 2013-11-16 at 17:03 +0100, Oleg Nesterov wrote:
> On 11/15, Andrew Morton wrote:
> >
> > Enable autofs4 to work in a "container". oz_pgrp is converted from pid_t
> > to struct pid and this is stored at mount time based on the "pgrp=" option
> > or if the option is missing then the current
On Sat, 2013-11-16 at 17:03 +0100, Oleg Nesterov wrote:
On 11/15, Andrew Morton wrote:
Enable autofs4 to work in a container. oz_pgrp is converted from pid_t
to struct pid and this is stored at mount time based on the pgrp= option
or if the option is missing then the current pgrp.
I
Hi all,
Once again this is long overdue so there are many changes.
The kernel patches are not being updated any more because of the
introduction of the vfs-automount changes to the VFS. If there
are needs in this area we will need to discuss how to deal with
them on the mailing list.
Known
Hi all,
Once again this is long overdue so there are many changes.
The kernel patches are not being updated any more because of the
introduction of the vfs-automount changes to the VFS. If there
are needs in this area we will need to discuss how to deal with
them on the mailing list.
Known
On Fri, 2013-10-11 at 07:29 -0600, David Ahern wrote:
> On 10/11/13 3:55 AM, Ian Kent wrote:
> > On Fri, 2013-10-11 at 10:06 +0800, Ian Kent wrote:
> >> On Thu, 2013-10-10 at 17:22 -0600, David Ahern wrote:
> >>> Running 3.12-rc3 just hit BUG in autofs4_expire_wait
On Fri, 2013-10-11 at 10:06 +0800, Ian Kent wrote:
> On Thu, 2013-10-10 at 17:22 -0600, David Ahern wrote:
> > Running 3.12-rc3 just hit BUG in autofs4_expire_wait
>
> It doesn't look like this could be due to Al's change to the locking in
> autos4_wait() and that the onl
On Fri, 2013-10-11 at 07:29 -0600, David Ahern wrote:
On 10/11/13 3:55 AM, Ian Kent wrote:
On Fri, 2013-10-11 at 10:06 +0800, Ian Kent wrote:
On Thu, 2013-10-10 at 17:22 -0600, David Ahern wrote:
Running 3.12-rc3 just hit BUG in autofs4_expire_wait
It doesn't look like this could be due
On Fri, 2013-10-11 at 10:06 +0800, Ian Kent wrote:
On Thu, 2013-10-10 at 17:22 -0600, David Ahern wrote:
Running 3.12-rc3 just hit BUG in autofs4_expire_wait
It doesn't look like this could be due to Al's change to the locking in
autos4_wait() and that the only change to autofs that I'm
On Thu, 2013-10-10 at 17:22 -0600, David Ahern wrote:
> Running 3.12-rc3 just hit BUG in autofs4_expire_wait
It doesn't look like this could be due to Al's change to the locking in
autos4_wait() and that the only change to autofs that I'm aware of.
Could you do a bisect please?
>
>
On Thu, 2013-10-10 at 17:22 -0600, David Ahern wrote:
Running 3.12-rc3 just hit BUG in autofs4_expire_wait
It doesn't look like this could be due to Al's change to the locking in
autos4_wait() and that the only change to autofs that I'm aware of.
Could you do a bisect please?
On Mon, 2013-09-09 at 06:45 -0400, Jeff Layton wrote:
> On Mon, 09 Sep 2013 15:18:00 +0800
> Ian Kent wrote:
>
> > On Sun, 2013-09-08 at 07:33 -0400, Jeff Layton wrote:
> > > On Sun, 08 Sep 2013 16:47:23 +0800
> > > Ian Kent wrote:
> > >
> >
On Mon, 2013-09-09 at 12:33 +1000, Stephen Rothwell wrote:
> Hi Al,
>
> After merging the vfs tree, today's linux-next build (powerpc
> ppc64_defconfig) failed like this:
>
> fs/autofs4/dev-ioctl.c: In function 'find_autofs_mount':
> fs/autofs4/dev-ioctl.c:193:2: error: implicit declaration of
On Sun, 2013-09-08 at 07:33 -0400, Jeff Layton wrote:
> On Sun, 08 Sep 2013 16:47:23 +0800
> Ian Kent wrote:
>
> > When reconnecting to automounts at startup an autofs ioctl is used
> > to find the device and inode of existing mounts so they can be used
> > to open a
901 - 1000 of 1322 matches
Mail list logo
