[PATCH 07/12] fs/locks: allow a lock request to block other requests.

is requeued or discarded, all blocked requests are woken. - When deadlock-detection looks for the lock which blocks a given request, we follow the chain of ->fl_blocker all the way to the top. Tested-by: kernel test robot Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/lock

[PATCH 06/12] fs/locks: use properly initialized file_lock when unlocking.

using flock_make_lock(), and changing it to allow a file_lock to be passed in, so memory allocation isn't always needed. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/locks.c | 28 ++-- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/fs

[PATCH 10/12] fs/locks: create a tree of dependent requests.

-and-tested-by: Martin Wilck Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/locks.c | 69 +++- 1 file changed, 63 insertions(+), 6 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index 91fc29912151..7a55d6feb12a 100644 --- a/fs

[PATCH 11/12] fs/locks: merge posix_unblock_lock() and locks_delete_block()

. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/cifs/file.c |2 +- fs/lockd/svclock.c |2 +- fs/locks.c | 36 +--- fs/nfsd/nfs4state.c |6 +++--- include/linux/fs.h |4 ++-- 5 files changed, 20 insertions(+), 30 deletions

[PATCH 10/12] fs/locks: create a tree of dependent requests.

-and-tested-by: Martin Wilck Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/locks.c | 69 +++- 1 file changed, 63 insertions(+), 6 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index 91fc29912151..7a55d6feb12a 100644 --- a/fs

[PATCH 11/12] fs/locks: merge posix_unblock_lock() and locks_delete_block()

. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/cifs/file.c |2 +- fs/lockd/svclock.c |2 +- fs/locks.c | 36 +--- fs/nfsd/nfs4state.c |6 +++--- include/linux/fs.h |4 ++-- 5 files changed, 20 insertions(+), 30 deletions

[PATCH 12/12] fs/locks: remove unnecessary white space.

- spaces before tabs, - spaces at the end of lines, - multiple blank lines, - blank lines before EXPORT_SYMBOL, can all go. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/locks.c | 33 - 1 file changed, 12 insertions(+), 21 deletions(-) diff

[PATCH 12/12] fs/locks: remove unnecessary white space.

- spaces before tabs, - spaces at the end of lines, - multiple blank lines, - blank lines before EXPORT_SYMBOL, can all go. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/locks.c | 33 - 1 file changed, 12 insertions(+), 21 deletions(-) diff

[PATCH 08/12] fs/locks: always delete_block after waiting.

to the new lock, though when merged with a pre-exiting lock, this might not happen. In all cases we don't want blocked locks to remain attached, so we remove them to be safe. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields Tested-by: syzbot+a4a3d526b4157113e...@syzkaller.appspotmail.com Tested

[PATCH 05/12] ocfs2: properly initial file_lock used for unlock.

Rather than assuming all-zeros is sufficient, use the available API to initialize the file_lock structure use for unlock. VFS-level changes will soon make it important that the list_heads in file_lock are always properly initialized. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields

[PATCH 03/12] NFS: use locks_copy_lock() to copy locks.

-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/nfs/nfs4proc.c |6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 867457d6dfbe..0ba2b0fb8ff3 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -6311,7 +6311,8 @@ static

[PATCH 04/12] gfs2: properly initial file_lock used for unlock.

Rather than assuming all-zeros is sufficient, use the available API to initialize the file_lock structure use for unlock. VFS-level changes will soon make it important that the list_heads in file_lock are always properly initialized. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields

[PATCH 09/12] fs/locks: change all *_conflict() functions to return bool.

. And convert some return (foo); to return foo; Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/locks.c | 27 +++ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index 2814662ecfe6..91fc29912151 100644 --- a/fs/locks.c

[PATCH 03/12] NFS: use locks_copy_lock() to copy locks.

-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/nfs/nfs4proc.c |6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 867457d6dfbe..0ba2b0fb8ff3 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -6311,7 +6311,8 @@ static

[PATCH 04/12] gfs2: properly initial file_lock used for unlock.

Rather than assuming all-zeros is sufficient, use the available API to initialize the file_lock structure use for unlock. VFS-level changes will soon make it important that the list_heads in file_lock are always properly initialized. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields

[PATCH 09/12] fs/locks: change all *_conflict() functions to return bool.

. And convert some return (foo); to return foo; Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/locks.c | 27 +++ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index 2814662ecfe6..91fc29912151 100644 --- a/fs/locks.c

[PATCH 08/12] fs/locks: always delete_block after waiting.

to the new lock, though when merged with a pre-exiting lock, this might not happen. In all cases we don't want blocked locks to remain attached, so we remove them to be safe. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields Tested-by: syzbot+a4a3d526b4157113e...@syzkaller.appspotmail.com Tested

[PATCH 05/12] ocfs2: properly initial file_lock used for unlock.

Rather than assuming all-zeros is sufficient, use the available API to initialize the file_lock structure use for unlock. VFS-level changes will soon make it important that the list_heads in file_lock are always properly initialized. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields

[PATCH 02/12] fs/locks: split out __locks_wake_up_blocks().

This functionality will be useful in future patches, so split it out from locks_wake_up_blocks(). Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/locks.c | 27 --- 1 file changed, 16 insertions(+), 11 deletions(-) diff --git a/fs/locks.c b/fs/locks.c

[PATCH 02/12] fs/locks: split out __locks_wake_up_blocks().

This functionality will be useful in future patches, so split it out from locks_wake_up_blocks(). Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/locks.c | 27 --- 1 file changed, 16 insertions(+), 11 deletions(-) diff --git a/fs/locks.c b/fs/locks.c

[PATCH 01/12] fs/locks: rename some lists and pointers.

that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/cifs/file.c |2 + fs/locks.c | 59 --- include/linux/fs.h |9 +- include

[PATCH 01/12] fs/locks: rename some lists and pointers.

that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown Reviewed-by: J. Bruce Fields --- fs/cifs/file.c |2 + fs/locks.c | 59 --- include/linux/fs.h |9 +- include

[PATCH 00/12 v6] fs/locks: avoid thundering-herd wake-ups

This series has the fixes for the recently reported performance regressions merged into the patches which caused them. It also has a couple of little fixes that have been mentioned on the list, and that Jeff had merged into his copy. Thanks, NeilBrown --- NeilBrown (12): fs/locks: rename

[PATCH 00/12 v6] fs/locks: avoid thundering-herd wake-ups

This series has the fixes for the recently reported performance regressions merged into the patches which caused them. It also has a couple of little fixes that have been mentioned on the list, and that Jeff had merged into his copy. Thanks, NeilBrown --- NeilBrown (12): fs/locks: rename

Re: [PATCH 02/25] iov_iter: Use accessor function [ver #2]

but as there are no users it is probably best to discard it. I discovered this because lustre uses (or rather "used") it. NeilBrown From: NeilBrown Date: Wed, 28 Nov 2018 12:38:30 +1100 Subject: [PATCH] iov_iter: discard iov_for_each() iov_for_each has no users and cannot compile as

Re: [PATCH 02/25] iov_iter: Use accessor function [ver #2]

but as there are no users it is probably best to discard it. I discovered this because lustre uses (or rather "used") it. NeilBrown From: NeilBrown Date: Wed, 28 Nov 2018 12:38:30 +1100 Subject: [PATCH] iov_iter: discard iov_for_each() iov_for_each has no users and cannot compile as

[PATCH] locks: fix performance regressions.

The kernel test robot reported two performance regressions caused by recent patches. Both appear to related to the global spinlock blocked_lock_lock being taken more often. This patch avoids taking that lock in the cases tested. Reported-by: kernel test robot Signed-off-by: NeilBrown --- Hi

[PATCH] locks: fix performance regressions.

The kernel test robot reported two performance regressions caused by recent patches. Both appear to related to the global spinlock blocked_lock_lock being taken more often. This patch avoids taking that lock in the cases tested. Reported-by: kernel test robot Signed-off-by: NeilBrown --- Hi

Re: [LKP] [fs/locks] 83b381078b: will-it-scale.per_thread_ops -62.5% regression

king a file that is only used > by that thread. Thanks for identifying that Bruce. This would certainly be a case where locks_delete_block() is now being called when it wasn't before. > > By the way, what's the X-axis on these graphs? (Or the y-axi

Re: [LKP] [fs/locks] 83b381078b: will-it-scale.per_thread_ops -62.5% regression

king a file that is only used > by that thread. Thanks for identifying that Bruce. This would certainly be a case where locks_delete_block() is now being called when it wasn't before. > > By the way, what's the X-axis on these graphs? (Or the y-axi

Re: [PATCH] staging: greybus: Parenthesis alignment

o_apbridgea_register_cport( data->connection, i2s_port, cportid, AUDIO_APBRIDGEA_DIRECTION_TX); but checkpatch doesn't like the '(' at the end of the line. Do you know why that is? Thanks, NeilBrown > > But if you are going to do this for multiple li

Re: [PATCH] staging: greybus: Parenthesis alignment

o_apbridgea_register_cport( data->connection, i2s_port, cportid, AUDIO_APBRIDGEA_DIRECTION_TX); but checkpatch doesn't like the '(' at the end of the line. Do you know why that is? Thanks, NeilBrown > > But if you are going to do this for multiple li

Re: [PATCH v6 1/3] staging: mt7621-mmc: Remove #if 0 blocks in sd.c

nd the "#endif /* end of --- */" at the end? Thanks, NeilBrown signature.asc Description: PGP signature

Re: [PATCH v6 1/3] staging: mt7621-mmc: Remove #if 0 blocks in sd.c

nd the "#endif /* end of --- */" at the end? Thanks, NeilBrown signature.asc Description: PGP signature

Re: KASAN: use-after-free Read in locks_delete_block

On Thu, Nov 15 2018, Dmitry Vyukov wrote: > On Wed, Nov 14, 2018 at 2:36 AM, Jeff Layton wrote: >> On Wed, 2018-11-14 at 07:40 +1100, NeilBrown wrote: >>> On Tue, Nov 13 2018, Jeff Layton wrote: >>> >>> > On Mon, 2018-11-12 at 12:34 -0800, syzbot wrote

Re: KASAN: use-after-free Read in locks_delete_block

On Thu, Nov 15 2018, Dmitry Vyukov wrote: > On Wed, Nov 14, 2018 at 2:36 AM, Jeff Layton wrote: >> On Wed, 2018-11-14 at 07:40 +1100, NeilBrown wrote: >>> On Tue, Nov 13 2018, Jeff Layton wrote: >>> >>> > On Mon, 2018-11-12 at 12:34 -0800, syzbot wrote

Re: KASAN: use-after-free Read in locks_delete_block

down in the mandatory locking code. This is with Neil's set > from last week. I haven't merged the series he sent the other day yet, > but they don't seem to be different in this regard. > > Looks like the fl_blocked list might have had an entry on it that was > freed without being rem

Re: KASAN: use-after-free Read in locks_delete_block

down in the mandatory locking code. This is with Neil's set > from last week. I haven't merged the series he sent the other day yet, > but they don't seem to be different in this regard. > > Looks like the fl_blocked list might have had an entry on it that was > freed without being rem

[PATCH 12/12] VFS: locks: remove unnecessary white space.

- spaces before tabs, - spaces at the end of lines, - multiple blank lines, - blank lines before EXPORT_SYMBOL, can all go. Signed-off-by: NeilBrown --- fs/locks.c | 33 - 1 file changed, 12 insertions(+), 21 deletions(-) diff --git a/fs/locks.c b/fs

[PATCH 12/12] VFS: locks: remove unnecessary white space.

- spaces before tabs, - spaces at the end of lines, - multiple blank lines, - blank lines before EXPORT_SYMBOL, can all go. Signed-off-by: NeilBrown --- fs/locks.c | 33 - 1 file changed, 12 insertions(+), 21 deletions(-) diff --git a/fs/locks.c b/fs

[PATCH 10/12] fs/locks: create a tree of dependent requests.

-and-tested-by: Martin Wilck Signed-off-by: NeilBrown --- fs/locks.c | 69 +++- 1 file changed, 63 insertions(+), 6 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index 74b24191d6e6..1006b566ddf5 100644 --- a/fs/locks.c +++ b/fs/locks.c

[PATCH 11/12] locks: merge posix_unblock_lock() and locks_delete_block()

. Signed-off-by: NeilBrown --- fs/cifs/file.c |2 +- fs/lockd/svclock.c |2 +- fs/locks.c | 36 +--- fs/nfsd/nfs4state.c |6 +++--- include/linux/fs.h |4 ++-- 5 files changed, 20 insertions(+), 30 deletions(-) diff --git a/fs/cifs/file.c

[PATCH 10/12] fs/locks: create a tree of dependent requests.

-and-tested-by: Martin Wilck Signed-off-by: NeilBrown --- fs/locks.c | 69 +++- 1 file changed, 63 insertions(+), 6 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index 74b24191d6e6..1006b566ddf5 100644 --- a/fs/locks.c +++ b/fs/locks.c

[PATCH 11/12] locks: merge posix_unblock_lock() and locks_delete_block()

. Signed-off-by: NeilBrown --- fs/cifs/file.c |2 +- fs/lockd/svclock.c |2 +- fs/locks.c | 36 +--- fs/nfsd/nfs4state.c |6 +++--- include/linux/fs.h |4 ++-- 5 files changed, 20 insertions(+), 30 deletions(-) diff --git a/fs/cifs/file.c

[PATCH 07/12] fs/locks: allow a lock request to block other requests.

is requeued or discarded, all blocked requests are woken. - When deadlock-detection looks for the lock which blocks a given request, we follow the chain of ->fl_blocker all the way to the top. Signed-off-by: NeilBrown --- fs/locks.c | 36 ++-- 1 file changed,

[PATCH 08/12] fs/locks: always delete_block after waiting.

to the new lock, though when merged with a pre-exiting lock, this might not happen. In all cases we don't want blocked locks to remain attached, so we remove them to be safe. Signed-off-by: NeilBrown --- fs/locks.c | 24 +--- 1 file changed, 9 insertions(+), 15 deletions

[PATCH 09/12] fs/locks: change all *_conflict() functions to return bool.

. And convert some return (foo); to return foo; Signed-off-by: NeilBrown --- fs/locks.c | 27 +++ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index 8d9d79e7a565..74b24191d6e6 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -795,47

[PATCH 07/12] fs/locks: allow a lock request to block other requests.

is requeued or discarded, all blocked requests are woken. - When deadlock-detection looks for the lock which blocks a given request, we follow the chain of ->fl_blocker all the way to the top. Signed-off-by: NeilBrown --- fs/locks.c | 36 ++-- 1 file changed,

[PATCH 08/12] fs/locks: always delete_block after waiting.

to the new lock, though when merged with a pre-exiting lock, this might not happen. In all cases we don't want blocked locks to remain attached, so we remove them to be safe. Signed-off-by: NeilBrown --- fs/locks.c | 24 +--- 1 file changed, 9 insertions(+), 15 deletions

[PATCH 09/12] fs/locks: change all *_conflict() functions to return bool.

. And convert some return (foo); to return foo; Signed-off-by: NeilBrown --- fs/locks.c | 27 +++ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index 8d9d79e7a565..74b24191d6e6 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -795,47

[PATCH 00/12 v5] locks: avoid thundering-herd wake-ups

e - I made a few little changes - flock_make_lock() now properly initialized a lock - using the code from Jeff - code actually compiles for ocfs2 and gfs2 Thanks, NeilBrown --- NeilBrown (12): fs/locks: rename some lists and pointers. fs/locks: split out __locks_wake

[PATCH 00/12 v5] locks: avoid thundering-herd wake-ups

e - I made a few little changes - flock_make_lock() now properly initialized a lock - using the code from Jeff - code actually compiles for ocfs2 and gfs2 Thanks, NeilBrown --- NeilBrown (12): fs/locks: rename some lists and pointers. fs/locks: split out __locks_wake

[PATCH 01/12] fs/locks: rename some lists and pointers.

that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown --- fs/cifs/file.c |2 + fs/locks.c | 59 --- include/linux/fs.h |9 +- include/trace/events/filelock.h | 16

[PATCH 06/12] locks: use properly initialized file_lock when unlocking.

using flock_make_lock(), and changing it to allow a file_lock to be passed in, so memory allocation isn't always needed. Signed-off-by: NeilBrown --- fs/locks.c | 28 ++-- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index

[PATCH 03/12] NFS: use locks_copy_lock() to copy locks.

-by: NeilBrown --- fs/nfs/nfs4proc.c |6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 867457d6dfbe..0ba2b0fb8ff3 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -6311,7 +6311,8 @@ static struct nfs4_unlockdata

[PATCH 05/12] ocfs2: properly initial file_lock used for unlock.

Rather than assuming all-zeros is sufficient, use the available API to initialize the file_lock structure use for unlock. VFS-level changes will soon make it important that the list_heads in file_lock are always properly initialized. Signed-off-by: NeilBrown --- fs/ocfs2/locks.c | 10

[PATCH 04/12] gfs2: properly initial file_lock used for unlock.

Rather than assuming all-zeros is sufficient, use the available API to initialize the file_lock structure use for unlock. VFS-level changes will soon make it important that the list_heads in file_lock are always properly initialized. Signed-off-by: NeilBrown --- fs/gfs2/file.c | 10

[PATCH 02/12] fs/locks: split out __locks_wake_up_blocks().

This functionality will be useful in future patches, so split it out from locks_wake_up_blocks(). Signed-off-by: NeilBrown --- fs/locks.c | 27 --- 1 file changed, 16 insertions(+), 11 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index c6df0c8b3d13..6adb22a21a92

[PATCH 03/12] NFS: use locks_copy_lock() to copy locks.

-by: NeilBrown --- fs/nfs/nfs4proc.c |6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 867457d6dfbe..0ba2b0fb8ff3 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -6311,7 +6311,8 @@ static struct nfs4_unlockdata

[PATCH 05/12] ocfs2: properly initial file_lock used for unlock.

Rather than assuming all-zeros is sufficient, use the available API to initialize the file_lock structure use for unlock. VFS-level changes will soon make it important that the list_heads in file_lock are always properly initialized. Signed-off-by: NeilBrown --- fs/ocfs2/locks.c | 10

[PATCH 04/12] gfs2: properly initial file_lock used for unlock.

Rather than assuming all-zeros is sufficient, use the available API to initialize the file_lock structure use for unlock. VFS-level changes will soon make it important that the list_heads in file_lock are always properly initialized. Signed-off-by: NeilBrown --- fs/gfs2/file.c | 10

[PATCH 02/12] fs/locks: split out __locks_wake_up_blocks().

This functionality will be useful in future patches, so split it out from locks_wake_up_blocks(). Signed-off-by: NeilBrown --- fs/locks.c | 27 --- 1 file changed, 16 insertions(+), 11 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index c6df0c8b3d13..6adb22a21a92

[PATCH 06/12] locks: use properly initialized file_lock when unlocking.

using flock_make_lock(), and changing it to allow a file_lock to be passed in, so memory allocation isn't always needed. Signed-off-by: NeilBrown --- fs/locks.c | 28 ++-- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index

[PATCH 01/12] fs/locks: rename some lists and pointers.

that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown --- fs/cifs/file.c |2 + fs/locks.c | 59 --- include/linux/fs.h |9 +- include/trace/events/filelock.h | 16

Re: [PATCH 10/12] fs/locks: create a tree of dependent requests.

On Thu, Nov 08 2018, J. Bruce Fields wrote: > On Fri, Nov 09, 2018 at 11:38:19AM +1100, NeilBrown wrote: >> On Thu, Nov 08 2018, J. Bruce Fields wrote: >> >> > On Mon, Nov 05, 2018 at 12:30:48PM +1100, NeilBrown wrote: >> >> When we find an existing

Re: [PATCH 10/12] fs/locks: create a tree of dependent requests.

On Thu, Nov 08 2018, J. Bruce Fields wrote: > On Fri, Nov 09, 2018 at 11:38:19AM +1100, NeilBrown wrote: >> On Thu, Nov 08 2018, J. Bruce Fields wrote: >> >> > On Mon, Nov 05, 2018 at 12:30:48PM +1100, NeilBrown wrote: >> >> When we find an existing

Re: [PATCH 22/23] SUNRPC: simplify auth_unix.

that's an argument in favor of building creds on demand > rather than keeping them in a hash table, isn't it. :-) I think it is. Thanks!!! NeilBrown > > >>> In addition, the comment near unx_marshal suggests we should >>> cache the marshaled on-the-wire vers

Re: [PATCH 22/23] SUNRPC: simplify auth_unix.

that's an argument in favor of building creds on demand > rather than keeping them in a hash table, isn't it. :-) I think it is. Thanks!!! NeilBrown > > >>> In addition, the comment near unx_marshal suggests we should >>> cache the marshaled on-the-wire vers

Re: [PATCH 10/12] fs/locks: create a tree of dependent requests.

On Thu, Nov 08 2018, J. Bruce Fields wrote: > On Mon, Nov 05, 2018 at 12:30:48PM +1100, NeilBrown wrote: >> When we find an existing lock which conflicts with a request, >> and the request wants to wait, we currently add the request >> to a list. When the lock is removed, th

Re: [PATCH 10/12] fs/locks: create a tree of dependent requests.

On Thu, Nov 08 2018, J. Bruce Fields wrote: > On Mon, Nov 05, 2018 at 12:30:48PM +1100, NeilBrown wrote: >> When we find an existing lock which conflicts with a request, >> and the request wants to wait, we currently add the request >> to a list. When the lock is removed, th

Re: [PATCH 01/12] fs/locks: rename some lists and pointers.

On Thu, Nov 08 2018, J. Bruce Fields wrote: > On Mon, Nov 05, 2018 at 12:30:47PM +1100, NeilBrown wrote: >> struct file lock contains an 'fl_next' pointer which >> is used to point to the lock that this request is blocked >> waiting for. So rename it to fl_blocker.

Re: [PATCH 01/12] fs/locks: rename some lists and pointers.

On Thu, Nov 08 2018, J. Bruce Fields wrote: > On Mon, Nov 05, 2018 at 12:30:47PM +1100, NeilBrown wrote: >> struct file lock contains an 'fl_next' pointer which >> is used to point to the lock that this request is blocked >> waiting for. So rename it to fl_blocker.

Re: [PATCH 22/23] SUNRPC: simplify auth_unix.

On Wed, Nov 07 2018, Chuck Lever wrote: > Hi Neil- > > >> On Nov 6, 2018, at 11:12 PM, NeilBrown wrote: >> >> 1/ discard 'struct unx_cred'. We don't need any data that >> is not already in 'struct rpc_cred'. >> 2/ Don't keep these creds in a hash t

Re: [PATCH 22/23] SUNRPC: simplify auth_unix.

On Wed, Nov 07 2018, Chuck Lever wrote: > Hi Neil- > > >> On Nov 6, 2018, at 11:12 PM, NeilBrown wrote: >> >> 1/ discard 'struct unx_cred'. We don't need any data that >> is not already in 'struct rpc_cred'. >> 2/ Don't keep these creds in a hash t

Re: [LKP] [locks] dee160df82: BUG:KASAN:null-ptr-deref_in_l

_after_hwframe+0x49/0xbe >> > > [ 42.063639] RIP: 0033:0x7f31c39fe250 >> > > [ 42.064798] Code: 73 01 c3 48 8b 0d 58 7d 20 00 f7 d8 64 89 01 48 83 >> > > c8 ff c3 66 0f 1f 44 00 00 83 3d 79 c1 20 00 00 75 10 b8 03 00 00 00 0f >> > > 05 <48&

Re: [LKP] [locks] dee160df82: BUG:KASAN:null-ptr-deref_in_l

_after_hwframe+0x49/0xbe >> > > [ 42.063639] RIP: 0033:0x7f31c39fe250 >> > > [ 42.064798] Code: 73 01 c3 48 8b 0d 58 7d 20 00 f7 d8 64 89 01 48 83 >> > > c8 ff c3 66 0f 1f 44 00 00 83 3d 79 c1 20 00 00 75 10 b8 03 00 00 00 0f >> > > 05 <48&

[PATCH 22/23] SUNRPC: simplify auth_unix.

table. As the lookup can happen during write-out, use a mempool to ensure forward progress. This means that we cannot compare two credentials for equality by comparing the pointers, but we never do that anyway. Signed-off-by: NeilBrown --- net/sunrpc/auth.c |1 net/sunrpc

[PATCH 21/23] SUNRPC: remove crbind rpc_cred operation

This now always just does get_rpccred(), so we don't need an operation pointer to know to do that. Signed-off-by: NeilBrown --- include/linux/sunrpc/auth.h|2 -- net/sunrpc/auth.c | 12 +--- net/sunrpc/auth_gss/auth_gss.c |2 -- net/sunrpc/auth_null.c

[PATCH 19/23] NFS/NFSD/SUNRPC: replace generic creds with 'struct cred'.

which is statically allocated and recognized where needed as having a special meaning. A look-up of a low-level cred will map this to a machine credential. Signed-off-by: NeilBrown --- fs/lockd/clntproc.c |6 + fs/nfs/blocklayout/blocklayout.c |2

[PATCH 23/23] SUNRPC discard cr_uid from struct rpc_cred.

Just use ->cr_cred->fsuid directly. Signed-off-by: NeilBrown --- include/linux/sunrpc/auth.h|2 -- net/sunrpc/auth.c |1 - net/sunrpc/auth_gss/auth_gss.c | 12 ++-- 3 files changed, 6 insertions(+), 9 deletions(-) diff --git a/include/linux/sunrpc/au

[PATCH 20/23] SUNRPC: remove generic cred code.

This is no longer used. Signed-off-by: NeilBrown --- include/linux/sunrpc/auth.h |6 - net/sunrpc/Makefile |2 net/sunrpc/auth.c | 18 net/sunrpc/auth_generic.c | 199 --- net/sunrpc/auth_null.c |2 5 files

[PATCH 22/23] SUNRPC: simplify auth_unix.

table. As the lookup can happen during write-out, use a mempool to ensure forward progress. This means that we cannot compare two credentials for equality by comparing the pointers, but we never do that anyway. Signed-off-by: NeilBrown --- net/sunrpc/auth.c |1 net/sunrpc

[PATCH 21/23] SUNRPC: remove crbind rpc_cred operation

This now always just does get_rpccred(), so we don't need an operation pointer to know to do that. Signed-off-by: NeilBrown --- include/linux/sunrpc/auth.h|2 -- net/sunrpc/auth.c | 12 +--- net/sunrpc/auth_gss/auth_gss.c |2 -- net/sunrpc/auth_null.c

[PATCH 19/23] NFS/NFSD/SUNRPC: replace generic creds with 'struct cred'.

which is statically allocated and recognized where needed as having a special meaning. A look-up of a low-level cred will map this to a machine credential. Signed-off-by: NeilBrown --- fs/lockd/clntproc.c |6 + fs/nfs/blocklayout/blocklayout.c |2

[PATCH 23/23] SUNRPC discard cr_uid from struct rpc_cred.

Just use ->cr_cred->fsuid directly. Signed-off-by: NeilBrown --- include/linux/sunrpc/auth.h|2 -- net/sunrpc/auth.c |1 - net/sunrpc/auth_gss/auth_gss.c | 12 ++-- 3 files changed, 6 insertions(+), 9 deletions(-) diff --git a/include/linux/sunrpc/au

[PATCH 20/23] SUNRPC: remove generic cred code.

This is no longer used. Signed-off-by: NeilBrown --- include/linux/sunrpc/auth.h |6 - net/sunrpc/Makefile |2 net/sunrpc/auth.c | 18 net/sunrpc/auth_generic.c | 199 --- net/sunrpc/auth_null.c |2 5 files

[PATCH 14/23] SUNRPC: add side channel to use non-generic cred for rpc call.

ave rpcauth_bindcred() notice it and bind it in place. Credit to kernel test robot for finding a bug in earlier version of this patch. Signed-off-by: NeilBrown --- include/linux/sunrpc/sched.h |2 ++ net/sunrpc/auth.c|6 +- net/sunrpc/clnt.c|2 +-

[PATCH 12/23] NFS/SUNRPC: don't lookup machine credential until rpcauth_bindcred().

ic 'struct rpc_cred' pointer is used. rpcauth_bindcred() recognizes this, finds the principal from the client, and binds the correct credential. Signed-off-by: NeilBrown --- fs/nfs/client.c | 11 ++- fs/nfs/nfs4_fs.h|9 + fs/nfs/nfs4state.c

[PATCH 17/23] NFS: change access cache to use 'struct cred'.

cache is large. This also keeps many fewer items pinned in the rpc cred cache, so the cred cache is less likely to get large. Signed-off-by: NeilBrown --- fs/nfs/dir.c | 44 +++- fs/nfs/nfs3proc.c |9 - fs/nfs/nfs4proc.c

[PATCH 16/23] SUNRPC: remove RPCAUTH_AUTH_NO_CRKEY_TIMEOUT

This is no longer used. Signed-off-by: NeilBrown --- include/linux/sunrpc/auth.h |3 --- net/sunrpc/auth_null.c |1 - net/sunrpc/auth_unix.c |1 - 3 files changed, 5 deletions(-) diff --git a/include/linux/sunrpc/auth.h b/include/linux/sunrpc/auth.h index 0bdc2f4957ff

[PATCH 15/23] NFS: move credential expiry tracking out of SUNRPC into NFS.

. Signed-off-by: NeilBrown --- fs/nfs/inode.c |2 + fs/nfs/write.c | 24 +++-- include/linux/nfs_fs.h |1 + include/linux/sunrpc/auth.h| 12 -- net/sunrpc/auth.c | 23 net/sunrpc/auth_generic.c | 75

[PATCH 14/23] SUNRPC: add side channel to use non-generic cred for rpc call.

ave rpcauth_bindcred() notice it and bind it in place. Credit to kernel test robot for finding a bug in earlier version of this patch. Signed-off-by: NeilBrown --- include/linux/sunrpc/sched.h |2 ++ net/sunrpc/auth.c|6 +- net/sunrpc/clnt.c|2 +-

[PATCH 12/23] NFS/SUNRPC: don't lookup machine credential until rpcauth_bindcred().

ic 'struct rpc_cred' pointer is used. rpcauth_bindcred() recognizes this, finds the principal from the client, and binds the correct credential. Signed-off-by: NeilBrown --- fs/nfs/client.c | 11 ++- fs/nfs/nfs4_fs.h|9 + fs/nfs/nfs4state.c

[PATCH 17/23] NFS: change access cache to use 'struct cred'.

cache is large. This also keeps many fewer items pinned in the rpc cred cache, so the cred cache is less likely to get large. Signed-off-by: NeilBrown --- fs/nfs/dir.c | 44 +++- fs/nfs/nfs3proc.c |9 - fs/nfs/nfs4proc.c

[PATCH 16/23] SUNRPC: remove RPCAUTH_AUTH_NO_CRKEY_TIMEOUT

This is no longer used. Signed-off-by: NeilBrown --- include/linux/sunrpc/auth.h |3 --- net/sunrpc/auth_null.c |1 - net/sunrpc/auth_unix.c |1 - 3 files changed, 5 deletions(-) diff --git a/include/linux/sunrpc/auth.h b/include/linux/sunrpc/auth.h index 0bdc2f4957ff

[PATCH 15/23] NFS: move credential expiry tracking out of SUNRPC into NFS.

. Signed-off-by: NeilBrown --- fs/nfs/inode.c |2 + fs/nfs/write.c | 24 +++-- include/linux/nfs_fs.h |1 + include/linux/sunrpc/auth.h| 12 -- net/sunrpc/auth.c | 23 net/sunrpc/auth_generic.c | 75

[PATCH 18/23] NFS: struct nfs_open_dir_context: convert rpc_cred pointer to cred.

Use the common 'struct cred' to pass credentials for readdir. Signed-off-by: NeilBrown --- fs/nfs/dir.c| 15 +-- fs/nfs/nfs3proc.c | 11 +-- fs/nfs/nfs4proc.c | 13 ++--- fs/nfs/proc.c | 11 +-- include/linux/nfs_fs.h

[PATCH 13/23] SUNRPC: introduce RPC_TASK_NULLCREDS to request auth_none

LL and set a task flag, similar to RPC_TASK_ROOTCREDS, which requests that NULL credentials be used by default. Signed-off-by: NeilBrown --- include/linux/sunrpc/sched.h |1 + net/sunrpc/auth.c|2 ++ net/sunrpc/clnt.c| 19 ++- 3 files changed, 9

[PATCH 18/23] NFS: struct nfs_open_dir_context: convert rpc_cred pointer to cred.

Use the common 'struct cred' to pass credentials for readdir. Signed-off-by: NeilBrown --- fs/nfs/dir.c| 15 +-- fs/nfs/nfs3proc.c | 11 +-- fs/nfs/nfs4proc.c | 13 ++--- fs/nfs/proc.c | 11 +-- include/linux/nfs_fs.h

[PATCH 13/23] SUNRPC: introduce RPC_TASK_NULLCREDS to request auth_none

LL and set a task flag, similar to RPC_TASK_ROOTCREDS, which requests that NULL credentials be used by default. Signed-off-by: NeilBrown --- include/linux/sunrpc/sched.h |1 + net/sunrpc/auth.c|2 ++ net/sunrpc/clnt.c| 19 ++- 3 files changed, 9

<    1   2   3   4   5   6   7   8   9   10   >