From: Aleksa Sarai <cyphar@cyphar.com>
To: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: [RFC][PATCHSET] sanitized pathwalk machinery (v2)
Date: Fri, 28 Feb 2020 12:24:51 +1100 [thread overview]
Message-ID: <20200228012451.upnq5r7fdctrk7pv@yavin> (raw)
In-Reply-To: <20200225012457.GA138294@ZenIV.linux.org.uk>
[-- Attachment #1: Type: text/plain, Size: 3039 bytes --]
On 2020-02-25, Al Viro <viro@zeniv.linux.org.uk> wrote:
> On Sun, Feb 23, 2020 at 01:12:21AM +0000, Al Viro wrote:
> > This is a slightly extended repost of the patchset posted on
> > Jan 19. Current branch is in vfs.git#work.do_last, the main
> > difference from the last time around being a bit of do_last()
> > untangling added in the end of series. #work.openat2 is already
> > in mainline, which simplifies the series - now it's a straight
> > branch with no merges.
>
> Whee... While trying to massage ".." handling towards the use of
> regular mount crossing semantics, I've found something interesting.
> Namely, if you start in a directory with overmounted parent,
> LOOKUP_NO_XDEV resolution of ../something will bloody well cross
> into the overmount.
Oh boy...
> Reason: follow_dotdot() (and its RCU counterpart) check for LOOKUP_NO_XDEV
> when crossing into underlying fs, but not when crossing into overmount
> of the parent.
>
> Interpretation of .. is basically
>
> loop: if we are in root // uncommon
> next = current position
> else if we are in root of a mounted filesystem // more rare
> move to underlying mountpoint
> goto loop
> else
> next = parent directory of current position // most common
>
> while next is overmounted // _VERY_ uncommon
> next = whatever's mounted on next
>
> move to next
>
> The second loop should've been sharing code with the normal mountpoint
> crossing. It doesn't, which has already lead to interesting inconsistencies
> (e.g. autofs generally expects ->d_manage() to be called before crossing
> into it; here it's not done). LOOKUP_NO_XDEV has just added one more...
You're quite right -- LOOKUP_NO_XDEV should block that and I missed it.
> Incidentally, another inconsistency is LOOKUP_BENEATH treatment in case
> when we have walked out of the subtree by way of e.g. procfs symlink and
> then ran into .. in the absolute root (that's
> if (!follow_up(&nd->path))
> break;
> in follow_dotdot()). Shouldn't that give the same reaction as ..
> in root (EXDEV on LOOKUP_BENEATH, that is)? It doesn't...
You can't go through procfs symlinks with LOOKUP_BENEATH, but if it's
possible to do that kind of jump then it should also be blocked (but I
would say that I'd prefer "block any kind of weird jump").
> Another one is about LOOKUP_NO_XDEV again: suppose you have process'
> root directly overmounted and cwd in the root of whatever's overmounting
> it. Resolution of .. will stay in cwd - we have no parent within the
> chroot jail we are in, so we move to whatever's overmounting that root.
> Which is the original location. Should we fail on LOOKUP_NO_XDEV here?
> Plain .. in the root of chroot jail (not overmounted by anything) does
> *not*...
I think LOOKUP_NO_XDEV should block that since you end up crossing a
mountpoint.
--
Aleksa Sarai
Senior Software Engineer (Containers)
SUSE Linux GmbH
<https://www.cyphar.com/>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]
next prev parent reply other threads:[~2020-02-28 1:25 UTC|newest]
Thread overview: 198+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-23 1:11 [RFC][PATCHSET] sanitized pathwalk machinery (v2) Al Viro
2020-02-23 1:15 ` [RFC][PATCH v2 01/34] do_add_mount(): lift lock_mount/unlock_mount into callers Al Viro
2020-02-23 1:15 ` [RFC][PATCH v2 02/34] fix automount/automount race properly Al Viro
2020-02-23 2:07 ` Linus Torvalds
2020-02-27 19:43 ` Al Viro
2020-02-27 20:00 ` Linus Torvalds
2020-02-23 1:15 ` [RFC][PATCH v2 03/34] follow_automount(): get rid of dead^Wstillborn code Al Viro
2020-02-23 1:15 ` [RFC][PATCH v2 04/34] follow_automount() doesn't need the entire nameidata Al Viro
2020-02-23 1:15 ` [RFC][PATCH v2 05/34] make build_open_flags() treat O_CREAT | O_EXCL as implying O_NOFOLLOW Al Viro
2020-02-23 1:15 ` [RFC][PATCH v2 06/34] handle_mounts(): start building a sane wrapper for follow_managed() Al Viro
2020-02-23 1:15 ` [RFC][PATCH v2 07/34] atomic_open(): saner calling conventions (return dentry on success) Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 08/34] lookup_open(): " Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 09/34] do_last(): collapse the call of path_to_nameidata() Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 10/34] handle_mounts(): pass dentry in, turn path into a pure out argument Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 11/34] lookup_fast(): consolidate the RCU success case Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 12/34] teach handle_mounts() to handle RCU mode Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 13/34] lookup_fast(): take mount traversal into callers Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 14/34] new step_into() flag: WALK_NOFOLLOW Al Viro
2020-02-23 2:14 ` Linus Torvalds
2020-02-23 22:07 ` Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 15/34] fold handle_mounts() into step_into() Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 16/34] LOOKUP_MOUNTPOINT: fold path_mountpointat() into path_lookupat() Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 17/34] expand the only remaining call of path_lookup_conditional() Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 18/34] merging pick_link() with get_link(), part 1 Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 19/34] merging pick_link() with get_link(), part 2 Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 20/34] merging pick_link() with get_link(), part 3 Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 21/34] merging pick_link() with get_link(), part 4 Al Viro
2020-02-23 2:19 ` Linus Torvalds
2020-02-23 1:16 ` [RFC][PATCH v2 22/34] merging pick_link() with get_link(), part 5 Al Viro
2020-02-23 2:22 ` Linus Torvalds
2020-02-23 1:16 ` [RFC][PATCH v2 23/34] merging pick_link() with get_link(), part 6 Al Viro
2020-02-23 2:24 ` Linus Torvalds
2020-02-23 1:16 ` [RFC][PATCH v2 24/34] finally fold get_link() into pick_link() Al Viro
2020-02-23 2:31 ` Linus Torvalds
2020-02-23 1:16 ` [RFC][PATCH v2 25/34] massage __follow_mount_rcu() a bit Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 26/34] new helper: traverse_mounts() Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 27/34] atomic_open(): return the right dentry in FMODE_OPENED case Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 28/34] atomic_open(): lift the call of may_open() into do_last() Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 29/34] do_last(): merge the may_open() calls Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 30/34] do_last(): don't bother with keeping got_write in FMODE_OPENED case Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 31/34] do_last(): rejoing the common path earlier in FMODE_{OPENED,CREATED} case Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 32/34] do_last(): simplify the liveness analysis past finish_open_created Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 33/34] do_last(): rejoin the common path even earlier in FMODE_{OPENED,CREATED} case Al Viro
2020-02-23 1:16 ` [RFC][PATCH v2 34/34] split the lookup-related parts of do_last() into a separate helper Al Viro
2020-02-25 1:24 ` [RFC][PATCHSET] sanitized pathwalk machinery (v2) Al Viro
2020-02-25 2:03 ` Al Viro
2020-02-28 1:24 ` Aleksa Sarai [this message]
2020-02-28 3:02 ` Al Viro
2020-03-01 21:51 ` [RFC][PATCHSET] sanitized pathwalk machinery (v3) Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 01/55] do_add_mount(): lift lock_mount/unlock_mount into callers Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 02/55] fix automount/automount race properly Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 03/55] follow_automount(): get rid of dead^Wstillborn code Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 04/55] follow_automount() doesn't need the entire nameidata Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 05/55] make build_open_flags() treat O_CREAT | O_EXCL as implying O_NOFOLLOW Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 06/55] handle_mounts(): start building a sane wrapper for follow_managed() Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 07/55] atomic_open(): saner calling conventions (return dentry on success) Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 08/55] lookup_open(): " Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 09/55] do_last(): collapse the call of path_to_nameidata() Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 10/55] handle_mounts(): pass dentry in, turn path into a pure out argument Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 11/55] lookup_fast(): consolidate the RCU success case Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 12/55] teach handle_mounts() to handle RCU mode Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 13/55] lookup_fast(): take mount traversal into callers Al Viro
2020-03-01 21:51 ` [RFC][PATCH v3 14/55] new step_into() flag: WALK_NOFOLLOW Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 15/55] fold handle_mounts() into step_into() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 16/55] LOOKUP_MOUNTPOINT: fold path_mountpointat() into path_lookupat() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 17/55] expand the only remaining call of path_lookup_conditional() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 18/55] merging pick_link() with get_link(), part 1 Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 19/55] merging pick_link() with get_link(), part 2 Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 20/55] merging pick_link() with get_link(), part 3 Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 21/55] merging pick_link() with get_link(), part 4 Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 22/55] merging pick_link() with get_link(), part 5 Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 23/55] merging pick_link() with get_link(), part 6 Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 24/55] finally fold get_link() into pick_link() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 25/55] sanitize handling of nd->last_type, kill LAST_BIND Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 26/55] namei: invert the meaning of WALK_FOLLOW Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 27/55] pick_link(): check for WALK_TRAILING, not LOOKUP_PARENT Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 28/55] link_path_walk(): simplify stack handling Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 29/55] massage __follow_mount_rcu() a bit Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 30/55] new helper: traverse_mounts() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 31/55] atomic_open(): return the right dentry in FMODE_OPENED case Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 32/55] atomic_open(): lift the call of may_open() into do_last() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 33/55] do_last(): merge the may_open() calls Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 34/55] do_last(): don't bother with keeping got_write in FMODE_OPENED case Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 35/55] do_last(): rejoing the common path earlier in FMODE_{OPENED,CREATED} case Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 36/55] do_last(): simplify the liveness analysis past finish_open_created Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 37/55] do_last(): rejoin the common path even earlier in FMODE_{OPENED,CREATED} case Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 38/55] split the lookup-related parts of do_last() into a separate helper Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 39/55] path_connected(): pass mount and dentry separately Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 40/55] path_parent_directory(): leave changing path->dentry to callers Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 41/55] follow_dotdot(): expand the call of path_parent_directory() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 42/55] follow_dotdot{,_rcu}(): lift switching nd->path to parent out of loop Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 43/55] follow_dotdot{,_rcu}(): lift LOOKUP_BENEATH checks " Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 44/55] move put_link() into handle_dots() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 45/55] handle_dots(): return ERR_PTR/NULL instead of -E.../0 Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 46/55] move follow_dotdot() and follow_dotdot_rcu() towards handle_dots() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 47/55] follow_dotdot{,_rcu}(): preparation to switch to step_into() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 48/55] follow_dotdot{,_rcu}(): switch to use of step_into() Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 49/55] lift all calls of step_into() out of follow_dotdot/follow_dotdot_rcu Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 50/55] follow_dotdot{,_rcu}(): massage loops Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 51/55] follow_dotdot_rcu(): be lazy about changing nd->path Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 52/55] follow_dotdot(): " Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 53/55] helper for mount rootwards traversal Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 54/55] non-RCU analogue of the previous commit Al Viro
2020-03-01 21:52 ` [RFC][PATCH v3 55/55] fs/namei.c: kill follow_mount() Al Viro
2020-03-01 22:34 ` [RFC][PATCHSET] sanitized pathwalk machinery (v3) Linus Torvalds
2020-03-02 0:39 ` Al Viro
2020-03-03 23:48 ` Eric W. Biederman
2020-03-04 0:24 ` Al Viro
2020-03-04 5:23 ` Eric W. Biederman
2020-03-04 6:55 ` Al Viro
2020-03-04 13:28 ` Matthew Wilcox
2020-03-04 16:20 ` Al Viro
2020-03-04 20:54 ` Al Viro
[not found] ` <20200304105946.4xseo3jokcnpptrj@yavin>
2020-03-04 21:00 ` Al Viro
2020-03-05 1:11 ` Matthew Wilcox
2020-03-13 23:53 ` [RFC][PATCHSET] sanitized pathwalk machinery (v4) Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 01/69] do_add_mount(): lift lock_mount/unlock_mount into callers Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 02/69] fix automount/automount race properly Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 03/69] follow_automount(): get rid of dead^Wstillborn code Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 04/69] follow_automount() doesn't need the entire nameidata Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 05/69] make build_open_flags() treat O_CREAT | O_EXCL as implying O_NOFOLLOW Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 06/69] handle_mounts(): start building a sane wrapper for follow_managed() Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 07/69] atomic_open(): saner calling conventions (return dentry on success) Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 08/69] lookup_open(): " Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 09/69] do_last(): collapse the call of path_to_nameidata() Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 10/69] handle_mounts(): pass dentry in, turn path into a pure out argument Al Viro
2020-03-13 23:52 ` [RFC][PATCH v4 11/69] lookup_fast(): consolidate the RCU success case Al Viro
2020-03-14 0:25 ` Linus Torvalds
2020-03-13 23:53 ` [RFC][PATCH v4 12/69] teach handle_mounts() to handle RCU mode Al Viro
2020-03-14 0:28 ` Linus Torvalds
2020-03-14 1:00 ` Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 13/69] lookup_fast(): take mount traversal into callers Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 14/69] step_into() callers: dismiss the symlink earlier Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 15/69] new step_into() flag: WALK_NOFOLLOW Al Viro
2020-03-14 0:32 ` Linus Torvalds
2020-03-14 1:06 ` Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 16/69] fold handle_mounts() into step_into() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 17/69] LOOKUP_MOUNTPOINT: fold path_mountpointat() into path_lookupat() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 18/69] expand the only remaining call of path_lookup_conditional() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 19/69] merging pick_link() with get_link(), part 1 Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 20/69] merging pick_link() with get_link(), part 2 Al Viro
2020-03-14 0:40 ` Linus Torvalds
2020-03-13 23:53 ` [RFC][PATCH v4 21/69] merging pick_link() with get_link(), part 3 Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 22/69] merging pick_link() with get_link(), part 4 Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 23/69] merging pick_link() with get_link(), part 5 Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 24/69] merging pick_link() with get_link(), part 6 Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 25/69] finally fold get_link() into pick_link() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 26/69] sanitize handling of nd->last_type, kill LAST_BIND Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 27/69] namei: invert the meaning of WALK_FOLLOW Al Viro
2020-03-14 0:42 ` Linus Torvalds
2020-03-13 23:53 ` [RFC][PATCH v4 28/69] pick_link(): check for WALK_TRAILING, not LOOKUP_PARENT Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 29/69] link_path_walk(): simplify stack handling Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 30/69] namei: have link_path_walk() maintain LOOKUP_PARENT Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 31/69] massage __follow_mount_rcu() a bit Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 32/69] new helper: traverse_mounts() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 33/69] atomic_open(): return the right dentry in FMODE_OPENED case Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 34/69] atomic_open(): lift the call of may_open() into do_last() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 35/69] do_last(): merge the may_open() calls Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 36/69] do_last(): don't bother with keeping got_write in FMODE_OPENED case Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 37/69] do_last(): rejoing the common path earlier in FMODE_{OPENED,CREATED} case Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 38/69] do_last(): simplify the liveness analysis past finish_open_created Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 39/69] do_last(): rejoin the common path even earlier in FMODE_{OPENED,CREATED} case Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 40/69] split the lookup-related parts of do_last() into a separate helper Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 41/69] path_connected(): pass mount and dentry separately Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 42/69] path_parent_directory(): leave changing path->dentry to callers Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 43/69] expand path_parent_directory() in its callers Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 44/69] follow_dotdot{,_rcu}(): lift switching nd->path to parent out of loop Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 45/69] follow_dotdot{,_rcu}(): lift LOOKUP_BENEATH checks " Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 46/69] move handle_dots(), follow_dotdot() and follow_dotdot_rcu() past step_into() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 47/69] handle_dots(), follow_dotdot{,_rcu}(): preparation to switch to step_into() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 48/69] follow_dotdot{,_rcu}(): switch to use of step_into() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 49/69] lift all calls of step_into() out of follow_dotdot/follow_dotdot_rcu Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 50/69] follow_dotdot{,_rcu}(): massage loops Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 51/69] follow_dotdot_rcu(): be lazy about changing nd->path Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 52/69] follow_dotdot(): " Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 53/69] helper for mount rootwards traversal Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 54/69] non-RCU analogue of the previous commit Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 55/69] fs/namei.c: kill follow_mount() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 56/69] pick_link(): more straightforward handling of allocation failures Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 57/69] pick_link(): pass it struct path already with normal refcounting rules Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 58/69] fold path_to_nameidata() into its only remaining caller Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 59/69] pick_link(): take reserving space on stack into a new helper Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 60/69] reserve_stack(): switch to __nd_alloc_stack() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 61/69] __nd_alloc_stack(): make it return bool Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 62/69] link_path_walk(): sample parent's i_uid and i_mode for the last component Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 63/69] take post-lookup part of do_last() out of loop Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 64/69] open_last_lookups(): consolidate fsnotify_create() calls Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 65/69] open_last_lookups(): don't abuse complete_walk() when all we want is unlazy Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 66/69] open_last_lookups(): lift O_EXCL|O_CREAT handling into do_open() Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 67/69] open_last_lookups(): move complete_walk() " Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 68/69] atomic_open(): no need to pass struct open_flags anymore Al Viro
2020-03-13 23:53 ` [RFC][PATCH v4 69/69] lookup_open(): don't bother with fallbacks to lookup+create Al Viro
[not found] ` <20200315094148.12872-1-hdanton@sina.com>
2020-03-15 14:23 ` [RFC][PATCH v4 22/69] merging pick_link() with get_link(), part 4 Al Viro
[not found] ` <20200315095323.220-1-hdanton@sina.com>
2020-03-15 14:24 ` [RFC][PATCH v4 23/69] merging pick_link() with get_link(), part 5 Al Viro
[not found] ` <20200315124007.1504-1-hdanton@sina.com>
2020-03-15 14:34 ` [RFC][PATCH v4 38/69] do_last(): simplify the liveness analysis past finish_open_created Al Viro
[not found] ` <20200315102905.12468-1-hdanton@sina.com>
2020-03-15 14:37 ` [RFC][PATCH v4 26/69] sanitize handling of nd->last_type, kill LAST_BIND Al Viro
2020-03-14 0:50 ` [RFC][PATCHSET] sanitized pathwalk machinery (v4) Linus Torvalds
2020-03-14 2:27 ` Al Viro
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200228012451.upnq5r7fdctrk7pv@yavin \
--to=cyphar@cyphar.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=viro@zeniv.linux.org.uk \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).