From: NeilBrown <neilb@suse.de>
To: Ingo Molnar <mingo@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org
Subject: [PATCH 3/9] XFS: use wait_var_event() when waiting of i_pincount.
Date: Mon, 19 Aug 2024 15:20:37 +1000 [thread overview]
Message-ID: <20240819053605.11706-4-neilb@suse.de> (raw)
In-Reply-To: <20240819053605.11706-1-neilb@suse.de>
__xfs_iunpin_wait() is nearly an open-coded version of wait_var_event().
The bit XFS_IPINNED in ->i_flags is never used (not set, cleared,
or tested). Its only role is in choosing a wait_queue.
The code is more transparent if we discard that flag bit and use
wait_var_event() to wait for the pincount to reach zero, which is
signalled by wake_up_var() - or more specifically
atomic_dec_and_wake_up_var()
In order to use io_schedule(), we actually use ___wait_var_event().
Signed-off-by: NeilBrown <neilb@suse.de>
---
fs/xfs/xfs_inode.c | 24 +++++-------------------
fs/xfs/xfs_inode.h | 2 --
fs/xfs/xfs_inode_item.c | 3 +--
3 files changed, 6 insertions(+), 23 deletions(-)
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 7dc6f326936c..a855295363b5 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -1899,29 +1899,15 @@ xfs_iunpin(
}
-static void
-__xfs_iunpin_wait(
- struct xfs_inode *ip)
-{
- wait_queue_head_t *wq = bit_waitqueue(&ip->i_flags, __XFS_IPINNED_BIT);
- DEFINE_WAIT_BIT(wait, &ip->i_flags, __XFS_IPINNED_BIT);
-
- xfs_iunpin(ip);
-
- do {
- prepare_to_wait(wq, &wait.wq_entry, TASK_UNINTERRUPTIBLE);
- if (xfs_ipincount(ip))
- io_schedule();
- } while (xfs_ipincount(ip));
- finish_wait(wq, &wait.wq_entry);
-}
-
void
xfs_iunpin_wait(
struct xfs_inode *ip)
{
- if (xfs_ipincount(ip))
- __xfs_iunpin_wait(ip);
+ if (xfs_ipincount(ip)) {
+ xfs_iunpin(ip);
+ ___wait_var_event(&ip->i_pincount, xfs_ipincount(ip) == 0,
+ TASK_UNINTERRUPTIBLE, 0, 0, io_schedule());
+ }
}
/*
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 51defdebef30..5062580034ec 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -337,8 +337,6 @@ static inline bool xfs_inode_has_bigrtalloc(struct xfs_inode *ip)
#define XFS_ITRUNCATED (1 << 5) /* truncated down so flush-on-close */
#define XFS_IDIRTY_RELEASE (1 << 6) /* dirty release already seen */
#define XFS_IFLUSHING (1 << 7) /* inode is being flushed */
-#define __XFS_IPINNED_BIT 8 /* wakeup key for zero pin count */
-#define XFS_IPINNED (1 << __XFS_IPINNED_BIT)
#define XFS_IEOFBLOCKS (1 << 9) /* has the preallocblocks tag set */
#define XFS_NEED_INACTIVE (1 << 10) /* see XFS_INACTIVATING below */
/*
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index b509cbd191f4..7fee751c7dfd 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -712,8 +712,7 @@ xfs_inode_item_unpin(
trace_xfs_inode_unpin(ip, _RET_IP_);
ASSERT(lip->li_buf || xfs_iflags_test(ip, XFS_ISTALE));
ASSERT(atomic_read(&ip->i_pincount) > 0);
- if (atomic_dec_and_test(&ip->i_pincount))
- wake_up_bit(&ip->i_flags, __XFS_IPINNED_BIT);
+ atomic_dec_and_wake_up_var(&ip->i_pincount);
}
STATIC uint
--
2.44.0
next prev parent reply other threads:[~2024-08-19 5:36 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-19 5:20 [PATCH 0/9 RFC] Make wake_up_{bit,var} less fragile NeilBrown
2024-08-19 5:20 ` [PATCH 1/9] i915: remove wake_up on I915_RESET_MODESET NeilBrown
2024-08-19 5:20 ` [PATCH 2/9] Introduce atomic_dec_and_wake_up_var() NeilBrown
2024-08-20 5:47 ` kernel test robot
2024-08-21 5:23 ` kernel test robot
2024-08-19 5:20 ` NeilBrown [this message]
2024-08-19 5:20 ` [PATCH 4/9] Use wait_var_event() instead of I_DIO_WAKEUP NeilBrown
2024-08-20 7:22 ` Christian Brauner
2024-08-20 19:12 ` Christian Brauner
2024-08-19 5:20 ` [PATCH 5/9] Block: switch bd_prepare_to_claim to use ___wait_var_event() NeilBrown
2024-08-20 4:18 ` Dave Chinner
2024-08-20 21:52 ` NeilBrown
2024-08-28 1:22 ` Dave Chinner
2024-08-28 5:20 ` NeilBrown
2024-08-21 7:10 ` kernel test robot
2024-08-19 5:20 ` [PATCH 6/9] block/pktdvd: switch congestion waiting to ___wait_var_event() NeilBrown
2024-08-19 5:20 ` [PATCH 7/9] Improve and expand wake_up_bit() interface NeilBrown
2024-08-19 5:20 ` [PATCH 8/9] Improve and extend wake_up_var() interface NeilBrown
2024-08-19 5:20 ` [PATCH 9/9] Use clear_and_wake_up_bit() where appropriate NeilBrown
2024-08-19 6:13 ` [PATCH 0/9 RFC] Make wake_up_{bit,var} less fragile Linus Torvalds
2024-08-20 21:47 ` NeilBrown
2024-08-20 21:58 ` Linus Torvalds
2024-08-20 22:15 ` NeilBrown
2024-08-20 22:24 ` Linus Torvalds
2024-08-19 8:16 ` Christian Brauner
2024-08-19 17:45 ` Linus Torvalds
2024-08-19 20:52 ` NeilBrown
2024-08-19 21:12 ` Linus Torvalds
2024-08-20 16:06 ` [PATCH RFC 0/5] inode: turn i_state into u32 Christian Brauner
2024-08-20 16:06 ` [PATCH RFC 1/5] fs: add i_state helpers Christian Brauner
2024-08-20 17:10 ` Linus Torvalds
2024-08-20 17:19 ` Linus Torvalds
2024-08-20 19:10 ` Christian Brauner
2024-08-20 19:08 ` Christian Brauner
2024-08-20 16:06 ` [PATCH RFC 2/5] writeback: port __I_SYNC to var event Christian Brauner
2024-08-20 16:06 ` [PATCH RFC 3/5] inode: port __I_NEW " Christian Brauner
2024-08-20 16:06 ` [PATCH RFC 4/5] inode: port __I_LRU_ISOLATING " Christian Brauner
2024-08-20 16:06 ` [PATCH RFC 5/5] inode: make i_state a u32 Christian Brauner
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=20240819053605.11706-4-neilb@suse.de \
--to=neilb@suse.de \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=torvalds@linux-foundation.org \
/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