From: Oleg Nesterov <oleg@redhat.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Boqun Feng <boqun.feng@gmail.com>,
David Howells <dhowells@redhat.com>,
Ingo Molnar <mingo@redhat.com>,
Li RongQing <lirongqing@baidu.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Waiman Long <longman@redhat.com>, Will Deacon <will@kernel.org>,
linux-kernel@vger.kernel.org
Subject: Re: [RFC 2/1] seqlock: make the read_seqbegin_or_lock() API more simple and less error-prone ?
Date: Wed, 1 Oct 2025 15:46:31 +0200 [thread overview]
Message-ID: <20251001134628.GD20441@redhat.com> (raw)
In-Reply-To: <20251001131337.GC20441@redhat.com>
On 10/01, Oleg Nesterov wrote:
>
> +static inline int xxx(seqlock_t *lock, int lockless, int *seq, unsigned long *flags)
> +{
> + if (lockless) {
> + *seq = read_seqbegin(lock);
> + return 1;
> + } else if (*seq & 1) {
> + if (flags)
> + read_sequnlock_excl_irqrestore(lock, *flags);
> + else
> + read_sequnlock_excl(lock);
> + return 0;
> + } else if (read_seqretry(lock, *seq)) {
> + if (flags)
> + read_seqlock_excl_irqsave(lock, *flags);
> + else
> + read_seqlock_excl(lock);
> + *seq = 1;
> + return 1;
> + } else {
> + return 0;
> + }
> +}
> +
> +#define __XXX(lock, lockless, seq, flags) \
> + for (int lockless = 1, seq; xxx(lock, lockless, &seq, flags); lockless = 0)
> +
> +#define XXX(lock, flags) \
> + __XXX(lock, __UNIQUE_ID(lockless), __UNIQUE_ID(seq), flags)
Note that __XXX() can have users too. See the patch below. Not that it makes a
lot of sense, just for example.
Oleg.
diff --git a/fs/d_path.c b/fs/d_path.c
index bb365511066b..b4cde188be4f 100644
--- a/fs/d_path.c
+++ b/fs/d_path.c
@@ -332,28 +332,23 @@ static char *__dentry_path(const struct dentry *d, struct prepend_buffer *p)
{
const struct dentry *dentry;
struct prepend_buffer b;
- int seq = 0;
rcu_read_lock();
-restart:
- dentry = d;
- b = *p;
- read_seqbegin_or_lock(&rename_lock, &seq);
- while (!IS_ROOT(dentry)) {
- const struct dentry *parent = dentry->d_parent;
+ __XXX(&rename_lock, lockless, seq, NULL) {
+ dentry = d;
+ b = *p;
+ while (!IS_ROOT(dentry)) {
+ const struct dentry *parent = dentry->d_parent;
- prefetch(parent);
- if (!prepend_name(&b, &dentry->d_name))
- break;
- dentry = parent;
- }
- if (!(seq & 1))
- rcu_read_unlock();
- if (need_seqretry(&rename_lock, seq)) {
- seq = 1;
- goto restart;
+ prefetch(parent);
+ if (!prepend_name(&b, &dentry->d_name))
+ break;
+ dentry = parent;
+ }
+ if (lockless)
+ rcu_read_unlock();
}
- done_seqretry(&rename_lock, seq);
+
if (b.len == p->len)
prepend_char(&b, '/');
return extract_string(&b);
next prev parent reply other threads:[~2025-10-01 13:48 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-28 16:19 [PATCH 0/1] documentation: seqlock: fix the wrong documentation of read_seqbegin_or_lock/need_seqretry Oleg Nesterov
2025-09-28 16:20 ` [PATCH 1/1] " Oleg Nesterov
2025-10-01 18:21 ` Waiman Long
2025-10-01 19:06 ` Oleg Nesterov
2025-10-01 19:24 ` Waiman Long
2025-10-01 19:34 ` Waiman Long
2025-10-02 11:01 ` Oleg Nesterov
2025-10-21 10:35 ` [tip: locking/core] " tip-bot2 for Oleg Nesterov
2025-09-28 16:20 ` [RFC 2/1] seqlock: make the read_seqbegin_or_lock() API more simple and less error-prone ? Oleg Nesterov
2025-09-29 0:41 ` [????] " Li,Rongqing
2025-09-29 6:47 ` Oleg Nesterov
2025-09-30 22:09 ` David Howells
2025-10-01 11:51 ` Oleg Nesterov
2025-10-01 13:02 ` Peter Zijlstra
2025-10-01 13:13 ` Oleg Nesterov
2025-10-01 13:46 ` Oleg Nesterov [this message]
2025-10-02 12:58 ` Oleg Nesterov
2025-10-05 14:47 ` [PATCH 0/5] seqlock: introduce SEQLOCK_READ_SECTION() Oleg Nesterov
2025-10-05 14:49 ` Oleg Nesterov
2025-10-05 14:50 ` [PATCH 1/5] " Oleg Nesterov
2025-10-05 15:34 ` Linus Torvalds
2025-10-05 16:07 ` Oleg Nesterov
2025-10-05 16:35 ` Linus Torvalds
2025-10-05 14:50 ` [PATCH 2/5] seqlock: change thread_group_cputime() to use SEQLOCK_READ_SECTION() Oleg Nesterov
2025-10-05 14:50 ` [PATCH 3/5] seqlock: change do_task_stat() " Oleg Nesterov
2025-10-05 14:50 ` [PATCH 4/5] seqlock: change do_io_accounting() " Oleg Nesterov
2025-10-05 14:50 ` [PATCH 5/5] seqlock: change __dentry_path() to use __SEQLOCK_READ_SECTION() Oleg Nesterov
2025-10-05 15:48 ` Linus Torvalds
2025-10-05 15:30 ` [PATCH 0/5] seqlock: introduce SEQLOCK_READ_SECTION() Al Viro
2025-10-05 17:40 ` Oleg Nesterov
2025-10-07 14:20 ` [PATCH 0/4] seqlock: introduce scoped_seqlock_read() and scoped_seqlock_read_irqsave() Oleg Nesterov
2025-10-07 14:21 ` [PATCH 1/4] " Oleg Nesterov
2025-10-07 16:35 ` Waiman Long
2025-10-07 17:18 ` Oleg Nesterov
2025-10-07 17:21 ` Waiman Long
2025-10-07 14:21 ` [PATCH 2/4] seqlock: change thread_group_cputime() to use scoped_seqlock_read_irqsave() Oleg Nesterov
2025-10-07 14:21 ` [PATCH 3/4] seqlock: change do_task_stat() " Oleg Nesterov
2025-10-07 14:21 ` [PATCH 4/4] seqlock: change do_io_accounting() " Oleg Nesterov
2025-10-07 15:38 ` [PATCH 0/4] seqlock: introduce scoped_seqlock_read() and scoped_seqlock_read_irqsave() Linus Torvalds
2025-10-07 16:34 ` Oleg Nesterov
2025-10-08 12:30 ` [PATCH v2 " Oleg Nesterov
2025-10-08 12:30 ` [PATCH v2 1/4] " Oleg Nesterov
2025-10-08 12:55 ` Peter Zijlstra
2025-10-08 12:59 ` Oleg Nesterov
2025-10-08 13:54 ` Peter Zijlstra
2025-10-08 16:05 ` Linus Torvalds
2025-10-08 16:55 ` Oleg Nesterov
2025-10-09 5:31 ` Linus Torvalds
2025-10-09 7:04 ` Linus Torvalds
2025-10-09 14:37 ` Oleg Nesterov
2025-10-09 16:18 ` Linus Torvalds
2025-10-09 19:50 ` Peter Zijlstra
2025-10-09 20:11 ` Peter Zijlstra
2025-10-09 20:24 ` Linus Torvalds
2025-10-09 22:12 ` Peter Zijlstra
2025-10-09 22:55 ` Linus Torvalds
2025-10-10 8:03 ` Peter Zijlstra
2025-10-10 12:32 ` Oleg Nesterov
2025-10-10 13:14 ` Oleg Nesterov
2025-10-13 9:03 ` Peter Zijlstra
2025-10-13 11:50 ` Oleg Nesterov
2025-10-10 15:30 ` Linus Torvalds
2025-10-09 23:20 ` Peter Zijlstra
2025-10-09 23:26 ` Linus Torvalds
2025-10-21 10:35 ` [tip: locking/core] seqlock: Introduce scoped_seqlock_read() tip-bot2 for Peter Zijlstra
2025-10-08 12:30 ` [PATCH v2 2/4] seqlock: change thread_group_cputime() to use scoped_seqlock_read_irqsave() Oleg Nesterov
2025-10-21 10:35 ` [tip: locking/core] seqlock: Change thread_group_cputime() to use scoped_seqlock_read() tip-bot2 for Oleg Nesterov
2025-10-08 12:30 ` [PATCH v2 3/4] seqlock: change do_task_stat() to use scoped_seqlock_read_irqsave() Oleg Nesterov
2025-10-21 10:35 ` [tip: locking/core] seqlock: Change do_task_stat() to use scoped_seqlock_read() tip-bot2 for Oleg Nesterov
2025-10-08 12:31 ` [PATCH v2 4/4] seqlock: change do_io_accounting() to use scoped_seqlock_read_irqsave() Oleg Nesterov
2025-10-21 10:35 ` [tip: locking/core] seqlock: Change do_io_accounting() to use scoped_seqlock_read() tip-bot2 for Oleg Nesterov
2025-10-08 12:56 ` [PATCH v2 0/4] seqlock: introduce scoped_seqlock_read() and scoped_seqlock_read_irqsave() Peter Zijlstra
2025-10-08 13:13 ` Oleg Nesterov
2025-10-08 13:55 ` Peter Zijlstra
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=20251001134628.GD20441@redhat.com \
--to=oleg@redhat.com \
--cc=boqun.feng@gmail.com \
--cc=dhowells@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lirongqing@baidu.com \
--cc=longman@redhat.com \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=torvalds@linux-foundation.org \
--cc=will@kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.