mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Vlastimil Babka <vbabka@suse.cz>
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>,
	Linus Torvalds <torvalds@linux-foundation.org>
Cc: bpf@vger.kernel.org, daniel@iogearbox.net, andrii@kernel.org,
	martin.lau@kernel.org, akpm@linux-foundation.org,
	peterz@infradead.org, bigeasy@linutronix.de, rostedt@goodmis.org,
	shakeel.butt@linux.dev, mhocko@suse.com, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH] locking/local_lock, mm: Replace localtry_ helpers with local_trylock_t type
Date: Tue, 1 Apr 2025 16:18:02 +0200	[thread overview]
Message-ID: <84d7adee-fa83-4a8b-8476-820212dc929e@suse.cz> (raw)
In-Reply-To: <20250401005134.14433-1-alexei.starovoitov@gmail.com>

On 4/1/25 02:51, Alexei Starovoitov wrote:
> From: Alexei Starovoitov <ast@kernel.org>
> 
> Partially revert commit 0aaddfb06882 ("locking/local_lock: Introduce localtry_lock_t").
> Remove localtry_*() helpers, since localtry_lock() name might
> be misinterpreted as "try lock".
> 
> Introduce local_trylock_irqsave() helper that only works

Introduce local_trylock[_irqsave]() helpers that only work

?

> with newly introduced local_trylock_t type.
> Note that attempt to use local_trylock_irqsave() with local_lock_t
> will cause compilation failure.
> 
> Usage and behavior in !PREEMPT_RT:
> 
> local_lock_t lock;                     // sizeof(lock) == 0

local_lock(&lock, ...);			// preempt disable

> local_lock_irqsave(&lock, ...);        // irq save
> if (local_trylock_irqsave(&lock, ...)) // compilation error
> 
> local_trylock_t lock;                  // sizeof(lock) == 4

ditto

> local_lock_irqsave(&lock, ...);        // irq save and acquired = 1
> if (local_trylock_irqsave(&lock, ...)) // if (!acquired) irq save
> 
> The existing local_lock_*() macros can be used either with
> local_lock_t or local_trylock_t.
> With local_trylock_t they set acquired = 1 while local_unlock_*() clears it.
> 
> In !PREEMPT_RT local_lock_irqsave(local_lock_t *) disables interrupts
> to protect critical section, but it doesn't prevent NMI, so the fully
> reentrant code cannot use local_lock_irqsave(local_lock_t *) for
> exclusive access.
> 
> The local_lock_irqsave(local_trylock_t *) helper disables interrupts
> and sets acquired=1, so local_trylock_irqsave(local_trylock_t *) from
> NMI attempting to acquire the same lock will return false.
> 
> In PREEMPT_RT local_lock_irqsave() maps to preemptible spin_lock().
> Map local_trylock_irqsave() to preemptible spin_trylock().
> When in hard IRQ or NMI return false right away, since
> spin_trylock() is not safe due to explicit locking in the underneath
> rt_spin_trylock() implementation. Removing this explicit locking and
> attempting only "trylock" is undesired due to PI implications.

And something like:

The local_trylock() without _irqsave can be used to avoid the cost of
disabling/enabling interrupts by only disabling preemption, so
local_trylock() in an interrupt attempting to acquire the same
lock will return false.

> Note there is no need to use local_inc for acquired variable,
> since it's a percpu variable with strict nesting scopes.
> 
> Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Acked-by: Vlastimil Babka <vbabka@suse.cz>

Is there a chance this refactoring will make it to -rc1? It would make
basing the further usage of the lock in mm and slab trees much easier.

But squash in the following fixups please:
----8<----
From bc9098ebb58a2958010428c9294547934852ffa2 Mon Sep 17 00:00:00 2001
From: Vlastimil Babka <vbabka@suse.cz>
Date: Tue, 1 Apr 2025 15:25:21 +0200
Subject: [PATCH] fixup! locking/local_lock, mm: Replace localtry_ helpers with
 local_trylock_t type

---
 include/linux/local_lock.h          |  5 ++---
 include/linux/local_lock_internal.h | 21 +++++++++++++++++++--
 2 files changed, 21 insertions(+), 5 deletions(-)

diff --git a/include/linux/local_lock.h b/include/linux/local_lock.h
index 9262109cca51..7ac9385cd475 100644
--- a/include/linux/local_lock.h
+++ b/include/linux/local_lock.h
@@ -52,15 +52,14 @@
 	__local_unlock_irqrestore(lock, flags)
 
 /**
- * local_trylock_irqsave - Try to acquire a per CPU local lock
+ * local_trylock - Try to acquire a per CPU local lock
  * @lock:	The lock variable
- * @flags:	Storage for interrupt flags
  *
  * The function can be used in any context such as NMI or HARDIRQ. Due to
  * locking constrains it will _always_ fail to acquire the lock in NMI or
  * HARDIRQ context on PREEMPT_RT.
  */
-#define local_trylock(lock, flags)	__local_trylock(lock, flags)
+#define local_trylock(lock)		__local_trylock(lock)
 
 /**
  * local_trylock_irqsave - Try to acquire a per CPU local lock, save and disable
diff --git a/include/linux/local_lock_internal.h b/include/linux/local_lock_internal.h
index cc79854206df..5634383c8e9e 100644
--- a/include/linux/local_lock_internal.h
+++ b/include/linux/local_lock_internal.h
@@ -23,7 +23,7 @@ typedef struct {
 #endif
 	/*
 	 * Same layout as local_lock_t with 'acquired' field at the end.
-	 * (local_trylock_t *) will be casted to (local_lock_t *).
+	 * (local_trylock_t *) will be cast to (local_lock_t *).
 	 */
 	int acquired;
 } local_trylock_t;
@@ -80,7 +80,7 @@ do {								\
 	lockdep_init_map_type(&(lock)->dep_map, #lock, &__key,  \
 			      0, LD_WAIT_CONFIG, LD_WAIT_INV,	\
 			      LD_LOCK_PERCPU);			\
-	local_lock_debug_init(lock);				\
+	local_lock_debug_init((local_lock_t *)lock);		\
 } while (0)
 
 #define __spinlock_nested_bh_init(lock)				\
@@ -128,6 +128,23 @@ do {								\
 		__local_lock_acquire(lock);			\
 	} while (0)
 
+#define __local_trylock(lock)					\
+	({							\
+		local_trylock_t *tl;				\
+								\
+		preempt_disable();				\
+		tl = this_cpu_ptr(lock);			\
+		if (READ_ONCE(tl->acquired) == 1) {		\
+			preempt_enable();			\
+			tl = NULL;				\
+		} else {					\
+			WRITE_ONCE(tl->acquired, 1);		\
+			local_trylock_acquire(			\
+				(local_lock_t *)tl);		\
+		}						\
+		!!tl;						\
+	})
+
 #define __local_trylock_irqsave(lock, flags)			\
 	({							\
 		local_trylock_t *tl;				\
-- 
2.49.0



  reply	other threads:[~2025-04-01 14:18 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-01  0:51 Alexei Starovoitov
2025-04-01 14:18 ` Vlastimil Babka [this message]
2025-04-01 20:55   ` Alexei Starovoitov
2025-04-02  7:30 ` Sebastian Andrzej Siewior
2025-04-02  9:02   ` Vlastimil Babka
2025-04-02 21:35     ` Alexei Starovoitov
2025-04-03  0:22       ` Alexei Starovoitov
2025-04-03  9:09       ` Vlastimil Babka
2025-04-03 14:44         ` Alexei Starovoitov

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=84d7adee-fa83-4a8b-8476-820212dc929e@suse.cz \
    --to=vbabka@suse.cz \
    --cc=akpm@linux-foundation.org \
    --cc=alexei.starovoitov@gmail.com \
    --cc=andrii@kernel.org \
    --cc=bigeasy@linutronix.de \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=martin.lau@kernel.org \
    --cc=mhocko@suse.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=shakeel.butt@linux.dev \
    --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

all inboxes | Powered by JetHome®