public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
To: bpf@vger.kernel.org, linux-kernel@vger.kernel.org
Cc: Barret Rhoden <brho@google.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Waiman Long <llong@redhat.com>,
	Alexei Starovoitov <ast@kernel.org>,
	Andrii Nakryiko <andrii@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Martin KaFai Lau <martin.lau@kernel.org>,
	Eduard Zingerman <eddyz87@gmail.com>,
	"Paul E. McKenney" <paulmck@kernel.org>,
	Tejun Heo <tj@kernel.org>, Josh Don <joshdon@google.com>,
	Dohyun Kim <dohyunkim@google.com>,
	kernel-team@meta.com
Subject: [PATCH bpf-next v1 06/22] rqspinlock: Drop PV and virtualization support
Date: Tue,  7 Jan 2025 05:59:48 -0800	[thread overview]
Message-ID: <20250107140004.2732830-7-memxor@gmail.com> (raw)
In-Reply-To: <20250107140004.2732830-1-memxor@gmail.com>

Changes to rqspinlock in subsequent commits will be algorithmic
modifications, which won't remain in agreement with the implementations
of paravirt spin lock and virt_spin_lock support. These future changes
include measures for terminating waiting loops in slow path after a
certain point. While using a fair lock like qspinlock directly inside
virtual machines leads to suboptimal performance under certain
conditions, we cannot use the existing virtualization support before we
make it resilient as well.  Therefore, drop it for now.

Reviewed-by: Barret Rhoden <brho@google.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
 kernel/locking/rqspinlock.c | 89 -------------------------------------
 1 file changed, 89 deletions(-)

diff --git a/kernel/locking/rqspinlock.c b/kernel/locking/rqspinlock.c
index b7920ae79410..fada0dca6f3b 100644
--- a/kernel/locking/rqspinlock.c
+++ b/kernel/locking/rqspinlock.c
@@ -11,8 +11,6 @@
  *          Peter Zijlstra <peterz@infradead.org>
  */
 
-#ifndef _GEN_PV_LOCK_SLOWPATH
-
 #include <linux/smp.h>
 #include <linux/bug.h>
 #include <linux/cpumask.h>
@@ -75,38 +73,9 @@
  * contexts: task, softirq, hardirq, nmi.
  *
  * Exactly fits one 64-byte cacheline on a 64-bit architecture.
- *
- * PV doubles the storage and uses the second cacheline for PV state.
  */
 static DEFINE_PER_CPU_ALIGNED(struct qnode, qnodes[_Q_MAX_NODES]);
 
-/*
- * Generate the native code for resilient_queued_spin_unlock_slowpath(); provide NOPs
- * for all the PV callbacks.
- */
-
-static __always_inline void __pv_init_node(struct mcs_spinlock *node) { }
-static __always_inline void __pv_wait_node(struct mcs_spinlock *node,
-					   struct mcs_spinlock *prev) { }
-static __always_inline void __pv_kick_node(struct qspinlock *lock,
-					   struct mcs_spinlock *node) { }
-static __always_inline u32  __pv_wait_head_or_lock(struct qspinlock *lock,
-						   struct mcs_spinlock *node)
-						   { return 0; }
-
-#define pv_enabled()		false
-
-#define pv_init_node		__pv_init_node
-#define pv_wait_node		__pv_wait_node
-#define pv_kick_node		__pv_kick_node
-#define pv_wait_head_or_lock	__pv_wait_head_or_lock
-
-#ifdef CONFIG_PARAVIRT_SPINLOCKS
-#define resilient_queued_spin_lock_slowpath	native_resilient_queued_spin_lock_slowpath
-#endif
-
-#endif /* _GEN_PV_LOCK_SLOWPATH */
-
 /**
  * resilient_queued_spin_lock_slowpath - acquire the queued spinlock
  * @lock: Pointer to queued spinlock structure
@@ -136,12 +105,6 @@ void __lockfunc resilient_queued_spin_lock_slowpath(struct qspinlock *lock, u32
 
 	BUILD_BUG_ON(CONFIG_NR_CPUS >= (1U << _Q_TAIL_CPU_BITS));
 
-	if (pv_enabled())
-		goto pv_queue;
-
-	if (virt_spin_lock(lock))
-		return;
-
 	/*
 	 * Wait for in-progress pending->locked hand-overs with a bounded
 	 * number of spins so that we guarantee forward progress.
@@ -212,7 +175,6 @@ void __lockfunc resilient_queued_spin_lock_slowpath(struct qspinlock *lock, u32
 	 */
 queue:
 	lockevent_inc(lock_slowpath);
-pv_queue:
 	node = this_cpu_ptr(&qnodes[0].mcs);
 	idx = node->count++;
 	tail = encode_tail(smp_processor_id(), idx);
@@ -251,7 +213,6 @@ void __lockfunc resilient_queued_spin_lock_slowpath(struct qspinlock *lock, u32
 
 	node->locked = 0;
 	node->next = NULL;
-	pv_init_node(node);
 
 	/*
 	 * We touched a (possibly) cold cacheline in the per-cpu queue node;
@@ -288,7 +249,6 @@ void __lockfunc resilient_queued_spin_lock_slowpath(struct qspinlock *lock, u32
 		/* Link @node into the waitqueue. */
 		WRITE_ONCE(prev->next, node);
 
-		pv_wait_node(node, prev);
 		arch_mcs_spin_lock_contended(&node->locked);
 
 		/*
@@ -312,23 +272,9 @@ void __lockfunc resilient_queued_spin_lock_slowpath(struct qspinlock *lock, u32
 	 * store-release that clears the locked bit and create lock
 	 * sequentiality; this is because the set_locked() function below
 	 * does not imply a full barrier.
-	 *
-	 * The PV pv_wait_head_or_lock function, if active, will acquire
-	 * the lock and return a non-zero value. So we have to skip the
-	 * atomic_cond_read_acquire() call. As the next PV queue head hasn't
-	 * been designated yet, there is no way for the locked value to become
-	 * _Q_SLOW_VAL. So both the set_locked() and the
-	 * atomic_cmpxchg_relaxed() calls will be safe.
-	 *
-	 * If PV isn't active, 0 will be returned instead.
-	 *
 	 */
-	if ((val = pv_wait_head_or_lock(lock, node)))
-		goto locked;
-
 	val = atomic_cond_read_acquire(&lock->val, !(VAL & _Q_LOCKED_PENDING_MASK));
 
-locked:
 	/*
 	 * claim the lock:
 	 *
@@ -341,11 +287,6 @@ void __lockfunc resilient_queued_spin_lock_slowpath(struct qspinlock *lock, u32
 	 */
 
 	/*
-	 * In the PV case we might already have _Q_LOCKED_VAL set, because
-	 * of lock stealing; therefore we must also allow:
-	 *
-	 * n,0,1 -> 0,0,1
-	 *
 	 * Note: at this point: (val & _Q_PENDING_MASK) == 0, because of the
 	 *       above wait condition, therefore any concurrent setting of
 	 *       PENDING will make the uncontended transition fail.
@@ -369,7 +310,6 @@ void __lockfunc resilient_queued_spin_lock_slowpath(struct qspinlock *lock, u32
 		next = smp_cond_load_relaxed(&node->next, (VAL));
 
 	arch_mcs_spin_unlock_contended(&next->locked);
-	pv_kick_node(lock, next);
 
 release:
 	trace_contention_end(lock, 0);
@@ -380,32 +320,3 @@ void __lockfunc resilient_queued_spin_lock_slowpath(struct qspinlock *lock, u32
 	__this_cpu_dec(qnodes[0].mcs.count);
 }
 EXPORT_SYMBOL(resilient_queued_spin_lock_slowpath);
-
-/*
- * Generate the paravirt code for resilient_queued_spin_unlock_slowpath().
- */
-#if !defined(_GEN_PV_LOCK_SLOWPATH) && defined(CONFIG_PARAVIRT_SPINLOCKS)
-#define _GEN_PV_LOCK_SLOWPATH
-
-#undef  pv_enabled
-#define pv_enabled()	true
-
-#undef pv_init_node
-#undef pv_wait_node
-#undef pv_kick_node
-#undef pv_wait_head_or_lock
-
-#undef  resilient_queued_spin_lock_slowpath
-#define resilient_queued_spin_lock_slowpath	__pv_resilient_queued_spin_lock_slowpath
-
-#include "qspinlock_paravirt.h"
-#include "rqspinlock.c"
-
-bool nopvspin;
-static __init int parse_nopvspin(char *arg)
-{
-	nopvspin = true;
-	return 0;
-}
-early_param("nopvspin", parse_nopvspin);
-#endif
-- 
2.43.5


  parent reply	other threads:[~2025-01-07 14:00 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-07 13:59 [PATCH bpf-next v1 00/22] Resilient Queued Spin Lock Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 01/22] locking: Move MCS struct definition to public header Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 02/22] locking: Move common qspinlock helpers to a private header Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 03/22] locking: Allow obtaining result of arch_mcs_spin_lock_contended Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 04/22] locking: Copy out qspinlock.c to rqspinlock.c Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 05/22] rqspinlock: Add rqspinlock.h header Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` Kumar Kartikeya Dwivedi [this message]
2025-01-07 13:59 ` [PATCH bpf-next v1 07/22] rqspinlock: Add support for timeouts Kumar Kartikeya Dwivedi
2025-01-07 14:50   ` Peter Zijlstra
2025-01-07 17:14     ` Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 08/22] rqspinlock: Protect pending bit owners from stalls Kumar Kartikeya Dwivedi
2025-01-07 14:51   ` Peter Zijlstra
2025-01-07 17:14     ` Kumar Kartikeya Dwivedi
2025-01-07 19:17       ` Peter Zijlstra
2025-01-07 19:22         ` Peter Zijlstra
2025-01-07 19:54           ` Kumar Kartikeya Dwivedi
2025-01-08  2:19   ` Waiman Long
2025-01-08 20:13     ` Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 09/22] rqspinlock: Protect waiters in queue " Kumar Kartikeya Dwivedi
2025-01-08  3:38   ` Waiman Long
2025-01-08 20:42     ` Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 10/22] rqspinlock: Protect waiters in trylock fallback " Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 11/22] rqspinlock: Add deadlock detection and recovery Kumar Kartikeya Dwivedi
2025-01-08 16:06   ` Waiman Long
2025-01-08 20:19     ` Kumar Kartikeya Dwivedi
2025-01-09  0:32       ` Waiman Long
2025-01-07 13:59 ` [PATCH bpf-next v1 12/22] rqspinlock: Add basic support for CONFIG_PARAVIRT Kumar Kartikeya Dwivedi
2025-01-08 16:27   ` Waiman Long
2025-01-08 20:32     ` Kumar Kartikeya Dwivedi
2025-01-09  0:48       ` Waiman Long
2025-01-09  2:42         ` Alexei Starovoitov
2025-01-09  2:58           ` Waiman Long
2025-01-09  3:37             ` Alexei Starovoitov
2025-01-09  3:46               ` Waiman Long
2025-01-09  3:53                 ` Alexei Starovoitov
2025-01-09  3:58                   ` Waiman Long
2025-01-07 13:59 ` [PATCH bpf-next v1 13/22] rqspinlock: Add helper to print a splat on timeout or deadlock Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 14/22] rqspinlock: Add macros for rqspinlock usage Kumar Kartikeya Dwivedi
2025-01-08 16:55   ` Waiman Long
2025-01-08 20:41     ` Kumar Kartikeya Dwivedi
2025-01-09  1:11       ` Waiman Long
2025-01-09  3:30         ` Alexei Starovoitov
2025-01-09  4:09           ` Waiman Long
2025-01-07 13:59 ` [PATCH bpf-next v1 15/22] rqspinlock: Add locktorture support Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 16/22] rqspinlock: Add entry to Makefile, MAINTAINERS Kumar Kartikeya Dwivedi
2025-01-07 13:59 ` [PATCH bpf-next v1 17/22] bpf: Convert hashtab.c to rqspinlock Kumar Kartikeya Dwivedi
2025-01-07 14:00 ` [PATCH bpf-next v1 18/22] bpf: Convert percpu_freelist.c " Kumar Kartikeya Dwivedi
2025-01-07 14:00 ` [PATCH bpf-next v1 19/22] bpf: Convert lpm_trie.c " Kumar Kartikeya Dwivedi
2025-01-07 14:00 ` [PATCH bpf-next v1 20/22] bpf: Introduce rqspinlock kfuncs Kumar Kartikeya Dwivedi
2025-01-08 10:23   ` kernel test robot
2025-01-08 10:23   ` kernel test robot
2025-01-08 10:44   ` kernel test robot
2025-01-07 14:00 ` [PATCH bpf-next v1 21/22] bpf: Implement verifier support for rqspinlock Kumar Kartikeya Dwivedi
2025-01-07 14:00 ` [PATCH bpf-next v1 22/22] selftests/bpf: Add tests " Kumar Kartikeya Dwivedi
2025-01-07 23:54 ` [PATCH bpf-next v1 00/22] Resilient Queued Spin Lock Linus Torvalds
2025-01-08  9:18   ` Peter Zijlstra
2025-01-08 20:12     ` Kumar Kartikeya Dwivedi
2025-01-08 20:30       ` Linus Torvalds
2025-01-08 21:06         ` Kumar Kartikeya Dwivedi
2025-01-08 21:30         ` Paul E. McKenney
2025-01-09 13:59       ` Waiman Long
2025-01-09 21:13         ` Kumar Kartikeya Dwivedi
2025-01-09 21:18           ` Waiman Long

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=20250107140004.2732830-7-memxor@gmail.com \
    --to=memxor@gmail.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=brho@google.com \
    --cc=daniel@iogearbox.net \
    --cc=dohyunkim@google.com \
    --cc=eddyz87@gmail.com \
    --cc=joshdon@google.com \
    --cc=kernel-team@meta.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=llong@redhat.com \
    --cc=martin.lau@kernel.org \
    --cc=paulmck@kernel.org \
    --cc=peterz@infradead.org \
    --cc=tj@kernel.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