From: Nicholas Piggin <npiggin@gmail.com>
To: linuxppc-dev@lists.ozlabs.org
Cc: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>,
Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
Nicholas Piggin <npiggin@gmail.com>,
"Nysal Jan K . A" <nysal@linux.ibm.com>
Subject: [PATCH 4/6] powerpc/qspinlock: don't propagate the not-sleepy state
Date: Mon, 16 Oct 2023 22:43:03 +1000 [thread overview]
Message-ID: <20231016124305.139923-5-npiggin@gmail.com> (raw)
In-Reply-To: <20231016124305.139923-1-npiggin@gmail.com>
To simplify things, don't propagate the not-sleepy condition back down
the queue. Instead, have the waiters clear their own node->sleepy when
finding the lock owner is not preempted.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/lib/qspinlock.c | 26 ++++++++------------------
1 file changed, 8 insertions(+), 18 deletions(-)
diff --git a/arch/powerpc/lib/qspinlock.c b/arch/powerpc/lib/qspinlock.c
index 0932d24a6b07..6bb627e90a32 100644
--- a/arch/powerpc/lib/qspinlock.c
+++ b/arch/powerpc/lib/qspinlock.c
@@ -350,7 +350,7 @@ static __always_inline bool yield_head_to_locked_owner(struct qspinlock *lock, u
return __yield_to_locked_owner(lock, val, paravirt, mustq);
}
-static __always_inline void propagate_sleepy(struct qnode *node, u32 val, bool *set_sleepy, bool paravirt)
+static __always_inline void propagate_sleepy(struct qnode *node, u32 val, bool paravirt)
{
struct qnode *next;
int owner;
@@ -359,18 +359,17 @@ static __always_inline void propagate_sleepy(struct qnode *node, u32 val, bool *
return;
if (!pv_yield_propagate_owner)
return;
- if (*set_sleepy)
- return;
next = READ_ONCE(node->next);
if (!next)
return;
+ if (next->sleepy)
+ return;
+
owner = get_owner_cpu(val);
- if (vcpu_is_preempted(owner)) {
+ if (vcpu_is_preempted(owner))
next->sleepy = 1;
- *set_sleepy = true;
- }
}
/* Called inside spin_begin() */
@@ -385,12 +384,7 @@ static __always_inline bool yield_to_prev(struct qspinlock *lock, struct qnode *
if (!pv_yield_propagate_owner)
goto yield_prev;
- if (!READ_ONCE(node->sleepy)) {
- /* Propagate back sleepy==false */
- if (node->next && node->next->sleepy)
- node->next->sleepy = 0;
- goto yield_prev;
- } else {
+ if (node->sleepy) {
u32 val = READ_ONCE(lock->val);
if (val & _Q_LOCKED_VAL) {
@@ -410,6 +404,7 @@ static __always_inline bool yield_to_prev(struct qspinlock *lock, struct qnode *
if (preempted)
return preempted;
}
+ node->sleepy = false;
}
yield_prev:
@@ -533,7 +528,6 @@ static __always_inline void queued_spin_lock_mcs_queue(struct qspinlock *lock, b
bool sleepy = false;
bool mustq = false;
int idx;
- bool set_sleepy = false;
int iters = 0;
BUILD_BUG_ON(CONFIG_NR_CPUS >= (1U << _Q_TAIL_CPU_BITS));
@@ -591,10 +585,6 @@ static __always_inline void queued_spin_lock_mcs_queue(struct qspinlock *lock, b
spec_barrier();
spin_end();
- /* Clear out stale propagated sleepy */
- if (paravirt && pv_yield_propagate_owner && node->sleepy)
- node->sleepy = 0;
-
smp_rmb(); /* acquire barrier for the mcs lock */
/*
@@ -636,7 +626,7 @@ static __always_inline void queued_spin_lock_mcs_queue(struct qspinlock *lock, b
}
}
- propagate_sleepy(node, val, &set_sleepy, paravirt);
+ propagate_sleepy(node, val, paravirt);
preempted = yield_head_to_locked_owner(lock, val, paravirt);
if (!maybe_stealers)
continue;
--
2.42.0
next prev parent reply other threads:[~2023-10-16 12:47 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-16 12:42 [PATCH 0/6] powerpc/qspinlock: Fix yield latency bug and other Nicholas Piggin
2023-10-16 12:43 ` [PATCH 1/6] powerpc/qspinlock: Fix stale propagated yield_cpu Nicholas Piggin
2023-10-16 12:43 ` [PATCH 2/6] powerpc/qspinlock: stop queued waiters trying to set lock sleepy Nicholas Piggin
2023-10-20 9:50 ` Nysal Jan K.A.
2023-10-16 12:43 ` [PATCH 3/6] powerpc/qspinlock: propagate owner preemptedness rather than CPU number Nicholas Piggin
2023-10-16 12:43 ` Nicholas Piggin [this message]
2023-10-16 12:43 ` [PATCH 5/6] powerpc/qspinlock: Propagate sleepy if previous waiter is preempted Nicholas Piggin
2023-10-16 12:43 ` [PATCH 6/6] powerpc/qspinlock: Rename yield_propagate_owner tunable Nicholas Piggin
2023-10-18 7:41 ` [PATCH 0/6] powerpc/qspinlock: Fix yield latency bug and other Shrikanth Hegde
2023-10-20 10:04 ` Nysal Jan K.A.
2023-10-20 12:01 ` (subset) " Michael Ellerman
2023-10-27 9:59 ` Michael Ellerman
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=20231016124305.139923-5-npiggin@gmail.com \
--to=npiggin@gmail.com \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=nysal@linux.ibm.com \
--cc=srikar@linux.vnet.ibm.com \
--cc=sshegde@linux.vnet.ibm.com \
/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).