From: "Paul E. McKenney" <paulmck@kernel.org>
To: rcu@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, kernel-team@meta.com,
rostedt@goodmis.org, "Paul E. McKenney" <paulmck@kernel.org>,
Z qiang <qiang.zhang1211@gmail.com>,
kernel test robot <oliver.sang@intel.com>,
Alexei Starovoitov <ast@kernel.org>,
Andrii Nakryiko <andrii@kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Kent Overstreet <kent.overstreet@linux.dev>,
bpf@vger.kernel.org
Subject: [PATCH rcu v2] 05/20] srcu: Make SRCU readers use ->srcu_ctrs for counter selection
Date: Thu, 30 Jan 2025 11:03:02 -0800 [thread overview]
Message-ID: <20250130190317.1652481-5-paulmck@kernel.org> (raw)
In-Reply-To: <1034ef54-b6b3-42bb-9bd8-4c37c164950d@paulmck-laptop>
This commit causes SRCU readers to use ->srcu_ctrs for counter
selection instead of ->srcu_idx. This takes another step towards
array-indexing-free SRCU readers.
[ paulmck: Apply kernel test robot feedback. ]
Co-developed-by: Z qiang <qiang.zhang1211@gmail.com>
Signed-off-by: Z qiang <qiang.zhang1211@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Tested-by: kernel test robot <oliver.sang@intel.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Kent Overstreet <kent.overstreet@linux.dev>
Cc: <bpf@vger.kernel.org>
---
include/linux/srcutree.h | 9 +++++----
kernel/rcu/srcutree.c | 23 +++++++++++++----------
2 files changed, 18 insertions(+), 14 deletions(-)
diff --git a/include/linux/srcutree.h b/include/linux/srcutree.h
index c794d599db5c..1b01ced61a45 100644
--- a/include/linux/srcutree.h
+++ b/include/linux/srcutree.h
@@ -101,6 +101,7 @@ struct srcu_usage {
*/
struct srcu_struct {
unsigned int srcu_idx; /* Current rdr array element. */
+ struct srcu_ctr __percpu *srcu_ctrp;
struct srcu_data __percpu *sda; /* Per-CPU srcu_data array. */
struct lockdep_map dep_map;
struct srcu_usage *srcu_sup; /* Update-side data. */
@@ -167,6 +168,7 @@ struct srcu_struct {
#define __SRCU_STRUCT_INIT(name, usage_name, pcpu_name) \
{ \
.sda = &pcpu_name, \
+ .srcu_ctrp = &pcpu_name.srcu_ctrs[0], \
__SRCU_STRUCT_INIT_COMMON(name, usage_name) \
}
@@ -222,13 +224,12 @@ void srcu_torture_stats_print(struct srcu_struct *ssp, char *tt, char *tf);
*/
static inline int __srcu_read_lock_lite(struct srcu_struct *ssp)
{
- int idx;
+ struct srcu_ctr __percpu *scp = READ_ONCE(ssp->srcu_ctrp);
RCU_LOCKDEP_WARN(!rcu_is_watching(), "RCU must be watching srcu_read_lock_lite().");
- idx = READ_ONCE(ssp->srcu_idx) & 0x1;
- this_cpu_inc(ssp->sda->srcu_ctrs[idx].srcu_locks.counter); /* Y */
+ this_cpu_inc(scp->srcu_locks.counter); /* Y */
barrier(); /* Avoid leaking the critical section. */
- return idx;
+ return scp - &ssp->sda->srcu_ctrs[0];
}
/*
diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c
index d4e9cd917a69..9af86ce2dd24 100644
--- a/kernel/rcu/srcutree.c
+++ b/kernel/rcu/srcutree.c
@@ -253,8 +253,10 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static)
atomic_set(&ssp->srcu_sup->srcu_barrier_cpu_cnt, 0);
INIT_DELAYED_WORK(&ssp->srcu_sup->work, process_srcu);
ssp->srcu_sup->sda_is_static = is_static;
- if (!is_static)
+ if (!is_static) {
ssp->sda = alloc_percpu(struct srcu_data);
+ ssp->srcu_ctrp = &ssp->sda->srcu_ctrs[0];
+ }
if (!ssp->sda)
goto err_free_sup;
init_srcu_struct_data(ssp);
@@ -742,12 +744,11 @@ EXPORT_SYMBOL_GPL(__srcu_check_read_flavor);
*/
int __srcu_read_lock(struct srcu_struct *ssp)
{
- int idx;
+ struct srcu_ctr __percpu *scp = READ_ONCE(ssp->srcu_ctrp);
- idx = READ_ONCE(ssp->srcu_idx) & 0x1;
- this_cpu_inc(ssp->sda->srcu_ctrs[idx].srcu_locks.counter);
+ this_cpu_inc(scp->srcu_locks.counter);
smp_mb(); /* B */ /* Avoid leaking the critical section. */
- return idx;
+ return scp - &ssp->sda->srcu_ctrs[0];
}
EXPORT_SYMBOL_GPL(__srcu_read_lock);
@@ -772,13 +773,12 @@ EXPORT_SYMBOL_GPL(__srcu_read_unlock);
*/
int __srcu_read_lock_nmisafe(struct srcu_struct *ssp)
{
- int idx;
- struct srcu_data *sdp = raw_cpu_ptr(ssp->sda);
+ struct srcu_ctr __percpu *scpp = READ_ONCE(ssp->srcu_ctrp);
+ struct srcu_ctr *scp = raw_cpu_ptr(scpp);
- idx = READ_ONCE(ssp->srcu_idx) & 0x1;
- atomic_long_inc(&sdp->srcu_ctrs[idx].srcu_locks);
+ atomic_long_inc(&scp->srcu_locks);
smp_mb__after_atomic(); /* B */ /* Avoid leaking the critical section. */
- return idx;
+ return scpp - &ssp->sda->srcu_ctrs[0];
}
EXPORT_SYMBOL_GPL(__srcu_read_lock_nmisafe);
@@ -1152,6 +1152,8 @@ static void srcu_flip(struct srcu_struct *ssp)
smp_mb(); /* E */ /* Pairs with B and C. */
WRITE_ONCE(ssp->srcu_idx, ssp->srcu_idx + 1); // Flip the counter.
+ WRITE_ONCE(ssp->srcu_ctrp,
+ &ssp->sda->srcu_ctrs[!(ssp->srcu_ctrp - &ssp->sda->srcu_ctrs[0])]);
/*
* Ensure that if the updater misses an __srcu_read_unlock()
@@ -2004,6 +2006,7 @@ static int srcu_module_coming(struct module *mod)
ssp->sda = alloc_percpu(struct srcu_data);
if (WARN_ON_ONCE(!ssp->sda))
return -ENOMEM;
+ ssp->srcu_ctrp = &ssp->sda->srcu_ctrs[0];
}
return 0;
}
--
2.40.1
next prev parent reply other threads:[~2025-01-30 19:03 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-01-16 20:21 [PATCH rcu 0/17] SRCU updates, including SRCU-fast Paul E. McKenney
2025-01-16 20:20 ` [PATCH rcu 01/17] srcu: Make Tiny SRCU able to operate in preemptible kernels Paul E. McKenney
2025-01-16 20:20 ` [PATCH rcu 02/17] srcu: Define SRCU_READ_FLAVOR_ALL in terms of symbols Paul E. McKenney
2025-01-16 20:20 ` [PATCH rcu 03/17] srcu: Use ->srcu_gp_seq for rcutorture reader batch Paul E. McKenney
2025-01-16 20:20 ` [PATCH rcu 04/17] srcu: Pull ->srcu_{un,}lock_count into a new srcu_ctr structure Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 05/17] srcu: Make SRCU readers use ->srcu_ctrs for counter selection Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 06/17] srcu: Make Tree SRCU updates independent of ->srcu_idx Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 07/17] srcu: Force synchronization for srcu_get_delay() Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 08/17] srcu: Rename srcu_check_read_flavor_lite() to srcu_check_read_flavor_force() Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 09/17] srcu: Add SRCU_READ_FLAVOR_SLOWGP to flag need for synchronize_rcu() Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 10/17] srcu: Pull pointer-to-integer conversion into __srcu_ptr_to_ctr() Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 11/17] srcu: Pull integer-to-pointer conversion into __srcu_ctr_to_ptr() Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 12/17] srcu: Move SRCU Tree/Tiny definitions from srcu.h Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 13/17] srcu: Add SRCU-fast readers Paul E. McKenney
2025-01-16 21:00 ` Alexei Starovoitov
2025-01-16 21:38 ` Peter Zijlstra
2025-01-16 21:55 ` Paul E. McKenney
2025-01-16 22:58 ` Alexei Starovoitov
2025-01-16 21:52 ` Andrii Nakryiko
2025-01-16 22:54 ` Paul E. McKenney
2025-01-16 22:57 ` Andrii Nakryiko
2025-01-17 0:07 ` Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 14/17] rcutorture: Add ability to test srcu_read_{,un}lock_fast() Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 15/17] refscale: Add srcu_read_lock_fast() support using "srcu-fast" Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 16/17] rcutorture: Make scenario SRCU-P use srcu_read_lock_fast() Paul E. McKenney
2025-01-16 20:21 ` [PATCH rcu 17/17] srcu: Fix srcu_read_unlock_{lite,nmisafe}() kernel-doc Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu 0/17] SRCU updates, including SRCU-fast Paul E. McKenney
2025-01-30 19:02 ` [PATCH rcu v2] 01/20] srcu: Make Tiny SRCU able to operate in preemptible kernels Paul E. McKenney
2025-01-30 19:02 ` [PATCH rcu v2] 02/20] srcu: Define SRCU_READ_FLAVOR_ALL in terms of symbols Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 03/20] srcu: Use ->srcu_gp_seq for rcutorture reader batch Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 04/20] srcu: Pull ->srcu_{un,}lock_count into a new srcu_ctr structure Paul E. McKenney
2025-01-30 19:03 ` Paul E. McKenney [this message]
2025-01-30 19:03 ` [PATCH rcu v2] 06/20] srcu: Make Tree SRCU updates independent of ->srcu_idx Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 07/20] srcu: Force synchronization for srcu_get_delay() Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 08/20] srcu: Rename srcu_check_read_flavor_lite() to srcu_check_read_flavor_force() Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 09/20] srcu: Add SRCU_READ_FLAVOR_SLOWGP to flag need for synchronize_rcu() Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 10/20] srcu: Pull pointer-to-integer conversion into __srcu_ptr_to_ctr() Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 11/20] srcu: Pull integer-to-pointer conversion into __srcu_ctr_to_ptr() Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 12/20] srcu: Move SRCU Tree/Tiny definitions from srcu.h Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 13/20] srcu: Add SRCU-fast readers Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 14/20] rcutorture: Add ability to test srcu_read_{,un}lock_fast() Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 15/20] refscale: Add srcu_read_lock_fast() support using "srcu-fast" Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 16/20] rcutorture: Make scenario SRCU-P use srcu_read_lock_fast() Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 17/20] srcu: Fix srcu_read_unlock_{lite,nmisafe}() kernel-doc Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 18/20] srcu: Document that srcu_{read_lock,down_read}() can share srcu_struct Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 19/20] srcu: Add srcu_down_read_fast() and srcu_up_read_fast() Paul E. McKenney
2025-01-30 19:03 ` [PATCH rcu v2] 20/20] srcu: Make SRCU-fast also be NMI-safe Paul E. McKenney
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=20250130190317.1652481-5-paulmck@kernel.org \
--to=paulmck@kernel.org \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=kent.overstreet@linux.dev \
--cc=kernel-team@meta.com \
--cc=linux-kernel@vger.kernel.org \
--cc=oliver.sang@intel.com \
--cc=peterz@infradead.org \
--cc=qiang.zhang1211@gmail.com \
--cc=rcu@vger.kernel.org \
--cc=rostedt@goodmis.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