[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <175097494507.406.15997424702190501157.tip-bot2@tip-bot2>
Date: Thu, 26 Jun 2025 21:55:45 -0000
From: "tip-bot2 for Yury Norov [NVIDIA]" <tip-bot2@...utronix.de>
To: linux-tip-commits@...r.kernel.org
Cc: "Yury Norov [NVIDIA]" <yury.norov@...il.com>,
Thomas Gleixner <tglx@...utronix.de>, x86@...nel.org,
linux-kernel@...r.kernel.org
Subject: [tip: smp/core] smp: Improve locality in smp_call_function_any()
The following commit has been merged into the smp/core branch of tip:
Commit-ID: 5f295519b42f100c735a1e8e1a70060e26f30c3f
Gitweb: https://git.kernel.org/tip/5f295519b42f100c735a1e8e1a70060e26f30c3f
Author: Yury Norov [NVIDIA] <yury.norov@...il.com>
AuthorDate: Sun, 22 Jun 2025 20:00:06 -04:00
Committer: Thomas Gleixner <tglx@...utronix.de>
CommitterDate: Thu, 26 Jun 2025 23:46:34 +02:00
smp: Improve locality in smp_call_function_any()
smp_call_function_any() tries to make a local call as it's the cheapest
option, or switches to a CPU in the same node. If it's not possible, the
algorithm gives up and searches for any CPU, in a numerical order.
Instead, it can search for the best CPU based on NUMA locality, including
the 2nd nearest hop (a set of equidistant nodes), and higher.
sched_numa_find_nth_cpu() does exactly that, and also helps to drop most
of the housekeeping code.
Signed-off-by: Yury Norov [NVIDIA] <yury.norov@...il.com>
Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
Link: https://lore.kernel.org/all/20250623000010.10124-2-yury.norov@gmail.com
---
kernel/smp.c | 19 +++----------------
1 file changed, 3 insertions(+), 16 deletions(-)
diff --git a/kernel/smp.c b/kernel/smp.c
index 974f3a3..7c8cfab 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -741,32 +741,19 @@ EXPORT_SYMBOL_GPL(smp_call_function_single_async);
*
* Selection preference:
* 1) current cpu if in @mask
- * 2) any cpu of current node if in @mask
- * 3) any other online cpu in @mask
+ * 2) nearest cpu in @mask, based on NUMA topology
*/
int smp_call_function_any(const struct cpumask *mask,
smp_call_func_t func, void *info, int wait)
{
unsigned int cpu;
- const struct cpumask *nodemask;
int ret;
/* Try for same CPU (cheapest) */
cpu = get_cpu();
- if (cpumask_test_cpu(cpu, mask))
- goto call;
-
- /* Try for same node. */
- nodemask = cpumask_of_node(cpu_to_node(cpu));
- for (cpu = cpumask_first_and(nodemask, mask); cpu < nr_cpu_ids;
- cpu = cpumask_next_and(cpu, nodemask, mask)) {
- if (cpu_online(cpu))
- goto call;
- }
+ if (!cpumask_test_cpu(cpu, mask))
+ cpu = sched_numa_find_nth_cpu(mask, 0, cpu_to_node(cpu));
- /* Any online will do: smp_call_function_single handles nr_cpu_ids. */
- cpu = cpumask_any_and(mask, cpu_online_mask);
-call:
ret = smp_call_function_single(cpu, func, info, wait);
put_cpu();
return ret;
Powered by blists - more mailing lists