mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Gabriele Monaco <gmonaco@redhat.com>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	Michael Jeanson <mjeanson@efficios.com>,
	Jens Axboe <axboe@kernel.dk>,
	"Paul E. McKenney" <paulmck@kernel.org>,
	"Gautham R. Shenoy" <gautham.shenoy@amd.com>,
	Florian Weimer <fweimer@redhat.com>,
	Tim Chen <tim.c.chen@intel.com>,
	Yury Norov <yury.norov@gmail.com>,
	Shrikanth Hegde <sshegde@linux.ibm.com>
Subject: [patch V3 07/20] cpumask: Introduce cpumask_weighted_or()
Date: Wed, 29 Oct 2025 14:09:07 +0100 (CET)	[thread overview]
Message-ID: <20251029124515.717519165@linutronix.de> (raw)
In-Reply-To: <20251029123717.886619142@linutronix.de>

CID management OR's two cpumasks and then calculates the weight on the
result. That's inefficient as that has to walk the same stuff twice. As
this is done with runqueue lock held, there is a real benefit of speeding
this up. Depending on the system this results in 10-20% less cycles spent
with runqueue lock held for a 4K cpumask.

Provide cpumask_weighted_or() and the corresponding bitmap functions which
return the weight of the OR result right away.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Yury Norov (NVIDIA) <yury.norov@gmail.com>
---
V3: Rename again - Yury
V2: Rename and use the BITMAP_WEIGHT() macro - Yury
---
 include/linux/bitmap.h  |   15 +++++++++++++++
 include/linux/cpumask.h |   16 ++++++++++++++++
 lib/bitmap.c            |    6 ++++++
 3 files changed, 37 insertions(+)

--- a/include/linux/bitmap.h
+++ b/include/linux/bitmap.h
@@ -45,6 +45,7 @@ struct device;
  *  bitmap_copy(dst, src, nbits)                *dst = *src
  *  bitmap_and(dst, src1, src2, nbits)          *dst = *src1 & *src2
  *  bitmap_or(dst, src1, src2, nbits)           *dst = *src1 | *src2
+ *  bitmap_weighted_or(dst, src1, src2, nbits)	*dst = *src1 | *src2. Returns Hamming Weight of dst
  *  bitmap_xor(dst, src1, src2, nbits)          *dst = *src1 ^ *src2
  *  bitmap_andnot(dst, src1, src2, nbits)       *dst = *src1 & ~(*src2)
  *  bitmap_complement(dst, src, nbits)          *dst = ~(*src)
@@ -165,6 +166,8 @@ bool __bitmap_and(unsigned long *dst, co
 		 const unsigned long *bitmap2, unsigned int nbits);
 void __bitmap_or(unsigned long *dst, const unsigned long *bitmap1,
 		 const unsigned long *bitmap2, unsigned int nbits);
+unsigned int __bitmap_weighted_or(unsigned long *dst, const unsigned long *bitmap1,
+				  const unsigned long *bitmap2, unsigned int nbits);
 void __bitmap_xor(unsigned long *dst, const unsigned long *bitmap1,
 		  const unsigned long *bitmap2, unsigned int nbits);
 bool __bitmap_andnot(unsigned long *dst, const unsigned long *bitmap1,
@@ -338,6 +341,18 @@ void bitmap_or(unsigned long *dst, const
 }
 
 static __always_inline
+unsigned int bitmap_weighted_or(unsigned long *dst, const unsigned long *src1,
+				const unsigned long *src2, unsigned int nbits)
+{
+	if (small_const_nbits(nbits)) {
+		*dst = *src1 | *src2;
+		return hweight_long(*dst & BITMAP_LAST_WORD_MASK(nbits));
+	} else {
+		return __bitmap_weighted_or(dst, src1, src2, nbits);
+	}
+}
+
+static __always_inline
 void bitmap_xor(unsigned long *dst, const unsigned long *src1,
 		const unsigned long *src2, unsigned int nbits)
 {
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -729,6 +729,22 @@ void cpumask_or(struct cpumask *dstp, co
 }
 
 /**
+ * cpumask_weighted_or - *dstp = *src1p | *src2p and return the weight of the result
+ * @dstp: the cpumask result
+ * @src1p: the first input
+ * @src2p: the second input
+ *
+ * Return: The number of bits set in the resulting cpumask @dstp
+ */
+static __always_inline
+unsigned int cpumask_weighted_or(struct cpumask *dstp, const struct cpumask *src1p,
+				 const struct cpumask *src2p)
+{
+	return bitmap_weighted_or(cpumask_bits(dstp), cpumask_bits(src1p),
+				  cpumask_bits(src2p), small_cpumask_bits);
+}
+
+/**
  * cpumask_xor - *dstp = *src1p ^ *src2p
  * @dstp: the cpumask result
  * @src1p: the first input
--- a/lib/bitmap.c
+++ b/lib/bitmap.c
@@ -355,6 +355,12 @@ unsigned int __bitmap_weight_andnot(cons
 }
 EXPORT_SYMBOL(__bitmap_weight_andnot);
 
+unsigned int __bitmap_weighted_or(unsigned long *dst, const unsigned long *bitmap1,
+				  const unsigned long *bitmap2, unsigned int bits)
+{
+	return BITMAP_WEIGHT(({dst[idx] = bitmap1[idx] | bitmap2[idx]; dst[idx]; }), bits);
+}
+
 void __bitmap_set(unsigned long *map, unsigned int start, int len)
 {
 	unsigned long *p = map + BIT_WORD(start);


  parent reply	other threads:[~2025-10-29 13:09 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-29 13:08 [patch V3 00/20] sched: Rewrite MM CID management Thomas Gleixner
2025-10-29 13:08 ` [patch V3 01/20] sched/mmcid: Revert the complex " Thomas Gleixner
2025-10-29 13:08 ` [patch V3 02/20] sched/mmcid: Use proper data structures Thomas Gleixner
2025-10-29 15:31   ` Mathieu Desnoyers
2025-10-29 13:08 ` [patch V3 03/20] sched/mmcid: Cacheline align MM CID storage Thomas Gleixner
2025-10-29 15:39   ` Mathieu Desnoyers
2025-10-29 21:09     ` Thomas Gleixner
2025-10-30 14:15       ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 04/20] sched: Fixup whitespace damage Thomas Gleixner
2025-10-29 15:42   ` Mathieu Desnoyers
2025-10-29 21:11     ` Thomas Gleixner
2025-10-29 13:09 ` [patch V3 05/20] sched/mmcid: Move scheduler code out of global header Thomas Gleixner
2025-10-29 15:43   ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 06/20] sched/mmcid: Prevent pointless work in mm_update_cpus_allowed() Thomas Gleixner
2025-10-29 15:45   ` Mathieu Desnoyers
2025-10-29 13:09 ` Thomas Gleixner [this message]
2025-10-29 15:49   ` [patch V3 07/20] cpumask: Introduce cpumask_weighted_or() Mathieu Desnoyers
2025-11-03  9:15   ` Shrikanth Hegde
2025-11-03 13:29     ` Thomas Gleixner
2025-11-10 16:11       ` Peter Zijlstra
2025-11-10 16:42         ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 08/20] sched/mmcid: Use cpumask_weighted_or() Thomas Gleixner
2025-10-29 15:51   ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 09/20] cpumask: Cache num_possible_cpus() Thomas Gleixner
2025-10-29 15:54   ` Mathieu Desnoyers
2025-10-29 21:11     ` Thomas Gleixner
2025-11-01 22:59       ` Thomas Gleixner
2025-11-03 10:06   ` Shrikanth Hegde
2025-11-03 13:28     ` Thomas Gleixner
2025-10-29 13:09 ` [patch V3 10/20] sched/mmcid: Convert mm CID mask to a bitmap Thomas Gleixner
2025-10-30 13:59   ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 11/20] signal: Move MMCID exit out of sighand lock Thomas Gleixner
2025-10-29 13:09 ` [patch V3 12/20] sched/mmcid: Move initialization out of line Thomas Gleixner
2025-10-30 14:02   ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 13/20] sched/mmcid: Provide precomputed maximal value Thomas Gleixner
2025-10-30 14:23   ` Mathieu Desnoyers
2025-10-31 15:06     ` Thomas Gleixner
2025-10-31 15:14       ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 14/20] sched/mmcid: Serialize sched_mm_cid_fork()/exit() with a mutex Thomas Gleixner
2025-10-30 14:25   ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 15/20] sched/mmcid: Introduce per task/CPU ownership infrastrcuture Thomas Gleixner
2025-10-30 14:45   ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 16/20] sched/mmcid: Provide new scheduler CID mechanism Thomas Gleixner
2025-10-30 15:09   ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 17/20] sched/mmcid: Provide CID ownership mode fixup functions Thomas Gleixner
2025-10-30 15:51   ` Mathieu Desnoyers
2025-10-31 16:54     ` Thomas Gleixner
2025-10-31 19:15       ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 18/20] irqwork: Move data struct to a types header Thomas Gleixner
2025-10-30 15:52   ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 19/20] sched/mmcid: Implement deferred mode change Thomas Gleixner
2025-10-30 15:56   ` Mathieu Desnoyers
2025-10-29 13:09 ` [patch V3 20/20] sched/mmcid: Switch over to the new mechanism Thomas Gleixner
2025-10-30 16:07   ` Mathieu Desnoyers
2025-10-31 16:57     ` Thomas Gleixner
2025-10-31 19:17       ` Mathieu Desnoyers
2025-11-24 12:10   ` Mark Brown
2025-11-24 12:27     ` Florian Weimer
2025-11-24 13:28       ` Mark Brown
2025-10-30  5:00 ` [patch V3 00/20] sched: Rewrite MM CID management Shrikanth Hegde
2025-10-30  6:40   ` Shrikanth Hegde
2025-10-31 19:36     ` Thomas Gleixner
2025-11-01  7:56       ` Shrikanth Hegde
2025-11-01 12:27         ` Thomas Gleixner
2025-11-10 17:09 ` Gabriele Monaco

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=20251029124515.717519165@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=axboe@kernel.dk \
    --cc=fweimer@redhat.com \
    --cc=gautham.shenoy@amd.com \
    --cc=gmonaco@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mjeanson@efficios.com \
    --cc=paulmck@kernel.org \
    --cc=peterz@infradead.org \
    --cc=sshegde@linux.ibm.com \
    --cc=tim.c.chen@intel.com \
    --cc=yury.norov@gmail.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

Powered by JetHome