linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Kiryl Shutsemau <kas@kernel.org>
To: Andrew Morton <akpm@linux-foundation.org>,
	Muchun Song <muchun.song@linux.dev>,
	David Hildenbrand <david@kernel.org>,
	Matthew Wilcox <willy@infradead.org>,
	Usama Arif <usamaarif642@gmail.com>,
	Frank van der Linden <fvdl@google.com>
Cc: Oscar Salvador <osalvador@suse.de>,
	Mike Rapoport <rppt@kernel.org>, Vlastimil Babka <vbabka@suse.cz>,
	Lorenzo Stoakes <lorenzo.stoakes@oracle.com>,
	Zi Yan <ziy@nvidia.com>, Baoquan He <bhe@redhat.com>,
	Michal Hocko <mhocko@suse.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Jonathan Corbet <corbet@lwn.net>,
	kernel-team@meta.com, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org,
	Kiryl Shutsemau <kas@kernel.org>
Subject: [PATCHv4 13/14] mm: Remove the branch from compound_head()
Date: Wed, 21 Jan 2026 16:22:50 +0000	[thread overview]
Message-ID: <20260121162253.2216580-14-kas@kernel.org> (raw)
In-Reply-To: <20260121162253.2216580-1-kas@kernel.org>

The compound_head() function is a hot path. For example, the zap path
calls it for every leaf page table entry.

Rewrite the helper function in a branchless manner to eliminate the risk
of CPU branch misprediction.

Signed-off-by: Kiryl Shutsemau <kas@kernel.org>
Reviewed-by: Muchun Song <muchun.song@linux.dev>
---
 include/linux/page-flags.h | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index f89702e101e8..95ac963b78af 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -224,25 +224,32 @@ static __always_inline bool compound_info_has_mask(void)
 static __always_inline unsigned long _compound_head(const struct page *page)
 {
 	unsigned long info = READ_ONCE(page->compound_info);
+	unsigned long mask;
+
+	if (!compound_info_has_mask()) {
+		/* Bit 0 encodes PageTail() */
+		if (info & 1)
+			return info - 1;
 
-	/* Bit 0 encodes PageTail() */
-	if (!(info & 1))
 		return (unsigned long)page;
-
-	/*
-	 * If compound_info_has_mask() is false, the rest of compound_info is
-	 * the pointer to the head page.
-	 */
-	if (!compound_info_has_mask())
-		return info - 1;
+	}
 
 	/*
 	 * If compoun_info_has_mask() is true the rest of the info encodes
 	 * the mask that converts the address of the tail page to the head page.
 	 *
 	 * No need to clear bit 0 in the mask as 'page' always has it clear.
+	 *
+	 * Let's do it in a branchless manner.
 	 */
-	return (unsigned long)page & info;
+
+	/* Non-tail: -1UL, Tail: 0 */
+	mask = (info & 1) - 1;
+
+	/* Non-tail: -1UL, Tail: info */
+	mask |= info;
+
+	return (unsigned long)page & mask;
 }
 
 #define compound_head(page)	((typeof(page))_compound_head(page))
-- 
2.51.2



  parent reply	other threads:[~2026-01-21 16:23 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-21 16:22 [PATCHv4 00/14] mm: Eliminate fake head pages from vmemmap optimization Kiryl Shutsemau
2026-01-21 16:22 ` [PATCHv4 01/14] mm: Move MAX_FOLIO_ORDER definition to mmzone.h Kiryl Shutsemau
2026-01-21 16:29   ` Zi Yan
2026-01-22  2:24   ` Muchun Song
2026-01-21 16:22 ` [PATCHv4 02/14] mm: Change the interface of prep_compound_tail() Kiryl Shutsemau
2026-01-21 16:32   ` Zi Yan
2026-01-21 16:22 ` [PATCHv4 03/14] mm: Rename the 'compound_head' field in the 'struct page' to 'compound_info' Kiryl Shutsemau
2026-01-21 16:34   ` Zi Yan
2026-01-21 16:22 ` [PATCHv4 04/14] mm: Move set/clear_compound_head() next to compound_head() Kiryl Shutsemau
2026-01-21 16:35   ` Zi Yan
2026-01-21 16:22 ` [PATCHv4 05/14] mm: Rework compound_head() for power-of-2 sizeof(struct page) Kiryl Shutsemau
2026-01-21 17:12   ` Zi Yan
2026-01-22 11:29     ` Kiryl Shutsemau
2026-01-22 11:52       ` Muchun Song
2026-01-21 16:22 ` [PATCHv4 06/14] mm: Make page_zonenum() use head page Kiryl Shutsemau
2026-01-21 16:28   ` Zi Yan
2026-01-21 16:22 ` [PATCHv4 07/14] mm/sparse: Check memmap alignment for compound_info_has_mask() Kiryl Shutsemau
2026-01-21 17:58   ` Zi Yan
2026-01-22 11:22     ` Kiryl Shutsemau
2026-01-22  3:10   ` Muchun Song
2026-01-22 11:28     ` Kiryl Shutsemau
2026-01-22 11:33       ` Muchun Song
2026-01-22 11:42         ` Muchun Song
2026-01-22 12:42           ` Kiryl Shutsemau
2026-01-22 14:02             ` Muchun Song
2026-01-22 17:59               ` Kiryl Shutsemau
2026-01-23  2:32                 ` Muchun Song
2026-01-23 12:07                   ` Kiryl Shutsemau
2026-01-21 16:22 ` [PATCHv4 08/14] mm/hugetlb: Refactor code around vmemmap_walk Kiryl Shutsemau
2026-01-22  8:08   ` Muchun Song
2026-01-21 16:22 ` [PATCHv4 09/14] mm/hugetlb: Remove fake head pages Kiryl Shutsemau
2026-01-22  7:00   ` Muchun Song
2026-01-27 14:51     ` Kiryl Shutsemau
2026-01-28  2:43       ` Muchun Song
2026-01-28 12:59         ` Kiryl Shutsemau
2026-01-29  3:04           ` Muchun Song
2026-01-21 16:22 ` [PATCHv4 10/14] mm: Drop fake head checks Kiryl Shutsemau
2026-01-21 18:16   ` Zi Yan
2026-01-22 12:48     ` Kiryl Shutsemau
2026-01-21 16:22 ` [PATCHv4 11/14] hugetlb: Remove VMEMMAP_SYNCHRONIZE_RCU Kiryl Shutsemau
2026-01-21 16:22 ` [PATCHv4 12/14] mm/hugetlb: Remove hugetlb_optimize_vmemmap_key static key Kiryl Shutsemau
2026-01-21 16:22 ` Kiryl Shutsemau [this message]
2026-01-21 18:21   ` [PATCHv4 13/14] mm: Remove the branch from compound_head() Zi Yan
2026-01-21 16:22 ` [PATCHv4 14/14] hugetlb: Update vmemmap_dedup.rst Kiryl Shutsemau
2026-01-22  2:22   ` Muchun Song
2026-01-21 18:44 ` [PATCHv4 00/14] mm: Eliminate fake head pages from vmemmap optimization Vlastimil Babka
2026-01-21 20:31   ` Zi Yan
2026-01-22 11:21     ` Kiryl Shutsemau

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=20260121162253.2216580-14-kas@kernel.org \
    --to=kas@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=bhe@redhat.com \
    --cc=corbet@lwn.net \
    --cc=david@kernel.org \
    --cc=fvdl@google.com \
    --cc=hannes@cmpxchg.org \
    --cc=kernel-team@meta.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=lorenzo.stoakes@oracle.com \
    --cc=mhocko@suse.com \
    --cc=muchun.song@linux.dev \
    --cc=osalvador@suse.de \
    --cc=rppt@kernel.org \
    --cc=usamaarif642@gmail.com \
    --cc=vbabka@suse.cz \
    --cc=willy@infradead.org \
    --cc=ziy@nvidia.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