linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Luiz Capitulino <luizcap@redhat.com>
To: linux-kernel@vger.kernel.org, linux-mm@kvack.org, david@kernel.org
Cc: ryan.roberts@arm.com, akpm@linux-foundation.org,
	lorenzo.stoakes@oracle.com
Subject: [PATCH 02/11] mm: introduce pgtable_has_pmd_leaves()
Date: Mon, 15 Dec 2025 16:16:44 -0500	[thread overview]
Message-ID: <49e5272a3fb756d1ffee0e4308bc9fa42fcdb81e.1765833318.git.luizcap@redhat.com> (raw)
In-Reply-To: <cover.1765833318.git.luizcap@redhat.com>

Currently, we have two helpers that check for PMD-sized pages but have
different names and slightly different semantics:

- has_transparent_hugepages(): the name suggests it checks if THP is
  enabled, but it actually checks if the CPU supports PMD-sized pages.
  It may perform a hardware check, so it can't be used in fast paths

- thp_disabled_by_hw(): the name suggests it checks if THP is disabled
  by the hardware, but it just returns a cached value acquired with
  has_transparent_hugepages(). This way, this helper can be called from
  fast paths

This commit introduces a new helper called pgtable_has_pmd_leaves()
which is intended to replace both has_transparent_hugepages() and
thp_disable_by_hw(). pgtable_has_pmd_leaves() has very clear semantics:
it returns true if the CPU supports PMD-sized pages and false otherwise.
It always returns a cached value, so it can be used in fast paths.

The new helper requires an initialization step that needs to be
performed very early during boot because there are
pgtable_has_pmd_leaves() users that want to call it from __setup()
handlers. This initialization is performed by init_arch_has_pmd_leaves(),
called early from start_kernel().

The next commits will convert users of both has_transparent_hugepages()
and thp_disabled_by_hw() to pgtable_has_pmd_leaves().

Signed-off-by: Luiz Capitulino <luizcap@redhat.com>
---
 include/linux/pgtable.h | 7 +++++++
 init/main.c             | 1 +
 mm/memory.c             | 8 ++++++++
 3 files changed, 16 insertions(+)

diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
index 652f287c1ef6..6733f90a1da4 100644
--- a/include/linux/pgtable.h
+++ b/include/linux/pgtable.h
@@ -2017,6 +2017,13 @@ static inline const char *pgtable_level_to_str(enum pgtable_level level)
 	}
 }
 
+extern bool __arch_has_pmd_leaves;
+static inline bool pgtable_has_pmd_leaves(void)
+{
+	return __arch_has_pmd_leaves;
+}
+void __init init_arch_has_pmd_leaves(void);
+
 #endif /* !__ASSEMBLY__ */
 
 #if !defined(MAX_POSSIBLE_PHYSMEM_BITS) && !defined(CONFIG_64BIT)
diff --git a/init/main.c b/init/main.c
index b84818ad9685..ad1209fffcde 100644
--- a/init/main.c
+++ b/init/main.c
@@ -1036,6 +1036,7 @@ void start_kernel(void)
 	smp_prepare_boot_cpu();	/* arch-specific boot-cpu hooks */
 	early_numa_node_init();
 	boot_cpu_hotplug_init();
+	init_arch_has_pmd_leaves();
 
 	print_kernel_cmdline(saved_command_line);
 	/* parameters may set static keys */
diff --git a/mm/memory.c b/mm/memory.c
index 2a55edc48a65..79bd59d5243f 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -177,6 +177,14 @@ static int __init init_zero_pfn(void)
 }
 early_initcall(init_zero_pfn);
 
+bool __arch_has_pmd_leaves __read_mostly;
+EXPORT_SYMBOL(__arch_has_pmd_leaves);
+
+void __init init_arch_has_pmd_leaves(void)
+{
+	__arch_has_pmd_leaves = has_transparent_hugepage();
+}
+
 void mm_trace_rss_stat(struct mm_struct *mm, int member)
 {
 	trace_rss_stat(mm, member);
-- 
2.52.0



  parent reply	other threads:[~2025-12-15 21:17 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-12-15 21:16 [PATCH 00/11] mm: thp: always enable mTHP support Luiz Capitulino
2025-12-15 21:16 ` [PATCH 01/11] docs: tmpfs: remove implementation detail reference Luiz Capitulino
2025-12-16  7:40   ` Baolin Wang
2025-12-15 21:16 ` Luiz Capitulino [this message]
2025-12-15 21:16 ` [PATCH 03/11] drivers: dax: use pgtable_has_pmd_leaves() Luiz Capitulino
2025-12-15 21:16 ` [PATCH 04/11] drivers: i915 selftest: " Luiz Capitulino
2025-12-15 21:16 ` [PATCH 05/11] drivers: nvdimm: " Luiz Capitulino
2025-12-15 21:16 ` [PATCH 06/11] mm: debug_vm_pgtable: " Luiz Capitulino
2025-12-15 21:16 ` [PATCH 07/11] mm: shmem: " Luiz Capitulino
2025-12-16  7:52   ` Baolin Wang
2025-12-16 13:47     ` Luiz Capitulino
2025-12-17  2:03       ` Baolin Wang
2025-12-17 18:50         ` Luiz Capitulino
2025-12-15 21:16 ` [PATCH 08/11] treewide: rename has_transparent_hugepage() to arch_has_pmd_leaves() Luiz Capitulino
2025-12-15 21:16 ` [PATCH 09/11] mm: replace thp_disabled_by_hw() with pgtable_has_pmd_leaves() Luiz Capitulino
2025-12-15 21:16 ` [PATCH 10/11] mm/thp: always enable mTHP support Luiz Capitulino
2025-12-15 21:16 ` [PATCH 11/11] mm/thp: x86: cleanup PSE feature bit usage Luiz Capitulino

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=49e5272a3fb756d1ffee0e4308bc9fa42fcdb81e.1765833318.git.luizcap@redhat.com \
    --to=luizcap@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=david@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=lorenzo.stoakes@oracle.com \
    --cc=ryan.roberts@arm.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