From: Will Deacon <will@kernel.org>
To: linux-kernel@vger.kernel.org
Cc: Nick Desaulniers <ndesaulniers@google.com>,
kernel-team@android.com, Jan Kara <jack@suse.cz>,
Minchan Kim <minchan@kernel.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Hugh Dickins <hughd@google.com>,
linux-mm@kvack.org, Vinayak Menon <vinmenon@codeaurora.org>,
"Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
Andrew Morton <akpm@linux-foundation.org>,
Will Deacon <will@kernel.org>,
linux-arm-kernel@lists.infradead.org
Subject: [PATCH v4 3/8] arm64: mm: Implement arch_wants_old_prefaulted_pte()
Date: Wed, 20 Jan 2021 17:36:07 +0000 [thread overview]
Message-ID: <20210120173612.20913-4-will@kernel.org> (raw)
In-Reply-To: <20210120173612.20913-1-will@kernel.org>
On CPUs with hardware AF/DBM, initialising prefaulted PTEs as 'old'
improves vmscan behaviour and does not appear to introduce any overhead
elsewhere.
Implement arch_wants_old_prefaulted_pte() to return 'true' if we detect
hardware access flag support at runtime. This can be extended in future
based on MIDR matching if necessary.
Cc: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
---
arch/arm64/include/asm/pgtable.h | 12 +++++++++++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
index 501562793ce2..e17b96d0e4b5 100644
--- a/arch/arm64/include/asm/pgtable.h
+++ b/arch/arm64/include/asm/pgtable.h
@@ -980,7 +980,17 @@ static inline bool arch_faults_on_old_pte(void)
return !cpu_has_hw_af();
}
-#define arch_faults_on_old_pte arch_faults_on_old_pte
+#define arch_faults_on_old_pte arch_faults_on_old_pte
+
+/*
+ * Experimentally, it's cheap to set the access flag in hardware and we
+ * benefit from prefaulting mappings as 'old' to start with.
+ */
+static inline bool arch_wants_old_prefaulted_pte(void)
+{
+ return !arch_faults_on_old_pte();
+}
+#define arch_wants_old_prefaulted_pte arch_wants_old_prefaulted_pte
#endif /* !__ASSEMBLY__ */
--
2.30.0.284.gd98b1dd5eaa7-goog
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
WARNING: multiple messages have this Message-ID (diff)
From: Will Deacon <will@kernel.org>
To: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org,
Will Deacon <will@kernel.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Jan Kara <jack@suse.cz>, Minchan Kim <minchan@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
"Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Vinayak Menon <vinmenon@codeaurora.org>,
Hugh Dickins <hughd@google.com>,
Nick Desaulniers <ndesaulniers@google.com>,
kernel-team@android.com
Subject: [PATCH v4 3/8] arm64: mm: Implement arch_wants_old_prefaulted_pte()
Date: Wed, 20 Jan 2021 17:36:07 +0000 [thread overview]
Message-ID: <20210120173612.20913-4-will@kernel.org> (raw)
In-Reply-To: <20210120173612.20913-1-will@kernel.org>
On CPUs with hardware AF/DBM, initialising prefaulted PTEs as 'old'
improves vmscan behaviour and does not appear to introduce any overhead
elsewhere.
Implement arch_wants_old_prefaulted_pte() to return 'true' if we detect
hardware access flag support at runtime. This can be extended in future
based on MIDR matching if necessary.
Cc: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
---
arch/arm64/include/asm/pgtable.h | 12 +++++++++++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
index 501562793ce2..e17b96d0e4b5 100644
--- a/arch/arm64/include/asm/pgtable.h
+++ b/arch/arm64/include/asm/pgtable.h
@@ -980,7 +980,17 @@ static inline bool arch_faults_on_old_pte(void)
return !cpu_has_hw_af();
}
-#define arch_faults_on_old_pte arch_faults_on_old_pte
+#define arch_faults_on_old_pte arch_faults_on_old_pte
+
+/*
+ * Experimentally, it's cheap to set the access flag in hardware and we
+ * benefit from prefaulting mappings as 'old' to start with.
+ */
+static inline bool arch_wants_old_prefaulted_pte(void)
+{
+ return !arch_faults_on_old_pte();
+}
+#define arch_wants_old_prefaulted_pte arch_wants_old_prefaulted_pte
#endif /* !__ASSEMBLY__ */
--
2.30.0.284.gd98b1dd5eaa7-goog
next prev parent reply other threads:[~2021-01-20 17:39 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-20 17:36 [PATCH v4 0/8] Create 'old' ptes for faultaround mappings on arm64 with hardware access flag Will Deacon
2021-01-20 17:36 ` Will Deacon
2021-01-20 17:36 ` [PATCH v4 1/8] mm: Cleanup faultaround and finish_fault() codepaths Will Deacon
2021-01-20 17:36 ` Will Deacon
2021-01-20 17:36 ` [PATCH v4 2/8] mm: Allow architectures to request 'old' entries when prefaulting Will Deacon
2021-01-20 17:36 ` Will Deacon
2021-01-20 17:36 ` Will Deacon [this message]
2021-01-20 17:36 ` [PATCH v4 3/8] arm64: mm: Implement arch_wants_old_prefaulted_pte() Will Deacon
2021-01-20 17:36 ` [PATCH v4 4/8] mm: Move immutable fields of 'struct vm_fault' into anonymous struct Will Deacon
2021-01-20 17:36 ` Will Deacon
2021-01-20 18:13 ` Nick Desaulniers
2021-01-20 18:13 ` Nick Desaulniers
2021-01-21 12:48 ` Will Deacon
2021-01-21 12:48 ` Will Deacon
2021-01-20 17:36 ` [PATCH v4 5/8] mm: Pass 'address' to map to do_set_pte() and drop FAULT_FLAG_PREFAULT Will Deacon
2021-01-20 17:36 ` Will Deacon
2021-01-20 17:36 ` [PATCH v4 6/8] mm: Avoid modifying vmf.address in __collapse_huge_page_swapin() Will Deacon
2021-01-20 17:36 ` Will Deacon
2021-01-20 17:36 ` [PATCH v4 7/8] mm: Use static initialisers for immutable fields of 'struct vm_fault' Will Deacon
2021-01-20 17:36 ` Will Deacon
2021-01-20 18:21 ` Nick Desaulniers
2021-01-20 18:21 ` Nick Desaulniers
2021-01-21 12:50 ` Will Deacon
2021-01-21 12:50 ` Will Deacon
2021-01-20 17:36 ` [PATCH v4 8/8] mm: Mark anonymous struct field of 'struct vm_fault' as 'const' Will Deacon
2021-01-20 17:36 ` Will Deacon
2021-01-20 18:27 ` Nick Desaulniers
2021-01-20 18:27 ` Nick Desaulniers
2021-01-20 19:02 ` Linus Torvalds
2021-01-20 19:02 ` Linus Torvalds
2021-01-21 13:11 ` Will Deacon
2021-01-21 13:11 ` Will Deacon
2021-01-21 19:24 ` Nick Desaulniers
2021-01-21 19:24 ` Nick Desaulniers
2021-01-21 21:28 ` Will Deacon
2021-01-21 21:28 ` Will Deacon
2021-01-22 19:10 ` Nick Desaulniers
2021-01-22 19:10 ` Nick Desaulniers
2021-01-22 19:27 ` Will Deacon
2021-01-22 19:27 ` Will Deacon
2021-01-22 17:47 ` Linus Torvalds
2021-01-22 17:47 ` Linus Torvalds
2021-01-26 23:08 ` [PATCH v4 0/8] Create 'old' ptes for faultaround mappings on arm64 with hardware access flag Will Deacon
2021-01-26 23:08 ` Will Deacon
2021-01-26 23:28 ` Hugh Dickins
2021-01-26 23:28 ` Hugh Dickins
2021-01-27 17:16 ` Will Deacon
2021-01-27 17:16 ` Will Deacon
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=20210120173612.20913-4-will@kernel.org \
--to=will@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=catalin.marinas@arm.com \
--cc=hughd@google.com \
--cc=jack@suse.cz \
--cc=kernel-team@android.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=minchan@kernel.org \
--cc=ndesaulniers@google.com \
--cc=torvalds@linux-foundation.org \
--cc=vinmenon@codeaurora.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.