From: Nadav Amit <nadav.amit@gmail.com>
To: linux-mm@kvack.org
Cc: linux-kernel@vger.kernel.org, Nadav Amit <namit@vmware.com>,
Andrea Arcangeli <aarcange@redhat.com>,
Andrew Cooper <andrew.cooper3@citrix.com>,
Andrew Morton <akpm@linux-foundation.org>,
Andy Lutomirski <luto@kernel.org>,
Dave Hansen <dave.hansen@linux.intel.com>,
Peter Xu <peterx@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Thomas Gleixner <tglx@linutronix.de>,
Will Deacon <will@kernel.org>, Yu Zhao <yuzhao@google.com>,
Nick Piggin <npiggin@gmail.com>,
x86@kernel.org
Subject: [PATCH v2 2/5] mm: avoid unnecessary flush on change_huge_pmd()
Date: Thu, 21 Oct 2021 05:21:09 -0700 [thread overview]
Message-ID: <20211021122112.592634-3-namit@vmware.com> (raw)
In-Reply-To: <20211021122112.592634-1-namit@vmware.com>
From: Nadav Amit <namit@vmware.com>
Calls to change_protection_range() on THP can trigger, at least on x86,
two TLB flushes for one page: one immediately, when pmdp_invalidate() is
called by change_huge_pmd(), and then another one later (that can be
batched) when change_protection_range() finishes.
The first TLB flush is only necessary to prevent the dirty bit (and with
a lesser importance the access bit) from changing while the PTE is
modified. However, this is not necessary as the x86 CPUs set the
dirty-bit atomically with an additional check that the PTE is (still)
present. One caveat is Intel's Knights Landing that has a bug and does
not do so.
Leverage this behavior to eliminate the unnecessary TLB flush in
change_huge_pmd(). Introduce a new arch specific pmdp_invalidate_ad()
that only invalidates the access and dirty bit from further changes.
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will@kernel.org>
Cc: Yu Zhao <yuzhao@google.com>
Cc: Nick Piggin <npiggin@gmail.com>
Cc: x86@kernel.org
Signed-off-by: Nadav Amit <namit@vmware.com>
---
arch/x86/include/asm/pgtable.h | 8 ++++++++
include/linux/pgtable.h | 5 +++++
mm/huge_memory.c | 7 ++++---
mm/pgtable-generic.c | 8 ++++++++
4 files changed, 25 insertions(+), 3 deletions(-)
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index 448cd01eb3ec..18c3366f8f4d 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -1146,6 +1146,14 @@ static inline pmd_t pmdp_establish(struct vm_area_struct *vma,
}
}
#endif
+
+#define __HAVE_ARCH_PMDP_INVALIDATE_AD
+static inline pmd_t pmdp_invalidate_ad(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp)
+{
+ return pmdp_establish(vma, address, pmdp, pmd_mkinvalid(*pmdp));
+}
+
/*
* Page table pages are page-aligned. The lower half of the top
* level is used for userspace and the top half for the kernel.
diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
index e24d2c992b11..622efe0a9ef0 100644
--- a/include/linux/pgtable.h
+++ b/include/linux/pgtable.h
@@ -561,6 +561,11 @@ extern pmd_t pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
pmd_t *pmdp);
#endif
+#ifndef __HAVE_ARCH_PMDP_INVALIDATE_AD
+extern pmd_t pmdp_invalidate_ad(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp);
+#endif
+
#ifndef __HAVE_ARCH_PTE_SAME
static inline int pte_same(pte_t pte_a, pte_t pte_b)
{
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index e5ea5f775d5c..435da011b1a2 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1795,10 +1795,11 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
* The race makes MADV_DONTNEED miss the huge pmd and don't clear it
* which may break userspace.
*
- * pmdp_invalidate() is required to make sure we don't miss
- * dirty/young flags set by hardware.
+ * pmdp_invalidate_ad() is required to make sure we don't miss
+ * dirty/young flags (which are also known as access/dirty) cannot be
+ * further modifeid by the hardware.
*/
- entry = pmdp_invalidate(vma, addr, pmd);
+ entry = pmdp_invalidate_ad(vma, addr, pmd);
entry = pmd_modify(entry, newprot);
if (preserve_write)
diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
index 4e640baf9794..b0ce6c7391bf 100644
--- a/mm/pgtable-generic.c
+++ b/mm/pgtable-generic.c
@@ -200,6 +200,14 @@ pmd_t pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
}
#endif
+#ifndef __HAVE_ARCH_PMDP_INVALIDATE_AD
+pmd_t pmdp_invalidate_ad(struct vm_area_struct *vma, unsigned long address,
+ pmd_t *pmdp)
+{
+ return pmdp_invalidate(vma, address, pmdp);
+}
+#endif
+
#ifndef pmdp_collapse_flush
pmd_t pmdp_collapse_flush(struct vm_area_struct *vma, unsigned long address,
pmd_t *pmdp)
--
2.25.1
next prev parent reply other threads:[~2021-10-21 19:52 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-21 12:21 [PATCH v2 0/5] mm/mprotect: avoid unnecessary TLB flushes Nadav Amit
2021-10-21 12:21 ` [PATCH v2 1/5] x86: Detection of Knights Landing A/D leak Nadav Amit
2021-10-26 15:54 ` Dave Hansen
2021-10-26 15:57 ` Nadav Amit
2021-10-21 12:21 ` Nadav Amit [this message]
2021-10-25 10:52 ` [PATCH v2 2/5] mm: avoid unnecessary flush on change_huge_pmd() Peter Zijlstra
2021-10-25 16:29 ` Nadav Amit
2021-10-26 16:06 ` Dave Hansen
2021-10-26 16:47 ` Nadav Amit
2021-10-26 16:53 ` Nadav Amit
2021-10-26 17:44 ` Nadav Amit
2021-10-26 18:44 ` Dave Hansen
2021-10-26 19:06 ` Nadav Amit
2021-10-26 19:40 ` Dave Hansen
2021-10-26 20:07 ` Nadav Amit
2021-10-26 20:47 ` Dave Hansen
2021-10-21 12:21 ` [PATCH v2 3/5] x86/mm: check exec permissions on fault Nadav Amit
2021-10-25 10:59 ` Peter Zijlstra
2021-10-25 11:13 ` Andrew Cooper
2021-10-25 14:23 ` Dave Hansen
2021-10-25 14:20 ` Dave Hansen
2021-10-25 16:19 ` Nadav Amit
2021-10-25 17:45 ` Dave Hansen
2021-10-25 17:51 ` Nadav Amit
2021-10-25 18:00 ` Dave Hansen
2021-10-21 12:21 ` [PATCH v2 4/5] mm/mprotect: use mmu_gather Nadav Amit
2021-10-21 12:21 ` [PATCH v2 5/5] mm/mprotect: do not flush on permission promotion Nadav Amit
2021-10-25 11:12 ` Peter Zijlstra
2021-10-25 16:27 ` Nadav Amit
2021-10-22 3:04 ` [PATCH v2 0/5] mm/mprotect: avoid unnecessary TLB flushes Andrew Morton
2021-10-22 21:58 ` Nadav Amit
2021-10-26 16:09 ` Dave Hansen
2021-10-25 10:50 ` Peter Zijlstra
2021-10-25 16:42 ` Nadav Amit
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=20211021122112.592634-3-namit@vmware.com \
--to=nadav.amit@gmail.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=andrew.cooper3@citrix.com \
--cc=dave.hansen@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@kernel.org \
--cc=namit@vmware.com \
--cc=npiggin@gmail.com \
--cc=peterx@redhat.com \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=will@kernel.org \
--cc=x86@kernel.org \
--cc=yuzhao@google.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