From: "Joel Fernandes (Google)" <joel@joelfernandes.org>
To: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org, Rich Felker <dalias@libc.org>,
linux-ia64@vger.kernel.org, linux-sh@vger.kernel.org,
Peter Zijlstra <peterz@infradead.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Dave Hansen <dave.hansen@linux.intel.com>,
Will Deacon <will.deacon@arm.com>,
mhocko@kernel.org, linux-mm@kvack.org, lokeshgidra@google.com,
"Joel Fernandes \(Google\)" <joel@joelfernandes.org>,
linux-riscv@lists.infradead.org, elfring@users.sourceforge.net,
Jonas Bonn <jonas@southpole.se>,
kvmarm@lists.cs.columbia.edu, dancol@google.com,
Yoshinori Sato <ysato@users.sourceforge.jp>,
sparclinux@vger.kernel.org, linux-xtensa@linux-xtensa.org,
linux-hexagon@vger.kernel.org, Helge Deller <deller@gmx.de>,
"maintainer:X86 ARCHITECTURE 32-BIT AND 64-BIT" <x86@kernel.org>,
hughd@google.com, "James E.J. Bottomley" <jejb@parisc-linux.org>,
kasan-dev@googlegroups.com, anton.ivanov@kot-begem
Subject: [PATCH 2/4] mm: speed up mremap by 500x on large regions (v2)
Date: Fri, 12 Oct 2018 18:31:58 -0700 [thread overview]
Message-ID: <20181013013200.206928-3-joel@joelfernandes.org> (raw)
In-Reply-To: <20181013013200.206928-1-joel@joelfernandes.org>
Android needs to mremap large regions of memory during memory management
related operations. The mremap system call can be really slow if THP is
not enabled. The bottleneck is move_page_tables, which is copying each
pte at a time, and can be really slow across a large map. Turning on THP
may not be a viable option, and is not for us. This patch speeds up the
performance for non-THP system by copying at the PMD level when possible.
The speed up is three orders of magnitude. On a 1GB mremap, the mremap
completion times drops from 160-250 millesconds to 380-400 microseconds.
Before:
Total mremap time for 1GB data: 242321014 nanoseconds.
Total mremap time for 1GB data: 196842467 nanoseconds.
Total mremap time for 1GB data: 167051162 nanoseconds.
After:
Total mremap time for 1GB data: 385781 nanoseconds.
Total mremap time for 1GB data: 388959 nanoseconds.
Total mremap time for 1GB data: 402813 nanoseconds.
Incase THP is enabled, the optimization is skipped. I also flush the
tlb every time we do this optimization since I couldn't find a way to
determine if the low-level PTEs are dirty. It is seen that the cost of
doing so is not much compared the improvement, on both x86-64 and arm64.
Cc: minchan@kernel.org
Cc: pantin@google.com
Cc: hughd@google.com
Cc: lokeshgidra@google.com
Cc: dancol@google.com
Cc: mhocko@kernel.org
Cc: kirill@shutemov.name
Cc: akpm@linux-foundation.org
Cc: kernel-team@android.com
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
---
arch/Kconfig | 5 ++++
mm/mremap.c | 65 ++++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 70 insertions(+)
diff --git a/arch/Kconfig b/arch/Kconfig
index 6801123932a5..9724fe39884f 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -518,6 +518,11 @@ config HAVE_IRQ_TIME_ACCOUNTING
Archs need to ensure they use a high enough resolution clock to
support irq time accounting and then call enable_sched_clock_irqtime().
+config HAVE_MOVE_PMD
+ bool
+ help
+ Archs that select this are able to move page tables at the PMD level.
+
config HAVE_ARCH_TRANSPARENT_HUGEPAGE
bool
diff --git a/mm/mremap.c b/mm/mremap.c
index 9e68a02a52b1..2fd163cff406 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c
@@ -191,6 +191,54 @@ static void move_ptes(struct vm_area_struct *vma, pmd_t *old_pmd,
drop_rmap_locks(vma);
}
+static bool move_normal_pmd(struct vm_area_struct *vma, unsigned long old_addr,
+ unsigned long new_addr, unsigned long old_end,
+ pmd_t *old_pmd, pmd_t *new_pmd, bool *need_flush)
+{
+ spinlock_t *old_ptl, *new_ptl;
+ struct mm_struct *mm = vma->vm_mm;
+
+ if ((old_addr & ~PMD_MASK) || (new_addr & ~PMD_MASK)
+ || old_end - old_addr < PMD_SIZE)
+ return false;
+
+ /*
+ * The destination pmd shouldn't be established, free_pgtables()
+ * should have release it.
+ */
+ if (WARN_ON(!pmd_none(*new_pmd)))
+ return false;
+
+ /*
+ * We don't have to worry about the ordering of src and dst
+ * ptlocks because exclusive mmap_sem prevents deadlock.
+ */
+ old_ptl = pmd_lock(vma->vm_mm, old_pmd);
+ if (old_ptl) {
+ pmd_t pmd;
+
+ new_ptl = pmd_lockptr(mm, new_pmd);
+ if (new_ptl != old_ptl)
+ spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
+
+ /* Clear the pmd */
+ pmd = *old_pmd;
+ pmd_clear(old_pmd);
+
+ VM_BUG_ON(!pmd_none(*new_pmd));
+
+ /* Set the new pmd */
+ set_pmd_at(mm, new_addr, new_pmd, pmd);
+ if (new_ptl != old_ptl)
+ spin_unlock(new_ptl);
+ spin_unlock(old_ptl);
+
+ *need_flush = true;
+ return true;
+ }
+ return false;
+}
+
unsigned long move_page_tables(struct vm_area_struct *vma,
unsigned long old_addr, struct vm_area_struct *new_vma,
unsigned long new_addr, unsigned long len,
@@ -239,7 +287,24 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
split_huge_pmd(vma, old_pmd, old_addr);
if (pmd_trans_unstable(old_pmd))
continue;
+ } else if (extent == PMD_SIZE && IS_ENABLED(CONFIG_HAVE_MOVE_PMD)) {
+ /*
+ * If the extent is PMD-sized, try to speed the move by
+ * moving at the PMD level if possible.
+ */
+ bool moved;
+
+ if (need_rmap_locks)
+ take_rmap_locks(vma);
+ moved = move_normal_pmd(vma, old_addr, new_addr,
+ old_end, old_pmd, new_pmd,
+ &need_flush);
+ if (need_rmap_locks)
+ drop_rmap_locks(vma);
+ if (moved)
+ continue;
}
+
if (pte_alloc(new_vma->vm_mm, new_pmd))
break;
next = (new_addr + PMD_SIZE) & PMD_MASK;
--
2.19.0.605.g01d371f741-goog
next prev parent reply other threads:[~2018-10-13 1:31 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-13 1:31 [PATCH 0/4] Add support for fast mremap Joel Fernandes (Google)
2018-10-13 1:31 ` [PATCH 1/4] treewide: remove unused address argument from pte_alloc functions (v2) Joel Fernandes (Google)
2018-10-24 8:37 ` Peter Zijlstra
2018-10-25 2:21 ` Joel Fernandes
2018-10-26 8:52 ` Peter Zijlstra
2018-10-25 10:47 ` Kirill A. Shutemov
2018-10-26 8:50 ` Peter Zijlstra
2018-10-13 1:31 ` Joel Fernandes (Google) [this message]
2018-10-15 9:42 ` [PATCH 2/4] mm: speed up mremap by 500x on large regions (v2) Christoph Hellwig
2018-10-15 22:33 ` Joel Fernandes
2018-10-16 11:29 ` Vlastimil Babka
2018-10-16 19:43 ` Joel Fernandes
2018-10-17 7:38 ` Vlastimil Babka
2018-10-24 10:12 ` Kirill A. Shutemov
2018-10-24 11:57 ` Balbir Singh
2018-10-24 12:57 ` Kirill A. Shutemov
2018-10-25 2:09 ` Joel Fernandes
2018-10-25 10:19 ` Kirill A. Shutemov
2018-10-26 21:11 ` Joel Fernandes
2018-10-29 10:28 ` Will Deacon
2018-10-25 2:13 ` Joel Fernandes
2018-10-27 10:21 ` Balbir Singh
2018-10-27 19:39 ` Joel Fernandes
2018-10-28 22:40 ` Balbir Singh
2018-10-13 1:31 ` [PATCH 3/4] arm64: select HAVE_MOVE_PMD for faster mremap (v1) Joel Fernandes (Google)
2018-10-13 1:32 ` [PATCH 4/4] x86: " Joel Fernandes (Google)
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=20181013013200.206928-3-joel@joelfernandes.org \
--to=joel@joelfernandes.org \
--cc=anton.ivanov@kot-begem \
--cc=catalin.marinas@arm.com \
--cc=dalias@libc.org \
--cc=dancol@google.com \
--cc=dave.hansen@linux.intel.com \
--cc=deller@gmx.de \
--cc=elfring@users.sourceforge.net \
--cc=hughd@google.com \
--cc=jejb@parisc-linux.org \
--cc=jonas@southpole.se \
--cc=kasan-dev@googlegroups.com \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=linux-hexagon@vger.kernel.org \
--cc=linux-ia64@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mips@linux-mips.org \
--cc=linux-mm@kvack.org \
--cc=linux-riscv@lists.infradead.org \
--cc=linux-sh@vger.kernel.org \
--cc=linux-xtensa@linux-xtensa.org \
--cc=lokeshgidra@google.com \
--cc=mhocko@kernel.org \
--cc=peterz@infradead.org \
--cc=sparclinux@vger.kernel.org \
--cc=will.deacon@arm.com \
--cc=x86@kernel.org \
--cc=ysato@users.sourceforge.jp \
/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;
as well as URLs for NNTP newsgroup(s).