linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Pavel Tatashin <pasha.tatashin@oracle.com>
To: linux-kernel@vger.kernel.org, sparclinux@vger.kernel.org,
	linux-mm@kvack.org, linuxppc-dev@lists.ozlabs.org,
	linux-s390@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	x86@kernel.org, kasan-dev@googlegroups.com,
	borntraeger@de.ibm.com, heiko.carstens@de.ibm.com,
	davem@davemloft.net, willy@infradead.org, mhocko@kernel.org,
	ard.biesheuvel@linaro.org, will.deacon@arm.com,
	catalin.marinas@arm.com, sam@ravnborg.org
Subject: [v6 09/15] sparc64: optimized struct page zeroing
Date: Mon,  7 Aug 2017 16:38:43 -0400	[thread overview]
Message-ID: <1502138329-123460-10-git-send-email-pasha.tatashin@oracle.com> (raw)
In-Reply-To: <1502138329-123460-1-git-send-email-pasha.tatashin@oracle.com>

Add an optimized mm_zero_struct_page(), so struct page's are zeroed without
calling memset(). We do eight to tent regular stores based on the size of
struct page. Compiler optimizes out the conditions of switch() statement.

Signed-off-by: Pavel Tatashin <pasha.tatashin@oracle.com>
Reviewed-by: Steven Sistare <steven.sistare@oracle.com>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Reviewed-by: Bob Picco <bob.picco@oracle.com>
---
 arch/sparc/include/asm/pgtable_64.h | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h
index 6fbd931f0570..cee5cc7ccc51 100644
--- a/arch/sparc/include/asm/pgtable_64.h
+++ b/arch/sparc/include/asm/pgtable_64.h
@@ -230,6 +230,36 @@ extern unsigned long _PAGE_ALL_SZ_BITS;
 extern struct page *mem_map_zero;
 #define ZERO_PAGE(vaddr)	(mem_map_zero)
 
+/* This macro must be updated when the size of struct page grows above 80
+ * or reduces below 64.
+ * The idea that compiler optimizes out switch() statement, and only
+ * leaves clrx instructions
+ */
+#define	mm_zero_struct_page(pp) do {					\
+	unsigned long *_pp = (void *)(pp);				\
+									\
+	 /* Check that struct page is either 64, 72, or 80 bytes */	\
+	BUILD_BUG_ON(sizeof(struct page) & 7);				\
+	BUILD_BUG_ON(sizeof(struct page) < 64);				\
+	BUILD_BUG_ON(sizeof(struct page) > 80);				\
+									\
+	switch (sizeof(struct page)) {					\
+	case 80:							\
+		_pp[9] = 0;	/* fallthrough */			\
+	case 72:							\
+		_pp[8] = 0;	/* fallthrough */			\
+	default:							\
+		_pp[7] = 0;						\
+		_pp[6] = 0;						\
+		_pp[5] = 0;						\
+		_pp[4] = 0;						\
+		_pp[3] = 0;						\
+		_pp[2] = 0;						\
+		_pp[1] = 0;						\
+		_pp[0] = 0;						\
+	}								\
+} while (0)
+
 /* PFNs are real physical page numbers.  However, mem_map only begins to record
  * per-page information starting at pfn_base.  This is to handle systems where
  * the first physical page in the machine is at some huge physical address,
-- 
2.14.0

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2017-08-07 20:39 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-08-07 20:38 [v6 00/15] complete deferred page initialization Pavel Tatashin
2017-08-07 20:38 ` [v6 01/15] x86/mm: reserve only exiting low pages Pavel Tatashin
2017-08-11  8:07   ` Michal Hocko
2017-08-11 15:24     ` Pasha Tatashin
2017-08-14 11:40       ` Michal Hocko
2017-08-14 13:30         ` Pasha Tatashin
2017-08-14 13:55   ` Michal Hocko
2017-08-17 15:37     ` Pasha Tatashin
2017-08-07 20:38 ` [v6 02/15] x86/mm: setting fields in deferred pages Pavel Tatashin
2017-08-11  9:02   ` Michal Hocko
2017-08-11 15:39     ` Pasha Tatashin
2017-08-14 11:43       ` Michal Hocko
2017-08-14 13:32         ` Pasha Tatashin
2017-08-07 20:38 ` [v6 03/15] sparc64/mm: " Pavel Tatashin
2017-08-07 20:38 ` [v6 04/15] mm: discard memblock data later Pavel Tatashin
2017-08-11  9:32   ` Michal Hocko
2017-08-11  9:50     ` Mel Gorman
2017-08-11 15:49     ` Pasha Tatashin
2017-08-11 16:04       ` Michal Hocko
2017-08-11 16:22         ` Pasha Tatashin
2017-08-14 11:36           ` Michal Hocko
2017-08-14 13:35             ` Pasha Tatashin
2017-08-11 19:00     ` Pasha Tatashin
2017-08-14 11:34       ` Michal Hocko
2017-08-14 13:39         ` Pasha Tatashin
2017-08-14 13:42           ` Michal Hocko
2017-08-07 20:38 ` [v6 05/15] mm: don't accessed uninitialized struct pages Pavel Tatashin
2017-08-11  9:37   ` Michal Hocko
2017-08-11 15:55     ` Pasha Tatashin
2017-08-14 11:47       ` Michal Hocko
2017-08-14 13:51         ` Pasha Tatashin
2017-08-17 15:28           ` Pasha Tatashin
2017-08-17 15:43             ` Michal Hocko
2017-08-15  9:33   ` Michal Hocko
2017-08-07 20:38 ` [v6 06/15] sparc64: simplify vmemmap_populate Pavel Tatashin
2017-08-07 20:38 ` [v6 07/15] mm: defining memblock_virt_alloc_try_nid_raw Pavel Tatashin
2017-08-11 12:39   ` Michal Hocko
2017-08-11 15:58     ` Pasha Tatashin
2017-08-11 16:06       ` Michal Hocko
2017-08-11 16:24         ` Pasha Tatashin
2017-08-07 20:38 ` [v6 08/15] mm: zero struct pages during initialization Pavel Tatashin
2017-08-11 12:50   ` Michal Hocko
2017-08-11 16:03     ` Pasha Tatashin
2017-08-07 20:38 ` Pavel Tatashin [this message]
2017-08-11 12:53   ` [v6 09/15] sparc64: optimized struct page zeroing Michal Hocko
2017-08-11 16:04     ` Pasha Tatashin
2017-08-07 20:38 ` [v6 10/15] x86/kasan: explicitly zero kasan shadow memory Pavel Tatashin
2017-08-07 20:38 ` [v6 11/15] arm64/kasan: " Pavel Tatashin
2017-08-08  9:07   ` Will Deacon
2017-08-08 11:49     ` Pasha Tatashin
2017-08-08 12:30       ` Will Deacon
2017-08-08 12:49         ` Pasha Tatashin
2017-08-08 13:15       ` David Laight
2017-08-08 13:30         ` Pasha Tatashin
2017-08-07 20:38 ` [v6 12/15] mm: explicitly zero pagetable memory Pavel Tatashin
2017-08-07 20:38 ` [v6 13/15] mm: stop zeroing memory during allocation in vmemmap Pavel Tatashin
2017-08-11 13:04   ` Michal Hocko
2017-08-11 16:11     ` Pasha Tatashin
2017-08-07 20:38 ` [v6 14/15] mm: optimize early system hash allocations Pavel Tatashin
2017-08-11 13:05   ` Michal Hocko
2017-08-11 16:13     ` Pasha Tatashin
2017-08-07 20:38 ` [v6 15/15] mm: debug for raw alloctor Pavel Tatashin
2017-08-11 13:08   ` Michal Hocko
2017-08-11 16:18     ` Pasha Tatashin
2017-08-14 11:50       ` Michal Hocko
2017-08-14 14:01         ` Pasha Tatashin
2017-08-15  9:36           ` Michal Hocko
2017-08-11  7:58 ` [v6 00/15] complete deferred page initialization Michal Hocko
2017-08-11 15:13   ` Pasha Tatashin
2017-08-11 15:22     ` Michal Hocko

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=1502138329-123460-10-git-send-email-pasha.tatashin@oracle.com \
    --to=pasha.tatashin@oracle.com \
    --cc=ard.biesheuvel@linaro.org \
    --cc=borntraeger@de.ibm.com \
    --cc=catalin.marinas@arm.com \
    --cc=davem@davemloft.net \
    --cc=heiko.carstens@de.ibm.com \
    --cc=kasan-dev@googlegroups.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-s390@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=mhocko@kernel.org \
    --cc=sam@ravnborg.org \
    --cc=sparclinux@vger.kernel.org \
    --cc=will.deacon@arm.com \
    --cc=willy@infradead.org \
    --cc=x86@kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox