Linux-Fsdevel Archive on lore.kernel.org
 help / color / Atom feed
From: Gao Xiang <gaoxiang25@huawei.com>
To: <chao@kernel.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	<devel@driverdev.osuosl.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	<linux-fsdevel@vger.kernel.org>, <linux-erofs@lists.ozlabs.org>,
	Miao Xie <miaoxie@huawei.com>, <weidu.du@huawei.com>,
	Fang Wei <fangwei1@huawei.com>, Gao Xiang <gaoxiang25@huawei.com>
Subject: [RFC PATCH 6/8] staging: erofs: introduce LZ4 decompression inplace
Date: Sat, 15 Jun 2019 02:16:17 +0800
Message-ID: <20190614181619.64905-7-gaoxiang25@huawei.com> (raw)
In-Reply-To: <20190614181619.64905-1-gaoxiang25@huawei.com>

compressed data will be usually loaded into last pages of
the extent (the last page for 4k) for in-place decompression
(more specifically, in-place IO), as ilustration below,

         start of compressed logical extent
           |                          end of this logical extent
           |                           |
     ______v___________________________v________
... |  page 6  |  page 7  |  page 8  |  page 9  | ...
    |__________|__________|__________|__________|
           .                         ^ .        ^
           .                         |compressed|
           .                         |   data   |
           .                           .        .
           |<          dstsize        >|<margin>|
                                       oend     iend
           op                        ip

Therefore, it's possible to do decompression inplace (thus no
memcpy at all) if the margin is sufficient and safe enough [1],
and it can be implemented only for fixed-size output compression
compared with fixed-size input compression.

No memcpy for most of in-place IO (about 99% of enwik9) after
decompression inplace is implemented and sequential read will
be improved of course (see the following patch for test results).

[1] https://github.com/lz4/lz4/commit/b17f578a919b7e6b078cede2d52be29dd48c8e8c
    https://github.com/lz4/lz4/commit/5997e139f53169fa3a1c1b4418d2452a90b01602

Signed-off-by: Gao Xiang <gaoxiang25@huawei.com>
---
 drivers/staging/erofs/compress.h     |  1 +
 drivers/staging/erofs/decompressor.c | 21 ++++++++++++++++++---
 drivers/staging/erofs/erofs_fs.h     |  3 ++-
 3 files changed, 21 insertions(+), 4 deletions(-)

diff --git a/drivers/staging/erofs/compress.h b/drivers/staging/erofs/compress.h
index ebeccb1f4eae..c43aa3374d28 100644
--- a/drivers/staging/erofs/compress.h
+++ b/drivers/staging/erofs/compress.h
@@ -17,6 +17,7 @@ enum {
 };
 
 struct z_erofs_decompress_req {
+	struct super_block *sb;
 	struct page **in, **out;
 
 	unsigned short pageofs_out;
diff --git a/drivers/staging/erofs/decompressor.c b/drivers/staging/erofs/decompressor.c
index 4f644bdea0bc..727bae6a3cb0 100644
--- a/drivers/staging/erofs/decompressor.c
+++ b/drivers/staging/erofs/decompressor.c
@@ -14,6 +14,9 @@
 #endif
 
 #define LZ4_MAX_DISTANCE_PAGES	DIV_ROUND_UP(LZ4_DISTANCE_MAX, PAGE_SIZE)
+#ifndef LZ4_DECOMPRESS_INPLACE_MARGIN
+#define LZ4_DECOMPRESS_INPLACE_MARGIN(srcsize)  (((srcsize) >> 8) + 32)
+#endif
 
 struct z_erofs_decompressor {
 	/*
@@ -131,9 +134,21 @@ static int lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out)
 	copied = false;
 	inlen = rq->inputsize - inputmargin;
 	if (rq->inplace_io) {
-		src = generic_copy_inplace_data(rq, src, inputmargin);
-		inputmargin = 0;
-		copied = true;
+		const uint oend = (rq->pageofs_out +
+				   rq->outputsize) & ~PAGE_MASK;
+		const uint nr = PAGE_ALIGN(rq->pageofs_out +
+					   rq->outputsize) >> PAGE_SHIFT;
+
+		if (rq->partial_decoding ||
+		    !(EROFS_SB(rq->sb)->requirements &
+		      EROFS_REQUIREMENT_LZ4_0PADDING) ||
+		    rq->out[nr - 1] != rq->in[0] ||
+		    rq->inputsize - oend <
+		      LZ4_DECOMPRESS_INPLACE_MARGIN(inlen)) {
+			src = generic_copy_inplace_data(rq, src, inputmargin);
+			inputmargin = 0;
+			copied = true;
+		}
 	}
 
 	ret = LZ4_decompress_safe_partial(src + inputmargin, out,
diff --git a/drivers/staging/erofs/erofs_fs.h b/drivers/staging/erofs/erofs_fs.h
index a05139f1df60..353322a3206c 100644
--- a/drivers/staging/erofs/erofs_fs.h
+++ b/drivers/staging/erofs/erofs_fs.h
@@ -21,7 +21,8 @@
  * Any bits that aren't in EROFS_ALL_REQUIREMENTS should be
  * incompatible with this kernel version.
  */
-#define EROFS_ALL_REQUIREMENTS  0
+#define EROFS_REQUIREMENT_LZ4_0PADDING	0x00000001
+#define EROFS_ALL_REQUIREMENTS		EROFS_REQUIREMENT_LZ4_0PADDING
 
 struct erofs_super_block {
 /*  0 */__le32 magic;           /* in the little endian */
-- 
2.17.1


  parent reply index

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-14 18:16 [RFC PATCH 0/8] staging: erofs: decompression inplace approach Gao Xiang
2019-06-14 18:16 ` [RFC PATCH 1/8] staging: erofs: add compacted ondisk compression indexes Gao Xiang
2019-06-14 18:16 ` [RFC PATCH 2/8] staging: erofs: add compacted compression indexes support Gao Xiang
2019-06-14 18:16 ` [RFC PATCH 3/8] staging: erofs: move per-CPU buffers implementation to utils.c Gao Xiang
2019-06-14 18:16 ` [RFC PATCH 4/8] staging: erofs: move stagingpage operations to compress.h Gao Xiang
2019-06-14 18:16 ` [RFC PATCH 5/8] staging: erofs: introduce generic decompression backend Gao Xiang
2019-06-14 18:16 ` Gao Xiang [this message]
2019-06-14 18:16 ` [RFC PATCH 7/8] staging: erofs: switch to new " Gao Xiang
2019-06-14 18:16 ` [RFC PATCH 8/8] staging: erofs: integrate decompression inplace Gao Xiang
2019-06-17 20:36 ` [RFC PATCH 0/8] staging: erofs: decompression inplace approach Greg Kroah-Hartman
2019-06-18  1:47   ` Gao Xiang
2019-06-18  5:47     ` Greg Kroah-Hartman
2019-06-18  6:18       ` Gao Xiang
2019-06-18  6:45         ` Greg Kroah-Hartman
2019-06-18  6:52           ` Gao Xiang
2019-06-18  7:05             ` Greg Kroah-Hartman
2019-06-18  7:13               ` Gao Xiang

Reply instructions:

You may reply publically 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=20190614181619.64905-7-gaoxiang25@huawei.com \
    --to=gaoxiang25@huawei.com \
    --cc=chao@kernel.org \
    --cc=devel@driverdev.osuosl.org \
    --cc=fangwei1@huawei.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-erofs@lists.ozlabs.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miaoxie@huawei.com \
    --cc=weidu.du@huawei.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

Linux-Fsdevel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-fsdevel/0 linux-fsdevel/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-fsdevel linux-fsdevel/ https://lore.kernel.org/linux-fsdevel \
		linux-fsdevel@vger.kernel.org linux-fsdevel@archiver.kernel.org
	public-inbox-index linux-fsdevel


Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-fsdevel


AGPL code for this site: git clone https://public-inbox.org/ public-inbox