[PATCH 0/2] Squashfs: add LZ4 compression

[PATCH 0/2] Squashfs: add LZ4 compression

[Phillip Lougher]

Hi,

The following is a repost of the Squashfs LZ4 compression patches
updated to the latest -rc kernel.  These patches were originally sent
last year, but they did not receive much interest at the time.

These patches are being reposted by popular demand.  I have received
many private and public requests to repost these patches and to get
them mainlined.

My intention is to submit them in the next kernel merge window.
If you want LZ4 support in Squashfs now is a good time to publically
support the inclusion of these patches.

Phillip Lougher (2):
      Squashfs: add LZ4 compression support
      Squashfs: Add LZ4 compression configuration option

 Documentation/filesystems/squashfs.txt |    8 +-
 fs/squashfs/Kconfig                    |   15 ++++
 fs/squashfs/Makefile                   |    1 +
 fs/squashfs/decompressor.c             |    7 ++
 fs/squashfs/decompressor.h             |    4 +
 fs/squashfs/lz4_wrapper.c              |  142 ++++++++++++++++++++++++++++++++
 fs/squashfs/squashfs_fs.h              |    1 +
 7 files changed, 174 insertions(+), 4 deletions(-)
 create mode 100644 fs/squashfs/lz4_wrapper.c

These patches are also available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-lz4.git master

Squashfs-tools 4.3 already supports LZ4 compression.

Phillip

[PATCH 1/2] Squashfs: add LZ4 compression support

[Phillip Lougher]

Add support for reading file systems compressed with the
LZ4 compression algorithm.

This patch adds the LZ4 decompressor wrapper code.

Signed-off-by: Phillip Lougher <phillip@squashfs.org.uk>
---
 fs/squashfs/lz4_wrapper.c |  142 +++++++++++++++++++++++++++++++++++++++++++++
 fs/squashfs/squashfs_fs.h |    1 +
 2 files changed, 143 insertions(+)
 create mode 100644 fs/squashfs/lz4_wrapper.c

diff --git a/fs/squashfs/lz4_wrapper.c b/fs/squashfs/lz4_wrapper.c
new file mode 100644
index 0000000..c31e2bc
--- /dev/null
+++ b/fs/squashfs/lz4_wrapper.c
@@ -0,0 +1,142 @@
+/*
+ * Copyright (c) 2013, 2014
+ * Phillip Lougher <phillip@squashfs.org.uk>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2. See
+ * the COPYING file in the top-level directory.
+ */
+
+#include <linux/buffer_head.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/lz4.h>
+
+#include "squashfs_fs.h"
+#include "squashfs_fs_sb.h"
+#include "squashfs.h"
+#include "decompressor.h"
+#include "page_actor.h"
+
+#define LZ4_LEGACY	1
+
+struct lz4_comp_opts {
+	__le32 version;
+	__le32 flags;
+};
+
+struct squashfs_lz4 {
+	void *input;
+	void *output;
+};
+
+
+static void *lz4_comp_opts(struct squashfs_sb_info *msblk,
+	void *buff, int len)
+{
+	struct lz4_comp_opts *comp_opts = buff;
+
+	/* LZ4 compressed filesystems always have compression options */
+	if (comp_opts == NULL || len < sizeof(*comp_opts))
+		return ERR_PTR(-EIO);
+
+	if (le32_to_cpu(comp_opts->version) != LZ4_LEGACY) {
+		/* LZ4 format currently used by the kernel is the 'legacy'
+		 * format */
+		ERROR("Unknown LZ4 version\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	return NULL;
+}
+
+
+static void *lz4_init(struct squashfs_sb_info *msblk, void *buff)
+{
+	int block_size = max_t(int, msblk->block_size, SQUASHFS_METADATA_SIZE);
+	struct squashfs_lz4 *stream;
+
+	stream = kzalloc(sizeof(*stream), GFP_KERNEL);
+	if (stream == NULL)
+		goto failed;
+	stream->input = vmalloc(block_size);
+	if (stream->input == NULL)
+		goto failed2;
+	stream->output = vmalloc(block_size);
+	if (stream->output == NULL)
+		goto failed3;
+
+	return stream;
+
+failed3:
+	vfree(stream->input);
+failed2:
+	kfree(stream);
+failed:
+	ERROR("Failed to initialise LZ4 decompressor\n");
+	return ERR_PTR(-ENOMEM);
+}
+
+
+static void lz4_free(void *strm)
+{
+	struct squashfs_lz4 *stream = strm;
+
+	if (stream) {
+		vfree(stream->input);
+		vfree(stream->output);
+	}
+	kfree(stream);
+}
+
+
+static int lz4_uncompress(struct squashfs_sb_info *msblk, void *strm,
+	struct buffer_head **bh, int b, int offset, int length,
+	struct squashfs_page_actor *output)
+{
+	struct squashfs_lz4 *stream = strm;
+	void *buff = stream->input, *data;
+	int avail, i, bytes = length, res;
+	size_t dest_len = output->length;
+
+	for (i = 0; i < b; i++) {
+		avail = min(bytes, msblk->devblksize - offset);
+		memcpy(buff, bh[i]->b_data + offset, avail);
+		buff += avail;
+		bytes -= avail;
+		offset = 0;
+		put_bh(bh[i]);
+	}
+
+	res = lz4_decompress_unknownoutputsize(stream->input, length,
+					stream->output, &dest_len);
+	if (res)
+		return -EIO;
+
+	bytes = dest_len;
+	data = squashfs_first_page(output);
+	buff = stream->output;
+	while (data) {
+		if (bytes <= PAGE_CACHE_SIZE) {
+			memcpy(data, buff, bytes);
+			break;
+		}
+		memcpy(data, buff, PAGE_CACHE_SIZE);
+		buff += PAGE_CACHE_SIZE;
+		bytes -= PAGE_CACHE_SIZE;
+		data = squashfs_next_page(output);
+	}
+	squashfs_finish_page(output);
+
+	return dest_len;
+}
+
+const struct squashfs_decompressor squashfs_lz4_comp_ops = {
+	.init = lz4_init,
+	.comp_opts = lz4_comp_opts,
+	.free = lz4_free,
+	.decompress = lz4_uncompress,
+	.id = LZ4_COMPRESSION,
+	.name = "lz4",
+	.supported = 1
+};
diff --git a/fs/squashfs/squashfs_fs.h b/fs/squashfs/squashfs_fs.h
index 4b2beda..506f4ba 100644
--- a/fs/squashfs/squashfs_fs.h
+++ b/fs/squashfs/squashfs_fs.h
@@ -240,6 +240,7 @@ struct meta_index {
 #define LZMA_COMPRESSION	2
 #define LZO_COMPRESSION		3
 #define XZ_COMPRESSION		4
+#define LZ4_COMPRESSION		5
 
 struct squashfs_super_block {
 	__le32			s_magic;
-- 
1.7.10.4

[PATCH 2/2] Squashfs: Add LZ4 compression configuration option

[Phillip Lougher]

Add the glue code, and also update the documentation.

Signed-off-by: Phillip Lougher <phillip@squashfs.org.uk>
---
 Documentation/filesystems/squashfs.txt |    8 ++++----
 fs/squashfs/Kconfig                    |   15 +++++++++++++++
 fs/squashfs/Makefile                   |    1 +
 fs/squashfs/decompressor.c             |    7 +++++++
 fs/squashfs/decompressor.h             |    4 ++++
 5 files changed, 31 insertions(+), 4 deletions(-)

diff --git a/Documentation/filesystems/squashfs.txt b/Documentation/filesystems/squashfs.txt
index 403c090..e5274f8 100644
--- a/Documentation/filesystems/squashfs.txt
+++ b/Documentation/filesystems/squashfs.txt
@@ -2,10 +2,10 @@ SQUASHFS 4.0 FILESYSTEM
 =======================
 
 Squashfs is a compressed read-only filesystem for Linux.
-It uses zlib/lzo/xz compression to compress files, inodes and directories.
-Inodes in the system are very small and all blocks are packed to minimise
-data overhead. Block sizes greater than 4K are supported up to a maximum
-of 1Mbytes (default block size 128K).
+It uses zlib, lz4, lzo, or xz compression to compress files, inodes and
+directories.  Inodes in the system are very small and all blocks are packed to
+minimise data overhead. Block sizes greater than 4K are supported up to a
+maximum of 1Mbytes (default block size 128K).
 
 Squashfs is intended for general read-only filesystem use, for archival
 use (i.e. in cases where a .tar.gz file may be used), and in constrained
diff --git a/fs/squashfs/Kconfig b/fs/squashfs/Kconfig
index b6fa865..ffb093e 100644
--- a/fs/squashfs/Kconfig
+++ b/fs/squashfs/Kconfig
@@ -120,6 +120,21 @@ config SQUASHFS_ZLIB
 
 	  If unsure, say Y.
 
+config SQUASHFS_LZ4
+	bool "Include support for LZ4 compressed file systems"
+	depends on SQUASHFS
+	select LZ4_DECOMPRESS
+	help
+	  Saying Y here includes support for reading Squashfs file systems
+	  compressed with LZ4 compression.  LZ4 compression is mainly
+	  aimed at embedded systems with slower CPUs where the overheads
+	  of zlib are too high.
+
+	  LZ4 is not the standard compression used in Squashfs and so most
+	  file systems will be readable without selecting this option.
+
+	  If unsure, say N.
+
 config SQUASHFS_LZO
 	bool "Include support for LZO compressed file systems"
 	depends on SQUASHFS
diff --git a/fs/squashfs/Makefile b/fs/squashfs/Makefile
index 4132520..246a6f3 100644
--- a/fs/squashfs/Makefile
+++ b/fs/squashfs/Makefile
@@ -11,6 +11,7 @@ squashfs-$(CONFIG_SQUASHFS_DECOMP_SINGLE) += decompressor_single.o
 squashfs-$(CONFIG_SQUASHFS_DECOMP_MULTI) += decompressor_multi.o
 squashfs-$(CONFIG_SQUASHFS_DECOMP_MULTI_PERCPU) += decompressor_multi_percpu.o
 squashfs-$(CONFIG_SQUASHFS_XATTR) += xattr.o xattr_id.o
+squashfs-$(CONFIG_SQUASHFS_LZ4) += lz4_wrapper.o
 squashfs-$(CONFIG_SQUASHFS_LZO) += lzo_wrapper.o
 squashfs-$(CONFIG_SQUASHFS_XZ) += xz_wrapper.o
 squashfs-$(CONFIG_SQUASHFS_ZLIB) += zlib_wrapper.o
diff --git a/fs/squashfs/decompressor.c b/fs/squashfs/decompressor.c
index ac22fe7..505d6aa 100644
--- a/fs/squashfs/decompressor.c
+++ b/fs/squashfs/decompressor.c
@@ -41,6 +41,12 @@ static const struct squashfs_decompressor squashfs_lzma_unsupported_comp_ops = {
 	NULL, NULL, NULL, NULL, LZMA_COMPRESSION, "lzma", 0
 };
 
+#ifndef CONFIG_SQUASHFS_LZ4
+static const struct squashfs_decompressor squashfs_lz4_comp_ops = {
+	NULL, NULL, NULL, LZ4_COMPRESSION, "lz4", 0
+};
+#endif
+
 #ifndef CONFIG_SQUASHFS_LZO
 static const struct squashfs_decompressor squashfs_lzo_comp_ops = {
 	NULL, NULL, NULL, NULL, LZO_COMPRESSION, "lzo", 0
@@ -65,6 +71,7 @@ static const struct squashfs_decompressor squashfs_unknown_comp_ops = {
 
 static const struct squashfs_decompressor *decompressor[] = {
 	&squashfs_zlib_comp_ops,
+	&squashfs_lz4_comp_ops,
 	&squashfs_lzo_comp_ops,
 	&squashfs_xz_comp_ops,
 	&squashfs_lzma_unsupported_comp_ops,
diff --git a/fs/squashfs/decompressor.h b/fs/squashfs/decompressor.h
index af09853..a25713c 100644
--- a/fs/squashfs/decompressor.h
+++ b/fs/squashfs/decompressor.h
@@ -46,6 +46,10 @@ static inline void *squashfs_comp_opts(struct squashfs_sb_info *msblk,
 extern const struct squashfs_decompressor squashfs_xz_comp_ops;
 #endif
 
+#ifdef CONFIG_SQUASHFS_LZ4
+extern const struct squashfs_decompressor squashfs_lz4_comp_ops;
+#endif
+
 #ifdef CONFIG_SQUASHFS_LZO
 extern const struct squashfs_decompressor squashfs_lzo_comp_ops;
 #endif
-- 
1.7.10.4

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Bruno Wolff III]

On Thu, Nov 27, 2014 at 08:00:47 +0000,
  Phillip Lougher <phillip@squashfs.org.uk> wrote:
>
>My intention is to submit them in the next kernel merge window.
>If you want LZ4 support in Squashfs now is a good time to publically
>support the inclusion of these patches.

Fedora has been supporting LZ4 functionallity in mksquashfs and unsquashfs 
for about 6 months (in F19+), but the Fedora kernel team won't add support 
in the kernel until LZ4 support is accepted for the upstream kernel.

Fwd: [PATCH 0/2] Squashfs: add LZ4 compression

[Marcin Szychowski]

---------- Forwarded message ----------
From: Marcin Szychowski <szycha@gmail.com>
Date: 2014-11-27 15:50 GMT+00:00
Subject: Re: [PATCH 0/2] Squashfs: add LZ4 compression
To: Bruno Wolff III <bruno@wolff.to>
Cc: Phillip Lougher <phillip@squashfs.org.uk>,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
texstar@gmail.com, martin@mvath.de, guanx.bac@gmail.com,
dave@vasilevsky.ca, blyons@students.naropa.edu, tokiclover@gmail.com,
afm404@gmail.com, hugochevrain@gmail.com


2014-11-27 13:37 GMT+00:00 Bruno Wolff III <bruno@wolff.to>:
>
> On Thu, Nov 27, 2014 at 08:00:47 +0000,
>  Phillip Lougher <phillip@squashfs.org.uk> wrote:
>>
>>
>> My intention is to submit them in the next kernel merge window.
>> If you want LZ4 support in Squashfs now is a good time to publically
>> support the inclusion of these patches.
>
>
> Fedora has been supporting LZ4 functionallity in mksquashfs and unsquashfs for about 6 months (in F19+), but the Fedora kernel team won't add support in the kernel until LZ4 support is accepted for the upstream kernel.

Hello everyone,

Squashfs is awesome piece of code.  It’s potential usage pattern
analysis could turn into pretty decent PhD thesis – from simple
archival purposes through embedded systems, to dynamic nearly-instant
multi-server software deployment systems, and many others.

On the other hand, LZ4 is fantastic compression algorithm with
well-known list of advantages.

As I wrote to Phillip earlier, I have been using squashfs+lz4 / aufs
to speed up my laptop (and reclaim some space, too).  That would not
be possible without LZ4: I could have gain some speed with LZO or disk
space with gzip or xz, but not both.  LZ4 is comparable with LZO in
terms of decompression speed and with gzip in terms of compression
ratios.  LZ4 is much faster in both compression and in particular
decompression speed than gzip (1-9) and xz (1-9).  It is just the best
choice for many storage-related purposes (see ZFS on Linux
[https://github.com/zfsonlinux/zfs/commit/9759c60f1a1503e48dc5c45a209c3edd5758319f]
and [http://wiki.illumos.org/display/illumos/LZ4+Compression])

I don’t quite get why does anyone need to be convinced to merge
advantages of two fine pieces of code, since both have been already in
the mainline kernel for some time now.  It’s like being forced to
choose between big car and safe car, or delicious food and
good-looking food.

It is not about forcing anyone to use something, it is about ease of choice.

Having squashfs with lz4 in mainline kernel sounds like next natural
step.  Please do not hesitate to take it.


-- 
Regards,
Marcin Szychowski

+48.791460067
http://adastor.pl

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Martin Vath]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1351 bytes --]

Hi:

> squashfs+lz4 / aufs to speed up my laptop

Squashfs proves its usefulness especially in combination
with aufs or overlayfs; I suppose that this combination
will become more popular with the spreading of overlayfs
(e.g. to save disk space etc.)

For users with such a combination, it is especially important
to get a fast (re)compression of huge directories.

The speed of LZ4 for compression in such a setting is really incredible.
Just for orientation for myself, I made a list of times/sizes
on some machines I had access to:

https://github.com/vaeth/squashmount/blob/master/compress.txt

(I do not claim that this is a scientific benchmark - just a
straightforward average over several runs).
The results for the kernel source and libreoffice are really
unbelievable, but I repeated them several times (and,
as you can see, for the kernel sources on different machines).

>From the user perspective, it is mainly important that
the files compressed in this way can be read by the kernel -
the decompression speed is here secondary, so I made no comparison
for this case, although lz4 is known to be very fast also
for decompression.

I would really like to see lz4 support included in the
squashfs kernel driver. I cannot imagine any negative consequences,
especially since squashfs and lz4 are in the kernel, anyway.

Sincerely
Martin Väth

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Dave Vasilevsky]

I'd like to support the inclusion of these patches.

There's enough demand for LZ4-compressed squashfs that I received an
unsolicited patch adding support to my FUSE implementation,
squashfuse: https://github.com/vasi/squashfuse/pull/2

At least a few people are currently using squashfuse to mount their
LZ4-compressed squashfs images. Ideally these users would be able to
use the in-kernel squashfs for improved performance.

Cheers,
Dave Vasilevsky

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Anthony F McInerney]

Thanks very much for keeping me in the loop.
I'd like to put a big +1 on this patch inclusion.
Looking forward to in-kernel support for LZ4 squashfs images.

Thanks
Anthony

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Guan Xin]

Hi Philip,

Thank you for your work on squashfs and your persistent effort on mainlining
squashfs-lz4 support!

I've been using squashfs with lz4 compression (together with aufs by Okajima)
for my desktop, laptops, home router and bluetooth music player for about
half a year. There was not a single issue. All I got were reduced latency
compared to lzo compression, and drastically increased compression speed.
The later is important to me because my application requires re-mksquashfs of
the whole system in order to make software updates permanent.

lz4 support (especially with -Xhc) is not only helpful, but also a
valuable addition
to squashfs.

Regards,
Guan

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Bruno Wolff III]

On Fri, Dec 12, 2014 at 13:23:19 +0100,
  toki clover <tokiclover@gmail.com> wrote:
>
>Now, I did not see any Linux FS devs activity/response to this... What
>a waste of time because if those patch don't make it for this merge
>window, rebasing/reposting will be, again, necessary.

The patches got pulled into linux-next. For XZ, the patches sat in 
linux-next for a full release cycle. I don't know whether or not 
if that is the plan for the lz4 patches.

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Phillip Lougher]

On 12/12/14 15:56, Bruno Wolff III wrote:
> On Fri, Dec 12, 2014 at 13:23:19 +0100,
>   toki clover <tokiclover@gmail.com> wrote:
>>
>> Now, I did not see any Linux FS devs activity/response to this... What
>> a waste of time because if those patch don't make it for this merge
>> window, rebasing/reposting will be, again, necessary.
>
> The patches got pulled into linux-next. For XZ, the patches sat in linux-next for a full release cycle. I don't know whether or not if that is the plan for the lz4 patches.
> .
>

The plan is to send out the merge request this weekend.

I've been too busy this week with my main job to do the merge request,
sorry (I'll probably be working past midnight again tonight, the
weeks leading up to Christmas are always really busy).

Phillip

[PATCH 0/2] Squashfs: add LZ4 compression

[Phillip Lougher]

Hi

Now that LZ4 compression support is in 3.11-rc1, I have written the
following two patches for Squashfs to use it.

Phillip Lougher (2):
  Squashfs: add LZ4 compression support
  Squashfs: Add LZ4 compression configuration option

 Documentation/filesystems/squashfs.txt |    8 +-
 fs/squashfs/Kconfig                    |   15 +++
 fs/squashfs/Makefile                   |    1 +
 fs/squashfs/decompressor.c             |    7 ++
 fs/squashfs/decompressor.h             |    4 +
 fs/squashfs/lz4_wrapper.c              |  163 ++++++++++++++++++++++++++++++++
 fs/squashfs/squashfs_fs.h              |    1 +
 7 files changed, 195 insertions(+), 4 deletions(-)
 create mode 100644 fs/squashfs/lz4_wrapper.c

These patches are also available in the git tree here:

browse: https://git.kernel.org/cgit/linux/kernel/git/pkl/squashfs-lz4.git
git clone: git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-lz4.git

LZ4 support has (obviously) also been added to the squashfs-tools
(Mksquashfs and Unsquashfs).  This is available from the Squashfs-tools
git repository here:

browse: https://git.kernel.org/cgit/fs/squashfs/squashfs-tools.git
git clone: git://git.kernel.org/pub/scm/fs/squashfs/squashfs-tools.git

When building the squashfs-tools edit the Makefile to enable LZ4
support (by default it is disabled).

LZ4 compression can be specified by using the -comp option, e.g.
% mksquashfs xxx img.sqsh -comp lz4

The use of LZ4 high compression can be specified using -Xhc, e.g.

% mksquashfs xxx img.sqsh -comp lz4 -Xhc

Phillip

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Gu Zheng]

Hi Phillip,
	Have some tests been carried out to confirm that Squashfs really 
can get benefit from LZ4 compression, comparing with lzo?

Thanks,
Gu

On 07/22/2013 10:21 AM, Phillip Lougher wrote:

> Hi
> 
> Now that LZ4 compression support is in 3.11-rc1, I have written the
> following two patches for Squashfs to use it.
> 
> Phillip Lougher (2):
>   Squashfs: add LZ4 compression support
>   Squashfs: Add LZ4 compression configuration option
> 
>  Documentation/filesystems/squashfs.txt |    8 +-
>  fs/squashfs/Kconfig                    |   15 +++
>  fs/squashfs/Makefile                   |    1 +
>  fs/squashfs/decompressor.c             |    7 ++
>  fs/squashfs/decompressor.h             |    4 +
>  fs/squashfs/lz4_wrapper.c              |  163 ++++++++++++++++++++++++++++++++
>  fs/squashfs/squashfs_fs.h              |    1 +
>  7 files changed, 195 insertions(+), 4 deletions(-)
>  create mode 100644 fs/squashfs/lz4_wrapper.c
> 
> These patches are also available in the git tree here:
> 
> browse: https://git.kernel.org/cgit/linux/kernel/git/pkl/squashfs-lz4.git
> git clone: git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-lz4.git
> 
> LZ4 support has (obviously) also been added to the squashfs-tools
> (Mksquashfs and Unsquashfs).  This is available from the Squashfs-tools
> git repository here:
> 
> browse: https://git.kernel.org/cgit/fs/squashfs/squashfs-tools.git
> git clone: git://git.kernel.org/pub/scm/fs/squashfs/squashfs-tools.git
> 
> When building the squashfs-tools edit the Makefile to enable LZ4
> support (by default it is disabled).
> 
> LZ4 compression can be specified by using the -comp option, e.g.
> % mksquashfs xxx img.sqsh -comp lz4
> 
> The use of LZ4 high compression can be specified using -Xhc, e.g.
> 
> % mksquashfs xxx img.sqsh -comp lz4 -Xhc
> 
> Phillip
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Phillip Lougher]

On 22 July 2013 04:05, Gu Zheng <guz.fnst@cn.fujitsu.com> wrote:
> Hi Phillip,
>         Have some tests been carried out to confirm that Squashfs really
> can get benefit from LZ4 compression, comparing with lzo?

This seems to be a loaded question, in that it seems to be trying to
reopen the "why add lz4 when we already have lzo" debate all over
again.  As LZ4 has been merged to mainline, this appears to be a
question that has already been answered.

As far as Squashfs is concerned,  I believe it is important to give
people the choice of using LZ4 to compress Squashfs filesystems now
its been mainlined.  As far as expected benefits are concerned,
Squashfs' use in embedded systems is very similar to compressing
kernels and initramfs data, in that it tends to be used to compress
root filesystems.  As such the benefits of using LZ4 in Squashfs
should be broadly similar to using LZ4 to compress kernels and
initramfs data.  Ultimately it is up to people to experiment and
choose whatever compression is best for their systems.

People are welcome to try the patches out and report their findings.

Phillip

>
> Thanks,
> Gu
>
> On 07/22/2013 10:21 AM, Phillip Lougher wrote:
>
>> Hi
>>
>> Now that LZ4 compression support is in 3.11-rc1, I have written the
>> following two patches for Squashfs to use it.
>>
>> Phillip Lougher (2):
>>   Squashfs: add LZ4 compression support
>>   Squashfs: Add LZ4 compression configuration option
>>
>>  Documentation/filesystems/squashfs.txt |    8 +-
>>  fs/squashfs/Kconfig                    |   15 +++
>>  fs/squashfs/Makefile                   |    1 +
>>  fs/squashfs/decompressor.c             |    7 ++
>>  fs/squashfs/decompressor.h             |    4 +
>>  fs/squashfs/lz4_wrapper.c              |  163 ++++++++++++++++++++++++++++++++
>>  fs/squashfs/squashfs_fs.h              |    1 +
>>  7 files changed, 195 insertions(+), 4 deletions(-)
>>  create mode 100644 fs/squashfs/lz4_wrapper.c
>>
>> These patches are also available in the git tree here:
>>
>> browse: https://git.kernel.org/cgit/linux/kernel/git/pkl/squashfs-lz4.git
>> git clone: git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-lz4.git
>>
>> LZ4 support has (obviously) also been added to the squashfs-tools
>> (Mksquashfs and Unsquashfs).  This is available from the Squashfs-tools
>> git repository here:
>>
>> browse: https://git.kernel.org/cgit/fs/squashfs/squashfs-tools.git
>> git clone: git://git.kernel.org/pub/scm/fs/squashfs/squashfs-tools.git
>>
>> When building the squashfs-tools edit the Makefile to enable LZ4
>> support (by default it is disabled).
>>
>> LZ4 compression can be specified by using the -comp option, e.g.
>> % mksquashfs xxx img.sqsh -comp lz4
>>
>> The use of LZ4 high compression can be specified using -Xhc, e.g.
>>
>> % mksquashfs xxx img.sqsh -comp lz4 -Xhc
>>
>> Phillip
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>> Please read the FAQ at  http://www.tux.org/lkml/
>>
>
>

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Gu Zheng]

On 07/22/2013 01:07 PM, Phillip Lougher wrote:

> On 22 July 2013 04:05, Gu Zheng <guz.fnst@cn.fujitsu.com> wrote:
>> Hi Phillip,
>>         Have some tests been carried out to confirm that Squashfs really
>> can get benefit from LZ4 compression, comparing with lzo?
> 
> This seems to be a loaded question, in that it seems to be trying to
> reopen the "why add lz4 when we already have lzo" debate all over
> again.  As LZ4 has been merged to mainline, this appears to be a
> question that has already been answered.

No, they are different. LZ4 can be merged to mainline, because we can
see the benefit(faster compressing speed under the enabled unaligned 
memory access) it brings to us comparing with lzo. 
But it's hard to say that it also really can bring benefit to Squashfs.

> 
> As far as Squashfs is concerned,  I believe it is important to give
> people the choice of using LZ4 to compress Squashfs filesystems now
> its been mainlined.  As far as expected benefits are concerned,
> Squashfs' use in embedded systems is very similar to compressing
> kernels and initramfs data, in that it tends to be used to compress
> root filesystems.  As such the benefits of using LZ4 in Squashfs
> should be broadly similar to using LZ4 to compress kernels and
> initramfs data.  Ultimately it is up to people to experiment and
> choose whatever compression is best for their systems.

Yes, but the premise is that using LZ4 in Squashfs is better than lzo
in at least one actual situation.

> 
> People are welcome to try the patches out and report their findings.

Let's look and see other guys' feedbacks, they are more persuasive.:)

Regards,
Gu

> 
> Phillip
> 
>>
>> Thanks,
>> Gu
>>
>> On 07/22/2013 10:21 AM, Phillip Lougher wrote:
>>
>>> Hi
>>>
>>> Now that LZ4 compression support is in 3.11-rc1, I have written the
>>> following two patches for Squashfs to use it.
>>>
>>> Phillip Lougher (2):
>>>   Squashfs: add LZ4 compression support
>>>   Squashfs: Add LZ4 compression configuration option
>>>
>>>  Documentation/filesystems/squashfs.txt |    8 +-
>>>  fs/squashfs/Kconfig                    |   15 +++
>>>  fs/squashfs/Makefile                   |    1 +
>>>  fs/squashfs/decompressor.c             |    7 ++
>>>  fs/squashfs/decompressor.h             |    4 +
>>>  fs/squashfs/lz4_wrapper.c              |  163 ++++++++++++++++++++++++++++++++
>>>  fs/squashfs/squashfs_fs.h              |    1 +
>>>  7 files changed, 195 insertions(+), 4 deletions(-)
>>>  create mode 100644 fs/squashfs/lz4_wrapper.c
>>>
>>> These patches are also available in the git tree here:
>>>
>>> browse: https://git.kernel.org/cgit/linux/kernel/git/pkl/squashfs-lz4.git
>>> git clone: git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-lz4.git
>>>
>>> LZ4 support has (obviously) also been added to the squashfs-tools
>>> (Mksquashfs and Unsquashfs).  This is available from the Squashfs-tools
>>> git repository here:
>>>
>>> browse: https://git.kernel.org/cgit/fs/squashfs/squashfs-tools.git
>>> git clone: git://git.kernel.org/pub/scm/fs/squashfs/squashfs-tools.git
>>>
>>> When building the squashfs-tools edit the Makefile to enable LZ4
>>> support (by default it is disabled).
>>>
>>> LZ4 compression can be specified by using the -comp option, e.g.
>>> % mksquashfs xxx img.sqsh -comp lz4
>>>
>>> The use of LZ4 high compression can be specified using -Xhc, e.g.
>>>
>>> % mksquashfs xxx img.sqsh -comp lz4 -Xhc
>>>
>>> Phillip
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>>> the body of a message to majordomo@vger.kernel.org
>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>> Please read the FAQ at  http://www.tux.org/lkml/
>>>
>>
>>
> 

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Phillip Lougher]

On 22 July 2013 07:04, Gu Zheng <guz.fnst@cn.fujitsu.com> wrote:
> On 07/22/2013 01:07 PM, Phillip Lougher wrote:
>
>> On 22 July 2013 04:05, Gu Zheng <guz.fnst@cn.fujitsu.com> wrote:
>>> Hi Phillip,
>>>         Have some tests been carried out to confirm that Squashfs really
>>> can get benefit from LZ4 compression, comparing with lzo?
>>
>> This seems to be a loaded question, in that it seems to be trying to
>> reopen the "why add lz4 when we already have lzo" debate all over
>> again.  As LZ4 has been merged to mainline, this appears to be a
>> question that has already been answered.
>
> No, they are different. LZ4 can be merged to mainline, because we can
> see the benefit(faster compressing speed under the enabled unaligned
> memory access) it brings to us comparing with lzo.

The users of  LZ4 that has been merged is for decompression of kernel
and initramfs, not compression.  So far I don't see any users of the
LZ4 compression code in the kernel.  So you're saying it was merged
for fast compression despite there being no users of it?

> But it's hard to say that it also really can bring benefit to Squashfs.

Prejudiced statement without any evidence, if that's your contention prove it.

>
>>
>> As far as Squashfs is concerned,  I believe it is important to give
>> people the choice of using LZ4 to compress Squashfs filesystems now
>> its been mainlined.  As far as expected benefits are concerned,
>> Squashfs' use in embedded systems is very similar to compressing
>> kernels and initramfs data, in that it tends to be used to compress
>> root filesystems.  As such the benefits of using LZ4 in Squashfs
>> should be broadly similar to using LZ4 to compress kernels and
>> initramfs data.  Ultimately it is up to people to experiment and
>> choose whatever compression is best for their systems.
>
> Yes, but the premise is that using LZ4 in Squashfs is better than lzo
> in at least one actual situation.
>

No.  All it takes is for LZ4 to be better/faster at decompressing the
kernel or initramfs to make it advantageous to use LZ4 for Squashfs,
thus avoiding the overhead of another decompressor in the kernel..

>>
>> People are welcome to try the patches out and report their findings.
>
> Let's look and see other guys' feedbacks, they are more persuasive.:)

Suit yourself, maybe when you start using evidence rather than
unsubstantiated assertions, you'll be more persuasive too.  So far all
I see is wind.

Phillip

>
> Regards,
> Gu
>
>>
>> Phillip
>>
>>>
>>> Thanks,
>>> Gu
>>>
>>> On 07/22/2013 10:21 AM, Phillip Lougher wrote:
>>>
>>>> Hi
>>>>
>>>> Now that LZ4 compression support is in 3.11-rc1, I have written the
>>>> following two patches for Squashfs to use it.
>>>>
>>>> Phillip Lougher (2):
>>>>   Squashfs: add LZ4 compression support
>>>>   Squashfs: Add LZ4 compression configuration option
>>>>
>>>>  Documentation/filesystems/squashfs.txt |    8 +-
>>>>  fs/squashfs/Kconfig                    |   15 +++
>>>>  fs/squashfs/Makefile                   |    1 +
>>>>  fs/squashfs/decompressor.c             |    7 ++
>>>>  fs/squashfs/decompressor.h             |    4 +
>>>>  fs/squashfs/lz4_wrapper.c              |  163 ++++++++++++++++++++++++++++++++
>>>>  fs/squashfs/squashfs_fs.h              |    1 +
>>>>  7 files changed, 195 insertions(+), 4 deletions(-)
>>>>  create mode 100644 fs/squashfs/lz4_wrapper.c
>>>>
>>>> These patches are also available in the git tree here:
>>>>
>>>> browse: https://git.kernel.org/cgit/linux/kernel/git/pkl/squashfs-lz4.git
>>>> git clone: git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-lz4.git
>>>>
>>>> LZ4 support has (obviously) also been added to the squashfs-tools
>>>> (Mksquashfs and Unsquashfs).  This is available from the Squashfs-tools
>>>> git repository here:
>>>>
>>>> browse: https://git.kernel.org/cgit/fs/squashfs/squashfs-tools.git
>>>> git clone: git://git.kernel.org/pub/scm/fs/squashfs/squashfs-tools.git
>>>>
>>>> When building the squashfs-tools edit the Makefile to enable LZ4
>>>> support (by default it is disabled).
>>>>
>>>> LZ4 compression can be specified by using the -comp option, e.g.
>>>> % mksquashfs xxx img.sqsh -comp lz4
>>>>
>>>> The use of LZ4 high compression can be specified using -Xhc, e.g.
>>>>
>>>> % mksquashfs xxx img.sqsh -comp lz4 -Xhc
>>>>
>>>> Phillip
>>>> --
>>>> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>>>> the body of a message to majordomo@vger.kernel.org
>>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>> Please read the FAQ at  http://www.tux.org/lkml/
>>>>
>>>
>>>
>>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Rob Landley]

On 07/22/2013 01:04:59 AM, Gu Zheng wrote:
> On 07/22/2013 01:07 PM, Phillip Lougher wrote:
> 
> > On 22 July 2013 04:05, Gu Zheng <guz.fnst@cn.fujitsu.com> wrote:
> >> Hi Phillip,
> >>         Have some tests been carried out to confirm that Squashfs  
> really
> >> can get benefit from LZ4 compression, comparing with lzo?
> >
> > This seems to be a loaded question, in that it seems to be trying to
> > reopen the "why add lz4 when we already have lzo" debate all over
> > again.  As LZ4 has been merged to mainline, this appears to be a
> > question that has already been answered.
> 
> No, they are different. LZ4 can be merged to mainline, because we can
> see the benefit(faster compressing speed under the enabled unaligned
> memory access) it brings to us comparing with lzo.
> But it's hard to say that it also really can bring benefit to  
> Squashfs.

A compression format was added to the kernel. Philip hooked up the code  
that was already in the kernel to a filesystem that was already in the  
kernel.

You consider this action controversial...

Rob

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Guan, Xin]

Gu Zheng <guz.fnst <at> cn.fujitsu.com> writes:

> But it's hard to say that it also really can bring benefit to Squashfs.

I did observe faster I/O in my daily use. The start up time of
applications on my squashfs+aufs root filesystem generally decreased by
10% to 20% compared to lzo. Not very much, but the absence of lag
with lz4 greatly improved user experience.


> Yes, but the premise is that using LZ4 in Squashfs is better than lzo
> in at least one actual situation.

I update my squashfs image (about 8GB compressed) ~ once every month.
Lz4 yields slightly worse compression ratio (+20%) than lzo but its
compression speed is about fifty (no typo) times faster.
In order to get rid of the endless waiting on "mksquashfs -comp lzo",
I had to port Phillip's patch (in particular the "lz4_wrapper.c") to
Linux 3.14.x so I can use lz4 compressed squashfs images everyday.


> Let's look and see other guys' feedbacks, they are more persuasive.:)

IMHO, a 50x boost of compression speed is persuasive.
Lz4 also helped me to get rid of occasional lags on file I/O,
greatly improved user experience with my Linux desktop.

Adding lz4 support to squashfs isn't complicated. Everyone can
review the patch in a couple of minutes.
So I vote on its inclusion in mainline.

Guan

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Guan, Xin]

Gu Zheng <guz.fnst <at> cn.fujitsu.com> writes:

> On 07/22/2013 01:07 PM, Phillip Lougher wrote:
> 
> But it's hard to say that it also really can bring benefit to Squashfs.

In addition to my previous message:

Tried HC mode of lz4 -- Compression ratio of my rootfs (the whole system) is
3~4% better than lzo; compression speed is 5x of lzo's.
Decompression, as already mentioned, was markedly faster than lzo.

Lz4 is a clear win over lzo from every aspect.

Guan

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Phillip Lougher]

On 23 July 2013 16:10, Bruno Wolff III <bruno@wolff.to> wrote:
> On Mon, Jul 22, 2013 at 03:21:01 +0100,
>   Phillip Lougher <phillip@squashfs.org.uk> wrote:
>>
>> Hi
>>
>> Now that LZ4 compression support is in 3.11-rc1, I have written the
>> following two patches for Squashfs to use it.
>
>
> If this gets accepted are you going to move the LZ4 changes to
> squashfs-tools  into the stable branch? I'd like to make LZ4 work in
> mksquashfs in at least rawhide once LZ4 is usable for squashfs file systems
> in the kernel. So I was wondering whether to expect this to appear in stable
> shortly or to expect to start doing builds using the master branch.

Hmm, good point.... I was intending to merge the LZ4 work onto stable
yesterday or today, but, I've received more resistance to the idea of
LZ4 in Squashfs than I expected....  I'm not sure that merging LZ4
onto stable now gives the right message, because if it does not go
into mainline, it will effectively be experimental or abandoned.

In otherwords I don't think it's wise yet to merge LZ4 onto stable,
not until at least there's some positive feedback on the mailing list.

Thoughts?  Maybe some positive feedback? :-)

A V2 of the patches will be coming ASAP hopefully with some performance stats.


Thanks

Phillip

Re: [PATCH 0/2] Squashfs: add LZ4 compression

[Bruno Wolff III]

On Tue, Jul 23, 2013 at 17:17:30 +0100,
   Phillip Lougher <phillip.lougher@gmail.com> wrote:
>
>In otherwords I don't think it's wise yet to merge LZ4 onto stable,
>not until at least there's some positive feedback on the mailing list.
>
>Thoughts?  Maybe some positive feedback? :-)

I think it makes sense to tie inclusion in the kernel with inclusion 
into the stable branch. If this gets into the 3.11 kernel, I'm going 
to want to get it into Fedora releases that use that kernel. This 
will initially be Fedora 20 (which will branch from rawhide in a couple 
of weeks) and will eventually be in Fedora 19 and probably Fedora 18. 
So I'd like to see it in stable, since I would feel less comfortable 
using the master branch.

If LZ4 support isn't in the kernel I think that there isn't a lot of 
need for support in Fedora since I don't think many people use it 
for archiving. It mostly gets used for live images. I'll make sure 
that LZ4 can be used for them, but I suspect most people will still 
prefer xz for them. But probably at least some people would prefer 
faster decompression rather than minimal size. As flash drives get 
bigger (for the same cost) more people will probably prefer speed over 
size.