Message-ID: <990d1aa9-86c4-4a7d-e8aa-f04006ec73bc@gmail.com>
Date:   Fri, 11 Aug 2017 07:45:24 -0400
From:   "Austin S. Hemmelgarn" <ahferroin7@...il.com>
To:     Nick Terrell <terrelln@...com>
Cc:     kernel-team@...com, Chris Mason <clm@...com>,
        Yann Collet <cyan@...com>, Adam Borowski <kilobyte@...band.pl>,
        David Sterba <dsterba@...e.cz>, linux-btrfs@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v5 3/5] btrfs: Add zstd support

On 2017-08-09 22:39, Nick Terrell wrote:
> Add zstd compression and decompression support to BtrFS. zstd at its
> fastest level compresses almost as well as zlib, while offering much
> faster compression and decompression, approaching lzo speeds.
> 
> I benchmarked btrfs with zstd compression against no compression, lzo
> compression, and zlib compression. I benchmarked two scenarios. Copying
> a set of files to btrfs, and then reading the files. Copying a tarball
> to btrfs, extracting it to btrfs, and then reading the extracted files.
> After every operation, I call `sync` and include the sync time.
> Between every pair of operations I unmount and remount the filesystem
> to avoid caching. The benchmark files can be found in the upstream
> zstd source repository under
> `contrib/linux-kernel/{btrfs-benchmark.sh,btrfs-extract-benchmark.sh}`
> [1] [2].
> 
> I ran the benchmarks on a Ubuntu 14.04 VM with 2 cores and 4 GiB of RAM.
> The VM is running on a MacBook Pro with a 3.1 GHz Intel Core i7 processor,
> 16 GB of RAM, and a SSD.
> 
> The first compression benchmark is copying 10 copies of the unzipped
> Silesia corpus [3] into a BtrFS filesystem mounted with
> `-o compress-force=Method`. The decompression benchmark times how long
> it takes to `tar` all 10 copies into `/dev/null`. The compression ratio is
> measured by comparing the output of `df` and `du`. See the benchmark file
> [1] for details. I benchmarked multiple zstd compression levels, although
> the patch uses zstd level 1.
> 
> | Method  | Ratio | Compression MB/s | Decompression speed |
> |---------|-------|------------------|---------------------|
> | None    |  0.99 |              504 |                 686 |
> | lzo     |  1.66 |              398 |                 442 |
> | zlib    |  2.58 |               65 |                 241 |
> | zstd 1  |  2.57 |              260 |                 383 |
> | zstd 3  |  2.71 |              174 |                 408 |
> | zstd 6  |  2.87 |               70 |                 398 |
> | zstd 9  |  2.92 |               43 |                 406 |
> | zstd 12 |  2.93 |               21 |                 408 |
> | zstd 15 |  3.01 |               11 |                 354 |
> 
> The next benchmark first copies `linux-4.11.6.tar` [4] to btrfs. Then it
> measures the compression ratio, extracts the tar, and deletes the tar.
> Then it measures the compression ratio again, and `tar`s the extracted
> files into `/dev/null`. See the benchmark file [2] for details.
> 
> | Method | Tar Ratio | Extract Ratio | Copy (s) | Extract (s)| Read (s) |
> |--------|-----------|---------------|----------|------------|----------|
> | None   |      0.97 |          0.78 |    0.981 |      5.501 |    8.807 |
> | lzo    |      2.06 |          1.38 |    1.631 |      8.458 |    8.585 |
> | zlib   |      3.40 |          1.86 |    7.750 |     21.544 |   11.744 |
> | zstd 1 |      3.57 |          1.85 |    2.579 |     11.479 |    9.389 |
> 
> [1] https://github.com/facebook/zstd/blob/dev/contrib/linux-kernel/btrfs-benchmark.sh
> [2] https://github.com/facebook/zstd/blob/dev/contrib/linux-kernel/btrfs-extract-benchmark.sh
> [3] http://sun.aei.polsl.pl/~sdeor/index.php?page=silesia
> [4] https://cdn.kernel.org/pub/linux/kernel/v4.x/linux-4.11.6.tar.xz
> 
> zstd source repository: https://github.com/facebook/zstd
> 
> Signed-off-by: Nick Terrell <terrelln@...com>
Considering how things went with the previous patch, I've been a bit 
more aggressive testing this one, but after now almost 72 hours of 
combined runtime for each of the architectures I have tests set up for 
with nothing breaking (well, nothing breaking that wasn't already 
breaking before this patch set, some of the raid56 tests are still 
failing semi-reliably as expected), I'd say this is reasonably stable 
and looks good overall.
> ---
> v2 -> v3:
> - Port upstream BtrFS commits e1ddce71d6, 389a6cfc2a, and 6acafd1eff
> - Change default compression level for BtrFS to 3
> 
> v3 -> v4:
> - Add missing includes, which fixes the aarch64 build
> - Fix minor linter warnings
> 
>   fs/btrfs/Kconfig           |   2 +
>   fs/btrfs/Makefile          |   2 +-
>   fs/btrfs/compression.c     |   1 +
>   fs/btrfs/compression.h     |   6 +-
>   fs/btrfs/ctree.h           |   1 +
>   fs/btrfs/disk-io.c         |   2 +
>   fs/btrfs/ioctl.c           |   6 +-
>   fs/btrfs/props.c           |   6 +
>   fs/btrfs/super.c           |  12 +-
>   fs/btrfs/sysfs.c           |   2 +
>   fs/btrfs/zstd.c            | 432 +++++++++++++++++++++++++++++++++++++++++++++
>   include/uapi/linux/btrfs.h |   8 +-
>   12 files changed, 468 insertions(+), 12 deletions(-)
>   create mode 100644 fs/btrfs/zstd.c
> 
> diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig
> index 80e9c18..a26c63b 100644
> --- a/fs/btrfs/Kconfig
> +++ b/fs/btrfs/Kconfig
> @@ -6,6 +6,8 @@ config BTRFS_FS
>   	select ZLIB_DEFLATE
>   	select LZO_COMPRESS
>   	select LZO_DECOMPRESS
> +	select ZSTD_COMPRESS
> +	select ZSTD_DECOMPRESS
>   	select RAID6_PQ
>   	select XOR_BLOCKS
>   	select SRCU
> diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile
> index 128ce17..962a95a 100644
> --- a/fs/btrfs/Makefile
> +++ b/fs/btrfs/Makefile
> @@ -6,7 +6,7 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
>   	   transaction.o inode.o file.o tree-defrag.o \
>   	   extent_map.o sysfs.o struct-funcs.o xattr.o ordered-data.o \
>   	   extent_io.o volumes.o async-thread.o ioctl.o locking.o orphan.o \
> -	   export.o tree-log.o free-space-cache.o zlib.o lzo.o \
> +	   export.o tree-log.o free-space-cache.o zlib.o lzo.o zstd.o \
>   	   compression.o delayed-ref.o relocation.o delayed-inode.o scrub.o \
>   	   reada.o backref.o ulist.o qgroup.o send.o dev-replace.o raid56.o \
>   	   uuid-tree.o props.o hash.o free-space-tree.o
> diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
> index d2ef9ac..4ff42d1 100644
> --- a/fs/btrfs/compression.c
> +++ b/fs/btrfs/compression.c
> @@ -704,6 +704,7 @@ static struct {
>   static const struct btrfs_compress_op * const btrfs_compress_op[] = {
>   	&btrfs_zlib_compress,
>   	&btrfs_lzo_compress,
> +	&btrfs_zstd_compress,
>   };
> 
>   void __init btrfs_init_compress(void)
> diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
> index 87f6d33..2269e00 100644
> --- a/fs/btrfs/compression.h
> +++ b/fs/btrfs/compression.h
> @@ -99,8 +99,9 @@ enum btrfs_compression_type {
>   	BTRFS_COMPRESS_NONE  = 0,
>   	BTRFS_COMPRESS_ZLIB  = 1,
>   	BTRFS_COMPRESS_LZO   = 2,
> -	BTRFS_COMPRESS_TYPES = 2,
> -	BTRFS_COMPRESS_LAST  = 3,
> +	BTRFS_COMPRESS_ZSTD  = 3,
> +	BTRFS_COMPRESS_TYPES = 3,
> +	BTRFS_COMPRESS_LAST  = 4,
>   };
> 
>   struct btrfs_compress_op {
> @@ -128,5 +129,6 @@ struct btrfs_compress_op {
> 
>   extern const struct btrfs_compress_op btrfs_zlib_compress;
>   extern const struct btrfs_compress_op btrfs_lzo_compress;
> +extern const struct btrfs_compress_op btrfs_zstd_compress;
> 
>   #endif
> diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
> index 3f3eb7b..845d77c 100644
> --- a/fs/btrfs/ctree.h
> +++ b/fs/btrfs/ctree.h
> @@ -270,6 +270,7 @@ struct btrfs_super_block {
>   	 BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS |		\
>   	 BTRFS_FEATURE_INCOMPAT_BIG_METADATA |		\
>   	 BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO |		\
> +	 BTRFS_FEATURE_INCOMPAT_COMPRESS_ZSTD |		\
>   	 BTRFS_FEATURE_INCOMPAT_RAID56 |		\
>   	 BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF |		\
>   	 BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA |	\
> diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
> index 080e2eb..04632f4 100644
> --- a/fs/btrfs/disk-io.c
> +++ b/fs/btrfs/disk-io.c
> @@ -2828,6 +2828,8 @@ int open_ctree(struct super_block *sb,
>   	features |= BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF;
>   	if (fs_info->compress_type == BTRFS_COMPRESS_LZO)
>   		features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO;
> +	else if (fs_info->compress_type == BTRFS_COMPRESS_ZSTD)
> +		features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_ZSTD;
> 
>   	if (features & BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA)
>   		btrfs_info(fs_info, "has skinny extents");
> diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
> index fa1b78c..b9963d9 100644
> --- a/fs/btrfs/ioctl.c
> +++ b/fs/btrfs/ioctl.c
> @@ -327,8 +327,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
> 
>   		if (fs_info->compress_type == BTRFS_COMPRESS_LZO)
>   			comp = "lzo";
> -		else
> +		else if (fs_info->compress_type == BTRFS_COMPRESS_ZLIB)
>   			comp = "zlib";
> +		else
> +			comp = "zstd";
>   		ret = btrfs_set_prop(inode, "btrfs.compression",
>   				     comp, strlen(comp), 0);
>   		if (ret)
> @@ -1466,6 +1468,8 @@ int btrfs_defrag_file(struct inode *inode, struct file *file,
> 
>   	if (range->compress_type == BTRFS_COMPRESS_LZO) {
>   		btrfs_set_fs_incompat(fs_info, COMPRESS_LZO);
> +	} else if (range->compress_type == BTRFS_COMPRESS_ZSTD) {
> +		btrfs_set_fs_incompat(fs_info, COMPRESS_ZSTD);
>   	}
> 
>   	ret = defrag_count;
> diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c
> index 4b23ae5..20631e9 100644
> --- a/fs/btrfs/props.c
> +++ b/fs/btrfs/props.c
> @@ -390,6 +390,8 @@ static int prop_compression_validate(const char *value, size_t len)
>   		return 0;
>   	else if (!strncmp("zlib", value, len))
>   		return 0;
> +	else if (!strncmp("zstd", value, len))
> +		return 0;
> 
>   	return -EINVAL;
>   }
> @@ -412,6 +414,8 @@ static int prop_compression_apply(struct inode *inode,
>   		type = BTRFS_COMPRESS_LZO;
>   	else if (!strncmp("zlib", value, len))
>   		type = BTRFS_COMPRESS_ZLIB;
> +	else if (!strncmp("zstd", value, len))
> +		type = BTRFS_COMPRESS_ZSTD;
>   	else
>   		return -EINVAL;
> 
> @@ -429,6 +433,8 @@ static const char *prop_compression_extract(struct inode *inode)
>   		return "zlib";
>   	case BTRFS_COMPRESS_LZO:
>   		return "lzo";
> +	case BTRFS_COMPRESS_ZSTD:
> +		return "zstd";
>   	}
> 
>   	return NULL;
> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
> index 12540b6..c370dea 100644
> --- a/fs/btrfs/super.c
> +++ b/fs/btrfs/super.c
> @@ -513,6 +513,14 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
>   				btrfs_clear_opt(info->mount_opt, NODATASUM);
>   				btrfs_set_fs_incompat(info, COMPRESS_LZO);
>   				no_compress = 0;
> +			} else if (strcmp(args[0].from, "zstd") == 0) {
> +				compress_type = "zstd";
> +				info->compress_type = BTRFS_COMPRESS_ZSTD;
> +				btrfs_set_opt(info->mount_opt, COMPRESS);
> +				btrfs_clear_opt(info->mount_opt, NODATACOW);
> +				btrfs_clear_opt(info->mount_opt, NODATASUM);
> +				btrfs_set_fs_incompat(info, COMPRESS_ZSTD);
> +				no_compress = 0;
>   			} else if (strncmp(args[0].from, "no", 2) == 0) {
>   				compress_type = "no";
>   				btrfs_clear_opt(info->mount_opt, COMPRESS);
> @@ -1227,8 +1235,10 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
>   	if (btrfs_test_opt(info, COMPRESS)) {
>   		if (info->compress_type == BTRFS_COMPRESS_ZLIB)
>   			compress_type = "zlib";
> -		else
> +		else if (info->compress_type == BTRFS_COMPRESS_LZO)
>   			compress_type = "lzo";
> +		else
> +			compress_type = "zstd";
>   		if (btrfs_test_opt(info, FORCE_COMPRESS))
>   			seq_printf(seq, ",compress-force=%s", compress_type);
>   		else
> diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
> index c2d5f35..2b6d37c 100644
> --- a/fs/btrfs/sysfs.c
> +++ b/fs/btrfs/sysfs.c
> @@ -200,6 +200,7 @@ BTRFS_FEAT_ATTR_INCOMPAT(mixed_backref, MIXED_BACKREF);
>   BTRFS_FEAT_ATTR_INCOMPAT(default_subvol, DEFAULT_SUBVOL);
>   BTRFS_FEAT_ATTR_INCOMPAT(mixed_groups, MIXED_GROUPS);
>   BTRFS_FEAT_ATTR_INCOMPAT(compress_lzo, COMPRESS_LZO);
> +BTRFS_FEAT_ATTR_INCOMPAT(compress_zstd, COMPRESS_ZSTD);
>   BTRFS_FEAT_ATTR_INCOMPAT(big_metadata, BIG_METADATA);
>   BTRFS_FEAT_ATTR_INCOMPAT(extended_iref, EXTENDED_IREF);
>   BTRFS_FEAT_ATTR_INCOMPAT(raid56, RAID56);
> @@ -212,6 +213,7 @@ static struct attribute *btrfs_supported_feature_attrs[] = {
>   	BTRFS_FEAT_ATTR_PTR(default_subvol),
>   	BTRFS_FEAT_ATTR_PTR(mixed_groups),
>   	BTRFS_FEAT_ATTR_PTR(compress_lzo),
> +	BTRFS_FEAT_ATTR_PTR(compress_zstd),
>   	BTRFS_FEAT_ATTR_PTR(big_metadata),
>   	BTRFS_FEAT_ATTR_PTR(extended_iref),
>   	BTRFS_FEAT_ATTR_PTR(raid56),
> diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c
> new file mode 100644
> index 0000000..607ce47
> --- /dev/null
> +++ b/fs/btrfs/zstd.c
> @@ -0,0 +1,432 @@
> +/*
> + * Copyright (c) 2016-present, Facebook, Inc.
> + * All rights reserved.
> + *
> + * This program is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU General Public
> + * License v2 as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> + * General Public License for more details.
> + */
> +#include <linux/bio.h>
> +#include <linux/err.h>
> +#include <linux/init.h>
> +#include <linux/kernel.h>
> +#include <linux/mm.h>
> +#include <linux/pagemap.h>
> +#include <linux/refcount.h>
> +#include <linux/sched.h>
> +#include <linux/slab.h>
> +#include <linux/zstd.h>
> +#include "compression.h"
> +
> +#define ZSTD_BTRFS_MAX_WINDOWLOG 17
> +#define ZSTD_BTRFS_MAX_INPUT (1 << ZSTD_BTRFS_MAX_WINDOWLOG)
> +#define ZSTD_BTRFS_DEFAULT_LEVEL 3
> +
> +static ZSTD_parameters zstd_get_btrfs_parameters(size_t src_len)
> +{
> +	ZSTD_parameters params = ZSTD_getParams(ZSTD_BTRFS_DEFAULT_LEVEL,
> +						src_len, 0);
> +
> +	if (params.cParams.windowLog > ZSTD_BTRFS_MAX_WINDOWLOG)
> +		params.cParams.windowLog = ZSTD_BTRFS_MAX_WINDOWLOG;
> +	WARN_ON(src_len > ZSTD_BTRFS_MAX_INPUT);
> +	return params;
> +}
> +
> +struct workspace {
> +	void *mem;
> +	size_t size;
> +	char *buf;
> +	struct list_head list;
> +};
> +
> +static void zstd_free_workspace(struct list_head *ws)
> +{
> +	struct workspace *workspace = list_entry(ws, struct workspace, list);
> +
> +	kvfree(workspace->mem);
> +	kfree(workspace->buf);
> +	kfree(workspace);
> +}
> +
> +static struct list_head *zstd_alloc_workspace(void)
> +{
> +	ZSTD_parameters params =
> +			zstd_get_btrfs_parameters(ZSTD_BTRFS_MAX_INPUT);
> +	struct workspace *workspace;
> +
> +	workspace = kzalloc(sizeof(*workspace), GFP_KERNEL);
> +	if (!workspace)
> +		return ERR_PTR(-ENOMEM);
> +
> +	workspace->size = max_t(size_t,
> +			ZSTD_CStreamWorkspaceBound(params.cParams),
> +			ZSTD_DStreamWorkspaceBound(ZSTD_BTRFS_MAX_INPUT));
> +	workspace->mem = kvmalloc(workspace->size, GFP_KERNEL);
> +	workspace->buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
> +	if (!workspace->mem || !workspace->buf)
> +		goto fail;
> +
> +	INIT_LIST_HEAD(&workspace->list);
> +
> +	return &workspace->list;
> +fail:
> +	zstd_free_workspace(&workspace->list);
> +	return ERR_PTR(-ENOMEM);
> +}
> +
> +static int zstd_compress_pages(struct list_head *ws,
> +		struct address_space *mapping,
> +		u64 start,
> +		struct page **pages,
> +		unsigned long *out_pages,
> +		unsigned long *total_in,
> +		unsigned long *total_out)
> +{
> +	struct workspace *workspace = list_entry(ws, struct workspace, list);
> +	ZSTD_CStream *stream;
> +	int ret = 0;
> +	int nr_pages = 0;
> +	struct page *in_page = NULL;  /* The current page to read */
> +	struct page *out_page = NULL; /* The current page to write to */
> +	ZSTD_inBuffer in_buf = { NULL, 0, 0 };
> +	ZSTD_outBuffer out_buf = { NULL, 0, 0 };
> +	unsigned long tot_in = 0;
> +	unsigned long tot_out = 0;
> +	unsigned long len = *total_out;
> +	const unsigned long nr_dest_pages = *out_pages;
> +	unsigned long max_out = nr_dest_pages * PAGE_SIZE;
> +	ZSTD_parameters params = zstd_get_btrfs_parameters(len);
> +
> +	*out_pages = 0;
> +	*total_out = 0;
> +	*total_in = 0;
> +
> +	/* Initialize the stream */
> +	stream = ZSTD_initCStream(params, len, workspace->mem,
> +			workspace->size);
> +	if (!stream) {
> +		pr_warn("BTRFS: ZSTD_initCStream failed\n");
> +		ret = -EIO;
> +		goto out;
> +	}
> +
> +	/* map in the first page of input data */
> +	in_page = find_get_page(mapping, start >> PAGE_SHIFT);
> +	in_buf.src = kmap(in_page);
> +	in_buf.pos = 0;
> +	in_buf.size = min_t(size_t, len, PAGE_SIZE);
> +
> +
> +	/* Allocate and map in the output buffer */
> +	out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
> +	if (out_page == NULL) {
> +		ret = -ENOMEM;
> +		goto out;
> +	}
> +	pages[nr_pages++] = out_page;
> +	out_buf.dst = kmap(out_page);
> +	out_buf.pos = 0;
> +	out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
> +
> +	while (1) {
> +		size_t ret2;
> +
> +		ret2 = ZSTD_compressStream(stream, &out_buf, &in_buf);
> +		if (ZSTD_isError(ret2)) {
> +			pr_debug("BTRFS: ZSTD_compressStream returned %d\n",
> +					ZSTD_getErrorCode(ret2));
> +			ret = -EIO;
> +			goto out;
> +		}
> +
> +		/* Check to see if we are making it bigger */
> +		if (tot_in + in_buf.pos > 8192 &&
> +				tot_in + in_buf.pos <
> +				tot_out + out_buf.pos) {
> +			ret = -E2BIG;
> +			goto out;
> +		}
> +
> +		/* We've reached the end of our output range */
> +		if (out_buf.pos >= max_out) {
> +			tot_out += out_buf.pos;
> +			ret = -E2BIG;
> +			goto out;
> +		}
> +
> +		/* Check if we need more output space */
> +		if (out_buf.pos == out_buf.size) {
> +			tot_out += PAGE_SIZE;
> +			max_out -= PAGE_SIZE;
> +			kunmap(out_page);
> +			if (nr_pages == nr_dest_pages) {
> +				out_page = NULL;
> +				ret = -E2BIG;
> +				goto out;
> +			}
> +			out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
> +			if (out_page == NULL) {
> +				ret = -ENOMEM;
> +				goto out;
> +			}
> +			pages[nr_pages++] = out_page;
> +			out_buf.dst = kmap(out_page);
> +			out_buf.pos = 0;
> +			out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
> +		}
> +
> +		/* We've reached the end of the input */
> +		if (in_buf.pos >= len) {
> +			tot_in += in_buf.pos;
> +			break;
> +		}
> +
> +		/* Check if we need more input */
> +		if (in_buf.pos == in_buf.size) {
> +			tot_in += PAGE_SIZE;
> +			kunmap(in_page);
> +			put_page(in_page);
> +
> +			start += PAGE_SIZE;
> +			len -= PAGE_SIZE;
> +			in_page = find_get_page(mapping, start >> PAGE_SHIFT);
> +			in_buf.src = kmap(in_page);
> +			in_buf.pos = 0;
> +			in_buf.size = min_t(size_t, len, PAGE_SIZE);
> +		}
> +	}
> +	while (1) {
> +		size_t ret2;
> +
> +		ret2 = ZSTD_endStream(stream, &out_buf);
> +		if (ZSTD_isError(ret2)) {
> +			pr_debug("BTRFS: ZSTD_endStream returned %d\n",
> +					ZSTD_getErrorCode(ret2));
> +			ret = -EIO;
> +			goto out;
> +		}
> +		if (ret2 == 0) {
> +			tot_out += out_buf.pos;
> +			break;
> +		}
> +		if (out_buf.pos >= max_out) {
> +			tot_out += out_buf.pos;
> +			ret = -E2BIG;
> +			goto out;
> +		}
> +
> +		tot_out += PAGE_SIZE;
> +		max_out -= PAGE_SIZE;
> +		kunmap(out_page);
> +		if (nr_pages == nr_dest_pages) {
> +			out_page = NULL;
> +			ret = -E2BIG;
> +			goto out;
> +		}
> +		out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
> +		if (out_page == NULL) {
> +			ret = -ENOMEM;
> +			goto out;
> +		}
> +		pages[nr_pages++] = out_page;
> +		out_buf.dst = kmap(out_page);
> +		out_buf.pos = 0;
> +		out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
> +	}
> +
> +	if (tot_out >= tot_in) {
> +		ret = -E2BIG;
> +		goto out;
> +	}
> +
> +	ret = 0;
> +	*total_in = tot_in;
> +	*total_out = tot_out;
> +out:
> +	*out_pages = nr_pages;
> +	/* Cleanup */
> +	if (in_page) {
> +		kunmap(in_page);
> +		put_page(in_page);
> +	}
> +	if (out_page)
> +		kunmap(out_page);
> +	return ret;
> +}
> +
> +static int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
> +{
> +	struct workspace *workspace = list_entry(ws, struct workspace, list);
> +	struct page **pages_in = cb->compressed_pages;
> +	u64 disk_start = cb->start;
> +	struct bio *orig_bio = cb->orig_bio;
> +	size_t srclen = cb->compressed_len;
> +	ZSTD_DStream *stream;
> +	int ret = 0;
> +	unsigned long page_in_index = 0;
> +	unsigned long total_pages_in = DIV_ROUND_UP(srclen, PAGE_SIZE);
> +	unsigned long buf_start;
> +	unsigned long total_out = 0;
> +	ZSTD_inBuffer in_buf = { NULL, 0, 0 };
> +	ZSTD_outBuffer out_buf = { NULL, 0, 0 };
> +
> +	stream = ZSTD_initDStream(
> +			ZSTD_BTRFS_MAX_INPUT, workspace->mem, workspace->size);
> +	if (!stream) {
> +		pr_debug("BTRFS: ZSTD_initDStream failed\n");
> +		ret = -EIO;
> +		goto done;
> +	}
> +
> +	in_buf.src = kmap(pages_in[page_in_index]);
> +	in_buf.pos = 0;
> +	in_buf.size = min_t(size_t, srclen, PAGE_SIZE);
> +
> +	out_buf.dst = workspace->buf;
> +	out_buf.pos = 0;
> +	out_buf.size = PAGE_SIZE;
> +
> +	while (1) {
> +		size_t ret2;
> +
> +		ret2 = ZSTD_decompressStream(stream, &out_buf, &in_buf);
> +		if (ZSTD_isError(ret2)) {
> +			pr_debug("BTRFS: ZSTD_decompressStream returned %d\n",
> +					ZSTD_getErrorCode(ret2));
> +			ret = -EIO;
> +			goto done;
> +		}
> +		buf_start = total_out;
> +		total_out += out_buf.pos;
> +		out_buf.pos = 0;
> +
> +		ret = btrfs_decompress_buf2page(out_buf.dst, buf_start,
> +				total_out, disk_start, orig_bio);
> +		if (ret == 0)
> +			break;
> +
> +		if (in_buf.pos >= srclen)
> +			break;
> +
> +		/* Check if we've hit the end of a frame */
> +		if (ret2 == 0)
> +			break;
> +
> +		if (in_buf.pos == in_buf.size) {
> +			kunmap(pages_in[page_in_index++]);
> +			if (page_in_index >= total_pages_in) {
> +				in_buf.src = NULL;
> +				ret = -EIO;
> +				goto done;
> +			}
> +			srclen -= PAGE_SIZE;
> +			in_buf.src = kmap(pages_in[page_in_index]);
> +			in_buf.pos = 0;
> +			in_buf.size = min_t(size_t, srclen, PAGE_SIZE);
> +		}
> +	}
> +	ret = 0;
> +	zero_fill_bio(orig_bio);
> +done:
> +	if (in_buf.src)
> +		kunmap(pages_in[page_in_index]);
> +	return ret;
> +}
> +
> +static int zstd_decompress(struct list_head *ws, unsigned char *data_in,
> +		struct page *dest_page,
> +		unsigned long start_byte,
> +		size_t srclen, size_t destlen)
> +{
> +	struct workspace *workspace = list_entry(ws, struct workspace, list);
> +	ZSTD_DStream *stream;
> +	int ret = 0;
> +	size_t ret2;
> +	ZSTD_inBuffer in_buf = { NULL, 0, 0 };
> +	ZSTD_outBuffer out_buf = { NULL, 0, 0 };
> +	unsigned long total_out = 0;
> +	unsigned long pg_offset = 0;
> +	char *kaddr;
> +
> +	stream = ZSTD_initDStream(
> +			ZSTD_BTRFS_MAX_INPUT, workspace->mem, workspace->size);
> +	if (!stream) {
> +		pr_warn("BTRFS: ZSTD_initDStream failed\n");
> +		ret = -EIO;
> +		goto finish;
> +	}
> +
> +	destlen = min_t(size_t, destlen, PAGE_SIZE);
> +
> +	in_buf.src = data_in;
> +	in_buf.pos = 0;
> +	in_buf.size = srclen;
> +
> +	out_buf.dst = workspace->buf;
> +	out_buf.pos = 0;
> +	out_buf.size = PAGE_SIZE;
> +
> +	ret2 = 1;
> +	while (pg_offset < destlen && in_buf.pos < in_buf.size) {
> +		unsigned long buf_start;
> +		unsigned long buf_offset;
> +		unsigned long bytes;
> +
> +		/* Check if the frame is over and we still need more input */
> +		if (ret2 == 0) {
> +			pr_debug("BTRFS: ZSTD_decompressStream ended early\n");
> +			ret = -EIO;
> +			goto finish;
> +		}
> +		ret2 = ZSTD_decompressStream(stream, &out_buf, &in_buf);
> +		if (ZSTD_isError(ret2)) {
> +			pr_debug("BTRFS: ZSTD_decompressStream returned %d\n",
> +					ZSTD_getErrorCode(ret2));
> +			ret = -EIO;
> +			goto finish;
> +		}
> +
> +		buf_start = total_out;
> +		total_out += out_buf.pos;
> +		out_buf.pos = 0;
> +
> +		if (total_out <= start_byte)
> +			continue;
> +
> +		if (total_out > start_byte && buf_start < start_byte)
> +			buf_offset = start_byte - buf_start;
> +		else
> +			buf_offset = 0;
> +
> +		bytes = min_t(unsigned long, destlen - pg_offset,
> +				out_buf.size - buf_offset);
> +
> +		kaddr = kmap_atomic(dest_page);
> +		memcpy(kaddr + pg_offset, out_buf.dst + buf_offset, bytes);
> +		kunmap_atomic(kaddr);
> +
> +		pg_offset += bytes;
> +	}
> +	ret = 0;
> +finish:
> +	if (pg_offset < destlen) {
> +		kaddr = kmap_atomic(dest_page);
> +		memset(kaddr + pg_offset, 0, destlen - pg_offset);
> +		kunmap_atomic(kaddr);
> +	}
> +	return ret;
> +}
> +
> +const struct btrfs_compress_op btrfs_zstd_compress = {
> +	.alloc_workspace = zstd_alloc_workspace,
> +	.free_workspace = zstd_free_workspace,
> +	.compress_pages = zstd_compress_pages,
> +	.decompress_bio = zstd_decompress_bio,
> +	.decompress = zstd_decompress,
> +};
> diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
> index 9aa74f3..378230c 100644
> --- a/include/uapi/linux/btrfs.h
> +++ b/include/uapi/linux/btrfs.h
> @@ -255,13 +255,7 @@ struct btrfs_ioctl_fs_info_args {
>   #define BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL	(1ULL << 1)
>   #define BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS	(1ULL << 2)
>   #define BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO	(1ULL << 3)
> -/*
> - * some patches floated around with a second compression method
> - * lets save that incompat here for when they do get in
> - * Note we don't actually support it, we're just reserving the
> - * number
> - */
> -#define BTRFS_FEATURE_INCOMPAT_COMPRESS_LZOv2	(1ULL << 4)
> +#define BTRFS_FEATURE_INCOMPAT_COMPRESS_ZSTD	(1ULL << 4)
> 
>   /*
>    * older kernels tried to do bigger metadata blocks, but the
> --
> 2.9.3
> 

Powered by blists - more mailing lists