From bc98a42c1f7d0f886c0c1b75a92a004976a46d9f Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 17 Jul 2017 08:45:34 +0100 Subject: VFS: Convert sb->s_flags & MS_RDONLY to sb_rdonly(sb) Firstly by applying the following with coccinelle's spatch: @@ expression SB; @@ -SB->s_flags & MS_RDONLY +sb_rdonly(SB) to effect the conversion to sb_rdonly(sb), then by applying: @@ expression A, SB; @@ ( -(!sb_rdonly(SB)) && A +!sb_rdonly(SB) && A | -A != (sb_rdonly(SB)) +A != sb_rdonly(SB) | -A == (sb_rdonly(SB)) +A == sb_rdonly(SB) | -!(sb_rdonly(SB)) +!sb_rdonly(SB) | -A && (sb_rdonly(SB)) +A && sb_rdonly(SB) | -A || (sb_rdonly(SB)) +A || sb_rdonly(SB) | -(sb_rdonly(SB)) != A +sb_rdonly(SB) != A | -(sb_rdonly(SB)) == A +sb_rdonly(SB) == A | -(sb_rdonly(SB)) && A +sb_rdonly(SB) && A | -(sb_rdonly(SB)) || A +sb_rdonly(SB) || A ) @@ expression A, B, SB; @@ ( -(sb_rdonly(SB)) ? 1 : 0 +sb_rdonly(SB) | -(sb_rdonly(SB)) ? A : B +sb_rdonly(SB) ? A : B ) to remove left over excess bracketage and finally by applying: @@ expression A, SB; @@ ( -(A & MS_RDONLY) != sb_rdonly(SB) +(bool)(A & MS_RDONLY) != sb_rdonly(SB) | -(A & MS_RDONLY) == sb_rdonly(SB) +(bool)(A & MS_RDONLY) == sb_rdonly(SB) ) to make comparisons against the result of sb_rdonly() (which is a bool) work correctly. Signed-off-by: David Howells --- fs/btrfs/disk-io.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 080e2ebb8aa0..4a9b40f604fa 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2478,7 +2478,7 @@ static int btrfs_replay_log(struct btrfs_fs_info *fs_info, return ret; } - if (fs_info->sb->s_flags & MS_RDONLY) { + if (sb_rdonly(fs_info->sb)) { ret = btrfs_commit_super(fs_info); if (ret) return ret; @@ -2874,7 +2874,7 @@ int open_ctree(struct super_block *sb, features = btrfs_super_compat_ro_flags(disk_super) & ~BTRFS_FEATURE_COMPAT_RO_SUPP; - if (!(sb->s_flags & MS_RDONLY) && features) { + if (!sb_rdonly(sb) && features) { btrfs_err(fs_info, "cannot mount read-write because of unsupported optional features (%llx)", features); @@ -3039,7 +3039,7 @@ retry_root_backup: btrfs_calc_num_tolerated_disk_barrier_failures(fs_info); if (fs_info->fs_devices->missing_devices > fs_info->num_tolerated_disk_barrier_failures && - !(sb->s_flags & MS_RDONLY)) { + !sb_rdonly(sb)) { btrfs_warn(fs_info, "missing devices (%llu) exceeds the limit (%d), writeable mount is not allowed", fs_info->fs_devices->missing_devices, @@ -3102,7 +3102,7 @@ retry_root_backup: if (ret) goto fail_qgroup; - if (!(sb->s_flags & MS_RDONLY)) { + if (!sb_rdonly(sb)) { ret = btrfs_cleanup_fs_roots(fs_info); if (ret) goto fail_qgroup; @@ -3128,7 +3128,7 @@ retry_root_backup: goto fail_qgroup; } - if (sb->s_flags & MS_RDONLY) + if (sb_rdonly(sb)) return 0; if (btrfs_test_opt(fs_info, CLEAR_CACHE) && @@ -3928,7 +3928,7 @@ void close_ctree(struct btrfs_fs_info *fs_info) cancel_work_sync(&fs_info->async_reclaim_work); - if (!(fs_info->sb->s_flags & MS_RDONLY)) { + if (!sb_rdonly(fs_info->sb)) { /* * If the cleaner thread is stopped and there are * block groups queued for removal, the deletion will be -- cgit v1.2.3 From 5c1aab1dd5445ed8bdcdbb575abc1b0d7ee5b2e7 Mon Sep 17 00:00:00 2001 From: Nick Terrell Date: Wed, 9 Aug 2017 19:39:02 -0700 Subject: btrfs: Add zstd support 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 Signed-off-by: Chris Mason --- 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 (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig index 80e9c18ea64f..a26c63b4ad68 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 128ce17a80b0..962a95aefb81 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 d2ef9ac2a630..4ff42d18a64d 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 87f6d3332163..2269e00854d8 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 3f3eb7b17cac..845d77c097d6 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 080e2ebb8aa0..04632f4de933 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 fa1b78cf25f6..b9963d94d727 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 4b23ae5d0e5c..20631e9273a0 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 12540b6104b5..c370deadb790 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 c2d5f3580b4c..2b6d37c09a81 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 000000000000..607ce47b483a --- /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 +#include +#include +#include +#include +#include +#include +#include +#include +#include +#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 9aa74f317747..378230c163d5 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 -- cgit v1.2.3 From 4330e183c9537df20952d4a9ee142c536fb8ae54 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Thu, 9 Mar 2017 09:34:37 +0800 Subject: btrfs: Do chunk level check for degraded rw mount Now use the btrfs_check_rw_degradable() to check if we can mount in the degraded mode. With this patch, we can mount in the following case: # mkfs.btrfs -f -m raid1 -d single /dev/sdb /dev/sdc # wipefs -a /dev/sdc # mount /dev/sdb /mnt/btrfs -o degraded As the single data chunk is only on sdb, so it's OK to mount as degraded, as missing one device is OK for RAID1. But still fail in the following case as expected: # mkfs.btrfs -f -m raid1 -d single /dev/sdb /dev/sdc # wipefs -a /dev/sdb # mount /dev/sdc /mnt/btrfs -o degraded As the data chunk is only in sdb, so it's not OK to mount it as degraded. Reported-by: Zhao Lei Reported-by: Anand Jain Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 080e2ebb8aa0..78fc7f5ab773 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3035,15 +3035,10 @@ retry_root_backup: btrfs_err(fs_info, "failed to read block groups: %d", ret); goto fail_sysfs; } - fs_info->num_tolerated_disk_barrier_failures = - btrfs_calc_num_tolerated_disk_barrier_failures(fs_info); - if (fs_info->fs_devices->missing_devices > - fs_info->num_tolerated_disk_barrier_failures && - !(sb->s_flags & MS_RDONLY)) { + + if (!(sb->s_flags & MS_RDONLY) && !btrfs_check_rw_degradable(fs_info)) { btrfs_warn(fs_info, -"missing devices (%llu) exceeds the limit (%d), writeable mount is not allowed", - fs_info->fs_devices->missing_devices, - fs_info->num_tolerated_disk_barrier_failures); + "writeable mount is not allowed due to too many missing devices"); goto fail_sysfs; } -- cgit v1.2.3 From d10b82fe29cf3730ab7c1a487f0f62f10af5d301 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Tue, 27 Jun 2017 17:28:40 +0800 Subject: btrfs: Allow barrier_all_devices to do chunk level device check The last user of num_tolerated_disk_barrier_failures is barrier_all_devices(). But it can be easily changed to the new per-chunk degradable check framework. Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 16 +++------------- 1 file changed, 3 insertions(+), 13 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 78fc7f5ab773..552098e1b65b 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3519,20 +3519,10 @@ static blk_status_t wait_dev_flush(struct btrfs_device *device) return bio->bi_status; } -static int check_barrier_error(struct btrfs_fs_devices *fsdevs) +static int check_barrier_error(struct btrfs_fs_info *fs_info) { - int dev_flush_error = 0; - struct btrfs_device *dev; - - list_for_each_entry_rcu(dev, &fsdevs->devices, dev_list) { - if (!dev->bdev || dev->last_flush_error) - dev_flush_error++; - } - - if (dev_flush_error > - fsdevs->fs_info->num_tolerated_disk_barrier_failures) + if (!btrfs_check_rw_degradable(fs_info)) return -EIO; - return 0; } @@ -3587,7 +3577,7 @@ static int barrier_all_devices(struct btrfs_fs_info *info) * to arrive at the volume status. So error checking * is being pushed to a separate loop. */ - return check_barrier_error(info->fs_devices); + return check_barrier_error(info); } return 0; } -- cgit v1.2.3 From bc3cce2378b9eeb420873b17664493e68fd07e8c Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Thu, 9 Mar 2017 09:34:41 +0800 Subject: btrfs: Cleanup num_tolerated_disk_barrier_failures As we use per-chunk degradable check, the global num_tolerated_disk_barrier_failures is of no use. We can now remove it. Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 2 -- fs/btrfs/disk-io.c | 54 ------------------------------------------------------ fs/btrfs/disk-io.h | 2 -- fs/btrfs/volumes.c | 17 ----------------- 4 files changed, 75 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 34ed3dabd43a..06033b5f1657 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1071,8 +1071,6 @@ struct btrfs_fs_info { /* next backup root to be overwritten */ int backup_root_index; - int num_tolerated_disk_barrier_failures; - /* device replace state */ struct btrfs_dev_replace dev_replace; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 552098e1b65b..2f366044d891 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3611,60 +3611,6 @@ int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags) return min_tolerated; } -int btrfs_calc_num_tolerated_disk_barrier_failures( - struct btrfs_fs_info *fs_info) -{ - struct btrfs_ioctl_space_info space; - struct btrfs_space_info *sinfo; - u64 types[] = {BTRFS_BLOCK_GROUP_DATA, - BTRFS_BLOCK_GROUP_SYSTEM, - BTRFS_BLOCK_GROUP_METADATA, - BTRFS_BLOCK_GROUP_DATA | BTRFS_BLOCK_GROUP_METADATA}; - int i; - int c; - int num_tolerated_disk_barrier_failures = - (int)fs_info->fs_devices->num_devices; - - for (i = 0; i < ARRAY_SIZE(types); i++) { - struct btrfs_space_info *tmp; - - sinfo = NULL; - rcu_read_lock(); - list_for_each_entry_rcu(tmp, &fs_info->space_info, list) { - if (tmp->flags == types[i]) { - sinfo = tmp; - break; - } - } - rcu_read_unlock(); - - if (!sinfo) - continue; - - down_read(&sinfo->groups_sem); - for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) { - u64 flags; - - if (list_empty(&sinfo->block_groups[c])) - continue; - - btrfs_get_block_group_info(&sinfo->block_groups[c], - &space); - if (space.total_bytes == 0 || space.used_bytes == 0) - continue; - flags = space.flags; - - num_tolerated_disk_barrier_failures = min( - num_tolerated_disk_barrier_failures, - btrfs_get_num_tolerated_disk_barrier_failures( - flags)); - } - up_read(&sinfo->groups_sem); - } - - return num_tolerated_disk_barrier_failures; -} - int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors) { struct list_head *head; diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index 0a634d3ffc16..b5d0c4cef819 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -142,8 +142,6 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans, int btree_lock_page_hook(struct page *page, void *data, void (*flush_fn)(void *)); int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags); -int btrfs_calc_num_tolerated_disk_barrier_failures( - struct btrfs_fs_info *fs_info); int __init btrfs_end_io_wq_init(void); void btrfs_end_io_wq_exit(void); diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 1bc822b5a4c0..2272ac48a8a9 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1987,9 +1987,6 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, free_fs_devices(cur_devices); } - fs_info->num_tolerated_disk_barrier_failures = - btrfs_calc_num_tolerated_disk_barrier_failures(fs_info); - out: mutex_unlock(&uuid_mutex); return ret; @@ -2487,8 +2484,6 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path "sysfs: failed to create fsid for sprout"); } - fs_info->num_tolerated_disk_barrier_failures = - btrfs_calc_num_tolerated_disk_barrier_failures(fs_info); ret = btrfs_commit_transaction(trans); if (seeding_dev) { @@ -3898,13 +3893,6 @@ int btrfs_balance(struct btrfs_balance_control *bctl, meta_target, data_target); } - if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) { - fs_info->num_tolerated_disk_barrier_failures = min( - btrfs_calc_num_tolerated_disk_barrier_failures(fs_info), - btrfs_get_num_tolerated_disk_barrier_failures( - bctl->sys.target)); - } - ret = insert_balance_item(fs_info, bctl); if (ret && ret != -EEXIST) goto out; @@ -3927,11 +3915,6 @@ int btrfs_balance(struct btrfs_balance_control *bctl, mutex_lock(&fs_info->balance_mutex); atomic_dec(&fs_info->balance_running); - if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) { - fs_info->num_tolerated_disk_barrier_failures = - btrfs_calc_num_tolerated_disk_barrier_failures(fs_info); - } - if (bargs) { memset(bargs, 0, sizeof(*bargs)); update_ioctl_balance_args(fs_info, 0, bargs); -- cgit v1.2.3 From a4f78750ef1882e59bb4f947e216cf61ef2d67d2 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Thu, 29 Jun 2017 18:37:49 +0200 Subject: btrfs: get fs_info from eb in btrfs_print_leaf, remove argument Signed-off-by: David Sterba --- fs/btrfs/ctree.c | 14 +++++++------- fs/btrfs/disk-io.c | 2 +- fs/btrfs/extent-tree.c | 6 +++--- fs/btrfs/print-tree.c | 6 ++++-- fs/btrfs/print-tree.h | 2 +- fs/btrfs/root-tree.c | 2 +- 6 files changed, 17 insertions(+), 15 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index 3f4daa9d6e2c..6d49db7d86be 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -4650,7 +4650,7 @@ void btrfs_truncate_item(struct btrfs_fs_info *fs_info, btrfs_mark_buffer_dirty(leaf); if (btrfs_leaf_free_space(fs_info, leaf) < 0) { - btrfs_print_leaf(fs_info, leaf); + btrfs_print_leaf(leaf); BUG(); } } @@ -4679,7 +4679,7 @@ void btrfs_extend_item(struct btrfs_fs_info *fs_info, struct btrfs_path *path, data_end = leaf_data_end(fs_info, leaf); if (btrfs_leaf_free_space(fs_info, leaf) < data_size) { - btrfs_print_leaf(fs_info, leaf); + btrfs_print_leaf(leaf); BUG(); } slot = path->slots[0]; @@ -4687,7 +4687,7 @@ void btrfs_extend_item(struct btrfs_fs_info *fs_info, struct btrfs_path *path, BUG_ON(slot < 0); if (slot >= nritems) { - btrfs_print_leaf(fs_info, leaf); + btrfs_print_leaf(leaf); btrfs_crit(fs_info, "slot %d too large, nritems %d", slot, nritems); BUG_ON(1); @@ -4718,7 +4718,7 @@ void btrfs_extend_item(struct btrfs_fs_info *fs_info, struct btrfs_path *path, btrfs_mark_buffer_dirty(leaf); if (btrfs_leaf_free_space(fs_info, leaf) < 0) { - btrfs_print_leaf(fs_info, leaf); + btrfs_print_leaf(leaf); BUG(); } } @@ -4757,7 +4757,7 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path, data_end = leaf_data_end(fs_info, leaf); if (btrfs_leaf_free_space(fs_info, leaf) < total_size) { - btrfs_print_leaf(fs_info, leaf); + btrfs_print_leaf(leaf); btrfs_crit(fs_info, "not enough freespace need %u have %d", total_size, btrfs_leaf_free_space(fs_info, leaf)); BUG(); @@ -4767,7 +4767,7 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path, unsigned int old_data = btrfs_item_end_nr(leaf, slot); if (old_data < data_end) { - btrfs_print_leaf(fs_info, leaf); + btrfs_print_leaf(leaf); btrfs_crit(fs_info, "slot %d old_data %d data_end %d", slot, old_data, data_end); BUG_ON(1); @@ -4811,7 +4811,7 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path, btrfs_mark_buffer_dirty(leaf); if (btrfs_leaf_free_space(fs_info, leaf) < 0) { - btrfs_print_leaf(fs_info, leaf); + btrfs_print_leaf(leaf); BUG(); } } diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 2f366044d891..a42ae7676759 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3984,7 +3984,7 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf) fs_info->dirty_metadata_batch); #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY if (btrfs_header_level(buf) == 0 && check_leaf(root, buf)) { - btrfs_print_leaf(fs_info, buf); + btrfs_print_leaf(buf); ASSERT(0); } #endif diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 08e620c43842..9f0563dbbd5f 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -6960,7 +6960,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, "umm, got %d back from search, was looking for %llu", ret, bytenr); if (ret > 0) - btrfs_print_leaf(info, path->nodes[0]); + btrfs_print_leaf(path->nodes[0]); } if (ret < 0) { btrfs_abort_transaction(trans, ret); @@ -6969,7 +6969,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, extent_slot = path->slots[0]; } } else if (WARN_ON(ret == -ENOENT)) { - btrfs_print_leaf(info, path->nodes[0]); + btrfs_print_leaf(path->nodes[0]); btrfs_err(info, "unable to find ref byte nr %llu parent %llu root %llu owner %llu offset %llu", bytenr, parent, root_objectid, owner_objectid, @@ -7006,7 +7006,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, btrfs_err(info, "umm, got %d back from search, was looking for %llu", ret, bytenr); - btrfs_print_leaf(info, path->nodes[0]); + btrfs_print_leaf(path->nodes[0]); } if (ret < 0) { btrfs_abort_transaction(trans, ret); diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c index fcae61e175f3..9c3911f4a100 100644 --- a/fs/btrfs/print-tree.c +++ b/fs/btrfs/print-tree.c @@ -161,8 +161,9 @@ static void print_uuid_item(struct extent_buffer *l, unsigned long offset, } } -void btrfs_print_leaf(struct btrfs_fs_info *fs_info, struct extent_buffer *l) +void btrfs_print_leaf(struct extent_buffer *l) { + struct btrfs_fs_info *fs_info; int i; u32 type, nr; struct btrfs_item *item; @@ -180,6 +181,7 @@ void btrfs_print_leaf(struct btrfs_fs_info *fs_info, struct extent_buffer *l) if (!l) return; + fs_info = l->fs_info; nr = btrfs_header_nritems(l); btrfs_info(fs_info, "leaf %llu total ptrs %d free space %d", @@ -329,7 +331,7 @@ void btrfs_print_tree(struct btrfs_fs_info *fs_info, struct extent_buffer *c) nr = btrfs_header_nritems(c); level = btrfs_header_level(c); if (level == 0) { - btrfs_print_leaf(fs_info, c); + btrfs_print_leaf(c); return; } btrfs_info(fs_info, diff --git a/fs/btrfs/print-tree.h b/fs/btrfs/print-tree.h index 4f2e0ea0e95a..689a52ee0cd1 100644 --- a/fs/btrfs/print-tree.h +++ b/fs/btrfs/print-tree.h @@ -18,6 +18,6 @@ #ifndef __PRINT_TREE_ #define __PRINT_TREE_ -void btrfs_print_leaf(struct btrfs_fs_info *fs_info, struct extent_buffer *l); +void btrfs_print_leaf(struct extent_buffer *l); void btrfs_print_tree(struct btrfs_fs_info *fs_info, struct extent_buffer *c); #endif diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index 460db0cb2d07..5b488af6f25e 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -151,7 +151,7 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root } if (ret != 0) { - btrfs_print_leaf(fs_info, path->nodes[0]); + btrfs_print_leaf(path->nodes[0]); btrfs_crit(fs_info, "unable to update root key %llu %u %llu", key->objectid, key->type, key->offset); BUG_ON(1); -- cgit v1.2.3 From abbb3b8ebfec577390873d33076d56bc6e758939 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Fri, 16 Jun 2017 00:50:33 +0200 Subject: btrfs: split write_dev_supers to two functions There are two independent parts, one that writes the superblocks and another that waits for completion. No functional changes, but cleanups, reformatting and comment updates. Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 129 ++++++++++++++++++++++++++++++----------------------- 1 file changed, 73 insertions(+), 56 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index a42ae7676759..b72711ad7121 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3373,19 +3373,17 @@ struct buffer_head *btrfs_read_dev_super(struct block_device *bdev) } /* - * this should be called twice, once with wait == 0 and - * once with wait == 1. When wait == 0 is done, all the buffer heads - * we write are pinned. + * Write superblock @sb to the @device. Do not wait for completion, all the + * buffer heads we write are pinned. * - * They are released when wait == 1 is done. - * max_mirrors must be the same for both runs, and it indicates how - * many supers on this one device should be written. + * Write @max_mirrors copies of the superblock, where 0 means default that fit + * the expected device size at commit time. Note that max_mirrors must be + * same for write and wait phases. * - * max_mirrors == 0 means to write them all. + * Return number of errors when buffer head is not found or submission fails. */ static int write_dev_supers(struct btrfs_device *device, - struct btrfs_super_block *sb, - int wait, int max_mirrors) + struct btrfs_super_block *sb, int max_mirrors) { struct buffer_head *bh; int i; @@ -3403,57 +3401,33 @@ static int write_dev_supers(struct btrfs_device *device, device->commit_total_bytes) break; - if (wait) { - bh = __find_get_block(device->bdev, bytenr / 4096, - BTRFS_SUPER_INFO_SIZE); - if (!bh) { - errors++; - continue; - } - wait_on_buffer(bh); - if (!buffer_uptodate(bh)) - errors++; + btrfs_set_super_bytenr(sb, bytenr); - /* drop our reference */ - brelse(bh); + crc = ~(u32)0; + crc = btrfs_csum_data((const char *)sb + BTRFS_CSUM_SIZE, crc, + BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE); + btrfs_csum_final(crc, sb->csum); - /* drop the reference from the wait == 0 run */ - brelse(bh); + /* One reference for us, and we leave it for the caller */ + bh = __getblk(device->bdev, bytenr / 4096, + BTRFS_SUPER_INFO_SIZE); + if (!bh) { + btrfs_err(device->fs_info, + "couldn't get super buffer head for bytenr %llu", + bytenr); + errors++; continue; - } else { - btrfs_set_super_bytenr(sb, bytenr); - - crc = ~(u32)0; - crc = btrfs_csum_data((const char *)sb + - BTRFS_CSUM_SIZE, crc, - BTRFS_SUPER_INFO_SIZE - - BTRFS_CSUM_SIZE); - btrfs_csum_final(crc, sb->csum); - - /* - * one reference for us, and we leave it for the - * caller - */ - bh = __getblk(device->bdev, bytenr / 4096, - BTRFS_SUPER_INFO_SIZE); - if (!bh) { - btrfs_err(device->fs_info, - "couldn't get super buffer head for bytenr %llu", - bytenr); - errors++; - continue; - } + } - memcpy(bh->b_data, sb, BTRFS_SUPER_INFO_SIZE); + memcpy(bh->b_data, sb, BTRFS_SUPER_INFO_SIZE); - /* one reference for submit_bh */ - get_bh(bh); + /* one reference for submit_bh */ + get_bh(bh); - set_buffer_uptodate(bh); - lock_buffer(bh); - bh->b_end_io = btrfs_end_buffer_write_sync; - bh->b_private = device; - } + set_buffer_uptodate(bh); + lock_buffer(bh); + bh->b_end_io = btrfs_end_buffer_write_sync; + bh->b_private = device; /* * we fua the first super. The others we allow @@ -3471,6 +3445,49 @@ static int write_dev_supers(struct btrfs_device *device, return errors < i ? 0 : -1; } +/* + * Wait for write completion of superblocks done by write_dev_supers, + * @max_mirrors same for write and wait phases. + * + * Return number of errors when buffer head is not found or not marked up to + * date. + */ +static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) +{ + struct buffer_head *bh; + int i; + int errors = 0; + u64 bytenr; + + if (max_mirrors == 0) + max_mirrors = BTRFS_SUPER_MIRROR_MAX; + + for (i = 0; i < max_mirrors; i++) { + bytenr = btrfs_sb_offset(i); + if (bytenr + BTRFS_SUPER_INFO_SIZE >= + device->commit_total_bytes) + break; + + bh = __find_get_block(device->bdev, bytenr / 4096, + BTRFS_SUPER_INFO_SIZE); + if (!bh) { + errors++; + continue; + } + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) + errors++; + + /* drop our reference */ + brelse(bh); + + /* drop the reference from the writing run */ + brelse(bh); + } + + return errors < i ? 0 : -1; +} + /* * endio for the write_dev_flush, this will wake anyone waiting * for the barrier when it is done @@ -3668,7 +3685,7 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors) flags = btrfs_super_flags(sb); btrfs_set_super_flags(sb, flags | BTRFS_HEADER_FLAG_WRITTEN); - ret = write_dev_supers(dev, sb, 0, max_mirrors); + ret = write_dev_supers(dev, sb, max_mirrors); if (ret) total_errors++; } @@ -3691,7 +3708,7 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors) if (!dev->in_fs_metadata || !dev->writeable) continue; - ret = write_dev_supers(dev, sb, 1, max_mirrors); + ret = wait_dev_supers(dev, max_mirrors); if (ret) total_errors++; } -- cgit v1.2.3 From 9f6d251033bc5bd0115c23fae8b6aac5f6263c2c Mon Sep 17 00:00:00 2001 From: David Sterba Date: Fri, 16 Jun 2017 01:48:05 +0200 Subject: btrfs: use named constant for bdev blocksize Superblock is read and written using buffer heads, we need to set the bdev blocksize. The magic constant has been hardcoded in several places, so replace it with a named constant. Signed-off-by: David Sterba --- fs/btrfs/check-integrity.c | 6 +++--- fs/btrfs/disk-io.c | 11 ++++++----- fs/btrfs/disk-io.h | 8 ++++++++ fs/btrfs/volumes.c | 6 +++--- 4 files changed, 20 insertions(+), 11 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index 11d37c94ce05..7d65d98d2790 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -795,12 +795,12 @@ static int btrfsic_process_superblock_dev_mirror( dev_bytenr = btrfs_sb_offset(superblock_mirror_num); if (dev_bytenr + BTRFS_SUPER_INFO_SIZE > device->commit_total_bytes) return -1; - bh = __bread(superblock_bdev, dev_bytenr / 4096, + bh = __bread(superblock_bdev, dev_bytenr / BTRFS_BDEV_BLOCKSIZE, BTRFS_SUPER_INFO_SIZE); if (NULL == bh) return -1; super_tmp = (struct btrfs_super_block *) - (bh->b_data + (dev_bytenr & 4095)); + (bh->b_data + (dev_bytenr & (BTRFS_BDEV_BLOCKSIZE - 1))); if (btrfs_super_bytenr(super_tmp) != dev_bytenr || btrfs_super_magic(super_tmp) != BTRFS_MAGIC || @@ -2758,7 +2758,7 @@ int btrfsic_submit_bh(int op, int op_flags, struct buffer_head *bh) (op == REQ_OP_WRITE) && bh->b_size > 0) { u64 dev_bytenr; - dev_bytenr = 4096 * bh->b_blocknr; + dev_bytenr = BTRFS_BDEV_BLOCKSIZE * bh->b_blocknr; if (dev_state->state->print_mask & BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH) pr_info("submit_bh(op=0x%x,0x%x, blocknr=%llu (bytenr %llu), size=%zu, data=%p, bdev=%p)\n", diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index b72711ad7121..9b1f4ef54438 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2694,8 +2694,8 @@ int open_ctree(struct super_block *sb, btrfs_init_balance(fs_info); btrfs_init_async_reclaim_work(&fs_info->async_reclaim_work); - sb->s_blocksize = 4096; - sb->s_blocksize_bits = blksize_bits(4096); + sb->s_blocksize = BTRFS_BDEV_BLOCKSIZE; + sb->s_blocksize_bits = blksize_bits(BTRFS_BDEV_BLOCKSIZE); btrfs_init_btree_inode(fs_info); @@ -3316,7 +3316,7 @@ int btrfs_read_dev_one_super(struct block_device *bdev, int copy_num, if (bytenr + BTRFS_SUPER_INFO_SIZE >= i_size_read(bdev->bd_inode)) return -EINVAL; - bh = __bread(bdev, bytenr / 4096, BTRFS_SUPER_INFO_SIZE); + bh = __bread(bdev, bytenr / BTRFS_BDEV_BLOCKSIZE, BTRFS_SUPER_INFO_SIZE); /* * If we fail to read from the underlying devices, as of now * the best option we have is to mark it EIO. @@ -3409,7 +3409,7 @@ static int write_dev_supers(struct btrfs_device *device, btrfs_csum_final(crc, sb->csum); /* One reference for us, and we leave it for the caller */ - bh = __getblk(device->bdev, bytenr / 4096, + bh = __getblk(device->bdev, bytenr / BTRFS_BDEV_BLOCKSIZE, BTRFS_SUPER_INFO_SIZE); if (!bh) { btrfs_err(device->fs_info, @@ -3468,7 +3468,8 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) device->commit_total_bytes) break; - bh = __find_get_block(device->bdev, bytenr / 4096, + bh = __find_get_block(device->bdev, + bytenr / BTRFS_BDEV_BLOCKSIZE, BTRFS_SUPER_INFO_SIZE); if (!bh) { errors++; diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index b5d0c4cef819..7f7c35d6347a 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -25,6 +25,14 @@ #define BTRFS_SUPER_MIRROR_MAX 3 #define BTRFS_SUPER_MIRROR_SHIFT 12 +/* + * Fixed blocksize for all devices, applies to specific ways of reading + * metadata like superblock. Must meet the set_blocksize requirements. + * + * Do not change. + */ +#define BTRFS_BDEV_BLOCKSIZE (4096) + enum btrfs_wq_endio_type { BTRFS_WQ_ENDIO_DATA = 0, BTRFS_WQ_ENDIO_METADATA = 1, diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 3a43a9cc6183..12bd04a4104f 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -303,7 +303,7 @@ btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder, if (flush) filemap_write_and_wait((*bdev)->bd_inode->i_mapping); - ret = set_blocksize(*bdev, 4096); + ret = set_blocksize(*bdev, BTRFS_BDEV_BLOCKSIZE); if (ret) { blkdev_put(*bdev, flags); goto error; @@ -2395,7 +2395,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path device->is_tgtdev_for_dev_replace = 0; device->mode = FMODE_EXCL; device->dev_stats_valid = 1; - set_blocksize(device->bdev, 4096); + set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE); if (seeding_dev) { sb->s_flags &= ~MS_RDONLY; @@ -2598,7 +2598,7 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info, device->is_tgtdev_for_dev_replace = 1; device->mode = FMODE_EXCL; device->dev_stats_valid = 1; - set_blocksize(device->bdev, 4096); + set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE); device->fs_devices = fs_info->fs_devices; list_add(&device->dev_list, &fs_info->fs_devices->devices); fs_info->fs_devices->num_devices++; -- cgit v1.2.3 From ea14b57fd1954fa3193e025224bbbeab7415c490 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Thu, 22 Jun 2017 02:19:11 +0200 Subject: btrfs: fix spelling of snapshotting Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 6 +++--- fs/btrfs/disk-io.c | 2 +- fs/btrfs/extent-tree.c | 22 +++++++++++----------- fs/btrfs/file.c | 10 +++++----- fs/btrfs/inode.c | 22 +++++++++++----------- fs/btrfs/ioctl.c | 10 +++++----- 6 files changed, 36 insertions(+), 36 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 6510f246f71e..874d814da371 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1258,7 +1258,7 @@ struct btrfs_root { */ int send_in_progress; struct btrfs_subvolume_writers *subv_writers; - atomic_t will_be_snapshoted; + atomic_t will_be_snapshotted; /* For qgroup metadata space reserve */ atomic64_t qgroup_meta_rsv; @@ -2773,8 +2773,8 @@ int btrfs_init_space_info(struct btrfs_fs_info *fs_info); int btrfs_delayed_refs_qgroup_accounting(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info); int __get_raid_index(u64 flags); -int btrfs_start_write_no_snapshoting(struct btrfs_root *root); -void btrfs_end_write_no_snapshoting(struct btrfs_root *root); +int btrfs_start_write_no_snapshotting(struct btrfs_root *root); +void btrfs_end_write_no_snapshotting(struct btrfs_root *root); void btrfs_wait_for_snapshot_creation(struct btrfs_root *root); void check_system_chunk(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, const u64 type); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 9b1f4ef54438..57a857142cda 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1343,7 +1343,7 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, atomic_set(&root->log_batch, 0); atomic_set(&root->orphan_inodes, 0); refcount_set(&root->refs, 1); - atomic_set(&root->will_be_snapshoted, 0); + atomic_set(&root->will_be_snapshotted, 0); atomic64_set(&root->qgroup_meta_rsv, 0); root->log_transid = 0; root->log_transid_committed = -1; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 42251c2eb7d2..7d7abc0d47b9 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -10989,14 +10989,14 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range) } /* - * btrfs_{start,end}_write_no_snapshoting() are similar to + * btrfs_{start,end}_write_no_snapshotting() are similar to * mnt_{want,drop}_write(), they are used to prevent some tasks from writing * data into the page cache through nocow before the subvolume is snapshoted, * but flush the data into disk after the snapshot creation, or to prevent - * operations while snapshoting is ongoing and that cause the snapshot to be + * operations while snapshotting is ongoing and that cause the snapshot to be * inconsistent (writes followed by expanding truncates for example). */ -void btrfs_end_write_no_snapshoting(struct btrfs_root *root) +void btrfs_end_write_no_snapshotting(struct btrfs_root *root) { percpu_counter_dec(&root->subv_writers->counter); /* @@ -11007,9 +11007,9 @@ void btrfs_end_write_no_snapshoting(struct btrfs_root *root) wake_up(&root->subv_writers->wait); } -int btrfs_start_write_no_snapshoting(struct btrfs_root *root) +int btrfs_start_write_no_snapshotting(struct btrfs_root *root) { - if (atomic_read(&root->will_be_snapshoted)) + if (atomic_read(&root->will_be_snapshotted)) return 0; percpu_counter_inc(&root->subv_writers->counter); @@ -11017,14 +11017,14 @@ int btrfs_start_write_no_snapshoting(struct btrfs_root *root) * Make sure counter is updated before we check for snapshot creation. */ smp_mb(); - if (atomic_read(&root->will_be_snapshoted)) { - btrfs_end_write_no_snapshoting(root); + if (atomic_read(&root->will_be_snapshotted)) { + btrfs_end_write_no_snapshotting(root); return 0; } return 1; } -static int wait_snapshoting_atomic_t(atomic_t *a) +static int wait_snapshotting_atomic_t(atomic_t *a) { schedule(); return 0; @@ -11035,11 +11035,11 @@ void btrfs_wait_for_snapshot_creation(struct btrfs_root *root) while (true) { int ret; - ret = btrfs_start_write_no_snapshoting(root); + ret = btrfs_start_write_no_snapshotting(root); if (ret) break; - wait_on_atomic_t(&root->will_be_snapshoted, - wait_snapshoting_atomic_t, + wait_on_atomic_t(&root->will_be_snapshotted, + wait_snapshotting_atomic_t, TASK_UNINTERRUPTIBLE); } } diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 9e75d8a39aac..58818cf7f82d 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1536,7 +1536,7 @@ static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos, u64 num_bytes; int ret; - ret = btrfs_start_write_no_snapshoting(root); + ret = btrfs_start_write_no_snapshotting(root); if (!ret) return -ENOSPC; @@ -1561,7 +1561,7 @@ static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos, NULL, NULL, NULL); if (ret <= 0) { ret = 0; - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); } else { *write_bytes = min_t(size_t, *write_bytes , num_bytes - pos + lockstart); @@ -1664,7 +1664,7 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, data_reserved, pos, write_bytes); else - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); break; } @@ -1767,7 +1767,7 @@ again: release_bytes = 0; if (only_release_metadata) - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); if (only_release_metadata && copied > 0) { lockstart = round_down(pos, @@ -1797,7 +1797,7 @@ again: if (release_bytes) { if (only_release_metadata) { - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); btrfs_delalloc_release_metadata(BTRFS_I(inode), release_bytes); } else { diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 9349a13b3d72..9ad9dda871ca 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1381,7 +1381,7 @@ next_slot: * we fall into common COW way. */ if (!nolock) { - err = btrfs_start_write_no_snapshoting(root); + err = btrfs_start_write_no_snapshotting(root); if (!err) goto out_check; } @@ -1393,12 +1393,12 @@ next_slot: if (csum_exist_in_range(fs_info, disk_bytenr, num_bytes)) { if (!nolock) - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); goto out_check; } if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr)) { if (!nolock) - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); goto out_check; } nocow = 1; @@ -1415,7 +1415,7 @@ out_check: if (extent_end <= start) { path->slots[0]++; if (!nolock && nocow) - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); if (nocow) btrfs_dec_nocow_writers(fs_info, disk_bytenr); goto next_slot; @@ -1438,7 +1438,7 @@ out_check: NULL); if (ret) { if (!nolock && nocow) - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); if (nocow) btrfs_dec_nocow_writers(fs_info, disk_bytenr); @@ -1459,7 +1459,7 @@ out_check: BTRFS_ORDERED_PREALLOC); if (IS_ERR(em)) { if (!nolock && nocow) - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); if (nocow) btrfs_dec_nocow_writers(fs_info, disk_bytenr); @@ -1499,7 +1499,7 @@ out_check: PAGE_UNLOCK | PAGE_SET_PRIVATE2); if (!nolock && nocow) - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); cur_offset = extent_end; /* @@ -5053,7 +5053,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) if (newsize > oldsize) { /* - * Don't do an expanding truncate while snapshoting is ongoing. + * Don't do an expanding truncate while snapshotting is ongoing. * This is to ensure the snapshot captures a fully consistent * state of this file - if the snapshot captures this expanding * truncation, it must capture all writes that happened before @@ -5062,13 +5062,13 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) btrfs_wait_for_snapshot_creation(root); ret = btrfs_cont_expand(inode, oldsize, newsize); if (ret) { - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); return ret; } trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); return PTR_ERR(trans); } @@ -5076,7 +5076,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL); pagecache_isize_extended(inode, oldsize, newsize); ret = btrfs_update_inode(trans, root, inode); - btrfs_end_write_no_snapshoting(root); + btrfs_end_write_no_snapshotting(root); btrfs_end_transaction(trans); } else { diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 4cfc3d4c0a37..7d144a676d95 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -607,7 +607,7 @@ fail_free: return ret; } -static void btrfs_wait_for_no_snapshoting_writes(struct btrfs_root *root) +static void btrfs_wait_for_no_snapshotting_writes(struct btrfs_root *root) { s64 writers; DEFINE_WAIT(wait); @@ -650,9 +650,9 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, goto free_pending; } - atomic_inc(&root->will_be_snapshoted); + atomic_inc(&root->will_be_snapshotted); smp_mb__after_atomic(); - btrfs_wait_for_no_snapshoting_writes(root); + btrfs_wait_for_no_snapshotting_writes(root); ret = btrfs_start_delalloc_inodes(root, 0); if (ret) @@ -723,8 +723,8 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, fail: btrfs_subvolume_release_metadata(fs_info, &pending_snapshot->block_rsv); dec_and_free: - if (atomic_dec_and_test(&root->will_be_snapshoted)) - wake_up_atomic_t(&root->will_be_snapshoted); + if (atomic_dec_and_test(&root->will_be_snapshotted)) + wake_up_atomic_t(&root->will_be_snapshotted); free_pending: kfree(pending_snapshot->root_item); btrfs_free_path(pending_snapshot->path); -- cgit v1.2.3 From 4958aa6821c7c4e9682806bc4fea92250d020829 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Thu, 22 Jun 2017 03:59:40 +0200 Subject: btrfs: drop chunk locks at the end of close_ctree The pinned chunks might be left over so we clean them but at this point of close_ctree, there's noone to race with, the locking can be removed. Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 57a857142cda..c2d3699a70bd 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3944,7 +3944,6 @@ void close_ctree(struct btrfs_fs_info *fs_info) __btrfs_free_block_rsv(root->orphan_block_rsv); root->orphan_block_rsv = NULL; - mutex_lock(&fs_info->chunk_mutex); while (!list_empty(&fs_info->pinned_chunks)) { struct extent_map *em; @@ -3953,7 +3952,6 @@ void close_ctree(struct btrfs_fs_info *fs_info) list_del_init(&em->list); free_extent_map(em); } - mutex_unlock(&fs_info->chunk_mutex); } int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid, -- cgit v1.2.3 From 44880fdc91bc4f6730e37f2cb6025b35c70b312d Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Sat, 29 Jul 2017 17:50:09 +0800 Subject: btrfs: use appropriate define for the fsid Though BTRFS_FSID_SIZE and BTRFS_UUID_SIZE are of the same size, we should use the matching constant for the fsid buffer. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/check-integrity.c | 2 +- fs/btrfs/disk-io.c | 6 +++--- fs/btrfs/scrub.c | 2 +- fs/btrfs/volumes.c | 16 ++++++++-------- 4 files changed, 13 insertions(+), 13 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index 7d65d98d2790..7b9aabc2cb78 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -1732,7 +1732,7 @@ static int btrfsic_test_for_metadata(struct btrfsic_state *state, num_pages = state->metablock_size >> PAGE_SHIFT; h = (struct btrfs_header *)datav[0]; - if (memcmp(h->fsid, fs_info->fsid, BTRFS_UUID_SIZE)) + if (memcmp(h->fsid, fs_info->fsid, BTRFS_FSID_SIZE)) return 1; for (i = 0; i < num_pages; i++) { diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index c2d3699a70bd..3715478fffed 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -529,7 +529,7 @@ static int check_tree_block_fsid(struct btrfs_fs_info *fs_info, struct extent_buffer *eb) { struct btrfs_fs_devices *fs_devices = fs_info->fs_devices; - u8 fsid[BTRFS_UUID_SIZE]; + u8 fsid[BTRFS_FSID_SIZE]; int ret = 1; read_extent_buffer(eb, fsid, btrfs_header_fsid(), BTRFS_FSID_SIZE); @@ -3681,7 +3681,7 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors) btrfs_set_stack_device_io_width(dev_item, dev->io_width); btrfs_set_stack_device_sector_size(dev_item, dev->sector_size); memcpy(dev_item->uuid, dev->uuid, BTRFS_UUID_SIZE); - memcpy(dev_item->fsid, dev->fs_devices->fsid, BTRFS_UUID_SIZE); + memcpy(dev_item->fsid, dev->fs_devices->fsid, BTRFS_FSID_SIZE); flags = btrfs_super_flags(sb); btrfs_set_super_flags(sb, flags | BTRFS_HEADER_FLAG_WRITTEN); @@ -4120,7 +4120,7 @@ static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info) ret = -EINVAL; } - if (memcmp(fs_info->fsid, sb->dev_item.fsid, BTRFS_UUID_SIZE) != 0) { + if (memcmp(fs_info->fsid, sb->dev_item.fsid, BTRFS_FSID_SIZE) != 0) { btrfs_err(fs_info, "dev_item UUID does not match fsid: %pU != %pU", fs_info->fsid, sb->dev_item.fsid); diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index f49b94ab3d2a..20c1ba19e665 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -1769,7 +1769,7 @@ static inline int scrub_check_fsid(u8 fsid[], struct btrfs_fs_devices *fs_devices = spage->dev->fs_devices; int ret; - ret = memcmp(fsid, fs_devices->fsid, BTRFS_UUID_SIZE); + ret = memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE); return !ret; } diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 12bd04a4104f..ed82edf987f4 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1719,7 +1719,7 @@ static int btrfs_add_device(struct btrfs_trans_handle *trans, ptr = btrfs_device_uuid(dev_item); write_extent_buffer(leaf, device->uuid, ptr, BTRFS_UUID_SIZE); ptr = btrfs_device_fsid(dev_item); - write_extent_buffer(leaf, fs_info->fsid, ptr, BTRFS_UUID_SIZE); + write_extent_buffer(leaf, fs_info->fsid, ptr, BTRFS_FSID_SIZE); btrfs_mark_buffer_dirty(leaf); ret = 0; @@ -2249,7 +2249,7 @@ static int btrfs_finish_sprout(struct btrfs_trans_handle *trans, struct btrfs_dev_item *dev_item; struct btrfs_device *device; struct btrfs_key key; - u8 fs_uuid[BTRFS_UUID_SIZE]; + u8 fs_uuid[BTRFS_FSID_SIZE]; u8 dev_uuid[BTRFS_UUID_SIZE]; u64 devid; int ret; @@ -2292,7 +2292,7 @@ next_slot: read_extent_buffer(leaf, dev_uuid, btrfs_device_uuid(dev_item), BTRFS_UUID_SIZE); read_extent_buffer(leaf, fs_uuid, btrfs_device_fsid(dev_item), - BTRFS_UUID_SIZE); + BTRFS_FSID_SIZE); device = btrfs_find_device(fs_info, devid, dev_uuid, fs_uuid); BUG_ON(!device); /* Logic error */ @@ -6243,7 +6243,7 @@ struct btrfs_device *btrfs_find_device(struct btrfs_fs_info *fs_info, u64 devid, cur_devices = fs_info->fs_devices; while (cur_devices) { if (!fsid || - !memcmp(cur_devices->fsid, fsid, BTRFS_UUID_SIZE)) { + !memcmp(cur_devices->fsid, fsid, BTRFS_FSID_SIZE)) { device = find_device(cur_devices, devid, uuid); if (device) return device; @@ -6518,7 +6518,7 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info, fs_devices = fs_info->fs_devices->seed; while (fs_devices) { - if (!memcmp(fs_devices->fsid, fsid, BTRFS_UUID_SIZE)) + if (!memcmp(fs_devices->fsid, fsid, BTRFS_FSID_SIZE)) return fs_devices; fs_devices = fs_devices->seed; @@ -6571,16 +6571,16 @@ static int read_one_dev(struct btrfs_fs_info *fs_info, struct btrfs_device *device; u64 devid; int ret; - u8 fs_uuid[BTRFS_UUID_SIZE]; + u8 fs_uuid[BTRFS_FSID_SIZE]; u8 dev_uuid[BTRFS_UUID_SIZE]; devid = btrfs_device_id(leaf, dev_item); read_extent_buffer(leaf, dev_uuid, btrfs_device_uuid(dev_item), BTRFS_UUID_SIZE); read_extent_buffer(leaf, fs_uuid, btrfs_device_fsid(dev_item), - BTRFS_UUID_SIZE); + BTRFS_FSID_SIZE); - if (memcmp(fs_uuid, fs_info->fsid, BTRFS_UUID_SIZE)) { + if (memcmp(fs_uuid, fs_info->fsid, BTRFS_FSID_SIZE)) { fs_devices = open_seed_devices(fs_info, fs_uuid); if (IS_ERR(fs_devices)) return PTR_ERR(fs_devices); -- cgit v1.2.3 From 43a0111103af2d358f6d2d671d8429ba169e53bb Mon Sep 17 00:00:00 2001 From: Lu Fengqi Date: Fri, 18 Aug 2017 16:38:07 +0800 Subject: btrfs: use btrfsic_submit_bio instead of submit_bio in write_dev_flush Although this bio has no data attached, it will reach this condition (bio->bi_opf & REQ_PREFLUSH) and then update the flush_gen of dev_state in __btrfsic_submit_bio. So we should still submit it through integrity checker. Otherwise, the integrity checker will throw the following warning when I mount a newly created btrfs filesystem. [10264.755497] btrfs: attempt to write superblock which references block M @29523968 (sdb1/1111654400/0) which is not flushed out of disk's write cache (block flush_gen=1, dev->flush_gen=0)! [10264.755498] btrfs: attempt to write superblock which references block M @29523968 (sdb1/37912576/0) which is not flushed out of disk's write cache (block flush_gen=1, dev->flush_gen=0)! Signed-off-by: Lu Fengqi Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 3715478fffed..195634098f21 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3517,7 +3517,7 @@ static void write_dev_flush(struct btrfs_device *device) init_completion(&device->flush_wait); bio->bi_private = &device->flush_wait; - submit_bio(bio); + btrfsic_submit_bio(bio); device->flush_bio_sent = 1; } -- cgit v1.2.3 From 583b723151794e2ff1691f1510b4e43710293875 Mon Sep 17 00:00:00 2001 From: Hans van Kranenburg Date: Fri, 28 Jul 2017 08:31:28 +0200 Subject: btrfs: Do not use data_alloc_cluster in ssd mode This patch provides a band aid to improve the 'out of the box' behaviour of btrfs for disks that are detected as being an ssd. In a general purpose mixed workload scenario, the current ssd mode causes overallocation of available raw disk space for data, while leaving behind increasing amounts of unused fragmented free space. This situation leads to early ENOSPC problems which are harming user experience and adoption of btrfs as a general purpose filesystem. This patch modifies the data extent allocation behaviour of the ssd mode to make it behave identical to nossd mode. The metadata behaviour and additional ssd_spread option stay untouched so far. Recommendations for future development are to reconsider the current oversimplified nossd / ssd distinction and the broken detection mechanism based on the rotational attribute in sysfs and provide experienced users with a more flexible way to choose allocator behaviour for data and metadata, optimized for certain use cases, while keeping sane 'out of the box' default settings. The internals of the current btrfs code have more potential than what currently gets exposed to the user to choose from. The SSD story... In the first year of btrfs development, around early 2008, btrfs gained a mount option which enables specific functionality for filesystems on solid state devices. The first occurance of this functionality is in commit e18e4809, labeled "Add mount -o ssd, which includes optimizations for seek free storage". The effect on allocating free space for doing (data) writes is to 'cluster' writes together, writing them out in contiguous space, as opposed to a 'tetris' way of putting all separate writes into any free space fragment that fits (which is what the -o nossd behaviour does). A somewhat simplified explanation of what happens is that, when for example, the 'cluster' size is set to 2MiB, when we do some writes, the data allocator will search for a free space block that is 2MiB big, and put the writes in there. The ssd mode itself might allow a 2MiB cluster to be composed of multiple free space extents with some existing data in between, while the additional ssd_spread mount option kills off this option and requires fully free space. The idea behind this is (commit 536ac8ae): "The [...] clusters make it more likely a given IO will completely overwrite the ssd block, so it doesn't have to do an internal rwm cycle."; ssd block meaning nand erase block. So, effectively this means applying a "locality based algorithm" and trying to outsmart the actual ssd. Since then, various changes have been made to the involved code, but the basic idea is still present, and gets activated whenever the ssd mount option is active. This also happens by default, when the rotational flag as seen at /sys/block//queue/rotational is set to 0. However, there's a number of problems with this approach. First, what the optimization is trying to do is outsmart the ssd by assuming there is a relation between the physical address space of the block device as seen by btrfs and the actual physical storage of the ssd, and then adjusting data placement. However, since the introduction of the Flash Translation Layer (FTL) which is a part of the internal controller of an ssd, these attempts are futile. The use of good quality FTL in consumer ssd products might have been limited in 2008, but this situation has changed drastically soon after that time. Today, even the flash memory in your automatic cat feeding machine or your grandma's wheelchair has a full featured one. Second, the behaviour as described above results in the filesystem being filled up with badly fragmented free space extents because of relatively small pieces of space that are freed up by deletes, but not selected again as part of a 'cluster'. Since the algorithm prefers allocating a new chunk over going back to tetris mode, the end result is a filesystem in which all raw space is allocated, but which is composed of underutilized chunks with a 'shotgun blast' pattern of fragmented free space. Usually, the next problematic thing that happens is the filesystem wanting to allocate new space for metadata, which causes the filesystem to fail in spectacular ways. Third, the default mount options you get for an ssd ('ssd' mode enabled, 'discard' not enabled), in combination with spreading out writes over the full address space and ignoring freed up space leads to worst case behaviour in providing information to the ssd itself, since it will never learn that all the free space left behind is actually free. There are two ways to let an ssd know previously written data does not have to be preserved, which are sending explicit signals using discard or fstrim, or by simply overwriting the space with new data. The worst case behaviour is the btrfs ssd_spread mount option in combination with not having discard enabled. It has a side effect of minimizing the reuse of free space previously written in. Fourth, the rotational flag in /sys/ does not reliably indicate if the device is a locally attached ssd. For example, iSCSI or NBD displays as non-rotational, while a loop device on an ssd shows up as rotational. The combination of the second and third problem effectively means that despite all the good intentions, the btrfs ssd mode reliably causes the ssd hardware and the filesystem structures and performance to be choked to death. The clickbait version of the title of this story would have been "Btrfs ssd optimizations considered harmful for ssds". The current nossd 'tetris' mode (even still without discard) allows a pattern of overwriting much more previously used space, causing many more implicit discards to happen because of the overwrite information the ssd gets. The actual location in the physical address space, as seen from the point of view of btrfs is irrelevant, because the actual writes to the low level flash are reordered anyway thanks to the FTL. Changes made in the code 1. Make ssd mode data allocation identical to tetris mode, like nossd. 2. Adjust and clean up filesystem mount messages so that we can easily identify if a kernel has this patch applied or not, when providing support to end users. Also, make better use of the *_and_info helpers to only trigger messages on actual state changes. Backporting notes Notes for whoever wants to backport this patch to their 4.9 LTS kernel: * First apply commit 951e7966 "btrfs: drop the nossd flag when remounting with -o ssd", or fixup the differences manually. * The rest of the conflicts are because of the fs_info refactoring. So, for example, instead of using fs_info, it's root->fs_info in extent-tree.c Signed-off-by: Hans van Kranenburg Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 6 +++--- fs/btrfs/disk-io.c | 6 ++---- fs/btrfs/extent-tree.c | 11 ++++++----- fs/btrfs/super.c | 16 +++++++++------- 4 files changed, 20 insertions(+), 19 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index a3db2b1738aa..02edcddbcc9c 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -470,8 +470,8 @@ struct btrfs_block_rsv { /* * free clusters are used to claim free space in relatively large chunks, - * allowing us to do less seeky writes. They are used for all metadata - * allocations and data allocations in ssd mode. + * allowing us to do less seeky writes. They are used for all metadata + * allocations. In ssd_spread mode they are also used for data allocations. */ struct btrfs_free_cluster { spinlock_t lock; @@ -967,7 +967,7 @@ struct btrfs_fs_info { struct reloc_control *reloc_ctl; - /* data_alloc_cluster is only used in ssd mode */ + /* data_alloc_cluster is only used in ssd_spread mode */ struct btrfs_free_cluster data_alloc_cluster; /* all metadata allocations go through this cluster */ diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 195634098f21..90b967ae46d0 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3053,11 +3053,9 @@ retry_root_backup: if (IS_ERR(fs_info->transaction_kthread)) goto fail_cleaner; - if (!btrfs_test_opt(fs_info, SSD) && - !btrfs_test_opt(fs_info, NOSSD) && + if (!btrfs_test_opt(fs_info, NOSSD) && !fs_info->fs_devices->rotating) { - btrfs_info(fs_info, "detected SSD devices, enabling SSD mode"); - btrfs_set_opt(fs_info->mount_opt, SSD); + btrfs_set_and_info(fs_info, SSD, "enabling ssd optimizations"); } /* diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index eff674bfd162..6a8da7c19182 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -6654,19 +6654,20 @@ fetch_cluster_info(struct btrfs_fs_info *fs_info, struct btrfs_space_info *space_info, u64 *empty_cluster) { struct btrfs_free_cluster *ret = NULL; - bool ssd = btrfs_test_opt(fs_info, SSD); *empty_cluster = 0; if (btrfs_mixed_space_info(space_info)) return ret; - if (ssd) - *empty_cluster = SZ_2M; if (space_info->flags & BTRFS_BLOCK_GROUP_METADATA) { ret = &fs_info->meta_alloc_cluster; - if (!ssd) + if (btrfs_test_opt(fs_info, SSD)) + *empty_cluster = SZ_2M; + else *empty_cluster = SZ_64K; - } else if ((space_info->flags & BTRFS_BLOCK_GROUP_DATA) && ssd) { + } else if ((space_info->flags & BTRFS_BLOCK_GROUP_DATA) && + btrfs_test_opt(fs_info, SSD_SPREAD)) { + *empty_cluster = SZ_2M; ret = &fs_info->data_alloc_cluster; } diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 8a9bcad3b06a..0b7a1d8cd08b 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -549,20 +549,22 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, break; case Opt_ssd: btrfs_set_and_info(info, SSD, - "use ssd allocation scheme"); + "enabling ssd optimizations"); btrfs_clear_opt(info->mount_opt, NOSSD); break; case Opt_ssd_spread: + btrfs_set_and_info(info, SSD, + "enabling ssd optimizations"); btrfs_set_and_info(info, SSD_SPREAD, - "use spread ssd allocation scheme"); - btrfs_set_opt(info->mount_opt, SSD); + "using spread ssd allocation scheme"); btrfs_clear_opt(info->mount_opt, NOSSD); break; case Opt_nossd: - btrfs_set_and_info(info, NOSSD, - "not using ssd allocation scheme"); - btrfs_clear_opt(info->mount_opt, SSD); - btrfs_clear_opt(info->mount_opt, SSD_SPREAD); + btrfs_set_opt(info->mount_opt, NOSSD); + btrfs_clear_and_info(info, SSD, + "not using ssd optimizations"); + btrfs_clear_and_info(info, SSD_SPREAD, + "not using spread ssd allocation scheme"); break; case Opt_barrier: btrfs_clear_and_info(info, NOBARRIER, -- cgit v1.2.3 From db95c876c568cef951fbbd4c0118cb5386e4bb99 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Fri, 18 Aug 2017 18:16:25 +0200 Subject: btrfs: submit superblock io with REQ_META and REQ_PRIO The superblock is also metadata of the filesystem so the relevant IO should be tagged as such. We also tag it as high priority, as it's the last block committed for metadata from a given transaction. Any delays would effectively block the whole transaction, also blocking any other operation holding the device_list_mutex. Reviewed-by: Josef Bacik Reviewed-by: Liu Bo Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 90b967ae46d0..27d458640536 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3433,9 +3433,10 @@ static int write_dev_supers(struct btrfs_device *device, */ if (i == 0) { ret = btrfsic_submit_bh(REQ_OP_WRITE, - REQ_SYNC | REQ_FUA, bh); + REQ_SYNC | REQ_FUA | REQ_META | REQ_PRIO, bh); } else { - ret = btrfsic_submit_bh(REQ_OP_WRITE, REQ_SYNC, bh); + ret = btrfsic_submit_bh(REQ_OP_WRITE, + REQ_SYNC | REQ_META | REQ_PRIO, bh); } if (ret) errors++; -- cgit v1.2.3 From 74d46992e0d9dee7f1f376de0d56d31614c8a17a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 23 Aug 2017 19:10:32 +0200 Subject: block: replace bi_bdev with a gendisk pointer and partitions index This way we don't need a block_device structure to submit I/O. The block_device has different life time rules from the gendisk and request_queue and is usually only available when the block device node is open. Other callers need to explicitly create one (e.g. the lightnvm passthrough code, or the new nvme multipathing code). For the actual I/O path all that we need is the gendisk, which exists once per block device. But given that the block layer also does partition remapping we additionally need a partition index, which is used for said remapping in generic_make_request. Note that all the block drivers generally want request_queue or sometimes the gendisk, so this removes a layer of indirection all over the stack. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- arch/powerpc/sysdev/axonram.c | 2 +- block/bio-integrity.c | 18 +++---- block/bio.c | 10 ++-- block/blk-core.c | 100 ++++++++++++++++++------------------ block/blk-flush.c | 2 +- block/blk-lib.c | 8 +-- block/blk-merge.c | 2 +- block/blk-zoned.c | 4 +- drivers/block/brd.c | 5 +- drivers/block/drbd/drbd_actlog.c | 2 +- drivers/block/drbd/drbd_bitmap.c | 2 +- drivers/block/drbd/drbd_int.h | 4 +- drivers/block/drbd/drbd_receiver.c | 4 +- drivers/block/drbd/drbd_req.c | 2 +- drivers/block/drbd/drbd_worker.c | 2 +- drivers/block/floppy.c | 2 +- drivers/block/pktcdvd.c | 11 ++-- drivers/block/xen-blkback/blkback.c | 4 +- drivers/md/bcache/debug.c | 2 +- drivers/md/bcache/io.c | 2 +- drivers/md/bcache/journal.c | 6 +-- drivers/md/bcache/request.c | 16 +++--- drivers/md/bcache/super.c | 6 +-- drivers/md/bcache/writeback.c | 5 +- drivers/md/dm-bio-record.h | 9 ++-- drivers/md/dm-bufio.c | 2 +- drivers/md/dm-cache-target.c | 4 +- drivers/md/dm-crypt.c | 4 +- drivers/md/dm-delay.c | 4 +- drivers/md/dm-era-target.c | 2 +- drivers/md/dm-flakey.c | 2 +- drivers/md/dm-integrity.c | 11 ++-- drivers/md/dm-io.c | 2 +- drivers/md/dm-linear.c | 2 +- drivers/md/dm-log-writes.c | 8 +-- drivers/md/dm-mpath.c | 2 +- drivers/md/dm-raid1.c | 12 ++--- drivers/md/dm-snap.c | 16 +++--- drivers/md/dm-stripe.c | 10 ++-- drivers/md/dm-switch.c | 2 +- drivers/md/dm-thin.c | 6 +-- drivers/md/dm-verity-target.c | 2 +- drivers/md/dm-zoned-metadata.c | 6 +-- drivers/md/dm-zoned-target.c | 4 +- drivers/md/dm.c | 10 ++-- drivers/md/faulty.c | 4 +- drivers/md/linear.c | 6 +-- drivers/md/md.c | 10 ++-- drivers/md/md.h | 9 +++- drivers/md/multipath.c | 8 +-- drivers/md/raid0.c | 7 ++- drivers/md/raid1.c | 34 ++++++------ drivers/md/raid10.c | 50 +++++++++--------- drivers/md/raid5-cache.c | 6 +-- drivers/md/raid5-ppl.c | 6 +-- drivers/md/raid5.c | 12 ++--- drivers/nvdimm/nd.h | 4 +- drivers/nvme/host/core.c | 11 +--- drivers/nvme/host/lightnvm.c | 15 +----- drivers/nvme/target/io-cmd.c | 6 +-- drivers/s390/block/dcssblk.c | 4 +- drivers/s390/block/xpram.c | 2 +- drivers/target/target_core_iblock.c | 4 +- fs/block_dev.c | 4 +- fs/btrfs/check-integrity.c | 12 ++--- fs/btrfs/disk-io.c | 2 +- fs/btrfs/extent_io.c | 6 +-- fs/btrfs/raid56.c | 8 +-- fs/btrfs/scrub.c | 12 ++--- fs/btrfs/volumes.c | 2 +- fs/buffer.c | 4 +- fs/crypto/bio.c | 2 +- fs/direct-io.c | 8 +-- fs/exofs/ore.c | 2 +- fs/ext4/page-io.c | 4 +- fs/ext4/readpage.c | 2 +- fs/f2fs/data.c | 5 +- fs/f2fs/segment.c | 2 +- fs/gfs2/lops.c | 2 +- fs/gfs2/meta_io.c | 2 +- fs/gfs2/ops_fstype.c | 2 +- fs/hfsplus/wrapper.c | 2 +- fs/iomap.c | 4 +- fs/jfs/jfs_logmgr.c | 4 +- fs/jfs/jfs_metapage.c | 4 +- fs/mpage.c | 2 +- fs/nfs/blocklayout/blocklayout.c | 2 +- fs/nilfs2/segbuf.c | 2 +- fs/ocfs2/cluster/heartbeat.c | 2 +- fs/xfs/xfs_aops.c | 2 +- fs/xfs/xfs_buf.c | 2 +- include/linux/bio.h | 18 +++++++ include/linux/blk_types.h | 3 +- include/trace/events/bcache.h | 6 +-- include/trace/events/block.h | 16 +++--- include/trace/events/f2fs.h | 2 +- kernel/power/swap.c | 5 +- kernel/trace/blktrace.c | 2 +- mm/page_io.c | 17 +++--- 99 files changed, 358 insertions(+), 357 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/arch/powerpc/sysdev/axonram.c b/arch/powerpc/sysdev/axonram.c index 2799706106c6..1e15deacccaf 100644 --- a/arch/powerpc/sysdev/axonram.c +++ b/arch/powerpc/sysdev/axonram.c @@ -110,7 +110,7 @@ axon_ram_irq_handler(int irq, void *dev) static blk_qc_t axon_ram_make_request(struct request_queue *queue, struct bio *bio) { - struct axon_ram_bank *bank = bio->bi_bdev->bd_disk->private_data; + struct axon_ram_bank *bank = bio->bi_disk->private_data; unsigned long phys_mem, phys_end; void *user_mem; struct bio_vec vec; diff --git a/block/bio-integrity.c b/block/bio-integrity.c index 5fa9a740fd99..fc71e6172869 100644 --- a/block/bio-integrity.c +++ b/block/bio-integrity.c @@ -146,7 +146,7 @@ int bio_integrity_add_page(struct bio *bio, struct page *page, iv = bip->bip_vec + bip->bip_vcnt; if (bip->bip_vcnt && - bvec_gap_to_prev(bdev_get_queue(bio->bi_bdev), + bvec_gap_to_prev(bio->bi_disk->queue, &bip->bip_vec[bip->bip_vcnt - 1], offset)) return 0; @@ -190,7 +190,7 @@ static inline unsigned int bio_integrity_bytes(struct blk_integrity *bi, static blk_status_t bio_integrity_process(struct bio *bio, struct bvec_iter *proc_iter, integrity_processing_fn *proc_fn) { - struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev); + struct blk_integrity *bi = blk_get_integrity(bio->bi_disk); struct blk_integrity_iter iter; struct bvec_iter bviter; struct bio_vec bv; @@ -199,7 +199,7 @@ static blk_status_t bio_integrity_process(struct bio *bio, void *prot_buf = page_address(bip->bip_vec->bv_page) + bip->bip_vec->bv_offset; - iter.disk_name = bio->bi_bdev->bd_disk->disk_name; + iter.disk_name = bio->bi_disk->disk_name; iter.interval = 1 << bi->interval_exp; iter.seed = proc_iter->bi_sector; iter.prot_buf = prot_buf; @@ -236,8 +236,8 @@ static blk_status_t bio_integrity_process(struct bio *bio, bool bio_integrity_prep(struct bio *bio) { struct bio_integrity_payload *bip; - struct blk_integrity *bi; - struct request_queue *q; + struct blk_integrity *bi = blk_get_integrity(bio->bi_disk); + struct request_queue *q = bio->bi_disk->queue; void *buf; unsigned long start, end; unsigned int len, nr_pages; @@ -245,11 +245,9 @@ bool bio_integrity_prep(struct bio *bio) unsigned int intervals; blk_status_t status; - bi = bdev_get_integrity(bio->bi_bdev); if (!bi) return true; - q = bdev_get_queue(bio->bi_bdev); if (bio_op(bio) != REQ_OP_READ && bio_op(bio) != REQ_OP_WRITE) return true; @@ -354,7 +352,7 @@ static void bio_integrity_verify_fn(struct work_struct *work) struct bio_integrity_payload *bip = container_of(work, struct bio_integrity_payload, bip_work); struct bio *bio = bip->bip_bio; - struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev); + struct blk_integrity *bi = blk_get_integrity(bio->bi_disk); struct bvec_iter iter = bio->bi_iter; /* @@ -411,7 +409,7 @@ bool __bio_integrity_endio(struct bio *bio) void bio_integrity_advance(struct bio *bio, unsigned int bytes_done) { struct bio_integrity_payload *bip = bio_integrity(bio); - struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev); + struct blk_integrity *bi = blk_get_integrity(bio->bi_disk); unsigned bytes = bio_integrity_bytes(bi, bytes_done >> 9); bip->bip_iter.bi_sector += bytes_done >> 9; @@ -428,7 +426,7 @@ EXPORT_SYMBOL(bio_integrity_advance); void bio_integrity_trim(struct bio *bio) { struct bio_integrity_payload *bip = bio_integrity(bio); - struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev); + struct blk_integrity *bi = blk_get_integrity(bio->bi_disk); bip->bip_iter.bi_size = bio_integrity_bytes(bi, bio_sectors(bio)); } diff --git a/block/bio.c b/block/bio.c index ecd1a9c7a301..6745759028da 100644 --- a/block/bio.c +++ b/block/bio.c @@ -593,10 +593,10 @@ void __bio_clone_fast(struct bio *bio, struct bio *bio_src) BUG_ON(bio->bi_pool && BVEC_POOL_IDX(bio)); /* - * most users will be overriding ->bi_bdev with a new target, + * most users will be overriding ->bi_disk with a new target, * so we don't set nor calculate new physical/hw segment counts here */ - bio->bi_bdev = bio_src->bi_bdev; + bio->bi_disk = bio_src->bi_disk; bio_set_flag(bio, BIO_CLONED); bio->bi_opf = bio_src->bi_opf; bio->bi_write_hint = bio_src->bi_write_hint; @@ -681,7 +681,7 @@ struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, bio = bio_alloc_bioset(gfp_mask, bio_segments(bio_src), bs); if (!bio) return NULL; - bio->bi_bdev = bio_src->bi_bdev; + bio->bi_disk = bio_src->bi_disk; bio->bi_opf = bio_src->bi_opf; bio->bi_write_hint = bio_src->bi_write_hint; bio->bi_iter.bi_sector = bio_src->bi_iter.bi_sector; @@ -1830,8 +1830,8 @@ again: goto again; } - if (bio->bi_bdev && bio_flagged(bio, BIO_TRACE_COMPLETION)) { - trace_block_bio_complete(bdev_get_queue(bio->bi_bdev), bio, + if (bio->bi_disk && bio_flagged(bio, BIO_TRACE_COMPLETION)) { + trace_block_bio_complete(bio->bi_disk->queue, bio, blk_status_to_errno(bio->bi_status)); bio_clear_flag(bio, BIO_TRACE_COMPLETION); } diff --git a/block/blk-core.c b/block/blk-core.c index d579501f24ba..fc1af9097dff 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -1910,40 +1910,15 @@ out_unlock: return BLK_QC_T_NONE; } -/* - * If bio->bi_dev is a partition, remap the location - */ -static inline void blk_partition_remap(struct bio *bio) -{ - struct block_device *bdev = bio->bi_bdev; - - /* - * Zone reset does not include bi_size so bio_sectors() is always 0. - * Include a test for the reset op code and perform the remap if needed. - */ - if (bdev != bdev->bd_contains && - (bio_sectors(bio) || bio_op(bio) == REQ_OP_ZONE_RESET)) { - struct hd_struct *p = bdev->bd_part; - - bio->bi_iter.bi_sector += p->start_sect; - bio->bi_bdev = bdev->bd_contains; - - trace_block_bio_remap(bdev_get_queue(bio->bi_bdev), bio, - bdev->bd_dev, - bio->bi_iter.bi_sector - p->start_sect); - } -} - static void handle_bad_sector(struct bio *bio) { char b[BDEVNAME_SIZE]; printk(KERN_INFO "attempt to access beyond end of device\n"); printk(KERN_INFO "%s: rw=%d, want=%Lu, limit=%Lu\n", - bdevname(bio->bi_bdev, b), - bio->bi_opf, + bio_devname(bio, b), bio->bi_opf, (unsigned long long)bio_end_sector(bio), - (long long)(i_size_read(bio->bi_bdev->bd_inode) >> 9)); + (long long)get_capacity(bio->bi_disk)); } #ifdef CONFIG_FAIL_MAKE_REQUEST @@ -1981,6 +1956,38 @@ static inline bool should_fail_request(struct hd_struct *part, #endif /* CONFIG_FAIL_MAKE_REQUEST */ +/* + * Remap block n of partition p to block n+start(p) of the disk. + */ +static inline int blk_partition_remap(struct bio *bio) +{ + struct hd_struct *p; + int ret = 0; + + /* + * Zone reset does not include bi_size so bio_sectors() is always 0. + * Include a test for the reset op code and perform the remap if needed. + */ + if (!bio->bi_partno || + (!bio_sectors(bio) && bio_op(bio) != REQ_OP_ZONE_RESET)) + return 0; + + rcu_read_lock(); + p = __disk_get_part(bio->bi_disk, bio->bi_partno); + if (likely(p && !should_fail_request(p, bio->bi_iter.bi_size))) { + bio->bi_iter.bi_sector += p->start_sect; + bio->bi_partno = 0; + trace_block_bio_remap(bio->bi_disk->queue, bio, part_devt(p), + bio->bi_iter.bi_sector - p->start_sect); + } else { + printk("%s: fail for partition %d\n", __func__, bio->bi_partno); + ret = -EIO; + } + rcu_read_unlock(); + + return ret; +} + /* * Check whether this bio extends beyond the end of the device. */ @@ -1992,7 +1999,7 @@ static inline int bio_check_eod(struct bio *bio, unsigned int nr_sectors) return 0; /* Test device or partition size, when known. */ - maxsector = i_size_read(bio->bi_bdev->bd_inode) >> 9; + maxsector = get_capacity(bio->bi_disk); if (maxsector) { sector_t sector = bio->bi_iter.bi_sector; @@ -2017,20 +2024,18 @@ generic_make_request_checks(struct bio *bio) int nr_sectors = bio_sectors(bio); blk_status_t status = BLK_STS_IOERR; char b[BDEVNAME_SIZE]; - struct hd_struct *part; might_sleep(); if (bio_check_eod(bio, nr_sectors)) goto end_io; - q = bdev_get_queue(bio->bi_bdev); + q = bio->bi_disk->queue; if (unlikely(!q)) { printk(KERN_ERR "generic_make_request: Trying to access " "nonexistent block-device %s (%Lu)\n", - bdevname(bio->bi_bdev, b), - (long long) bio->bi_iter.bi_sector); + bio_devname(bio, b), (long long)bio->bi_iter.bi_sector); goto end_io; } @@ -2042,17 +2047,11 @@ generic_make_request_checks(struct bio *bio) if ((bio->bi_opf & REQ_NOWAIT) && !queue_is_rq_based(q)) goto not_supported; - part = bio->bi_bdev->bd_part; - if (should_fail_request(part, bio->bi_iter.bi_size) || - should_fail_request(&part_to_disk(part)->part0, - bio->bi_iter.bi_size)) + if (should_fail_request(&bio->bi_disk->part0, bio->bi_iter.bi_size)) goto end_io; - /* - * If this device has partitions, remap block n - * of partition p to block n+start(p) of the disk. - */ - blk_partition_remap(bio); + if (blk_partition_remap(bio)) + goto end_io; if (bio_check_eod(bio, nr_sectors)) goto end_io; @@ -2081,16 +2080,16 @@ generic_make_request_checks(struct bio *bio) goto not_supported; break; case REQ_OP_WRITE_SAME: - if (!bdev_write_same(bio->bi_bdev)) + if (!q->limits.max_write_same_sectors) goto not_supported; break; case REQ_OP_ZONE_REPORT: case REQ_OP_ZONE_RESET: - if (!bdev_is_zoned(bio->bi_bdev)) + if (!blk_queue_is_zoned(q)) goto not_supported; break; case REQ_OP_WRITE_ZEROES: - if (!bdev_write_zeroes_sectors(bio->bi_bdev)) + if (!q->limits.max_write_zeroes_sectors) goto not_supported; break; default: @@ -2197,7 +2196,7 @@ blk_qc_t generic_make_request(struct bio *bio) bio_list_init(&bio_list_on_stack[0]); current->bio_list = bio_list_on_stack; do { - struct request_queue *q = bdev_get_queue(bio->bi_bdev); + struct request_queue *q = bio->bi_disk->queue; if (likely(blk_queue_enter(q, bio->bi_opf & REQ_NOWAIT) == 0)) { struct bio_list lower, same; @@ -2215,7 +2214,7 @@ blk_qc_t generic_make_request(struct bio *bio) bio_list_init(&lower); bio_list_init(&same); while ((bio = bio_list_pop(&bio_list_on_stack[0])) != NULL) - if (q == bdev_get_queue(bio->bi_bdev)) + if (q == bio->bi_disk->queue) bio_list_add(&same, bio); else bio_list_add(&lower, bio); @@ -2258,7 +2257,7 @@ blk_qc_t submit_bio(struct bio *bio) unsigned int count; if (unlikely(bio_op(bio) == REQ_OP_WRITE_SAME)) - count = bdev_logical_block_size(bio->bi_bdev) >> 9; + count = queue_logical_block_size(bio->bi_disk->queue); else count = bio_sectors(bio); @@ -2275,8 +2274,7 @@ blk_qc_t submit_bio(struct bio *bio) current->comm, task_pid_nr(current), op_is_write(bio_op(bio)) ? "WRITE" : "READ", (unsigned long long)bio->bi_iter.bi_sector, - bdevname(bio->bi_bdev, b), - count); + bio_devname(bio, b), count); } } @@ -3049,8 +3047,8 @@ void blk_rq_bio_prep(struct request_queue *q, struct request *rq, rq->__data_len = bio->bi_iter.bi_size; rq->bio = rq->biotail = bio; - if (bio->bi_bdev) - rq->rq_disk = bio->bi_bdev->bd_disk; + if (bio->bi_disk) + rq->rq_disk = bio->bi_disk; } #if ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE diff --git a/block/blk-flush.c b/block/blk-flush.c index ed5fe322abba..83b7d5b41c79 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -525,7 +525,7 @@ int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask, return -ENXIO; bio = bio_alloc(gfp_mask, 0); - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH; ret = submit_bio_wait(bio); diff --git a/block/blk-lib.c b/block/blk-lib.c index 3fe0aec90597..e01adb5145b3 100644 --- a/block/blk-lib.c +++ b/block/blk-lib.c @@ -77,7 +77,7 @@ int __blkdev_issue_discard(struct block_device *bdev, sector_t sector, bio = next_bio(bio, 0, gfp_mask); bio->bi_iter.bi_sector = sector; - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio_set_op_attrs(bio, op, 0); bio->bi_iter.bi_size = req_sects << 9; @@ -168,7 +168,7 @@ static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector, while (nr_sects) { bio = next_bio(bio, 1, gfp_mask); bio->bi_iter.bi_sector = sector; - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_vcnt = 1; bio->bi_io_vec->bv_page = page; bio->bi_io_vec->bv_offset = 0; @@ -241,7 +241,7 @@ static int __blkdev_issue_write_zeroes(struct block_device *bdev, while (nr_sects) { bio = next_bio(bio, 0, gfp_mask); bio->bi_iter.bi_sector = sector; - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_opf = REQ_OP_WRITE_ZEROES; if (flags & BLKDEV_ZERO_NOUNMAP) bio->bi_opf |= REQ_NOUNMAP; @@ -323,7 +323,7 @@ int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector, bio = next_bio(bio, __blkdev_sectors_to_bio_pages(nr_sects), gfp_mask); bio->bi_iter.bi_sector = sector; - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio_set_op_attrs(bio, REQ_OP_WRITE, 0); while (nr_sects != 0) { diff --git a/block/blk-merge.c b/block/blk-merge.c index 05f116bfb99d..aa524cad5bea 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -786,7 +786,7 @@ bool blk_rq_merge_ok(struct request *rq, struct bio *bio) return false; /* must be same device and not a special request */ - if (rq->rq_disk != bio->bi_bdev->bd_disk || req_no_special_merge(rq)) + if (rq->rq_disk != bio->bi_disk || req_no_special_merge(rq)) return false; /* only merge integrity protected bio into ditto rq */ diff --git a/block/blk-zoned.c b/block/blk-zoned.c index 3bd15d8095b1..ff57fb51b338 100644 --- a/block/blk-zoned.c +++ b/block/blk-zoned.c @@ -116,7 +116,7 @@ int blkdev_report_zones(struct block_device *bdev, if (!bio) return -ENOMEM; - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_iter.bi_sector = blk_zone_start(q, sector); bio_set_op_attrs(bio, REQ_OP_ZONE_REPORT, 0); @@ -234,7 +234,7 @@ int blkdev_reset_zones(struct block_device *bdev, bio = bio_alloc(gfp_mask, 0); bio->bi_iter.bi_sector = sector; - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio_set_op_attrs(bio, REQ_OP_ZONE_RESET, 0); ret = submit_bio_wait(bio); diff --git a/drivers/block/brd.c b/drivers/block/brd.c index 104b71c0490d..006e1cb7e6f0 100644 --- a/drivers/block/brd.c +++ b/drivers/block/brd.c @@ -294,14 +294,13 @@ out: static blk_qc_t brd_make_request(struct request_queue *q, struct bio *bio) { - struct block_device *bdev = bio->bi_bdev; - struct brd_device *brd = bdev->bd_disk->private_data; + struct brd_device *brd = bio->bi_disk->private_data; struct bio_vec bvec; sector_t sector; struct bvec_iter iter; sector = bio->bi_iter.bi_sector; - if (bio_end_sector(bio) > get_capacity(bdev->bd_disk)) + if (bio_end_sector(bio) > get_capacity(bio->bi_disk)) goto io_error; bio_for_each_segment(bvec, bio, iter) { diff --git a/drivers/block/drbd/drbd_actlog.c b/drivers/block/drbd/drbd_actlog.c index e02c45cd3c5a..5f0eaee8c8a7 100644 --- a/drivers/block/drbd/drbd_actlog.c +++ b/drivers/block/drbd/drbd_actlog.c @@ -151,7 +151,7 @@ static int _drbd_md_sync_page_io(struct drbd_device *device, op_flags |= REQ_SYNC; bio = bio_alloc_drbd(GFP_NOIO); - bio->bi_bdev = bdev->md_bdev; + bio_set_dev(bio, bdev->md_bdev); bio->bi_iter.bi_sector = sector; err = -EIO; if (bio_add_page(bio, device->md_io.page, size, 0) != size) diff --git a/drivers/block/drbd/drbd_bitmap.c b/drivers/block/drbd/drbd_bitmap.c index 809fd245c3dc..bd97908c766f 100644 --- a/drivers/block/drbd/drbd_bitmap.c +++ b/drivers/block/drbd/drbd_bitmap.c @@ -1019,7 +1019,7 @@ static void bm_page_io_async(struct drbd_bm_aio_ctx *ctx, int page_nr) __must_ho bm_store_page_idx(page, page_nr); } else page = b->bm_pages[page_nr]; - bio->bi_bdev = device->ldev->md_bdev; + bio_set_dev(bio, device->ldev->md_bdev); bio->bi_iter.bi_sector = on_disk_sector; /* bio_add_page of a single page to an empty bio will always succeed, * according to api. Do we want to assert that? */ diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h index d17b6e6393c7..819f9d0bc875 100644 --- a/drivers/block/drbd/drbd_int.h +++ b/drivers/block/drbd/drbd_int.h @@ -1628,8 +1628,8 @@ static inline void drbd_generic_make_request(struct drbd_device *device, int fault_type, struct bio *bio) { __release(local); - if (!bio->bi_bdev) { - drbd_err(device, "drbd_generic_make_request: bio->bi_bdev == NULL\n"); + if (!bio->bi_disk) { + drbd_err(device, "drbd_generic_make_request: bio->bi_disk == NULL\n"); bio->bi_status = BLK_STS_IOERR; bio_endio(bio); return; diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c index c7e95e6380fb..ece6e5d7dc3f 100644 --- a/drivers/block/drbd/drbd_receiver.c +++ b/drivers/block/drbd/drbd_receiver.c @@ -1265,7 +1265,7 @@ static void submit_one_flush(struct drbd_device *device, struct issue_flush_cont octx->device = device; octx->ctx = ctx; - bio->bi_bdev = device->ldev->backing_bdev; + bio_set_dev(bio, device->ldev->backing_bdev); bio->bi_private = octx; bio->bi_end_io = one_flush_endio; bio->bi_opf = REQ_OP_FLUSH | REQ_PREFLUSH; @@ -1548,7 +1548,7 @@ next_bio: } /* > peer_req->i.sector, unless this is the first bio */ bio->bi_iter.bi_sector = sector; - bio->bi_bdev = device->ldev->backing_bdev; + bio_set_dev(bio, device->ldev->backing_bdev); bio_set_op_attrs(bio, op, op_flags); bio->bi_private = peer_req; bio->bi_end_io = drbd_peer_request_endio; diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c index 8d6b5d137b5e..447c975f5481 100644 --- a/drivers/block/drbd/drbd_req.c +++ b/drivers/block/drbd/drbd_req.c @@ -1179,7 +1179,7 @@ drbd_submit_req_private_bio(struct drbd_request *req) else type = DRBD_FAULT_DT_RD; - bio->bi_bdev = device->ldev->backing_bdev; + bio_set_dev(bio, device->ldev->backing_bdev); /* State may have changed since we grabbed our reference on the * ->ldev member. Double check, and short-circuit to endio. diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c index 1d8726a8df34..c268d886c4f0 100644 --- a/drivers/block/drbd/drbd_worker.c +++ b/drivers/block/drbd/drbd_worker.c @@ -1513,7 +1513,7 @@ int w_restart_disk_io(struct drbd_work *w, int cancel) drbd_al_begin_io(device, &req->i); drbd_req_make_private_bio(req, req->master_bio); - req->private_bio->bi_bdev = device->ldev->backing_bdev; + bio_set_dev(req->private_bio, device->ldev->backing_bdev); generic_make_request(req->private_bio); return 0; diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 9c00f29e40c1..60c086a53609 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -4134,7 +4134,7 @@ static int __floppy_read_block_0(struct block_device *bdev, int drive) cbdata.drive = drive; bio_init(&bio, &bio_vec, 1); - bio.bi_bdev = bdev; + bio_set_dev(&bio, bdev); bio_add_page(&bio, page, size, 0); bio.bi_iter.bi_sector = 0; diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index 6b8b097abbb9..67974796c350 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -1028,7 +1028,7 @@ static void pkt_gather_data(struct pktcdvd_device *pd, struct packet_data *pkt) bio = pkt->r_bios[f]; bio_reset(bio); bio->bi_iter.bi_sector = pkt->sector + f * (CD_FRAMESIZE >> 9); - bio->bi_bdev = pd->bdev; + bio_set_dev(bio, pd->bdev); bio->bi_end_io = pkt_end_io_read; bio->bi_private = pkt; @@ -1122,7 +1122,7 @@ static int pkt_start_recovery(struct packet_data *pkt) pkt->sector = new_sector; bio_reset(pkt->bio); - pkt->bio->bi_bdev = pd->bdev; + bio_set_set(pkt->bio, pd->bdev); bio_set_op_attrs(pkt->bio, REQ_OP_WRITE, 0); pkt->bio->bi_iter.bi_sector = new_sector; pkt->bio->bi_iter.bi_size = pkt->frames * CD_FRAMESIZE; @@ -1267,7 +1267,7 @@ static void pkt_start_write(struct pktcdvd_device *pd, struct packet_data *pkt) bio_reset(pkt->w_bio); pkt->w_bio->bi_iter.bi_sector = pkt->sector; - pkt->w_bio->bi_bdev = pd->bdev; + bio_set_dev(pkt->w_bio, pd->bdev); pkt->w_bio->bi_end_io = pkt_end_io_packet_write; pkt->w_bio->bi_private = pkt; @@ -2314,7 +2314,7 @@ static void pkt_make_request_read(struct pktcdvd_device *pd, struct bio *bio) psd->pd = pd; psd->bio = bio; - cloned_bio->bi_bdev = pd->bdev; + bio_set_dev(cloned_bio, pd->bdev); cloned_bio->bi_private = psd; cloned_bio->bi_end_io = pkt_end_io_read_cloned; pd->stats.secs_r += bio_sectors(bio); @@ -2415,8 +2415,7 @@ static blk_qc_t pkt_make_request(struct request_queue *q, struct bio *bio) pd = q->queuedata; if (!pd) { - pr_err("%s incorrect request queue\n", - bdevname(bio->bi_bdev, b)); + pr_err("%s incorrect request queue\n", bio_devname(bio, b)); goto end_io; } diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index 5f3a813e7ae0..987d665e82de 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -1363,7 +1363,7 @@ static int dispatch_rw_block_io(struct xen_blkif_ring *ring, goto fail_put_bio; biolist[nbio++] = bio; - bio->bi_bdev = preq.bdev; + bio_set_dev(bio, preq.bdev); bio->bi_private = pending_req; bio->bi_end_io = end_block_io_op; bio->bi_iter.bi_sector = preq.sector_number; @@ -1382,7 +1382,7 @@ static int dispatch_rw_block_io(struct xen_blkif_ring *ring, goto fail_put_bio; biolist[nbio++] = bio; - bio->bi_bdev = preq.bdev; + bio_set_dev(bio, preq.bdev); bio->bi_private = pending_req; bio->bi_end_io = end_block_io_op; bio_set_op_attrs(bio, operation, operation_flags); diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c index 35a5a7210e51..61076eda2e6d 100644 --- a/drivers/md/bcache/debug.c +++ b/drivers/md/bcache/debug.c @@ -49,7 +49,7 @@ void bch_btree_verify(struct btree *b) v->keys.ops = b->keys.ops; bio = bch_bbio_alloc(b->c); - bio->bi_bdev = PTR_CACHE(b->c, &b->key, 0)->bdev; + bio_set_dev(bio, PTR_CACHE(b->c, &b->key, 0)->bdev); bio->bi_iter.bi_sector = PTR_OFFSET(&b->key, 0); bio->bi_iter.bi_size = KEY_SIZE(&v->key) << 9; bio->bi_opf = REQ_OP_READ | REQ_META; diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c index 6a9b85095e7b..7e871bdc0097 100644 --- a/drivers/md/bcache/io.c +++ b/drivers/md/bcache/io.c @@ -34,7 +34,7 @@ void __bch_submit_bbio(struct bio *bio, struct cache_set *c) struct bbio *b = container_of(bio, struct bbio, bio); bio->bi_iter.bi_sector = PTR_OFFSET(&b->key, 0); - bio->bi_bdev = PTR_CACHE(c, &b->key, 0)->bdev; + bio_set_dev(bio, PTR_CACHE(c, &b->key, 0)->bdev); b->submit_time_us = local_clock_us(); closure_bio_submit(bio, bio->bi_private); diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c index 0352d05e495c..7e1d1c3ba33a 100644 --- a/drivers/md/bcache/journal.c +++ b/drivers/md/bcache/journal.c @@ -53,7 +53,7 @@ reread: left = ca->sb.bucket_size - offset; bio_reset(bio); bio->bi_iter.bi_sector = bucket + offset; - bio->bi_bdev = ca->bdev; + bio_set_dev(bio, ca->bdev); bio->bi_iter.bi_size = len << 9; bio->bi_end_io = journal_read_endio; @@ -452,7 +452,7 @@ static void do_journal_discard(struct cache *ca) bio_set_op_attrs(bio, REQ_OP_DISCARD, 0); bio->bi_iter.bi_sector = bucket_to_sector(ca->set, ca->sb.d[ja->discard_idx]); - bio->bi_bdev = ca->bdev; + bio_set_dev(bio, ca->bdev); bio->bi_iter.bi_size = bucket_bytes(ca); bio->bi_end_io = journal_discard_endio; @@ -623,7 +623,7 @@ static void journal_write_unlocked(struct closure *cl) bio_reset(bio); bio->bi_iter.bi_sector = PTR_OFFSET(k, i); - bio->bi_bdev = ca->bdev; + bio_set_dev(bio, ca->bdev); bio->bi_iter.bi_size = sectors << 9; bio->bi_end_io = journal_write_endio; diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index 72eb97176403..0e1463d0c334 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -607,7 +607,7 @@ static void request_endio(struct bio *bio) static void bio_complete(struct search *s) { if (s->orig_bio) { - struct request_queue *q = bdev_get_queue(s->orig_bio->bi_bdev); + struct request_queue *q = s->orig_bio->bi_disk->queue; generic_end_io_acct(q, bio_data_dir(s->orig_bio), &s->d->disk->part0, s->start_time); @@ -735,7 +735,7 @@ static void cached_dev_read_done(struct closure *cl) if (s->iop.bio) { bio_reset(s->iop.bio); s->iop.bio->bi_iter.bi_sector = s->cache_miss->bi_iter.bi_sector; - s->iop.bio->bi_bdev = s->cache_miss->bi_bdev; + bio_copy_dev(s->iop.bio, s->cache_miss); s->iop.bio->bi_iter.bi_size = s->insert_bio_sectors << 9; bch_bio_map(s->iop.bio, NULL); @@ -794,7 +794,7 @@ static int cached_dev_cache_miss(struct btree *b, struct search *s, !(bio->bi_opf & REQ_META) && s->iop.c->gc_stats.in_use < CUTOFF_CACHE_READA) reada = min_t(sector_t, dc->readahead >> 9, - bdev_sectors(bio->bi_bdev) - bio_end_sector(bio)); + get_capacity(bio->bi_disk) - bio_end_sector(bio)); s->insert_bio_sectors = min(sectors, bio_sectors(bio) + reada); @@ -820,7 +820,7 @@ static int cached_dev_cache_miss(struct btree *b, struct search *s, goto out_submit; cache_bio->bi_iter.bi_sector = miss->bi_iter.bi_sector; - cache_bio->bi_bdev = miss->bi_bdev; + bio_copy_dev(cache_bio, miss); cache_bio->bi_iter.bi_size = s->insert_bio_sectors << 9; cache_bio->bi_end_io = request_endio; @@ -919,7 +919,7 @@ static void cached_dev_write(struct cached_dev *dc, struct search *s) struct bio *flush = bio_alloc_bioset(GFP_NOIO, 0, dc->disk.bio_split); - flush->bi_bdev = bio->bi_bdev; + bio_copy_dev(flush, bio); flush->bi_end_io = request_endio; flush->bi_private = cl; flush->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH; @@ -956,13 +956,13 @@ static blk_qc_t cached_dev_make_request(struct request_queue *q, struct bio *bio) { struct search *s; - struct bcache_device *d = bio->bi_bdev->bd_disk->private_data; + struct bcache_device *d = bio->bi_disk->private_data; struct cached_dev *dc = container_of(d, struct cached_dev, disk); int rw = bio_data_dir(bio); generic_start_io_acct(q, rw, bio_sectors(bio), &d->disk->part0); - bio->bi_bdev = dc->bdev; + bio_set_dev(bio, dc->bdev); bio->bi_iter.bi_sector += dc->sb.data_offset; if (cached_dev_get(dc)) { @@ -1072,7 +1072,7 @@ static blk_qc_t flash_dev_make_request(struct request_queue *q, { struct search *s; struct closure *cl; - struct bcache_device *d = bio->bi_bdev->bd_disk->private_data; + struct bcache_device *d = bio->bi_disk->private_data; int rw = bio_data_dir(bio); generic_start_io_acct(q, rw, bio_sectors(bio), &d->disk->part0); diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index 8352fad765f6..974d832e54a6 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -257,7 +257,7 @@ void bch_write_bdev_super(struct cached_dev *dc, struct closure *parent) closure_init(cl, parent); bio_reset(bio); - bio->bi_bdev = dc->bdev; + bio_set_dev(bio, dc->bdev); bio->bi_end_io = write_bdev_super_endio; bio->bi_private = dc; @@ -303,7 +303,7 @@ void bcache_write_super(struct cache_set *c) SET_CACHE_SYNC(&ca->sb, CACHE_SYNC(&c->sb)); bio_reset(bio); - bio->bi_bdev = ca->bdev; + bio_set_dev(bio, ca->bdev); bio->bi_end_io = write_super_endio; bio->bi_private = ca; @@ -508,7 +508,7 @@ static void prio_io(struct cache *ca, uint64_t bucket, int op, closure_init_stack(cl); bio->bi_iter.bi_sector = bucket * ca->sb.bucket_size; - bio->bi_bdev = ca->bdev; + bio_set_dev(bio, ca->bdev); bio->bi_iter.bi_size = bucket_bytes(ca); bio->bi_end_io = prio_endio; diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c index 42c66e76f05e..c49022a8dc9d 100644 --- a/drivers/md/bcache/writeback.c +++ b/drivers/md/bcache/writeback.c @@ -181,7 +181,7 @@ static void write_dirty(struct closure *cl) dirty_init(w); bio_set_op_attrs(&io->bio, REQ_OP_WRITE, 0); io->bio.bi_iter.bi_sector = KEY_START(&w->key); - io->bio.bi_bdev = io->dc->bdev; + bio_set_dev(&io->bio, io->dc->bdev); io->bio.bi_end_io = dirty_endio; closure_bio_submit(&io->bio, cl); @@ -250,8 +250,7 @@ static void read_dirty(struct cached_dev *dc) dirty_init(w); bio_set_op_attrs(&io->bio, REQ_OP_READ, 0); io->bio.bi_iter.bi_sector = PTR_OFFSET(&w->key, 0); - io->bio.bi_bdev = PTR_CACHE(dc->disk.c, - &w->key, 0)->bdev; + bio_set_dev(&io->bio, PTR_CACHE(dc->disk.c, &w->key, 0)->bdev); io->bio.bi_end_io = read_dirty_endio; if (bio_alloc_pages(&io->bio, GFP_KERNEL)) diff --git a/drivers/md/dm-bio-record.h b/drivers/md/dm-bio-record.h index dd3646111561..c82578af56a5 100644 --- a/drivers/md/dm-bio-record.h +++ b/drivers/md/dm-bio-record.h @@ -18,21 +18,24 @@ */ struct dm_bio_details { - struct block_device *bi_bdev; + struct gendisk *bi_disk; + u8 bi_partno; unsigned long bi_flags; struct bvec_iter bi_iter; }; static inline void dm_bio_record(struct dm_bio_details *bd, struct bio *bio) { - bd->bi_bdev = bio->bi_bdev; + bd->bi_disk = bio->bi_disk; + bd->bi_partno = bio->bi_partno; bd->bi_flags = bio->bi_flags; bd->bi_iter = bio->bi_iter; } static inline void dm_bio_restore(struct dm_bio_details *bd, struct bio *bio) { - bio->bi_bdev = bd->bi_bdev; + bio->bi_disk = bd->bi_disk; + bio->bi_partno = bd->bi_partno; bio->bi_flags = bd->bi_flags; bio->bi_iter = bd->bi_iter; } diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c index 44f4a8ac95bd..9601225e0ae9 100644 --- a/drivers/md/dm-bufio.c +++ b/drivers/md/dm-bufio.c @@ -616,7 +616,7 @@ static void use_inline_bio(struct dm_buffer *b, int rw, sector_t sector, bio_init(&b->bio, b->bio_vec, DM_BUFIO_INLINE_VECS); b->bio.bi_iter.bi_sector = sector; - b->bio.bi_bdev = b->c->bdev; + bio_set_dev(&b->bio, b->c->bdev); b->bio.bi_end_io = inline_endio; /* * Use of .bi_private isn't a problem here because diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c index c5ea03fc7ee1..dcac25c2be7a 100644 --- a/drivers/md/dm-cache-target.c +++ b/drivers/md/dm-cache-target.c @@ -833,7 +833,7 @@ static bool is_discarded_oblock(struct cache *cache, dm_oblock_t b) *--------------------------------------------------------------*/ static void remap_to_origin(struct cache *cache, struct bio *bio) { - bio->bi_bdev = cache->origin_dev->bdev; + bio_set_dev(bio, cache->origin_dev->bdev); } static void remap_to_cache(struct cache *cache, struct bio *bio, @@ -842,7 +842,7 @@ static void remap_to_cache(struct cache *cache, struct bio *bio, sector_t bi_sector = bio->bi_iter.bi_sector; sector_t block = from_cblock(cblock); - bio->bi_bdev = cache->cache_dev->bdev; + bio_set_dev(bio, cache->cache_dev->bdev); if (!block_size_is_power_of_two(cache)) bio->bi_iter.bi_sector = (block * cache->sectors_per_block) + diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 73c2e270cda6..ca99147208a9 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -1544,7 +1544,7 @@ static void clone_init(struct dm_crypt_io *io, struct bio *clone) clone->bi_private = io; clone->bi_end_io = crypt_endio; - clone->bi_bdev = cc->dev->bdev; + bio_set_dev(clone, cc->dev->bdev); clone->bi_opf = io->base_bio->bi_opf; } @@ -2793,7 +2793,7 @@ static int crypt_map(struct dm_target *ti, struct bio *bio) */ if (unlikely(bio->bi_opf & REQ_PREFLUSH || bio_op(bio) == REQ_OP_DISCARD)) { - bio->bi_bdev = cc->dev->bdev; + bio_set_dev(bio, cc->dev->bdev); if (bio_sectors(bio)) bio->bi_iter.bi_sector = cc->start + dm_target_offset(ti, bio->bi_iter.bi_sector); diff --git a/drivers/md/dm-delay.c b/drivers/md/dm-delay.c index ae3158795d26..2209a9700acd 100644 --- a/drivers/md/dm-delay.c +++ b/drivers/md/dm-delay.c @@ -282,7 +282,7 @@ static int delay_map(struct dm_target *ti, struct bio *bio) struct delay_c *dc = ti->private; if ((bio_data_dir(bio) == WRITE) && (dc->dev_write)) { - bio->bi_bdev = dc->dev_write->bdev; + bio_set_dev(bio, dc->dev_write->bdev); if (bio_sectors(bio)) bio->bi_iter.bi_sector = dc->start_write + dm_target_offset(ti, bio->bi_iter.bi_sector); @@ -290,7 +290,7 @@ static int delay_map(struct dm_target *ti, struct bio *bio) return delay_bio(dc, dc->write_delay, bio); } - bio->bi_bdev = dc->dev_read->bdev; + bio_set_dev(bio, dc->dev_read->bdev); bio->bi_iter.bi_sector = dc->start_read + dm_target_offset(ti, bio->bi_iter.bi_sector); diff --git a/drivers/md/dm-era-target.c b/drivers/md/dm-era-target.c index e7ba89f98d8d..ba84b8d62cd0 100644 --- a/drivers/md/dm-era-target.c +++ b/drivers/md/dm-era-target.c @@ -1192,7 +1192,7 @@ static dm_block_t get_block(struct era *era, struct bio *bio) static void remap_to_origin(struct era *era, struct bio *bio) { - bio->bi_bdev = era->origin_dev->bdev; + bio_set_dev(bio, era->origin_dev->bdev); } /*---------------------------------------------------------------- diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c index e2c7234931bc..7146c2d9762d 100644 --- a/drivers/md/dm-flakey.c +++ b/drivers/md/dm-flakey.c @@ -274,7 +274,7 @@ static void flakey_map_bio(struct dm_target *ti, struct bio *bio) { struct flakey_c *fc = ti->private; - bio->bi_bdev = fc->dev->bdev; + bio_set_dev(bio, fc->dev->bdev); if (bio_sectors(bio) || bio_op(bio) == REQ_OP_ZONE_RESET) bio->bi_iter.bi_sector = flakey_map_sector(ti, bio->bi_iter.bi_sector); diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c index 3acce09bba35..27c0f223f8ea 100644 --- a/drivers/md/dm-integrity.c +++ b/drivers/md/dm-integrity.c @@ -250,7 +250,8 @@ struct dm_integrity_io { struct completion *completion; - struct block_device *orig_bi_bdev; + struct gendisk *orig_bi_disk; + u8 orig_bi_partno; bio_end_io_t *orig_bi_end_io; struct bio_integrity_payload *orig_bi_integrity; struct bvec_iter orig_bi_iter; @@ -1164,7 +1165,8 @@ static void integrity_end_io(struct bio *bio) struct dm_integrity_io *dio = dm_per_bio_data(bio, sizeof(struct dm_integrity_io)); bio->bi_iter = dio->orig_bi_iter; - bio->bi_bdev = dio->orig_bi_bdev; + bio->bi_disk = dio->orig_bi_disk; + bio->bi_partno = dio->orig_bi_partno; if (dio->orig_bi_integrity) { bio->bi_integrity = dio->orig_bi_integrity; bio->bi_opf |= REQ_INTEGRITY; @@ -1681,8 +1683,9 @@ sleep: dio->orig_bi_iter = bio->bi_iter; - dio->orig_bi_bdev = bio->bi_bdev; - bio->bi_bdev = ic->dev->bdev; + dio->orig_bi_disk = bio->bi_disk; + dio->orig_bi_partno = bio->bi_partno; + bio_set_dev(bio, ic->dev->bdev); dio->orig_bi_integrity = bio_integrity(bio); bio->bi_integrity = NULL; diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c index 25039607f3cb..b4357ed4d541 100644 --- a/drivers/md/dm-io.c +++ b/drivers/md/dm-io.c @@ -347,7 +347,7 @@ static void do_region(int op, int op_flags, unsigned region, bio = bio_alloc_bioset(GFP_NOIO, num_bvecs, io->client->bios); bio->bi_iter.bi_sector = where->sector + (where->count - remaining); - bio->bi_bdev = where->bdev; + bio_set_dev(bio, where->bdev); bio->bi_end_io = endio; bio_set_op_attrs(bio, op, op_flags); store_io_and_region_in_bio(bio, io, region); diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c index 41971a090e34..405eca206d67 100644 --- a/drivers/md/dm-linear.c +++ b/drivers/md/dm-linear.c @@ -88,7 +88,7 @@ static void linear_map_bio(struct dm_target *ti, struct bio *bio) { struct linear_c *lc = ti->private; - bio->bi_bdev = lc->dev->bdev; + bio_set_dev(bio, lc->dev->bdev); if (bio_sectors(bio) || bio_op(bio) == REQ_OP_ZONE_RESET) bio->bi_iter.bi_sector = linear_map_sector(ti, bio->bi_iter.bi_sector); diff --git a/drivers/md/dm-log-writes.c b/drivers/md/dm-log-writes.c index a1da0eb58a93..534a254eb977 100644 --- a/drivers/md/dm-log-writes.c +++ b/drivers/md/dm-log-writes.c @@ -198,7 +198,7 @@ static int write_metadata(struct log_writes_c *lc, void *entry, } bio->bi_iter.bi_size = 0; bio->bi_iter.bi_sector = sector; - bio->bi_bdev = lc->logdev->bdev; + bio_set_dev(bio, lc->logdev->bdev); bio->bi_end_io = log_end_io; bio->bi_private = lc; bio_set_op_attrs(bio, REQ_OP_WRITE, 0); @@ -263,7 +263,7 @@ static int log_one_block(struct log_writes_c *lc, } bio->bi_iter.bi_size = 0; bio->bi_iter.bi_sector = sector; - bio->bi_bdev = lc->logdev->bdev; + bio_set_dev(bio, lc->logdev->bdev); bio->bi_end_io = log_end_io; bio->bi_private = lc; bio_set_op_attrs(bio, REQ_OP_WRITE, 0); @@ -285,7 +285,7 @@ static int log_one_block(struct log_writes_c *lc, } bio->bi_iter.bi_size = 0; bio->bi_iter.bi_sector = sector; - bio->bi_bdev = lc->logdev->bdev; + bio_set_dev(bio, lc->logdev->bdev); bio->bi_end_io = log_end_io; bio->bi_private = lc; bio_set_op_attrs(bio, REQ_OP_WRITE, 0); @@ -539,7 +539,7 @@ static void normal_map_bio(struct dm_target *ti, struct bio *bio) { struct log_writes_c *lc = ti->private; - bio->bi_bdev = lc->dev->bdev; + bio_set_dev(bio, lc->dev->bdev); } static int log_writes_map(struct dm_target *ti, struct bio *bio) diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index 0e8ab5bb3575..573046bd5c46 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -566,7 +566,7 @@ static int __multipath_map_bio(struct multipath *m, struct bio *bio, struct dm_m mpio->nr_bytes = nr_bytes; bio->bi_status = 0; - bio->bi_bdev = pgpath->path.dev->bdev; + bio_set_dev(bio, pgpath->path.dev->bdev); bio->bi_opf |= REQ_FAILFAST_TRANSPORT; if (pgpath->pg->ps.type->start_io) diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index a4fbd911d566..c0b82136b2d1 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -145,7 +145,7 @@ static void dispatch_bios(void *context, struct bio_list *bio_list) struct dm_raid1_bio_record { struct mirror *m; - /* if details->bi_bdev == NULL, details were not saved */ + /* if details->bi_disk == NULL, details were not saved */ struct dm_bio_details details; region_t write_region; }; @@ -464,7 +464,7 @@ static sector_t map_sector(struct mirror *m, struct bio *bio) static void map_bio(struct mirror *m, struct bio *bio) { - bio->bi_bdev = m->dev->bdev; + bio_set_dev(bio, m->dev->bdev); bio->bi_iter.bi_sector = map_sector(m, bio); } @@ -1199,7 +1199,7 @@ static int mirror_map(struct dm_target *ti, struct bio *bio) struct dm_raid1_bio_record *bio_record = dm_per_bio_data(bio, sizeof(struct dm_raid1_bio_record)); - bio_record->details.bi_bdev = NULL; + bio_record->details.bi_disk = NULL; if (rw == WRITE) { /* Save region for mirror_end_io() handler */ @@ -1266,7 +1266,7 @@ static int mirror_end_io(struct dm_target *ti, struct bio *bio, goto out; if (unlikely(*error)) { - if (!bio_record->details.bi_bdev) { + if (!bio_record->details.bi_disk) { /* * There wasn't enough memory to record necessary * information for a retry or there was no other @@ -1291,7 +1291,7 @@ static int mirror_end_io(struct dm_target *ti, struct bio *bio, bd = &bio_record->details; dm_bio_restore(bd, bio); - bio_record->details.bi_bdev = NULL; + bio_record->details.bi_disk = NULL; bio->bi_status = 0; queue_bio(ms, bio, rw); @@ -1301,7 +1301,7 @@ static int mirror_end_io(struct dm_target *ti, struct bio *bio, } out: - bio_record->details.bi_bdev = NULL; + bio_record->details.bi_disk = NULL; return DM_ENDIO_DONE; } diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 1ba41048b438..1113b42e1eda 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -1663,7 +1663,7 @@ __find_pending_exception(struct dm_snapshot *s, static void remap_exception(struct dm_snapshot *s, struct dm_exception *e, struct bio *bio, chunk_t chunk) { - bio->bi_bdev = s->cow->bdev; + bio_set_dev(bio, s->cow->bdev); bio->bi_iter.bi_sector = chunk_to_sector(s->store, dm_chunk_number(e->new_chunk) + (chunk - e->old_chunk)) + @@ -1681,7 +1681,7 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio) init_tracked_chunk(bio); if (bio->bi_opf & REQ_PREFLUSH) { - bio->bi_bdev = s->cow->bdev; + bio_set_dev(bio, s->cow->bdev); return DM_MAPIO_REMAPPED; } @@ -1769,7 +1769,7 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio) goto out; } } else { - bio->bi_bdev = s->origin->bdev; + bio_set_dev(bio, s->origin->bdev); track_chunk(s, bio, chunk); } @@ -1802,9 +1802,9 @@ static int snapshot_merge_map(struct dm_target *ti, struct bio *bio) if (bio->bi_opf & REQ_PREFLUSH) { if (!dm_bio_get_target_bio_nr(bio)) - bio->bi_bdev = s->origin->bdev; + bio_set_dev(bio, s->origin->bdev); else - bio->bi_bdev = s->cow->bdev; + bio_set_dev(bio, s->cow->bdev); return DM_MAPIO_REMAPPED; } @@ -1824,7 +1824,7 @@ static int snapshot_merge_map(struct dm_target *ti, struct bio *bio) chunk >= s->first_merging_chunk && chunk < (s->first_merging_chunk + s->num_merging_chunks)) { - bio->bi_bdev = s->origin->bdev; + bio_set_dev(bio, s->origin->bdev); bio_list_add(&s->bios_queued_during_merge, bio); r = DM_MAPIO_SUBMITTED; goto out_unlock; @@ -1838,7 +1838,7 @@ static int snapshot_merge_map(struct dm_target *ti, struct bio *bio) } redirect_to_origin: - bio->bi_bdev = s->origin->bdev; + bio_set_dev(bio, s->origin->bdev); if (bio_data_dir(bio) == WRITE) { up_write(&s->lock); @@ -2285,7 +2285,7 @@ static int origin_map(struct dm_target *ti, struct bio *bio) struct dm_origin *o = ti->private; unsigned available_sectors; - bio->bi_bdev = o->dev->bdev; + bio_set_dev(bio, o->dev->bdev); if (unlikely(bio->bi_opf & REQ_PREFLUSH)) return DM_MAPIO_REMAPPED; diff --git a/drivers/md/dm-stripe.c b/drivers/md/dm-stripe.c index a0375530b07f..ab50d7c4377f 100644 --- a/drivers/md/dm-stripe.c +++ b/drivers/md/dm-stripe.c @@ -270,7 +270,7 @@ static int stripe_map_range(struct stripe_c *sc, struct bio *bio, stripe_map_range_sector(sc, bio_end_sector(bio), target_stripe, &end); if (begin < end) { - bio->bi_bdev = sc->stripe[target_stripe].dev->bdev; + bio_set_dev(bio, sc->stripe[target_stripe].dev->bdev); bio->bi_iter.bi_sector = begin + sc->stripe[target_stripe].physical_start; bio->bi_iter.bi_size = to_bytes(end - begin); @@ -291,7 +291,7 @@ static int stripe_map(struct dm_target *ti, struct bio *bio) if (bio->bi_opf & REQ_PREFLUSH) { target_bio_nr = dm_bio_get_target_bio_nr(bio); BUG_ON(target_bio_nr >= sc->stripes); - bio->bi_bdev = sc->stripe[target_bio_nr].dev->bdev; + bio_set_dev(bio, sc->stripe[target_bio_nr].dev->bdev); return DM_MAPIO_REMAPPED; } if (unlikely(bio_op(bio) == REQ_OP_DISCARD) || @@ -306,7 +306,7 @@ static int stripe_map(struct dm_target *ti, struct bio *bio) &stripe, &bio->bi_iter.bi_sector); bio->bi_iter.bi_sector += sc->stripe[stripe].physical_start; - bio->bi_bdev = sc->stripe[stripe].dev->bdev; + bio_set_dev(bio, sc->stripe[stripe].dev->bdev); return DM_MAPIO_REMAPPED; } @@ -430,9 +430,7 @@ static int stripe_end_io(struct dm_target *ti, struct bio *bio, return DM_ENDIO_DONE; memset(major_minor, 0, sizeof(major_minor)); - sprintf(major_minor, "%d:%d", - MAJOR(disk_devt(bio->bi_bdev->bd_disk)), - MINOR(disk_devt(bio->bi_bdev->bd_disk))); + sprintf(major_minor, "%d:%d", MAJOR(bio_dev(bio)), MINOR(bio_dev(bio))); /* * Test to see which stripe drive triggered the event diff --git a/drivers/md/dm-switch.c b/drivers/md/dm-switch.c index 871c18fe000d..2dcea4c56f37 100644 --- a/drivers/md/dm-switch.c +++ b/drivers/md/dm-switch.c @@ -322,7 +322,7 @@ static int switch_map(struct dm_target *ti, struct bio *bio) sector_t offset = dm_target_offset(ti, bio->bi_iter.bi_sector); unsigned path_nr = switch_get_path_nr(sctx, offset); - bio->bi_bdev = sctx->path_list[path_nr].dmdev->bdev; + bio_set_dev(bio, sctx->path_list[path_nr].dmdev->bdev); bio->bi_iter.bi_sector = sctx->path_list[path_nr].start + offset; return DM_MAPIO_REMAPPED; diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c index 9dec2f8cc739..69d88aee3055 100644 --- a/drivers/md/dm-thin.c +++ b/drivers/md/dm-thin.c @@ -679,7 +679,7 @@ static void remap(struct thin_c *tc, struct bio *bio, dm_block_t block) struct pool *pool = tc->pool; sector_t bi_sector = bio->bi_iter.bi_sector; - bio->bi_bdev = tc->pool_dev->bdev; + bio_set_dev(bio, tc->pool_dev->bdev); if (block_size_is_power_of_two(pool)) bio->bi_iter.bi_sector = (block << pool->sectors_per_block_shift) | @@ -691,7 +691,7 @@ static void remap(struct thin_c *tc, struct bio *bio, dm_block_t block) static void remap_to_origin(struct thin_c *tc, struct bio *bio) { - bio->bi_bdev = tc->origin_dev->bdev; + bio_set_dev(bio, tc->origin_dev->bdev); } static int bio_triggers_commit(struct thin_c *tc, struct bio *bio) @@ -3313,7 +3313,7 @@ static int pool_map(struct dm_target *ti, struct bio *bio) * As this is a singleton target, ti->begin is always zero. */ spin_lock_irqsave(&pool->lock, flags); - bio->bi_bdev = pt->data_dev->bdev; + bio_set_dev(bio, pt->data_dev->bdev); r = DM_MAPIO_REMAPPED; spin_unlock_irqrestore(&pool->lock, flags); diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c index b46705ebf01f..1c5b6185c79d 100644 --- a/drivers/md/dm-verity-target.c +++ b/drivers/md/dm-verity-target.c @@ -637,7 +637,7 @@ static int verity_map(struct dm_target *ti, struct bio *bio) struct dm_verity *v = ti->private; struct dm_verity_io *io; - bio->bi_bdev = v->data_dev->bdev; + bio_set_dev(bio, v->data_dev->bdev); bio->bi_iter.bi_sector = verity_map_sector(v, bio->bi_iter.bi_sector); if (((unsigned)bio->bi_iter.bi_sector | bio_sectors(bio)) & diff --git a/drivers/md/dm-zoned-metadata.c b/drivers/md/dm-zoned-metadata.c index a4fa2ada6883..70485de37b66 100644 --- a/drivers/md/dm-zoned-metadata.c +++ b/drivers/md/dm-zoned-metadata.c @@ -409,7 +409,7 @@ static struct dmz_mblock *dmz_fetch_mblock(struct dmz_metadata *zmd, } bio->bi_iter.bi_sector = dmz_blk2sect(block); - bio->bi_bdev = zmd->dev->bdev; + bio_set_dev(bio, zmd->dev->bdev); bio->bi_private = mblk; bio->bi_end_io = dmz_mblock_bio_end_io; bio_set_op_attrs(bio, REQ_OP_READ, REQ_META | REQ_PRIO); @@ -564,7 +564,7 @@ static void dmz_write_mblock(struct dmz_metadata *zmd, struct dmz_mblock *mblk, set_bit(DMZ_META_WRITING, &mblk->state); bio->bi_iter.bi_sector = dmz_blk2sect(block); - bio->bi_bdev = zmd->dev->bdev; + bio_set_dev(bio, zmd->dev->bdev); bio->bi_private = mblk; bio->bi_end_io = dmz_mblock_bio_end_io; bio_set_op_attrs(bio, REQ_OP_WRITE, REQ_META | REQ_PRIO); @@ -586,7 +586,7 @@ static int dmz_rdwr_block(struct dmz_metadata *zmd, int op, sector_t block, return -ENOMEM; bio->bi_iter.bi_sector = dmz_blk2sect(block); - bio->bi_bdev = zmd->dev->bdev; + bio_set_dev(bio, zmd->dev->bdev); bio_set_op_attrs(bio, op, REQ_SYNC | REQ_META | REQ_PRIO); bio_add_page(bio, page, DMZ_BLOCK_SIZE, 0); ret = submit_bio_wait(bio); diff --git a/drivers/md/dm-zoned-target.c b/drivers/md/dm-zoned-target.c index b08bbbd4d902..b87c1741da4b 100644 --- a/drivers/md/dm-zoned-target.c +++ b/drivers/md/dm-zoned-target.c @@ -238,7 +238,7 @@ static void dmz_submit_write_bio(struct dmz_target *dmz, struct dm_zone *zone, struct dmz_bioctx *bioctx = dm_per_bio_data(bio, sizeof(struct dmz_bioctx)); /* Setup and submit the BIO */ - bio->bi_bdev = dmz->dev->bdev; + bio_set_dev(bio, dmz->dev->bdev); bio->bi_iter.bi_sector = dmz_start_sect(dmz->metadata, zone) + dmz_blk2sect(chunk_block); atomic_inc(&bioctx->ref); generic_make_request(bio); @@ -586,7 +586,7 @@ static int dmz_map(struct dm_target *ti, struct bio *bio) (unsigned long long)dmz_chunk_block(dmz->dev, dmz_bio_block(bio)), (unsigned int)dmz_bio_blocks(bio)); - bio->bi_bdev = dev->bdev; + bio_set_dev(bio, dev->bdev); if (!nr_sectors && bio_op(bio) != REQ_OP_WRITE) return DM_MAPIO_REMAPPED; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 8612a2d1ccd9..b28b9ce8f4ff 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -851,10 +851,10 @@ static void clone_endio(struct bio *bio) if (unlikely(error == BLK_STS_TARGET)) { if (bio_op(bio) == REQ_OP_WRITE_SAME && - !bdev_get_queue(bio->bi_bdev)->limits.max_write_same_sectors) + !bio->bi_disk->queue->limits.max_write_same_sectors) disable_write_same(md); if (bio_op(bio) == REQ_OP_WRITE_ZEROES && - !bdev_get_queue(bio->bi_bdev)->limits.max_write_zeroes_sectors) + !bio->bi_disk->queue->limits.max_write_zeroes_sectors) disable_write_zeroes(md); } @@ -1215,8 +1215,8 @@ static void __map_bio(struct dm_target_io *tio) break; case DM_MAPIO_REMAPPED: /* the bio has been remapped so dispatch it */ - trace_block_bio_remap(bdev_get_queue(clone->bi_bdev), clone, - tio->io->bio->bi_bdev->bd_dev, sector); + trace_block_bio_remap(clone->bi_disk->queue, clone, + bio_dev(tio->io->bio), sector); generic_make_request(clone); break; case DM_MAPIO_KILL: @@ -1796,7 +1796,7 @@ static struct mapped_device *alloc_dev(int minor) goto bad; bio_init(&md->flush_bio, NULL, 0); - md->flush_bio.bi_bdev = md->bdev; + bio_set_dev(&md->flush_bio, md->bdev); md->flush_bio.bi_opf = REQ_OP_WRITE | REQ_PREFLUSH | REQ_SYNC; dm_stats_init(&md->stats); diff --git a/drivers/md/faulty.c b/drivers/md/faulty.c index 06a64d5d8c6c..38264b38420f 100644 --- a/drivers/md/faulty.c +++ b/drivers/md/faulty.c @@ -216,12 +216,12 @@ static bool faulty_make_request(struct mddev *mddev, struct bio *bio) if (failit) { struct bio *b = bio_clone_fast(bio, GFP_NOIO, mddev->bio_set); - b->bi_bdev = conf->rdev->bdev; + bio_set_dev(b, conf->rdev->bdev); b->bi_private = bio; b->bi_end_io = faulty_fail; bio = b; } else - bio->bi_bdev = conf->rdev->bdev; + bio_set_dev(bio, conf->rdev->bdev); generic_make_request(bio); return true; diff --git a/drivers/md/linear.c b/drivers/md/linear.c index 5f1eb9189542..c464fb48039a 100644 --- a/drivers/md/linear.c +++ b/drivers/md/linear.c @@ -275,17 +275,17 @@ static bool linear_make_request(struct mddev *mddev, struct bio *bio) bio = split; } - bio->bi_bdev = tmp_dev->rdev->bdev; + bio_set_dev(bio, tmp_dev->rdev->bdev); bio->bi_iter.bi_sector = bio->bi_iter.bi_sector - start_sector + data_offset; if (unlikely((bio_op(bio) == REQ_OP_DISCARD) && - !blk_queue_discard(bdev_get_queue(bio->bi_bdev)))) { + !blk_queue_discard(bio->bi_disk->queue))) { /* Just ignore it */ bio_endio(bio); } else { if (mddev->gendisk) - trace_block_bio_remap(bdev_get_queue(bio->bi_bdev), + trace_block_bio_remap(bio->bi_disk->queue, bio, disk_devt(mddev->gendisk), bio_sector); mddev_check_writesame(mddev, bio); diff --git a/drivers/md/md.c b/drivers/md/md.c index c99634612fc4..0afdc1bfd7cb 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -422,7 +422,7 @@ static void submit_flushes(struct work_struct *ws) bi = bio_alloc_mddev(GFP_NOIO, 0, mddev); bi->bi_end_io = md_end_flush; bi->bi_private = rdev; - bi->bi_bdev = rdev->bdev; + bio_set_dev(bi, rdev->bdev); bi->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH; atomic_inc(&mddev->flush_pending); submit_bio(bi); @@ -772,7 +772,7 @@ void md_super_write(struct mddev *mddev, struct md_rdev *rdev, atomic_inc(&rdev->nr_pending); - bio->bi_bdev = rdev->meta_bdev ? rdev->meta_bdev : rdev->bdev; + bio_set_dev(bio, rdev->meta_bdev ? rdev->meta_bdev : rdev->bdev); bio->bi_iter.bi_sector = sector; bio_add_page(bio, page, size, 0); bio->bi_private = rdev; @@ -803,8 +803,10 @@ int sync_page_io(struct md_rdev *rdev, sector_t sector, int size, struct bio *bio = md_bio_alloc_sync(rdev->mddev); int ret; - bio->bi_bdev = (metadata_op && rdev->meta_bdev) ? - rdev->meta_bdev : rdev->bdev; + if (metadata_op && rdev->meta_bdev) + bio_set_dev(bio, rdev->meta_bdev); + else + bio_set_dev(bio, rdev->bdev); bio_set_op_attrs(bio, op, op_flags); if (metadata_op) bio->bi_iter.bi_sector = sector + rdev->sb_start; diff --git a/drivers/md/md.h b/drivers/md/md.h index 09db03455801..c0d436fb88f0 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -509,6 +509,11 @@ static inline void md_sync_acct(struct block_device *bdev, unsigned long nr_sect atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io); } +static inline void md_sync_acct_bio(struct bio *bio, unsigned long nr_sectors) +{ + atomic_add(nr_sectors, &bio->bi_disk->sync_io); +} + struct md_personality { char *name; @@ -721,14 +726,14 @@ static inline void mddev_clear_unsupported_flags(struct mddev *mddev, static inline void mddev_check_writesame(struct mddev *mddev, struct bio *bio) { if (bio_op(bio) == REQ_OP_WRITE_SAME && - !bdev_get_queue(bio->bi_bdev)->limits.max_write_same_sectors) + !bio->bi_disk->queue->limits.max_write_same_sectors) mddev->queue->limits.max_write_same_sectors = 0; } static inline void mddev_check_write_zeroes(struct mddev *mddev, struct bio *bio) { if (bio_op(bio) == REQ_OP_WRITE_ZEROES && - !bdev_get_queue(bio->bi_bdev)->limits.max_write_zeroes_sectors) + !bio->bi_disk->queue->limits.max_write_zeroes_sectors) mddev->queue->limits.max_write_zeroes_sectors = 0; } #endif /* _MD_MD_H */ diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index 23a162ba6c56..b68e0666b9b0 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -134,7 +134,7 @@ static bool multipath_make_request(struct mddev *mddev, struct bio * bio) __bio_clone_fast(&mp_bh->bio, bio); mp_bh->bio.bi_iter.bi_sector += multipath->rdev->data_offset; - mp_bh->bio.bi_bdev = multipath->rdev->bdev; + bio_set_dev(&mp_bh->bio, multipath->rdev->bdev); mp_bh->bio.bi_opf |= REQ_FAILFAST_TRANSPORT; mp_bh->bio.bi_end_io = multipath_end_request; mp_bh->bio.bi_private = mp_bh; @@ -345,17 +345,17 @@ static void multipathd(struct md_thread *thread) if ((mp_bh->path = multipath_map (conf))<0) { pr_err("multipath: %s: unrecoverable IO read error for block %llu\n", - bdevname(bio->bi_bdev,b), + bio_devname(bio, b), (unsigned long long)bio->bi_iter.bi_sector); multipath_end_bh_io(mp_bh, BLK_STS_IOERR); } else { pr_err("multipath: %s: redirecting sector %llu to another IO path\n", - bdevname(bio->bi_bdev,b), + bio_devname(bio, b), (unsigned long long)bio->bi_iter.bi_sector); *bio = *(mp_bh->master_bio); bio->bi_iter.bi_sector += conf->multipaths[mp_bh->path].rdev->data_offset; - bio->bi_bdev = conf->multipaths[mp_bh->path].rdev->bdev; + bio_set_dev(bio, conf->multipaths[mp_bh->path].rdev->bdev); bio->bi_opf |= REQ_FAILFAST_TRANSPORT; bio->bi_end_io = multipath_end_request; bio->bi_private = mp_bh; diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c index 94d9ae9b0fd0..05a4521b832f 100644 --- a/drivers/md/raid0.c +++ b/drivers/md/raid0.c @@ -588,14 +588,13 @@ static bool raid0_make_request(struct mddev *mddev, struct bio *bio) zone = find_zone(mddev->private, §or); tmp_dev = map_sector(mddev, zone, sector, §or); - bio->bi_bdev = tmp_dev->bdev; + bio_set_dev(bio, tmp_dev->bdev); bio->bi_iter.bi_sector = sector + zone->dev_start + tmp_dev->data_offset; if (mddev->gendisk) - trace_block_bio_remap(bdev_get_queue(bio->bi_bdev), - bio, disk_devt(mddev->gendisk), - bio_sector); + trace_block_bio_remap(bio->bi_disk->queue, bio, + disk_devt(mddev->gendisk), bio_sector); mddev_check_writesame(mddev, bio); mddev_check_write_zeroes(mddev, bio); generic_make_request(bio); diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index f50958ded9f0..baf5e358d22a 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -786,13 +786,13 @@ static void flush_bio_list(struct r1conf *conf, struct bio *bio) while (bio) { /* submit pending writes */ struct bio *next = bio->bi_next; - struct md_rdev *rdev = (void*)bio->bi_bdev; + struct md_rdev *rdev = (void *)bio->bi_disk; bio->bi_next = NULL; - bio->bi_bdev = rdev->bdev; + bio_set_dev(bio, rdev->bdev); if (test_bit(Faulty, &rdev->flags)) { bio_io_error(bio); } else if (unlikely((bio_op(bio) == REQ_OP_DISCARD) && - !blk_queue_discard(bdev_get_queue(bio->bi_bdev)))) + !blk_queue_discard(bio->bi_disk->queue))) /* Just ignore it */ bio_endio(bio); else @@ -1273,7 +1273,7 @@ static void raid1_read_request(struct mddev *mddev, struct bio *bio, read_bio->bi_iter.bi_sector = r1_bio->sector + mirror->rdev->data_offset; - read_bio->bi_bdev = mirror->rdev->bdev; + bio_set_dev(read_bio, mirror->rdev->bdev); read_bio->bi_end_io = raid1_end_read_request; bio_set_op_attrs(read_bio, op, do_sync); if (test_bit(FailFast, &mirror->rdev->flags) && @@ -1282,9 +1282,8 @@ static void raid1_read_request(struct mddev *mddev, struct bio *bio, read_bio->bi_private = r1_bio; if (mddev->gendisk) - trace_block_bio_remap(bdev_get_queue(read_bio->bi_bdev), - read_bio, disk_devt(mddev->gendisk), - r1_bio->sector); + trace_block_bio_remap(read_bio->bi_disk->queue, read_bio, + disk_devt(mddev->gendisk), r1_bio->sector); generic_make_request(read_bio); } @@ -1496,7 +1495,7 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio, mbio->bi_iter.bi_sector = (r1_bio->sector + conf->mirrors[i].rdev->data_offset); - mbio->bi_bdev = conf->mirrors[i].rdev->bdev; + bio_set_dev(mbio, conf->mirrors[i].rdev->bdev); mbio->bi_end_io = raid1_end_write_request; mbio->bi_opf = bio_op(bio) | (bio->bi_opf & (REQ_SYNC | REQ_FUA)); if (test_bit(FailFast, &conf->mirrors[i].rdev->flags) && @@ -1508,11 +1507,11 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio, atomic_inc(&r1_bio->remaining); if (mddev->gendisk) - trace_block_bio_remap(bdev_get_queue(mbio->bi_bdev), + trace_block_bio_remap(mbio->bi_disk->queue, mbio, disk_devt(mddev->gendisk), r1_bio->sector); /* flush_pending_writes() needs access to the rdev so...*/ - mbio->bi_bdev = (void*)conf->mirrors[i].rdev; + mbio->bi_disk = (void *)conf->mirrors[i].rdev; cb = blk_check_plugged(raid1_unplug, mddev, sizeof(*plug)); if (cb) @@ -1990,8 +1989,7 @@ static int fix_sync_read_error(struct r1bio *r1_bio) * Don't fail devices as that won't really help. */ pr_crit_ratelimited("md/raid1:%s: %s: unrecoverable I/O read error for block %llu\n", - mdname(mddev), - bdevname(bio->bi_bdev, b), + mdname(mddev), bio_devname(bio, b), (unsigned long long)r1_bio->sector); for (d = 0; d < conf->raid_disks * 2; d++) { rdev = conf->mirrors[d].rdev; @@ -2082,7 +2080,7 @@ static void process_checks(struct r1bio *r1_bio) b->bi_status = status; b->bi_iter.bi_sector = r1_bio->sector + conf->mirrors[i].rdev->data_offset; - b->bi_bdev = conf->mirrors[i].rdev->bdev; + bio_set_dev(b, conf->mirrors[i].rdev->bdev); b->bi_end_io = end_sync_read; rp->raid_bio = r1_bio; b->bi_private = rp; @@ -2350,7 +2348,7 @@ static int narrow_write_error(struct r1bio *r1_bio, int i) bio_trim(wbio, sector - r1_bio->sector, sectors); wbio->bi_iter.bi_sector += rdev->data_offset; - wbio->bi_bdev = rdev->bdev; + bio_set_dev(wbio, rdev->bdev); if (submit_bio_wait(wbio) < 0) /* failure! */ @@ -2440,7 +2438,6 @@ static void handle_read_error(struct r1conf *conf, struct r1bio *r1_bio) struct mddev *mddev = conf->mddev; struct bio *bio; struct md_rdev *rdev; - dev_t bio_dev; sector_t bio_sector; clear_bit(R1BIO_ReadError, &r1_bio->state); @@ -2454,7 +2451,6 @@ static void handle_read_error(struct r1conf *conf, struct r1bio *r1_bio) */ bio = r1_bio->bios[r1_bio->read_disk]; - bio_dev = bio->bi_bdev->bd_dev; bio_sector = conf->mirrors[r1_bio->read_disk].rdev->data_offset + r1_bio->sector; bio_put(bio); r1_bio->bios[r1_bio->read_disk] = NULL; @@ -2727,7 +2723,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr, if (bio->bi_end_io) { atomic_inc(&rdev->nr_pending); bio->bi_iter.bi_sector = sector_nr + rdev->data_offset; - bio->bi_bdev = rdev->bdev; + bio_set_dev(bio, rdev->bdev); if (test_bit(FailFast, &rdev->flags)) bio->bi_opf |= MD_FAILFAST; } @@ -2853,7 +2849,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr, bio = r1_bio->bios[i]; if (bio->bi_end_io == end_sync_read) { read_targets--; - md_sync_acct(bio->bi_bdev, nr_sectors); + md_sync_acct_bio(bio, nr_sectors); if (read_targets == 1) bio->bi_opf &= ~MD_FAILFAST; generic_make_request(bio); @@ -2862,7 +2858,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr, } else { atomic_set(&r1_bio->remaining, 1); bio = r1_bio->bios[r1_bio->read_disk]; - md_sync_acct(bio->bi_bdev, nr_sectors); + md_sync_acct_bio(bio, nr_sectors); if (read_targets == 1) bio->bi_opf &= ~MD_FAILFAST; generic_make_request(bio); diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index f55d4cc085f6..d1f948e371e0 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -901,13 +901,13 @@ static void flush_pending_writes(struct r10conf *conf) while (bio) { /* submit pending writes */ struct bio *next = bio->bi_next; - struct md_rdev *rdev = (void*)bio->bi_bdev; + struct md_rdev *rdev = (void*)bio->bi_disk; bio->bi_next = NULL; - bio->bi_bdev = rdev->bdev; + bio_set_dev(bio, rdev->bdev); if (test_bit(Faulty, &rdev->flags)) { bio_io_error(bio); } else if (unlikely((bio_op(bio) == REQ_OP_DISCARD) && - !blk_queue_discard(bdev_get_queue(bio->bi_bdev)))) + !blk_queue_discard(bio->bi_disk->queue))) /* Just ignore it */ bio_endio(bio); else @@ -1085,13 +1085,13 @@ static void raid10_unplug(struct blk_plug_cb *cb, bool from_schedule) while (bio) { /* submit pending writes */ struct bio *next = bio->bi_next; - struct md_rdev *rdev = (void*)bio->bi_bdev; + struct md_rdev *rdev = (void*)bio->bi_disk; bio->bi_next = NULL; - bio->bi_bdev = rdev->bdev; + bio_set_dev(bio, rdev->bdev); if (test_bit(Faulty, &rdev->flags)) { bio_io_error(bio); } else if (unlikely((bio_op(bio) == REQ_OP_DISCARD) && - !blk_queue_discard(bdev_get_queue(bio->bi_bdev)))) + !blk_queue_discard(bio->bi_disk->queue))) /* Just ignore it */ bio_endio(bio); else @@ -1200,7 +1200,7 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio, read_bio->bi_iter.bi_sector = r10_bio->devs[slot].addr + choose_data_offset(r10_bio, rdev); - read_bio->bi_bdev = rdev->bdev; + bio_set_dev(read_bio, rdev->bdev); read_bio->bi_end_io = raid10_end_read_request; bio_set_op_attrs(read_bio, op, do_sync); if (test_bit(FailFast, &rdev->flags) && @@ -1209,7 +1209,7 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio, read_bio->bi_private = r10_bio; if (mddev->gendisk) - trace_block_bio_remap(bdev_get_queue(read_bio->bi_bdev), + trace_block_bio_remap(read_bio->bi_disk->queue, read_bio, disk_devt(mddev->gendisk), r10_bio->sector); generic_make_request(read_bio); @@ -1249,7 +1249,7 @@ static void raid10_write_one_disk(struct mddev *mddev, struct r10bio *r10_bio, mbio->bi_iter.bi_sector = (r10_bio->devs[n_copy].addr + choose_data_offset(r10_bio, rdev)); - mbio->bi_bdev = rdev->bdev; + bio_set_dev(mbio, rdev->bdev); mbio->bi_end_io = raid10_end_write_request; bio_set_op_attrs(mbio, op, do_sync | do_fua); if (!replacement && test_bit(FailFast, @@ -1259,11 +1259,11 @@ static void raid10_write_one_disk(struct mddev *mddev, struct r10bio *r10_bio, mbio->bi_private = r10_bio; if (conf->mddev->gendisk) - trace_block_bio_remap(bdev_get_queue(mbio->bi_bdev), + trace_block_bio_remap(mbio->bi_disk->queue, mbio, disk_devt(conf->mddev->gendisk), r10_bio->sector); /* flush_pending_writes() needs access to the rdev so...*/ - mbio->bi_bdev = (void *)rdev; + mbio->bi_disk = (void *)rdev; atomic_inc(&r10_bio->remaining); @@ -2094,7 +2094,7 @@ static void sync_request_write(struct mddev *mddev, struct r10bio *r10_bio) if (test_bit(FailFast, &conf->mirrors[d].rdev->flags)) tbio->bi_opf |= MD_FAILFAST; tbio->bi_iter.bi_sector += conf->mirrors[d].rdev->data_offset; - tbio->bi_bdev = conf->mirrors[d].rdev->bdev; + bio_set_dev(tbio, conf->mirrors[d].rdev->bdev); generic_make_request(tbio); } @@ -2552,7 +2552,7 @@ static int narrow_write_error(struct r10bio *r10_bio, int i) wsector = r10_bio->devs[i].addr + (sector - r10_bio->sector); wbio->bi_iter.bi_sector = wsector + choose_data_offset(r10_bio, rdev); - wbio->bi_bdev = rdev->bdev; + bio_set_dev(wbio, rdev->bdev); bio_set_op_attrs(wbio, REQ_OP_WRITE, 0); if (submit_bio_wait(wbio) < 0) @@ -2575,7 +2575,6 @@ static void handle_read_error(struct mddev *mddev, struct r10bio *r10_bio) struct bio *bio; struct r10conf *conf = mddev->private; struct md_rdev *rdev = r10_bio->devs[slot].rdev; - dev_t bio_dev; sector_t bio_last_sector; /* we got a read error. Maybe the drive is bad. Maybe just @@ -2587,7 +2586,6 @@ static void handle_read_error(struct mddev *mddev, struct r10bio *r10_bio) * frozen. */ bio = r10_bio->devs[slot].bio; - bio_dev = bio->bi_bdev->bd_dev; bio_last_sector = r10_bio->devs[slot].addr + rdev->data_offset + r10_bio->sectors; bio_put(bio); r10_bio->devs[slot].bio = NULL; @@ -2950,7 +2948,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, /* Again, very different code for resync and recovery. * Both must result in an r10bio with a list of bios that - * have bi_end_io, bi_sector, bi_bdev set, + * have bi_end_io, bi_sector, bi_disk set, * and bi_private set to the r10bio. * For recovery, we may actually create several r10bios * with 2 bios in each, that correspond to the bios in the main one. @@ -3095,7 +3093,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, from_addr = r10_bio->devs[j].addr; bio->bi_iter.bi_sector = from_addr + rdev->data_offset; - bio->bi_bdev = rdev->bdev; + bio_set_dev(bio, rdev->bdev); atomic_inc(&rdev->nr_pending); /* and we write to 'i' (if not in_sync) */ @@ -3117,7 +3115,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, bio_set_op_attrs(bio, REQ_OP_WRITE, 0); bio->bi_iter.bi_sector = to_addr + mrdev->data_offset; - bio->bi_bdev = mrdev->bdev; + bio_set_dev(bio, mrdev->bdev); atomic_inc(&r10_bio->remaining); } else r10_bio->devs[1].bio->bi_end_io = NULL; @@ -3143,7 +3141,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, bio_set_op_attrs(bio, REQ_OP_WRITE, 0); bio->bi_iter.bi_sector = to_addr + mreplace->data_offset; - bio->bi_bdev = mreplace->bdev; + bio_set_dev(bio, mreplace->bdev); atomic_inc(&r10_bio->remaining); break; } @@ -3289,7 +3287,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, if (test_bit(FailFast, &rdev->flags)) bio->bi_opf |= MD_FAILFAST; bio->bi_iter.bi_sector = sector + rdev->data_offset; - bio->bi_bdev = rdev->bdev; + bio_set_dev(bio, rdev->bdev); count++; rdev = rcu_dereference(conf->mirrors[d].replacement); @@ -3311,7 +3309,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, if (test_bit(FailFast, &rdev->flags)) bio->bi_opf |= MD_FAILFAST; bio->bi_iter.bi_sector = sector + rdev->data_offset; - bio->bi_bdev = rdev->bdev; + bio_set_dev(bio, rdev->bdev); count++; rcu_read_unlock(); } @@ -3367,7 +3365,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, r10_bio->sectors = nr_sectors; if (bio->bi_end_io == end_sync_read) { - md_sync_acct(bio->bi_bdev, nr_sectors); + md_sync_acct_bio(bio, nr_sectors); bio->bi_status = 0; generic_make_request(bio); } @@ -4383,7 +4381,7 @@ read_more: read_bio = bio_alloc_mddev(GFP_KERNEL, RESYNC_PAGES, mddev); - read_bio->bi_bdev = rdev->bdev; + bio_set_dev(read_bio, rdev->bdev); read_bio->bi_iter.bi_sector = (r10_bio->devs[r10_bio->read_slot].addr + rdev->data_offset); read_bio->bi_private = r10_bio; @@ -4417,7 +4415,7 @@ read_more: if (!rdev2 || test_bit(Faulty, &rdev2->flags)) continue; - b->bi_bdev = rdev2->bdev; + bio_set_dev(b, rdev2->bdev); b->bi_iter.bi_sector = r10_bio->devs[s/2].addr + rdev2->new_data_offset; b->bi_end_io = end_reshape_write; @@ -4449,7 +4447,7 @@ read_more: r10_bio->sectors = nr_sectors; /* Now submit the read */ - md_sync_acct(read_bio->bi_bdev, r10_bio->sectors); + md_sync_acct_bio(read_bio, r10_bio->sectors); atomic_inc(&r10_bio->remaining); read_bio->bi_next = NULL; generic_make_request(read_bio); @@ -4511,7 +4509,7 @@ static void reshape_request_write(struct mddev *mddev, struct r10bio *r10_bio) } atomic_inc(&rdev->nr_pending); rcu_read_unlock(); - md_sync_acct(b->bi_bdev, r10_bio->sectors); + md_sync_acct_bio(b, r10_bio->sectors); atomic_inc(&r10_bio->remaining); b->bi_next = NULL; generic_make_request(b); diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c index bfa1e907c472..f253a9c583c1 100644 --- a/drivers/md/raid5-cache.c +++ b/drivers/md/raid5-cache.c @@ -728,7 +728,7 @@ static struct bio *r5l_bio_alloc(struct r5l_log *log) struct bio *bio = bio_alloc_bioset(GFP_NOIO, BIO_MAX_PAGES, log->bs); bio_set_op_attrs(bio, REQ_OP_WRITE, 0); - bio->bi_bdev = log->rdev->bdev; + bio_set_dev(bio, log->rdev->bdev); bio->bi_iter.bi_sector = log->rdev->data_offset + log->log_start; return bio; @@ -1291,7 +1291,7 @@ void r5l_flush_stripe_to_raid(struct r5l_log *log) if (!do_flush) return; bio_reset(&log->flush_bio); - log->flush_bio.bi_bdev = log->rdev->bdev; + bio_set_dev(&log->flush_bio, log->rdev->bdev); log->flush_bio.bi_end_io = r5l_log_flush_endio; log->flush_bio.bi_opf = REQ_OP_WRITE | REQ_PREFLUSH; submit_bio(&log->flush_bio); @@ -1669,7 +1669,7 @@ static int r5l_recovery_fetch_ra_pool(struct r5l_log *log, sector_t offset) { bio_reset(ctx->ra_bio); - ctx->ra_bio->bi_bdev = log->rdev->bdev; + bio_set_dev(ctx->ra_bio, log->rdev->bdev); bio_set_op_attrs(ctx->ra_bio, REQ_OP_READ, 0); ctx->ra_bio->bi_iter.bi_sector = log->rdev->data_offset + offset; diff --git a/drivers/md/raid5-ppl.c b/drivers/md/raid5-ppl.c index 44ad5baf3206..1e237c40d6fa 100644 --- a/drivers/md/raid5-ppl.c +++ b/drivers/md/raid5-ppl.c @@ -415,7 +415,7 @@ static void ppl_submit_iounit_bio(struct ppl_io_unit *io, struct bio *bio) pr_debug("%s: seq: %llu size: %u sector: %llu dev: %s\n", __func__, io->seq, bio->bi_iter.bi_size, (unsigned long long)bio->bi_iter.bi_sector, - bdevname(bio->bi_bdev, b)); + bio_devname(bio, b)); submit_bio(bio); } @@ -453,7 +453,7 @@ static void ppl_submit_iounit(struct ppl_io_unit *io) bio->bi_end_io = ppl_log_endio; bio->bi_opf = REQ_OP_WRITE | REQ_FUA; - bio->bi_bdev = log->rdev->bdev; + bio_set_dev(bio, log->rdev->bdev); bio->bi_iter.bi_sector = log->rdev->ppl.sector; bio_add_page(bio, io->header_page, PAGE_SIZE, 0); @@ -468,7 +468,7 @@ static void ppl_submit_iounit(struct ppl_io_unit *io) bio = bio_alloc_bioset(GFP_NOIO, BIO_MAX_PAGES, ppl_conf->bs); bio->bi_opf = prev->bi_opf; - bio->bi_bdev = prev->bi_bdev; + bio_copy_dev(bio, prev); bio->bi_iter.bi_sector = bio_end_sector(prev); bio_add_page(bio, sh->ppl_page, PAGE_SIZE, 0); diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index d687aeb1b538..3ae8bbceb6c4 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -1096,7 +1096,7 @@ again: set_bit(STRIPE_IO_STARTED, &sh->state); - bi->bi_bdev = rdev->bdev; + bio_set_dev(bi, rdev->bdev); bio_set_op_attrs(bi, op, op_flags); bi->bi_end_io = op_is_write(op) ? raid5_end_write_request @@ -1145,7 +1145,7 @@ again: set_bit(R5_DOUBLE_LOCKED, &sh->dev[i].flags); if (conf->mddev->gendisk) - trace_block_bio_remap(bdev_get_queue(bi->bi_bdev), + trace_block_bio_remap(bi->bi_disk->queue, bi, disk_devt(conf->mddev->gendisk), sh->dev[i].sector); if (should_defer && op_is_write(op)) @@ -1160,7 +1160,7 @@ again: set_bit(STRIPE_IO_STARTED, &sh->state); - rbi->bi_bdev = rrdev->bdev; + bio_set_dev(rbi, rrdev->bdev); bio_set_op_attrs(rbi, op, op_flags); BUG_ON(!op_is_write(op)); rbi->bi_end_io = raid5_end_write_request; @@ -1193,7 +1193,7 @@ again: if (op == REQ_OP_DISCARD) rbi->bi_vcnt = 0; if (conf->mddev->gendisk) - trace_block_bio_remap(bdev_get_queue(rbi->bi_bdev), + trace_block_bio_remap(rbi->bi_disk->queue, rbi, disk_devt(conf->mddev->gendisk), sh->dev[i].sector); if (should_defer && op_is_write(op)) @@ -5233,7 +5233,7 @@ static int raid5_read_one_chunk(struct mddev *mddev, struct bio *raid_bio) atomic_inc(&rdev->nr_pending); rcu_read_unlock(); raid_bio->bi_next = (void*)rdev; - align_bi->bi_bdev = rdev->bdev; + bio_set_dev(align_bi, rdev->bdev); bio_clear_flag(align_bi, BIO_SEG_VALID); if (is_badblock(rdev, align_bi->bi_iter.bi_sector, @@ -5255,7 +5255,7 @@ static int raid5_read_one_chunk(struct mddev *mddev, struct bio *raid_bio) spin_unlock_irq(&conf->device_lock); if (mddev->gendisk) - trace_block_bio_remap(bdev_get_queue(align_bi->bi_bdev), + trace_block_bio_remap(align_bi->bi_disk->queue, align_bi, disk_devt(mddev->gendisk), raid_bio->bi_iter.bi_sector); generic_make_request(align_bi); diff --git a/drivers/nvdimm/nd.h b/drivers/nvdimm/nd.h index 73062da3177f..a87f793f2945 100644 --- a/drivers/nvdimm/nd.h +++ b/drivers/nvdimm/nd.h @@ -390,7 +390,7 @@ int nd_region_activate(struct nd_region *nd_region); void __nd_iostat_start(struct bio *bio, unsigned long *start); static inline bool nd_iostat_start(struct bio *bio, unsigned long *start) { - struct gendisk *disk = bio->bi_bdev->bd_disk; + struct gendisk *disk = bio->bi_disk; if (!blk_queue_io_stat(disk->queue)) return false; @@ -402,7 +402,7 @@ static inline bool nd_iostat_start(struct bio *bio, unsigned long *start) } static inline void nd_iostat_end(struct bio *bio, unsigned long start) { - struct gendisk *disk = bio->bi_bdev->bd_disk; + struct gendisk *disk = bio->bi_disk; generic_end_io_acct(disk->queue, bio_data_dir(bio), &disk->part0, start); diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index c49f1f8b2e57..f03452db7938 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -613,11 +613,7 @@ int __nvme_submit_user_cmd(struct request_queue *q, struct nvme_command *cmd, if (!disk) goto submit; - bio->bi_bdev = bdget_disk(disk, 0); - if (!bio->bi_bdev) { - ret = -ENODEV; - goto out_unmap; - } + bio->bi_disk = disk; if (meta_buffer && meta_len) { struct bio_integrity_payload *bip; @@ -668,11 +664,8 @@ int __nvme_submit_user_cmd(struct request_queue *q, struct nvme_command *cmd, out_free_meta: kfree(meta); out_unmap: - if (bio) { - if (disk && bio->bi_bdev) - bdput(bio->bi_bdev); + if (bio) blk_rq_unmap_user(bio); - } out: blk_mq_free_request(req); return ret; diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c index be8541335e31..c1a28569e843 100644 --- a/drivers/nvme/host/lightnvm.c +++ b/drivers/nvme/host/lightnvm.c @@ -643,17 +643,9 @@ static int nvme_nvm_submit_user_cmd(struct request_queue *q, vcmd->ph_rw.metadata = cpu_to_le64(metadata_dma); } - if (!disk) - goto submit; - - bio->bi_bdev = bdget_disk(disk, 0); - if (!bio->bi_bdev) { - ret = -ENODEV; - goto err_meta; - } + bio->bi_disk = disk; } -submit: blk_execute_rq(q, NULL, rq, 0); if (nvme_req(rq)->flags & NVME_REQ_CANCELLED) @@ -673,11 +665,8 @@ err_meta: if (meta_buf && meta_len) dma_pool_free(dev->dma_pool, metadata, metadata_dma); err_map: - if (bio) { - if (disk && bio->bi_bdev) - bdput(bio->bi_bdev); + if (bio) blk_rq_unmap_user(bio); - } err_ppa: if (ppa_buf && ppa_len) dma_pool_free(dev->dma_pool, ppa_list, ppa_dma); diff --git a/drivers/nvme/target/io-cmd.c b/drivers/nvme/target/io-cmd.c index 3b4d47a6abdb..0d4c23dc4532 100644 --- a/drivers/nvme/target/io-cmd.c +++ b/drivers/nvme/target/io-cmd.c @@ -68,7 +68,7 @@ static void nvmet_execute_rw(struct nvmet_req *req) nvmet_inline_bio_init(req); bio = &req->inline_bio; - bio->bi_bdev = req->ns->bdev; + bio_set_dev(bio, req->ns->bdev); bio->bi_iter.bi_sector = sector; bio->bi_private = req; bio->bi_end_io = nvmet_bio_done; @@ -80,7 +80,7 @@ static void nvmet_execute_rw(struct nvmet_req *req) struct bio *prev = bio; bio = bio_alloc(GFP_KERNEL, min(sg_cnt, BIO_MAX_PAGES)); - bio->bi_bdev = req->ns->bdev; + bio_set_dev(bio, req->ns->bdev); bio->bi_iter.bi_sector = sector; bio_set_op_attrs(bio, op, op_flags); @@ -104,7 +104,7 @@ static void nvmet_execute_flush(struct nvmet_req *req) nvmet_inline_bio_init(req); bio = &req->inline_bio; - bio->bi_bdev = req->ns->bdev; + bio_set_dev(bio, req->ns->bdev); bio->bi_private = req; bio->bi_end_io = nvmet_bio_done; bio->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH; diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c index 68bae4f6bd88..7abb240847c0 100644 --- a/drivers/s390/block/dcssblk.c +++ b/drivers/s390/block/dcssblk.c @@ -856,14 +856,14 @@ dcssblk_make_request(struct request_queue *q, struct bio *bio) blk_queue_split(q, &bio); bytes_done = 0; - dev_info = bio->bi_bdev->bd_disk->private_data; + dev_info = bio->bi_disk->private_data; if (dev_info == NULL) goto fail; if ((bio->bi_iter.bi_sector & 7) != 0 || (bio->bi_iter.bi_size & 4095) != 0) /* Request is not page-aligned. */ goto fail; - if (bio_end_sector(bio) > get_capacity(bio->bi_bdev->bd_disk)) { + if (bio_end_sector(bio) > get_capacity(bio->bi_disk)) { /* Request beyond end of DCSS segment. */ goto fail; } diff --git a/drivers/s390/block/xpram.c b/drivers/s390/block/xpram.c index a48f0d40c1d2..571a0709e1e5 100644 --- a/drivers/s390/block/xpram.c +++ b/drivers/s390/block/xpram.c @@ -183,7 +183,7 @@ static unsigned long xpram_highest_page_index(void) */ static blk_qc_t xpram_make_request(struct request_queue *q, struct bio *bio) { - xpram_device_t *xdev = bio->bi_bdev->bd_disk->private_data; + xpram_device_t *xdev = bio->bi_disk->private_data; struct bio_vec bvec; struct bvec_iter iter; unsigned int index; diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c index ee7c7fa55dad..07c814c42648 100644 --- a/drivers/target/target_core_iblock.c +++ b/drivers/target/target_core_iblock.c @@ -338,7 +338,7 @@ iblock_get_bio(struct se_cmd *cmd, sector_t lba, u32 sg_num, int op, return NULL; } - bio->bi_bdev = ib_dev->ibd_bd; + bio_set_dev(bio, ib_dev->ibd_bd); bio->bi_private = cmd; bio->bi_end_io = &iblock_bio_done; bio->bi_iter.bi_sector = lba; @@ -395,7 +395,7 @@ iblock_execute_sync_cache(struct se_cmd *cmd) bio = bio_alloc(GFP_KERNEL, 0); bio->bi_end_io = iblock_end_io_flush; - bio->bi_bdev = ib_dev->ibd_bd; + bio_set_dev(bio, ib_dev->ibd_bd); bio->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH; if (!immed) bio->bi_private = cmd; diff --git a/fs/block_dev.c b/fs/block_dev.c index d29d1c70f833..bb715b2fcfb8 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -223,7 +223,7 @@ __blkdev_direct_IO_simple(struct kiocb *iocb, struct iov_iter *iter, } bio_init(&bio, vecs, nr_pages); - bio.bi_bdev = bdev; + bio_set_dev(&bio, bdev); bio.bi_iter.bi_sector = pos >> 9; bio.bi_write_hint = iocb->ki_hint; bio.bi_private = current; @@ -362,7 +362,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) blk_start_plug(&plug); for (;;) { - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_iter.bi_sector = pos >> 9; bio->bi_write_hint = iocb->ki_hint; bio->bi_private = dio; diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index 9d3854839038..fb07e3c22b9a 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -1635,7 +1635,7 @@ static int btrfsic_read_block(struct btrfsic_state *state, unsigned int j; bio = btrfs_io_bio_alloc(num_pages - i); - bio->bi_bdev = block_ctx->dev->bdev; + bio_set_dev(bio, block_ctx->dev->bdev); bio->bi_iter.bi_sector = dev_bytenr >> 9; bio_set_op_attrs(bio, REQ_OP_READ, 0); @@ -2803,7 +2803,7 @@ static void __btrfsic_submit_bio(struct bio *bio) mutex_lock(&btrfsic_mutex); /* since btrfsic_submit_bio() is also called before * btrfsic_mount(), this might return NULL */ - dev_state = btrfsic_dev_state_lookup(bio->bi_bdev->bd_dev); + dev_state = btrfsic_dev_state_lookup(bio_dev(bio)); if (NULL != dev_state && (bio_op(bio) == REQ_OP_WRITE) && bio_has_data(bio)) { unsigned int i = 0; @@ -2819,10 +2819,10 @@ static void __btrfsic_submit_bio(struct bio *bio) bio_is_patched = 0; if (dev_state->state->print_mask & BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH) - pr_info("submit_bio(rw=%d,0x%x, bi_vcnt=%u, bi_sector=%llu (bytenr %llu), bi_bdev=%p)\n", + pr_info("submit_bio(rw=%d,0x%x, bi_vcnt=%u, bi_sector=%llu (bytenr %llu), bi_disk=%p)\n", bio_op(bio), bio->bi_opf, segs, (unsigned long long)bio->bi_iter.bi_sector, - dev_bytenr, bio->bi_bdev); + dev_bytenr, bio->bi_disk); mapped_datav = kmalloc_array(segs, sizeof(*mapped_datav), GFP_NOFS); @@ -2851,8 +2851,8 @@ static void __btrfsic_submit_bio(struct bio *bio) } else if (NULL != dev_state && (bio->bi_opf & REQ_PREFLUSH)) { if (dev_state->state->print_mask & BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH) - pr_info("submit_bio(rw=%d,0x%x FLUSH, bdev=%p)\n", - bio_op(bio), bio->bi_opf, bio->bi_bdev); + pr_info("submit_bio(rw=%d,0x%x FLUSH, disk=%p)\n", + bio_op(bio), bio->bi_opf, bio->bi_disk); if (!dev_state->dummy_block_for_bio_bh_flush.is_iodone) { if ((dev_state->state->print_mask & (BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH | diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 080e2ebb8aa0..0640c27e63e9 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3499,7 +3499,7 @@ static void write_dev_flush(struct btrfs_device *device) bio_reset(bio); bio->bi_end_io = btrfs_end_empty_barrier; - bio->bi_bdev = device->bdev; + bio_set_dev(bio, device->bdev); bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_PREFLUSH; init_completion(&device->flush_wait); bio->bi_private = &device->flush_wait; diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 0aff9b278c19..42b12a85ab49 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2033,7 +2033,7 @@ int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start, bio_put(bio); return -EIO; } - bio->bi_bdev = dev->bdev; + bio_set_dev(bio, dev->bdev); bio->bi_opf = REQ_OP_WRITE | REQ_SYNC; bio_add_page(bio, page, length, pg_offset); @@ -2335,7 +2335,7 @@ struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio, bio = btrfs_io_bio_alloc(1); bio->bi_end_io = endio_func; bio->bi_iter.bi_sector = failrec->logical >> 9; - bio->bi_bdev = fs_info->fs_devices->latest_bdev; + bio_set_dev(bio, fs_info->fs_devices->latest_bdev); bio->bi_iter.bi_size = 0; bio->bi_private = data; @@ -2675,7 +2675,7 @@ struct bio *btrfs_bio_alloc(struct block_device *bdev, u64 first_byte) struct bio *bio; bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, btrfs_bioset); - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_iter.bi_sector = first_byte >> 9; btrfs_io_bio_init(btrfs_io_bio(bio)); return bio; diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index 208638384cd2..d268cb633735 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -1090,7 +1090,8 @@ static int rbio_add_io_page(struct btrfs_raid_bio *rbio, */ if (last_end == disk_start && stripe->dev->bdev && !last->bi_status && - last->bi_bdev == stripe->dev->bdev) { + last->bi_disk == stripe->dev->bdev->bd_disk && + last->bi_partno == stripe->dev->bdev->bd_partno) { ret = bio_add_page(last, page, PAGE_SIZE, 0); if (ret == PAGE_SIZE) return 0; @@ -1100,7 +1101,7 @@ static int rbio_add_io_page(struct btrfs_raid_bio *rbio, /* put a new bio on the list */ bio = btrfs_io_bio_alloc(bio_max_len >> PAGE_SHIFT ?: 1); bio->bi_iter.bi_size = 0; - bio->bi_bdev = stripe->dev->bdev; + bio_set_dev(bio, stripe->dev->bdev); bio->bi_iter.bi_sector = disk_start >> 9; bio_add_page(bio, page, PAGE_SIZE, 0); @@ -1347,7 +1348,8 @@ static int find_bio_stripe(struct btrfs_raid_bio *rbio, stripe_start = stripe->physical; if (physical >= stripe_start && physical < stripe_start + rbio->stripe_len && - bio->bi_bdev == stripe->dev->bdev) { + bio->bi_disk == stripe->dev->bdev->bd_disk && + bio->bi_partno == stripe->dev->bdev->bd_partno) { return i; } } diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 6f1e4c984b94..b0b71e8e4c36 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -1738,7 +1738,7 @@ static void scrub_recheck_block(struct btrfs_fs_info *fs_info, WARN_ON(!page->page); bio = btrfs_io_bio_alloc(1); - bio->bi_bdev = page->dev->bdev; + bio_set_dev(bio, page->dev->bdev); bio_add_page(bio, page->page, PAGE_SIZE, 0); if (!retry_failed_mirror && scrub_is_page_on_raid56(page)) { @@ -1826,7 +1826,7 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad, } bio = btrfs_io_bio_alloc(1); - bio->bi_bdev = page_bad->dev->bdev; + bio_set_dev(bio, page_bad->dev->bdev); bio->bi_iter.bi_sector = page_bad->physical >> 9; bio_set_op_attrs(bio, REQ_OP_WRITE, 0); @@ -1921,7 +1921,7 @@ again: bio->bi_private = sbio; bio->bi_end_io = scrub_wr_bio_end_io; - bio->bi_bdev = sbio->dev->bdev; + bio_set_dev(bio, sbio->dev->bdev); bio->bi_iter.bi_sector = sbio->physical >> 9; bio_set_op_attrs(bio, REQ_OP_WRITE, 0); sbio->status = 0; @@ -1964,7 +1964,7 @@ static void scrub_wr_submit(struct scrub_ctx *sctx) sbio = sctx->wr_curr_bio; sctx->wr_curr_bio = NULL; - WARN_ON(!sbio->bio->bi_bdev); + WARN_ON(!sbio->bio->bi_disk); scrub_pending_bio_inc(sctx); /* process all writes in a single worker thread. Then the block layer * orders the requests before sending them to the driver which @@ -2321,7 +2321,7 @@ again: bio->bi_private = sbio; bio->bi_end_io = scrub_bio_end_io; - bio->bi_bdev = sbio->dev->bdev; + bio_set_dev(bio, sbio->dev->bdev); bio->bi_iter.bi_sector = sbio->physical >> 9; bio_set_op_attrs(bio, REQ_OP_READ, 0); sbio->status = 0; @@ -4627,7 +4627,7 @@ static int write_page_nocow(struct scrub_ctx *sctx, bio = btrfs_io_bio_alloc(1); bio->bi_iter.bi_size = 0; bio->bi_iter.bi_sector = physical_for_dev_replace >> 9; - bio->bi_bdev = dev->bdev; + bio_set_dev(bio, dev->bdev); bio->bi_opf = REQ_OP_WRITE | REQ_SYNC; ret = bio_add_page(bio, page, PAGE_SIZE, 0); if (ret != PAGE_SIZE) { diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e8b9a269fdde..f9f0f474a64f 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -6188,7 +6188,7 @@ static void submit_stripe_bio(struct btrfs_bio *bbio, struct bio *bio, rcu_read_unlock(); } #endif - bio->bi_bdev = dev->bdev; + bio_set_dev(bio, dev->bdev); btrfs_bio_counter_inc_noblocked(fs_info); diff --git a/fs/buffer.c b/fs/buffer.c index 5715dac7821f..50e51a67dc78 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -3057,7 +3057,7 @@ void guard_bio_eod(int op, struct bio *bio) struct bio_vec *bvec = &bio->bi_io_vec[bio->bi_vcnt - 1]; unsigned truncated_bytes; - maxsector = i_size_read(bio->bi_bdev->bd_inode) >> 9; + maxsector = get_capacity(bio->bi_disk); if (!maxsector) return; @@ -3116,7 +3116,7 @@ static int submit_bh_wbc(int op, int op_flags, struct buffer_head *bh, } bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9); - bio->bi_bdev = bh->b_bdev; + bio_set_dev(bio, bh->b_bdev); bio->bi_write_hint = write_hint; bio_add_page(bio, bh->b_page, bh->b_size, bh_offset(bh)); diff --git a/fs/crypto/bio.c b/fs/crypto/bio.c index 6181e9526860..483784d5eb73 100644 --- a/fs/crypto/bio.c +++ b/fs/crypto/bio.c @@ -115,7 +115,7 @@ int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk, err = -ENOMEM; goto errout; } - bio->bi_bdev = inode->i_sb->s_bdev; + bio_set_dev(bio, inode->i_sb->s_bdev); bio->bi_iter.bi_sector = pblk << (inode->i_sb->s_blocksize_bits - 9); bio_set_op_attrs(bio, REQ_OP_WRITE, 0); diff --git a/fs/direct-io.c b/fs/direct-io.c index 08cf27811e5a..5fa2211e49ae 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -111,7 +111,7 @@ struct dio { int op; int op_flags; blk_qc_t bio_cookie; - struct block_device *bio_bdev; + struct gendisk *bio_disk; struct inode *inode; loff_t i_size; /* i_size when submitted */ dio_iodone_t *end_io; /* IO completion function */ @@ -377,7 +377,7 @@ dio_bio_alloc(struct dio *dio, struct dio_submit *sdio, */ bio = bio_alloc(GFP_KERNEL, nr_vecs); - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_iter.bi_sector = first_sector; bio_set_op_attrs(bio, dio->op, dio->op_flags); if (dio->is_async) @@ -412,7 +412,7 @@ static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio) if (dio->is_async && dio->op == REQ_OP_READ && dio->should_dirty) bio_set_pages_dirty(bio); - dio->bio_bdev = bio->bi_bdev; + dio->bio_disk = bio->bi_disk; if (sdio->submit_io) { sdio->submit_io(bio, dio->inode, sdio->logical_offset_in_bio); @@ -458,7 +458,7 @@ static struct bio *dio_await_one(struct dio *dio) dio->waiter = current; spin_unlock_irqrestore(&dio->bio_lock, flags); if (!(dio->iocb->ki_flags & IOCB_HIPRI) || - !blk_mq_poll(bdev_get_queue(dio->bio_bdev), dio->bio_cookie)) + !blk_mq_poll(dio->bio_disk->queue, dio->bio_cookie)) io_schedule(); /* wake up sets us TASK_RUNNING */ spin_lock_irqsave(&dio->bio_lock, flags); diff --git a/fs/exofs/ore.c b/fs/exofs/ore.c index 8bb72807e70d..3c6a9c156b7a 100644 --- a/fs/exofs/ore.c +++ b/fs/exofs/ore.c @@ -869,7 +869,7 @@ static int _write_mirror(struct ore_io_state *ios, int cur_comp) goto out; } - bio->bi_bdev = NULL; + bio->bi_disk = NULL; bio->bi_next = NULL; per_dev->offset = master_dev->offset; per_dev->length = master_dev->length; diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index c2fce4478cca..55ad7dd149d0 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -300,7 +300,7 @@ static void ext4_end_bio(struct bio *bio) char b[BDEVNAME_SIZE]; if (WARN_ONCE(!io_end, "io_end is NULL: %s: sector %Lu len %u err %d\n", - bdevname(bio->bi_bdev, b), + bio_devname(bio, b), (long long) bio->bi_iter.bi_sector, (unsigned) bio_sectors(bio), bio->bi_status)) { @@ -375,7 +375,7 @@ static int io_submit_init_bio(struct ext4_io_submit *io, return -ENOMEM; wbc_init_bio(io->io_wbc, bio); bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9); - bio->bi_bdev = bh->b_bdev; + bio_set_dev(bio, bh->b_bdev); bio->bi_end_io = ext4_end_bio; bio->bi_private = ext4_get_io_end(io->io_end); io->io_bio = bio; diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c index 40a5497b0f60..04c90643af7a 100644 --- a/fs/ext4/readpage.c +++ b/fs/ext4/readpage.c @@ -254,7 +254,7 @@ int ext4_mpage_readpages(struct address_space *mapping, fscrypt_release_ctx(ctx); goto set_error_page; } - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_iter.bi_sector = blocks[0] << (blkbits - 9); bio->bi_end_io = mpage_end_io; bio->bi_private = ctx; diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 87c1f4150c64..a791aac4c5af 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -142,7 +142,7 @@ struct block_device *f2fs_target_device(struct f2fs_sb_info *sbi, } } if (bio) { - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_iter.bi_sector = SECTOR_FROM_BLOCK(blk_addr); } return bdev; @@ -161,7 +161,8 @@ int f2fs_target_device_index(struct f2fs_sb_info *sbi, block_t blkaddr) static bool __same_bdev(struct f2fs_sb_info *sbi, block_t blk_addr, struct bio *bio) { - return f2fs_target_device(sbi, blk_addr, NULL) == bio->bi_bdev; + struct block_device *b = f2fs_target_device(sbi, blk_addr, NULL); + return bio->bi_disk == b->bd_disk && bio->bi_partno == b->bd_partno; } /* diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index f964b68718c1..6f8fc4a6e701 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -447,7 +447,7 @@ static int __submit_flush_wait(struct f2fs_sb_info *sbi, int ret; bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_PREFLUSH; - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); ret = submit_bio_wait(bio); bio_put(bio); diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 3010f9edd177..720c19ada0f9 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -265,7 +265,7 @@ static struct bio *gfs2_log_alloc_bio(struct gfs2_sbd *sdp, u64 blkno) bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES); bio->bi_iter.bi_sector = blkno * (sb->s_blocksize >> 9); - bio->bi_bdev = sb->s_bdev; + bio_set_dev(bio, sb->s_bdev); bio->bi_end_io = gfs2_end_log_write; bio->bi_private = sdp; diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index fabe1614f879..39433a173baa 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -221,7 +221,7 @@ static void gfs2_submit_bhs(int op, int op_flags, struct buffer_head *bhs[], bio = bio_alloc(GFP_NOIO, num); bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9); - bio->bi_bdev = bh->b_bdev; + bio_set_dev(bio, bh->b_bdev); while (num > 0) { bh = *bhs; if (!bio_add_page(bio, bh->b_page, bh->b_size, bh_offset(bh))) { diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index e76058d34b74..8155e16076e1 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -242,7 +242,7 @@ static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector, int silent) bio = bio_alloc(GFP_NOFS, 1); bio->bi_iter.bi_sector = sector * (sb->s_blocksize >> 9); - bio->bi_bdev = sb->s_bdev; + bio_set_dev(bio, sb->s_bdev); bio_add_page(bio, page, PAGE_SIZE, 0); bio->bi_end_io = end_bio_io_page; diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c index e254fa0f0697..10032b919a85 100644 --- a/fs/hfsplus/wrapper.c +++ b/fs/hfsplus/wrapper.c @@ -65,7 +65,7 @@ int hfsplus_submit_bio(struct super_block *sb, sector_t sector, bio = bio_alloc(GFP_NOIO, 1); bio->bi_iter.bi_sector = sector; - bio->bi_bdev = sb->s_bdev; + bio_set_dev(bio, sb->s_bdev); bio_set_op_attrs(bio, op, op_flags); if (op != WRITE && data) diff --git a/fs/iomap.c b/fs/iomap.c index 039266128b7f..77be8850997b 100644 --- a/fs/iomap.c +++ b/fs/iomap.c @@ -805,7 +805,7 @@ iomap_dio_zero(struct iomap_dio *dio, struct iomap *iomap, loff_t pos, struct bio *bio; bio = bio_alloc(GFP_KERNEL, 1); - bio->bi_bdev = iomap->bdev; + bio_set_dev(bio, iomap->bdev); bio->bi_iter.bi_sector = iomap->blkno + ((pos - iomap->offset) >> 9); bio->bi_private = dio; @@ -884,7 +884,7 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length, return 0; bio = bio_alloc(GFP_KERNEL, nr_pages); - bio->bi_bdev = iomap->bdev; + bio_set_dev(bio, iomap->bdev); bio->bi_iter.bi_sector = iomap->blkno + ((pos - iomap->offset) >> 9); bio->bi_write_hint = dio->iocb->ki_hint; diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index a21f0e9eecd4..0e5d412c0b01 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -1995,7 +1995,7 @@ static int lbmRead(struct jfs_log * log, int pn, struct lbuf ** bpp) bio = bio_alloc(GFP_NOFS, 1); bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9); - bio->bi_bdev = log->bdev; + bio_set_dev(bio, log->bdev); bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset); BUG_ON(bio->bi_iter.bi_size != LOGPSIZE); @@ -2139,7 +2139,7 @@ static void lbmStartIO(struct lbuf * bp) bio = bio_alloc(GFP_NOFS, 1); bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9); - bio->bi_bdev = log->bdev; + bio_set_dev(bio, log->bdev); bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset); BUG_ON(bio->bi_iter.bi_size != LOGPSIZE); diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 65120a471729..1c4b9ad4d7ab 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -430,7 +430,7 @@ static int metapage_writepage(struct page *page, struct writeback_control *wbc) len = min(xlen, (int)JFS_SBI(inode->i_sb)->nbperpage); bio = bio_alloc(GFP_NOFS, 1); - bio->bi_bdev = inode->i_sb->s_bdev; + bio_set_dev(bio, inode->i_sb->s_bdev); bio->bi_iter.bi_sector = pblock << (inode->i_blkbits - 9); bio->bi_end_io = metapage_write_end_io; bio->bi_private = page; @@ -510,7 +510,7 @@ static int metapage_readpage(struct file *fp, struct page *page) submit_bio(bio); bio = bio_alloc(GFP_NOFS, 1); - bio->bi_bdev = inode->i_sb->s_bdev; + bio_set_dev(bio, inode->i_sb->s_bdev); bio->bi_iter.bi_sector = pblock << (inode->i_blkbits - 9); bio->bi_end_io = metapage_read_end_io; diff --git a/fs/mpage.c b/fs/mpage.c index 2e4c41ccb5c9..37bb77c1302c 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -83,7 +83,7 @@ mpage_alloc(struct block_device *bdev, } if (bio) { - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_iter.bi_sector = first_sector; } return bio; diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c index d8863a804b15..995d707537da 100644 --- a/fs/nfs/blocklayout/blocklayout.c +++ b/fs/nfs/blocklayout/blocklayout.c @@ -130,7 +130,7 @@ bl_alloc_init_bio(int npg, struct block_device *bdev, sector_t disk_sector, if (bio) { bio->bi_iter.bi_sector = disk_sector; - bio->bi_bdev = bdev; + bio_set_dev(bio, bdev); bio->bi_end_io = end_io; bio->bi_private = par; } diff --git a/fs/nilfs2/segbuf.c b/fs/nilfs2/segbuf.c index e73c86d9855c..6c5009cc4e6f 100644 --- a/fs/nilfs2/segbuf.c +++ b/fs/nilfs2/segbuf.c @@ -400,7 +400,7 @@ static struct bio *nilfs_alloc_seg_bio(struct the_nilfs *nilfs, sector_t start, bio = bio_alloc(GFP_NOIO, nr_vecs); } if (likely(bio)) { - bio->bi_bdev = nilfs->ns_bdev; + bio_set_dev(bio, nilfs->ns_bdev); bio->bi_iter.bi_sector = start << (nilfs->ns_blocksize_bits - 9); } diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index ffe003982d95..6aea15746a56 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -554,7 +554,7 @@ static struct bio *o2hb_setup_one_bio(struct o2hb_region *reg, /* Must put everything in 512 byte sectors for the bio... */ bio->bi_iter.bi_sector = (reg->hr_start_block + cs) << (bits - 9); - bio->bi_bdev = reg->hr_bdev; + bio_set_dev(bio, reg->hr_bdev); bio->bi_private = wc; bio->bi_end_io = o2hb_bio_end_io; bio_set_op_attrs(bio, op, op_flags); diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 6bf120bb1a17..c8ca03a5a08f 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -517,7 +517,7 @@ xfs_init_bio_from_bh( struct buffer_head *bh) { bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9); - bio->bi_bdev = bh->b_bdev; + bio_set_dev(bio, bh->b_bdev); } static struct xfs_ioend * diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index 72f038492ba8..b1c9711e79a4 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -1281,7 +1281,7 @@ next_chunk: nr_pages = min(total_nr_pages, BIO_MAX_PAGES); bio = bio_alloc(GFP_NOIO, nr_pages); - bio->bi_bdev = bp->b_target->bt_bdev; + bio_set_dev(bio, bp->b_target->bt_bdev); bio->bi_iter.bi_sector = sector; bio->bi_end_io = xfs_buf_bio_end_io; bio->bi_private = bp; diff --git a/include/linux/bio.h b/include/linux/bio.h index 9276788a9b24..a8fe7935332f 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -494,6 +494,24 @@ extern struct bio_vec *bvec_alloc(gfp_t, int, unsigned long *, mempool_t *); extern void bvec_free(mempool_t *, struct bio_vec *, unsigned int); extern unsigned int bvec_nr_vecs(unsigned short idx); +#define bio_set_dev(bio, bdev) \ +do { \ + (bio)->bi_disk = (bdev)->bd_disk; \ + (bio)->bi_partno = (bdev)->bd_partno; \ +} while (0) + +#define bio_copy_dev(dst, src) \ +do { \ + (dst)->bi_disk = (src)->bi_disk; \ + (dst)->bi_partno = (src)->bi_partno; \ +} while (0) + +#define bio_dev(bio) \ + disk_devt((bio)->bi_disk) + +#define bio_devname(bio, buf) \ + __bdevname(bio_dev(bio), (buf)) + #ifdef CONFIG_BLK_CGROUP int bio_associate_blkcg(struct bio *bio, struct cgroup_subsys_state *blkcg_css); int bio_associate_current(struct bio *bio); diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index d2eb87c84d82..a2d2aa709cef 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -48,7 +48,8 @@ struct blk_issue_stat { */ struct bio { struct bio *bi_next; /* request queue link */ - struct block_device *bi_bdev; + struct gendisk *bi_disk; + u8 bi_partno; blk_status_t bi_status; unsigned int bi_opf; /* bottom bits req flags, * top bits REQ_OP. Use diff --git a/include/trace/events/bcache.h b/include/trace/events/bcache.h index df3e9ae5ad8d..daf749138ff8 100644 --- a/include/trace/events/bcache.h +++ b/include/trace/events/bcache.h @@ -21,7 +21,7 @@ DECLARE_EVENT_CLASS(bcache_request, ), TP_fast_assign( - __entry->dev = bio->bi_bdev->bd_dev; + __entry->dev = bio_dev(bio); __entry->orig_major = d->disk->major; __entry->orig_minor = d->disk->first_minor; __entry->sector = bio->bi_iter.bi_sector; @@ -98,7 +98,7 @@ DECLARE_EVENT_CLASS(bcache_bio, ), TP_fast_assign( - __entry->dev = bio->bi_bdev->bd_dev; + __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio->bi_iter.bi_size >> 9; blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); @@ -133,7 +133,7 @@ TRACE_EVENT(bcache_read, ), TP_fast_assign( - __entry->dev = bio->bi_bdev->bd_dev; + __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio->bi_iter.bi_size >> 9; blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); diff --git a/include/trace/events/block.h b/include/trace/events/block.h index d0dbe60d8a6d..f815aaaef755 100644 --- a/include/trace/events/block.h +++ b/include/trace/events/block.h @@ -236,8 +236,7 @@ TRACE_EVENT(block_bio_bounce, ), TP_fast_assign( - __entry->dev = bio->bi_bdev ? - bio->bi_bdev->bd_dev : 0; + __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio_sectors(bio); blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); @@ -274,7 +273,7 @@ TRACE_EVENT(block_bio_complete, ), TP_fast_assign( - __entry->dev = bio->bi_bdev->bd_dev; + __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio_sectors(bio); __entry->error = error; @@ -302,7 +301,7 @@ DECLARE_EVENT_CLASS(block_bio_merge, ), TP_fast_assign( - __entry->dev = bio->bi_bdev->bd_dev; + __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio_sectors(bio); blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); @@ -369,7 +368,7 @@ TRACE_EVENT(block_bio_queue, ), TP_fast_assign( - __entry->dev = bio->bi_bdev->bd_dev; + __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio_sectors(bio); blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); @@ -397,7 +396,8 @@ DECLARE_EVENT_CLASS(block_get_rq, ), TP_fast_assign( - __entry->dev = bio ? bio->bi_bdev->bd_dev : 0; + __entry->dev = bio ? bio_dev(bio) : 0; + __entry->dev = bio_dev(bio); __entry->sector = bio ? bio->bi_iter.bi_sector : 0; __entry->nr_sector = bio ? bio_sectors(bio) : 0; blk_fill_rwbs(__entry->rwbs, @@ -532,7 +532,7 @@ TRACE_EVENT(block_split, ), TP_fast_assign( - __entry->dev = bio->bi_bdev->bd_dev; + __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->new_sector = new_sector; blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); @@ -573,7 +573,7 @@ TRACE_EVENT(block_bio_remap, ), TP_fast_assign( - __entry->dev = bio->bi_bdev->bd_dev; + __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio_sectors(bio); __entry->old_dev = dev; diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index 6f77a2755abb..bc4dd7837e4c 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -829,7 +829,7 @@ DECLARE_EVENT_CLASS(f2fs__bio, TP_fast_assign( __entry->dev = sb->s_dev; - __entry->target = bio->bi_bdev->bd_dev; + __entry->target = bio_dev(bio); __entry->op = bio_op(bio); __entry->op_flags = bio->bi_opf; __entry->type = type; diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 57d22571f306..d7cdc426ee38 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -242,8 +242,7 @@ static void hib_end_io(struct bio *bio) if (bio->bi_status) { printk(KERN_ALERT "Read-error on swap-device (%u:%u:%Lu)\n", - imajor(bio->bi_bdev->bd_inode), - iminor(bio->bi_bdev->bd_inode), + MAJOR(bio_dev(bio)), MINOR(bio_dev(bio)), (unsigned long long)bio->bi_iter.bi_sector); } @@ -270,7 +269,7 @@ static int hib_submit_io(int op, int op_flags, pgoff_t page_off, void *addr, bio = bio_alloc(__GFP_RECLAIM | __GFP_HIGH, 1); bio->bi_iter.bi_sector = page_off * (PAGE_SIZE >> 9); - bio->bi_bdev = hib_resume_bdev; + bio_set_dev(bio, hib_resume_bdev); bio_set_op_attrs(bio, op, op_flags); if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) { diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 7724de18d2fe..2a685b45b73b 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -963,7 +963,7 @@ static void blk_add_trace_bio_remap(void *ignore, return; r.device_from = cpu_to_be32(dev); - r.device_to = cpu_to_be32(bio->bi_bdev->bd_dev); + r.device_to = cpu_to_be32(bio_dev(bio)); r.sector_from = cpu_to_be64(from); __blk_add_trace(bt, bio->bi_iter.bi_sector, bio->bi_iter.bi_size, diff --git a/mm/page_io.c b/mm/page_io.c index b6c4ac388209..9cf1bc751d79 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -31,7 +31,10 @@ static struct bio *get_swap_bio(gfp_t gfp_flags, bio = bio_alloc(gfp_flags, 1); if (bio) { - bio->bi_iter.bi_sector = map_swap_page(page, &bio->bi_bdev); + struct block_device *bdev; + + bio->bi_iter.bi_sector = map_swap_page(page, &bdev); + bio_set_dev(bio, bdev); bio->bi_iter.bi_sector <<= PAGE_SHIFT - 9; bio->bi_end_io = end_io; @@ -57,8 +60,7 @@ void end_swap_bio_write(struct bio *bio) */ set_page_dirty(page); pr_alert("Write-error on swap-device (%u:%u:%llu)\n", - imajor(bio->bi_bdev->bd_inode), - iminor(bio->bi_bdev->bd_inode), + MAJOR(bio_dev(bio)), MINOR(bio_dev(bio)), (unsigned long long)bio->bi_iter.bi_sector); ClearPageReclaim(page); } @@ -123,8 +125,7 @@ static void end_swap_bio_read(struct bio *bio) SetPageError(page); ClearPageUptodate(page); pr_alert("Read-error on swap-device (%u:%u:%llu)\n", - imajor(bio->bi_bdev->bd_inode), - iminor(bio->bi_bdev->bd_inode), + MAJOR(bio_dev(bio)), MINOR(bio_dev(bio)), (unsigned long long)bio->bi_iter.bi_sector); goto out; } @@ -338,7 +339,7 @@ int swap_readpage(struct page *page, bool do_poll) int ret = 0; struct swap_info_struct *sis = page_swap_info(page); blk_qc_t qc; - struct block_device *bdev; + struct gendisk *disk; VM_BUG_ON_PAGE(!PageSwapCache(page), page); VM_BUG_ON_PAGE(!PageLocked(page), page); @@ -377,7 +378,7 @@ int swap_readpage(struct page *page, bool do_poll) ret = -ENOMEM; goto out; } - bdev = bio->bi_bdev; + disk = bio->bi_disk; bio->bi_private = current; bio_set_op_attrs(bio, REQ_OP_READ, 0); count_vm_event(PSWPIN); @@ -388,7 +389,7 @@ int swap_readpage(struct page *page, bool do_poll) if (!READ_ONCE(bio->bi_private)) break; - if (!blk_mq_poll(bdev_get_queue(bdev), qc)) + if (!blk_mq_poll(disk->queue, qc)) break; } __set_current_state(TASK_RUNNING); -- cgit v1.2.3 From 58efbc9f5463308c43d812fd0748a4dee44c8a16 Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Tue, 22 Aug 2017 23:45:59 -0700 Subject: Btrfs: fix blk_status_t/errno confusion This fixes several instances of blk_status_t and bare errno ints being mixed up, some of which are real bugs. In the normal case, 0 matches BLK_STS_OK, so we don't observe any effects of the missing conversion, but in case of errors or passes through the repair/retry paths, the errors get mixed up. The changes were identified using 'sparse', we don't have reports of the buggy behaviour. Fixes: 4e4cbee93d56 ("block: switch bios to blk_status_t") Signed-off-by: Omar Sandoval Reviewed-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 4 ++-- fs/btrfs/inode.c | 70 +++++++++++++++++++++++++++++------------------------- fs/btrfs/raid56.c | 34 +++++++++++++------------- fs/btrfs/volumes.c | 10 ++++---- fs/btrfs/volumes.h | 6 ++--- 5 files changed, 64 insertions(+), 60 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 080e2ebb8aa0..f45b61fe9a9a 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3516,7 +3516,7 @@ static blk_status_t wait_dev_flush(struct btrfs_device *device) struct bio *bio = device->flush_bio; if (!device->flush_bio_sent) - return 0; + return BLK_STS_OK; device->flush_bio_sent = 0; wait_for_completion_io(&device->flush_wait); @@ -3563,7 +3563,7 @@ static int barrier_all_devices(struct btrfs_fs_info *info) continue; write_dev_flush(dev); - dev->last_flush_error = 0; + dev->last_flush_error = BLK_STS_OK; } /* wait for all the barriers */ diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 95c212037095..24bcd5cd9cf2 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7924,11 +7924,12 @@ err: return ret; } -static inline int submit_dio_repair_bio(struct inode *inode, struct bio *bio, - int mirror_num) +static inline blk_status_t submit_dio_repair_bio(struct inode *inode, + struct bio *bio, + int mirror_num) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - int ret; + blk_status_t ret; BUG_ON(bio_op(bio) == REQ_OP_WRITE); @@ -7980,10 +7981,10 @@ static int btrfs_check_dio_repairable(struct inode *inode, return 1; } -static int dio_read_error(struct inode *inode, struct bio *failed_bio, - struct page *page, unsigned int pgoff, - u64 start, u64 end, int failed_mirror, - bio_end_io_t *repair_endio, void *repair_arg) +static blk_status_t dio_read_error(struct inode *inode, struct bio *failed_bio, + struct page *page, unsigned int pgoff, + u64 start, u64 end, int failed_mirror, + bio_end_io_t *repair_endio, void *repair_arg) { struct io_failure_record *failrec; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; @@ -7993,18 +7994,19 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, int read_mode = 0; int segs; int ret; + blk_status_t status; BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE); ret = btrfs_get_io_failure_record(inode, start, end, &failrec); if (ret) - return ret; + return errno_to_blk_status(ret); ret = btrfs_check_dio_repairable(inode, failed_bio, failrec, failed_mirror); if (!ret) { free_io_failure(failure_tree, io_tree, failrec); - return -EIO; + return BLK_STS_IOERR; } segs = bio_segments(failed_bio); @@ -8022,13 +8024,13 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, "Repair DIO Read Error: submitting new dio read[%#x] to this_mirror=%d, in_validation=%d\n", read_mode, failrec->this_mirror, failrec->in_validation); - ret = submit_dio_repair_bio(inode, bio, failrec->this_mirror); - if (ret) { + status = submit_dio_repair_bio(inode, bio, failrec->this_mirror); + if (status) { free_io_failure(failure_tree, io_tree, failrec); bio_put(bio); } - return ret; + return status; } struct btrfs_retry_complete { @@ -8065,8 +8067,8 @@ end: bio_put(bio); } -static int __btrfs_correct_data_nocsum(struct inode *inode, - struct btrfs_io_bio *io_bio) +static blk_status_t __btrfs_correct_data_nocsum(struct inode *inode, + struct btrfs_io_bio *io_bio) { struct btrfs_fs_info *fs_info; struct bio_vec bvec; @@ -8076,8 +8078,8 @@ static int __btrfs_correct_data_nocsum(struct inode *inode, unsigned int pgoff; u32 sectorsize; int nr_sectors; - int ret; - int err = 0; + blk_status_t ret; + blk_status_t err = BLK_STS_OK; fs_info = BTRFS_I(inode)->root->fs_info; sectorsize = fs_info->sectorsize; @@ -8183,11 +8185,12 @@ static blk_status_t __btrfs_subio_endio_read(struct inode *inode, int csum_pos; bool uptodate = (err == 0); int ret; + blk_status_t status; fs_info = BTRFS_I(inode)->root->fs_info; sectorsize = fs_info->sectorsize; - err = 0; + err = BLK_STS_OK; start = io_bio->logical; done.inode = inode; io_bio->bio.bi_iter = io_bio->iter; @@ -8209,12 +8212,12 @@ try_again: done.start = start; init_completion(&done.done); - ret = dio_read_error(inode, &io_bio->bio, bvec.bv_page, - pgoff, start, start + sectorsize - 1, - io_bio->mirror_num, - btrfs_retry_endio, &done); - if (ret) { - err = errno_to_blk_status(ret); + status = dio_read_error(inode, &io_bio->bio, bvec.bv_page, + pgoff, start, start + sectorsize - 1, + io_bio->mirror_num, btrfs_retry_endio, + &done); + if (status) { + err = status; goto next; } @@ -8250,7 +8253,7 @@ static blk_status_t btrfs_subio_endio_read(struct inode *inode, if (unlikely(err)) return __btrfs_correct_data_nocsum(inode, io_bio); else - return 0; + return BLK_STS_OK; } else { return __btrfs_subio_endio_read(inode, io_bio, err); } @@ -8423,9 +8426,9 @@ static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode, return 0; } -static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, - u64 file_offset, int skip_sum, - int async_submit) +static inline blk_status_t +__btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, u64 file_offset, + int skip_sum, int async_submit) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_dio_private *dip = bio->bi_private; @@ -8488,6 +8491,7 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, int clone_offset = 0; int clone_len; int ret; + blk_status_t status; map_length = orig_bio->bi_iter.bi_size; submit_len = map_length; @@ -8537,9 +8541,9 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, */ atomic_inc(&dip->pending_bios); - ret = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, - async_submit); - if (ret) { + status = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, + async_submit); + if (status) { bio_put(bio); atomic_dec(&dip->pending_bios); goto out_err; @@ -8557,9 +8561,9 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, } while (submit_len > 0); submit: - ret = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, - async_submit); - if (!ret) + status = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, + async_submit); + if (!status) return 0; bio_put(bio); diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index 208638384cd2..2cf6ba40f7c4 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -905,7 +905,7 @@ static void raid_write_end_io(struct bio *bio) if (!atomic_dec_and_test(&rbio->stripes_pending)) return; - err = 0; + err = BLK_STS_OK; /* OK, we have read all the stripes we need to. */ max_errors = (rbio->operation == BTRFS_RBIO_PARITY_SCRUB) ? @@ -1324,7 +1324,7 @@ write_data: return; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); } /* @@ -1475,7 +1475,7 @@ static void raid_rmw_end_io(struct bio *bio) cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); } static void async_rmw_stripe(struct btrfs_raid_bio *rbio) @@ -1579,7 +1579,7 @@ static int raid56_rmw_stripe(struct btrfs_raid_bio *rbio) return 0; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); return -EIO; finish: @@ -1795,12 +1795,12 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio) void **pointers; int faila = -1, failb = -1; struct page *page; - int err; + blk_status_t err; int i; pointers = kcalloc(rbio->real_stripes, sizeof(void *), GFP_NOFS); if (!pointers) { - err = -ENOMEM; + err = BLK_STS_RESOURCE; goto cleanup_io; } @@ -1856,7 +1856,7 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio) * a bad data or Q stripe. * TODO, we should redo the xor here. */ - err = -EIO; + err = BLK_STS_IOERR; goto cleanup; } /* @@ -1882,7 +1882,7 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio) if (rbio->bbio->raid_map[failb] == RAID6_Q_STRIPE) { if (rbio->bbio->raid_map[faila] == RAID5_P_STRIPE) { - err = -EIO; + err = BLK_STS_IOERR; goto cleanup; } /* @@ -1954,13 +1954,13 @@ pstripe: } } - err = 0; + err = BLK_STS_OK; cleanup: kfree(pointers); cleanup_io: if (rbio->operation == BTRFS_RBIO_READ_REBUILD) { - if (err == 0) + if (err == BLK_STS_OK) cache_rbio_pages(rbio); else clear_bit(RBIO_CACHE_READY_BIT, &rbio->flags); @@ -1968,7 +1968,7 @@ cleanup_io: rbio_orig_end_io(rbio, err); } else if (rbio->operation == BTRFS_RBIO_REBUILD_MISSING) { rbio_orig_end_io(rbio, err); - } else if (err == 0) { + } else if (err == BLK_STS_OK) { rbio->faila = -1; rbio->failb = -1; @@ -2005,7 +2005,7 @@ static void raid_recover_end_io(struct bio *bio) return; if (atomic_read(&rbio->error) > rbio->bbio->max_errors) - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); else __raid_recover_end_io(rbio); } @@ -2104,7 +2104,7 @@ out: cleanup: if (rbio->operation == BTRFS_RBIO_READ_REBUILD || rbio->operation == BTRFS_RBIO_REBUILD_MISSING) - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); return -EIO; } @@ -2431,7 +2431,7 @@ submit_write: nr_data = bio_list_size(&bio_list); if (!nr_data) { /* Every parity is right */ - rbio_orig_end_io(rbio, 0); + rbio_orig_end_io(rbio, BLK_STS_OK); return; } @@ -2451,7 +2451,7 @@ submit_write: return; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); } static inline int is_data_stripe(struct btrfs_raid_bio *rbio, int stripe) @@ -2519,7 +2519,7 @@ static void validate_rbio_for_parity_scrub(struct btrfs_raid_bio *rbio) return; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); } /* @@ -2633,7 +2633,7 @@ static void raid56_parity_scrub_stripe(struct btrfs_raid_bio *rbio) return; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); return; finish: diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e8b9a269fdde..bd679bc7a1a9 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -6212,8 +6212,8 @@ static void bbio_error(struct btrfs_bio *bbio, struct bio *bio, u64 logical) } } -int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, - int mirror_num, int async_submit) +blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, + int mirror_num, int async_submit) { struct btrfs_device *dev; struct bio *first_bio = bio; @@ -6233,7 +6233,7 @@ int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, &map_length, &bbio, mirror_num, 1); if (ret) { btrfs_bio_counter_dec(fs_info); - return ret; + return errno_to_blk_status(ret); } total_devs = bbio->num_stripes; @@ -6256,7 +6256,7 @@ int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, } btrfs_bio_counter_dec(fs_info); - return ret; + return errno_to_blk_status(ret); } if (map_length < length) { @@ -6283,7 +6283,7 @@ int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, dev_nr, async_submit); } btrfs_bio_counter_dec(fs_info); - return 0; + return BLK_STS_OK; } struct btrfs_device *btrfs_find_device(struct btrfs_fs_info *fs_info, u64 devid, diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 6f45fd60d15a..93277fc60930 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -74,7 +74,7 @@ struct btrfs_device { int missing; int can_discard; int is_tgtdev_for_dev_replace; - int last_flush_error; + blk_status_t last_flush_error; int flush_bio_sent; #ifdef __BTRFS_NEED_DEVICE_DATA_ORDERED @@ -416,8 +416,8 @@ int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, u64 type); void btrfs_mapping_init(struct btrfs_mapping_tree *tree); void btrfs_mapping_tree_free(struct btrfs_mapping_tree *tree); -int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, - int mirror_num, int async_submit); +blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, + int mirror_num, int async_submit); int btrfs_open_devices(struct btrfs_fs_devices *fs_devices, fmode_t flags, void *holder); int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder, -- cgit v1.2.3 From fed3b381145e2e7c66b0b3f640851e1633ebd07f Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Wed, 13 Sep 2017 12:25:21 -0600 Subject: Btrfs: do not backup tree roots when fsync It doesn't make sense to backup tree roots when doing fsync, since during fsync those tree roots have not been consistent on disk. Signed-off-by: Liu Bo Reviewed-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 27d458640536..0f2271815eb6 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3641,7 +3641,14 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors) u64 flags; do_barriers = !btrfs_test_opt(fs_info, NOBARRIER); - backup_super_roots(fs_info); + + /* + * max_mirrors == 0 indicates we're from commit_transaction, + * not from fsync where the tree roots in fs_info have not + * been consistent on disk. + */ + if (max_mirrors == 0) + backup_super_roots(fs_info); sb = fs_info->super_for_commit; dev_item = &sb->dev_item; -- cgit v1.2.3 From ee87cf5ed9fe8e41d057f5f5c7f49c18b6d3f94c Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 1 Aug 2017 18:35:08 +0800 Subject: btrfs: copy fsid to super_block s_uuid We didn't copy fsid to struct super_block.s_uuid so Overlay disables index feature with btrfs as the lower FS. kernel: overlayfs: fs on '/lower' does not support file handles, falling back to index=off. Fix this by publishing the fsid through struct super_block.s_uuid. [ dsterba: I think that setting s_uuid is the last missing bit. Overlay needs the file handle encoding support from the lower filesystem, which is supported. Filling the whole filesystem id is correct, the subvolume id is encoded in the file handle buffer from inside btrfs_encode_fh. ] Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index dfdab849037b..0a98567f8cfc 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2901,6 +2901,7 @@ int open_ctree(struct super_block *sb, sb->s_blocksize = sectorsize; sb->s_blocksize_bits = blksize_bits(sectorsize); + memcpy(&sb->s_uuid, fs_info->fsid, BTRFS_FSID_SIZE); mutex_lock(&fs_info->chunk_mutex); ret = btrfs_read_sys_array(fs_info); -- cgit v1.2.3 From 6300463b14c1c2665674eb8f15843e5bb7a7ff84 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Mon, 21 Aug 2017 15:49:59 -0600 Subject: Btrfs: make plug in writing meta blocks really work We have started plug in btrfs_write_and_wait_marked_extents() but the generated IOs actually go to device's schedule IO list where the work is doing in another task, thus the started plug doesn't make any sense. And since we wait for IOs immediately after writing meta blocks, it's the same case as writing log tree, doing sync submit can merge more IOs. Signed-off-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 6 ++++-- fs/btrfs/transaction.c | 2 ++ 2 files changed, 6 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 0a98567f8cfc..23442747363e 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1005,8 +1005,10 @@ static blk_status_t __btree_submit_bio_done(void *private_data, struct bio *bio, return ret; } -static int check_async_write(unsigned long bio_flags) +static int check_async_write(struct btrfs_inode *bi, unsigned long bio_flags) { + if (atomic_read(&bi->sync_writers)) + return 0; if (bio_flags & EXTENT_BIO_TREE_LOG) return 0; #ifdef CONFIG_X86 @@ -1022,7 +1024,7 @@ static blk_status_t btree_submit_bio_hook(void *private_data, struct bio *bio, { struct inode *inode = private_data; struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - int async = check_async_write(bio_flags); + int async = check_async_write(BTRFS_I(inode), bio_flags); blk_status_t ret; if (bio_op(bio) != REQ_OP_WRITE) { diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index f615d59b0489..9c5f126064bd 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -950,6 +950,7 @@ int btrfs_write_marked_extents(struct btrfs_fs_info *fs_info, u64 start = 0; u64 end; + atomic_inc(&BTRFS_I(fs_info->btree_inode)->sync_writers); while (!find_first_extent_bit(dirty_pages, start, &start, &end, mark, &cached_state)) { bool wait_writeback = false; @@ -985,6 +986,7 @@ int btrfs_write_marked_extents(struct btrfs_fs_info *fs_info, cond_resched(); start = end + 1; } + atomic_dec(&BTRFS_I(fs_info->btree_inode)->sync_writers); return werr; } -- cgit v1.2.3 From 18fdc67900c5bfd1eeb41cfa50ea6f2eb7266f73 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Wed, 13 Sep 2017 12:18:22 -0600 Subject: Btrfs: remove bio_flags which indicates a meta block of log-tree Since both committing transaction and writing log-tree are doing plugging on metadata IO, we can unify to use %sync_writers to benefit both cases, instead of checking bio_flags while writing meta blocks of log-tree. We can remove this bio_flags because in order to write dirty blocks, log tree also uses btrfs_write_marked_extents(), inside which we have enabled %sync_writers, therefore, every write goes in a synchronous way, so does checksuming. Please also note that, bio_flags is applied per-context while %sync_writers is applied per-inode, so this might incur some overhead, ie. 1) while log tree is flushing its dirty blocks via btrfs_write_marked_extents(), in which %sync_writers is increased by one. 2) in the meantime, some writeback operations may happen upon btrfs's metadata inode, so these writes go synchronously, too. However, AFAICS, the overhead is not a big one while the win is that we unify the two places that needs synchronous way and remove a special hack/flag. This removes the bio_flags related stuff for writing log-tree. Signed-off-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 6 ++---- fs/btrfs/extent_io.c | 13 ++----------- fs/btrfs/extent_io.h | 1 - 3 files changed, 4 insertions(+), 16 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 23442747363e..90310d9e647a 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1005,12 +1005,10 @@ static blk_status_t __btree_submit_bio_done(void *private_data, struct bio *bio, return ret; } -static int check_async_write(struct btrfs_inode *bi, unsigned long bio_flags) +static int check_async_write(struct btrfs_inode *bi) { if (atomic_read(&bi->sync_writers)) return 0; - if (bio_flags & EXTENT_BIO_TREE_LOG) - return 0; #ifdef CONFIG_X86 if (static_cpu_has(X86_FEATURE_XMM4_2)) return 0; @@ -1024,7 +1022,7 @@ static blk_status_t btree_submit_bio_hook(void *private_data, struct bio *bio, { struct inode *inode = private_data; struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - int async = check_async_write(BTRFS_I(inode), bio_flags); + int async = check_async_write(BTRFS_I(inode)); blk_status_t ret; if (bio_op(bio) != REQ_OP_WRITE) { diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 970190cd347e..0dfcef37352b 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -109,7 +109,6 @@ struct extent_page_data { struct bio *bio; struct extent_io_tree *tree; get_extent_t *get_extent; - unsigned long bio_flags; /* tells writepage not to lock the state bits for this range * it still does the unlocking @@ -3715,7 +3714,6 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, u64 offset = eb->start; u32 nritems; unsigned long i, num_pages; - unsigned long bio_flags = 0; unsigned long start, end; unsigned int write_flags = wbc_to_write_flags(wbc) | REQ_META; int ret = 0; @@ -3723,8 +3721,6 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, clear_bit(EXTENT_BUFFER_WRITE_ERR, &eb->bflags); num_pages = num_extent_pages(eb->start, eb->len); atomic_set(&eb->io_pages, num_pages); - if (btrfs_header_owner(eb) == BTRFS_TREE_LOG_OBJECTID) - bio_flags = EXTENT_BIO_TREE_LOG; /* set btree blocks beyond nritems with 0 to avoid stale content. */ nritems = btrfs_header_nritems(eb); @@ -3751,8 +3747,7 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, p, offset >> 9, PAGE_SIZE, 0, bdev, &epd->bio, end_bio_extent_buffer_writepage, - 0, epd->bio_flags, bio_flags, false); - epd->bio_flags = bio_flags; + 0, 0, 0, false); if (ret) { set_btree_ioerr(p); if (PageWriteback(p)) @@ -3789,7 +3784,6 @@ int btree_write_cache_pages(struct address_space *mapping, .tree = tree, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, - .bio_flags = 0, }; int ret = 0; int done = 0; @@ -4062,7 +4056,7 @@ static void flush_epd_write_bio(struct extent_page_data *epd) if (epd->bio) { int ret; - ret = submit_one_bio(epd->bio, 0, epd->bio_flags); + ret = submit_one_bio(epd->bio, 0, 0); BUG_ON(ret < 0); /* -ENOMEM */ epd->bio = NULL; } @@ -4085,7 +4079,6 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page, .get_extent = get_extent, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, - .bio_flags = 0, }; ret = __extent_writepage(page, wbc, &epd); @@ -4110,7 +4103,6 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode, .get_extent = get_extent, .extent_locked = 1, .sync_io = mode == WB_SYNC_ALL, - .bio_flags = 0, }; struct writeback_control wbc_writepages = { .sync_mode = mode, @@ -4150,7 +4142,6 @@ int extent_writepages(struct extent_io_tree *tree, .get_extent = get_extent, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, - .bio_flags = 0, }; ret = extent_write_cache_pages(mapping, wbc, __extent_writepage, &epd, diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index faffa28ba707..861dacb371c7 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -33,7 +33,6 @@ * type for this bio */ #define EXTENT_BIO_COMPRESSED 1 -#define EXTENT_BIO_TREE_LOG 2 #define EXTENT_BIO_FLAG_SHIFT 16 /* these are bit numbers for test/set bit */ -- cgit v1.2.3 From c3267bbaa9cae09b62960eafe33ad19196803285 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Wed, 23 Aug 2017 16:57:56 +0900 Subject: btrfs: Refactor check_leaf function for later expansion Current check_leaf() function does a good job checking key order and item offset/size. However it only checks from slot 0 to the last but one slot, this is good but makes later expansion hard. So this refactoring iterates from slot 0 to the last slot. For key comparison, it uses a key with all 0 as initial key, so all valid keys should be larger than that. And for item size/offset checks, it compares current item end with previous item offset. For slot 0, use leaf end as a special case. This makes later item/key offset checks and item size checks easier to be implemented. Also, makes check_leaf() to return -EUCLEAN other than -EIO to indicate error. Signed-off-by: Qu Wenruo Reviewed-by: Nikolay Borisov Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 50 +++++++++++++++++++++++++++----------------------- 1 file changed, 27 insertions(+), 23 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 90310d9e647a..6c97ae1e7288 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -553,8 +553,9 @@ static noinline int check_leaf(struct btrfs_root *root, struct extent_buffer *leaf) { struct btrfs_fs_info *fs_info = root->fs_info; + /* No valid key type is 0, so all key should be larger than this key */ + struct btrfs_key prev_key = {0, 0, 0}; struct btrfs_key key; - struct btrfs_key leaf_key; u32 nritems = btrfs_header_nritems(leaf); int slot; @@ -587,7 +588,7 @@ static noinline int check_leaf(struct btrfs_root *root, CORRUPT("non-root leaf's nritems is 0", leaf, check_root, 0); free_extent_buffer(eb); - return -EIO; + return -EUCLEAN; } free_extent_buffer(eb); } @@ -597,28 +598,23 @@ static noinline int check_leaf(struct btrfs_root *root, if (nritems == 0) return 0; - /* Check the 0 item */ - if (btrfs_item_offset_nr(leaf, 0) + btrfs_item_size_nr(leaf, 0) != - BTRFS_LEAF_DATA_SIZE(fs_info)) { - CORRUPT("invalid item offset size pair", leaf, root, 0); - return -EIO; - } - /* - * Check to make sure each items keys are in the correct order and their - * offsets make sense. We only have to loop through nritems-1 because - * we check the current slot against the next slot, which verifies the - * next slot's offset+size makes sense and that the current's slot - * offset is correct. + * Check the following things to make sure this is a good leaf, and + * leaf users won't need to bother with similar sanity checks: + * + * 1) key order + * 2) item offset and size + * No overlap, no hole, all inside the leaf. */ - for (slot = 0; slot < nritems - 1; slot++) { - btrfs_item_key_to_cpu(leaf, &leaf_key, slot); - btrfs_item_key_to_cpu(leaf, &key, slot + 1); + for (slot = 0; slot < nritems; slot++) { + u32 item_end_expected; + + btrfs_item_key_to_cpu(leaf, &key, slot); /* Make sure the keys are in the right order */ - if (btrfs_comp_cpu_keys(&leaf_key, &key) >= 0) { + if (btrfs_comp_cpu_keys(&prev_key, &key) >= 0) { CORRUPT("bad key order", leaf, root, slot); - return -EIO; + return -EUCLEAN; } /* @@ -626,10 +622,14 @@ static noinline int check_leaf(struct btrfs_root *root, * item data starts at the end of the leaf and grows towards the * front. */ - if (btrfs_item_offset_nr(leaf, slot) != - btrfs_item_end_nr(leaf, slot + 1)) { + if (slot == 0) + item_end_expected = BTRFS_LEAF_DATA_SIZE(fs_info); + else + item_end_expected = btrfs_item_offset_nr(leaf, + slot - 1); + if (btrfs_item_end_nr(leaf, slot) != item_end_expected) { CORRUPT("slot offset bad", leaf, root, slot); - return -EIO; + return -EUCLEAN; } /* @@ -640,8 +640,12 @@ static noinline int check_leaf(struct btrfs_root *root, if (btrfs_item_end_nr(leaf, slot) > BTRFS_LEAF_DATA_SIZE(fs_info)) { CORRUPT("slot end outside of leaf", leaf, root, slot); - return -EIO; + return -EUCLEAN; } + + prev_key.objectid = key.objectid; + prev_key.type = key.type; + prev_key.offset = key.offset; } return 0; -- cgit v1.2.3 From 7f43d4affb2a254d421ab20b0cf65ac2569909fb Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Wed, 23 Aug 2017 16:57:57 +0900 Subject: btrfs: Check if item pointer overlaps with the item itself Function check_leaf() checks if any item pointer points outside of the leaf, but it doesn't check if the pointer overlaps with the item itself. Normally only the last item may be the victim, but adding such check is never a bad idea anyway. Signed-off-by: Qu Wenruo Reviewed-by: Nikolay Borisov Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 6c97ae1e7288..d1770b3e0385 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -643,6 +643,13 @@ static noinline int check_leaf(struct btrfs_root *root, return -EUCLEAN; } + /* Also check if the item pointer overlaps with btrfs item. */ + if (btrfs_item_nr_offset(slot) + sizeof(struct btrfs_item) > + btrfs_item_ptr_offset(leaf, slot)) { + CORRUPT("slot overlap with its data", leaf, root, slot); + return -EUCLEAN; + } + prev_key.objectid = key.objectid; prev_key.type = key.type; prev_key.offset = key.offset; -- cgit v1.2.3 From 40c3c40947324d9f40bf47830c92c59a9bbadf4a Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Wed, 23 Aug 2017 16:57:58 +0900 Subject: btrfs: Add sanity check for EXTENT_DATA when reading out leaf Add extra checks for item with EXTENT_DATA type. This checks the following thing: 0) Key offset All key offsets must be aligned to sectorsize. Inline extent must have 0 for key offset. 1) Item size Uncompressed inline file extent size must match item size. (Compressed inline file extent has no information about its on-disk size.) Regular/preallocated file extent size must be a fixed value. 2) Every member of regular file extent item Including alignment for bytenr and offset, possible value for compression/encryption/type. 3) Type/compression/encode must be one of the valid values. This should be the most comprehensive and strict check in the context of btrfs_item for EXTENT_DATA. Signed-off-by: Qu Wenruo Reviewed-by: Nikolay Borisov Reviewed-by: David Sterba [ switch to BTRFS_FILE_EXTENT_TYPES, similar to what BTRFS_COMPRESS_TYPES does ] Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 103 ++++++++++++++++++++++++++++++++++++++++ include/uapi/linux/btrfs_tree.h | 1 + 2 files changed, 104 insertions(+) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index d1770b3e0385..b863d41f7d0a 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -549,6 +549,100 @@ static int check_tree_block_fsid(struct btrfs_fs_info *fs_info, btrfs_header_level(eb) == 0 ? "leaf" : "node", \ reason, btrfs_header_bytenr(eb), root->objectid, slot) +static int check_extent_data_item(struct btrfs_root *root, + struct extent_buffer *leaf, + struct btrfs_key *key, int slot) +{ + struct btrfs_file_extent_item *fi; + u32 sectorsize = root->fs_info->sectorsize; + u32 item_size = btrfs_item_size_nr(leaf, slot); + + if (!IS_ALIGNED(key->offset, sectorsize)) { + CORRUPT("unaligned key offset for file extent", + leaf, root, slot); + return -EUCLEAN; + } + + fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); + + if (btrfs_file_extent_type(leaf, fi) > BTRFS_FILE_EXTENT_TYPES) { + CORRUPT("invalid file extent type", leaf, root, slot); + return -EUCLEAN; + } + + /* + * Support for new compression/encrption must introduce incompat flag, + * and must be caught in open_ctree(). + */ + if (btrfs_file_extent_compression(leaf, fi) > BTRFS_COMPRESS_TYPES) { + CORRUPT("invalid file extent compression", leaf, root, slot); + return -EUCLEAN; + } + if (btrfs_file_extent_encryption(leaf, fi)) { + CORRUPT("invalid file extent encryption", leaf, root, slot); + return -EUCLEAN; + } + if (btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE) { + /* Inline extent must have 0 as key offset */ + if (key->offset) { + CORRUPT("inline extent has non-zero key offset", + leaf, root, slot); + return -EUCLEAN; + } + + /* Compressed inline extent has no on-disk size, skip it */ + if (btrfs_file_extent_compression(leaf, fi) != + BTRFS_COMPRESS_NONE) + return 0; + + /* Uncompressed inline extent size must match item size */ + if (item_size != BTRFS_FILE_EXTENT_INLINE_DATA_START + + btrfs_file_extent_ram_bytes(leaf, fi)) { + CORRUPT("plaintext inline extent has invalid size", + leaf, root, slot); + return -EUCLEAN; + } + return 0; + } + + /* Regular or preallocated extent has fixed item size */ + if (item_size != sizeof(*fi)) { + CORRUPT( + "regluar or preallocated extent data item size is invalid", + leaf, root, slot); + return -EUCLEAN; + } + if (!IS_ALIGNED(btrfs_file_extent_ram_bytes(leaf, fi), sectorsize) || + !IS_ALIGNED(btrfs_file_extent_disk_bytenr(leaf, fi), sectorsize) || + !IS_ALIGNED(btrfs_file_extent_disk_num_bytes(leaf, fi), sectorsize) || + !IS_ALIGNED(btrfs_file_extent_offset(leaf, fi), sectorsize) || + !IS_ALIGNED(btrfs_file_extent_num_bytes(leaf, fi), sectorsize)) { + CORRUPT( + "regular or preallocated extent data item has unaligned value", + leaf, root, slot); + return -EUCLEAN; + } + + return 0; +} + +/* + * Common point to switch the item-specific validation. + */ +static int check_leaf_item(struct btrfs_root *root, + struct extent_buffer *leaf, + struct btrfs_key *key, int slot) +{ + int ret = 0; + + switch (key->type) { + case BTRFS_EXTENT_DATA_KEY: + ret = check_extent_data_item(root, leaf, key, slot); + break; + } + return ret; +} + static noinline int check_leaf(struct btrfs_root *root, struct extent_buffer *leaf) { @@ -605,9 +699,13 @@ static noinline int check_leaf(struct btrfs_root *root, * 1) key order * 2) item offset and size * No overlap, no hole, all inside the leaf. + * 3) item content + * If possible, do comprehensive sanity check. + * NOTE: All checks must only rely on the item data itself. */ for (slot = 0; slot < nritems; slot++) { u32 item_end_expected; + int ret; btrfs_item_key_to_cpu(leaf, &key, slot); @@ -650,6 +748,11 @@ static noinline int check_leaf(struct btrfs_root *root, return -EUCLEAN; } + /* Check if the item size and content meet other criteria */ + ret = check_leaf_item(root, leaf, &key, slot); + if (ret < 0) + return ret; + prev_key.objectid = key.objectid; prev_key.type = key.type; prev_key.offset = key.offset; diff --git a/include/uapi/linux/btrfs_tree.h b/include/uapi/linux/btrfs_tree.h index 10689e1fdf11..3142645a27f5 100644 --- a/include/uapi/linux/btrfs_tree.h +++ b/include/uapi/linux/btrfs_tree.h @@ -732,6 +732,7 @@ struct btrfs_balance_item { #define BTRFS_FILE_EXTENT_INLINE 0 #define BTRFS_FILE_EXTENT_REG 1 #define BTRFS_FILE_EXTENT_PREALLOC 2 +#define BTRFS_FILE_EXTENT_TYPES 2 struct btrfs_file_extent_item { /* -- cgit v1.2.3 From 4b865cab96fe2a30ed512cf667b354bd291b3b0a Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Wed, 23 Aug 2017 16:57:59 +0900 Subject: btrfs: Add checker for EXTENT_CSUM EXTENT_CSUM checker is a relatively easy one, only needs to check: 1) Objectid Fixed to BTRFS_EXTENT_CSUM_OBJECTID 2) Key offset alignment Must be aligned to sectorsize 3) Item size alignedment Must be aligned to csum size Signed-off-by: Qu Wenruo Reviewed-by: Nikolay Borisov Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index b863d41f7d0a..c8633f2abdf1 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -626,6 +626,27 @@ static int check_extent_data_item(struct btrfs_root *root, return 0; } +static int check_csum_item(struct btrfs_root *root, struct extent_buffer *leaf, + struct btrfs_key *key, int slot) +{ + u32 sectorsize = root->fs_info->sectorsize; + u32 csumsize = btrfs_super_csum_size(root->fs_info->super_copy); + + if (key->objectid != BTRFS_EXTENT_CSUM_OBJECTID) { + CORRUPT("invalid objectid for csum item", leaf, root, slot); + return -EUCLEAN; + } + if (!IS_ALIGNED(key->offset, sectorsize)) { + CORRUPT("unaligned key offset for csum item", leaf, root, slot); + return -EUCLEAN; + } + if (!IS_ALIGNED(btrfs_item_size_nr(leaf, slot), csumsize)) { + CORRUPT("unaligned csum item size", leaf, root, slot); + return -EUCLEAN; + } + return 0; +} + /* * Common point to switch the item-specific validation. */ @@ -639,6 +660,9 @@ static int check_leaf_item(struct btrfs_root *root, case BTRFS_EXTENT_DATA_KEY: ret = check_extent_data_item(root, leaf, key, slot); break; + case BTRFS_EXTENT_CSUM_KEY: + ret = check_csum_item(root, leaf, key, slot); + break; } return ret; } -- cgit v1.2.3 From 557ea5dd003d371536f6b4e8f7c8209a2b6fd4e3 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 9 Oct 2017 01:51:02 +0000 Subject: btrfs: Move leaf and node validation checker to tree-checker.c It's no doubt the comprehensive tree block checker will become larger, so moving them into their own files is quite reasonable. Signed-off-by: Qu Wenruo [ wording adjustments ] Signed-off-by: David Sterba --- fs/btrfs/Makefile | 2 +- fs/btrfs/disk-io.c | 285 +------------------------------------------- fs/btrfs/tree-checker.c | 309 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/btrfs/tree-checker.h | 26 ++++ 4 files changed, 340 insertions(+), 282 deletions(-) create mode 100644 fs/btrfs/tree-checker.c create mode 100644 fs/btrfs/tree-checker.h (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile index 962a95aefb81..88255e133ade 100644 --- a/fs/btrfs/Makefile +++ b/fs/btrfs/Makefile @@ -9,7 +9,7 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.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 + uuid-tree.o props.o hash.o free-space-tree.o tree-checker.o btrfs-$(CONFIG_BTRFS_FS_POSIX_ACL) += acl.o btrfs-$(CONFIG_BTRFS_FS_CHECK_INTEGRITY) += check-integrity.o diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index c8633f2abdf1..09407384e996 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -50,6 +50,7 @@ #include "sysfs.h" #include "qgroup.h" #include "compression.h" +#include "tree-checker.h" #ifdef CONFIG_X86 #include @@ -543,284 +544,6 @@ static int check_tree_block_fsid(struct btrfs_fs_info *fs_info, return ret; } -#define CORRUPT(reason, eb, root, slot) \ - btrfs_crit(root->fs_info, \ - "corrupt %s, %s: block=%llu, root=%llu, slot=%d", \ - btrfs_header_level(eb) == 0 ? "leaf" : "node", \ - reason, btrfs_header_bytenr(eb), root->objectid, slot) - -static int check_extent_data_item(struct btrfs_root *root, - struct extent_buffer *leaf, - struct btrfs_key *key, int slot) -{ - struct btrfs_file_extent_item *fi; - u32 sectorsize = root->fs_info->sectorsize; - u32 item_size = btrfs_item_size_nr(leaf, slot); - - if (!IS_ALIGNED(key->offset, sectorsize)) { - CORRUPT("unaligned key offset for file extent", - leaf, root, slot); - return -EUCLEAN; - } - - fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); - - if (btrfs_file_extent_type(leaf, fi) > BTRFS_FILE_EXTENT_TYPES) { - CORRUPT("invalid file extent type", leaf, root, slot); - return -EUCLEAN; - } - - /* - * Support for new compression/encrption must introduce incompat flag, - * and must be caught in open_ctree(). - */ - if (btrfs_file_extent_compression(leaf, fi) > BTRFS_COMPRESS_TYPES) { - CORRUPT("invalid file extent compression", leaf, root, slot); - return -EUCLEAN; - } - if (btrfs_file_extent_encryption(leaf, fi)) { - CORRUPT("invalid file extent encryption", leaf, root, slot); - return -EUCLEAN; - } - if (btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE) { - /* Inline extent must have 0 as key offset */ - if (key->offset) { - CORRUPT("inline extent has non-zero key offset", - leaf, root, slot); - return -EUCLEAN; - } - - /* Compressed inline extent has no on-disk size, skip it */ - if (btrfs_file_extent_compression(leaf, fi) != - BTRFS_COMPRESS_NONE) - return 0; - - /* Uncompressed inline extent size must match item size */ - if (item_size != BTRFS_FILE_EXTENT_INLINE_DATA_START + - btrfs_file_extent_ram_bytes(leaf, fi)) { - CORRUPT("plaintext inline extent has invalid size", - leaf, root, slot); - return -EUCLEAN; - } - return 0; - } - - /* Regular or preallocated extent has fixed item size */ - if (item_size != sizeof(*fi)) { - CORRUPT( - "regluar or preallocated extent data item size is invalid", - leaf, root, slot); - return -EUCLEAN; - } - if (!IS_ALIGNED(btrfs_file_extent_ram_bytes(leaf, fi), sectorsize) || - !IS_ALIGNED(btrfs_file_extent_disk_bytenr(leaf, fi), sectorsize) || - !IS_ALIGNED(btrfs_file_extent_disk_num_bytes(leaf, fi), sectorsize) || - !IS_ALIGNED(btrfs_file_extent_offset(leaf, fi), sectorsize) || - !IS_ALIGNED(btrfs_file_extent_num_bytes(leaf, fi), sectorsize)) { - CORRUPT( - "regular or preallocated extent data item has unaligned value", - leaf, root, slot); - return -EUCLEAN; - } - - return 0; -} - -static int check_csum_item(struct btrfs_root *root, struct extent_buffer *leaf, - struct btrfs_key *key, int slot) -{ - u32 sectorsize = root->fs_info->sectorsize; - u32 csumsize = btrfs_super_csum_size(root->fs_info->super_copy); - - if (key->objectid != BTRFS_EXTENT_CSUM_OBJECTID) { - CORRUPT("invalid objectid for csum item", leaf, root, slot); - return -EUCLEAN; - } - if (!IS_ALIGNED(key->offset, sectorsize)) { - CORRUPT("unaligned key offset for csum item", leaf, root, slot); - return -EUCLEAN; - } - if (!IS_ALIGNED(btrfs_item_size_nr(leaf, slot), csumsize)) { - CORRUPT("unaligned csum item size", leaf, root, slot); - return -EUCLEAN; - } - return 0; -} - -/* - * Common point to switch the item-specific validation. - */ -static int check_leaf_item(struct btrfs_root *root, - struct extent_buffer *leaf, - struct btrfs_key *key, int slot) -{ - int ret = 0; - - switch (key->type) { - case BTRFS_EXTENT_DATA_KEY: - ret = check_extent_data_item(root, leaf, key, slot); - break; - case BTRFS_EXTENT_CSUM_KEY: - ret = check_csum_item(root, leaf, key, slot); - break; - } - return ret; -} - -static noinline int check_leaf(struct btrfs_root *root, - struct extent_buffer *leaf) -{ - struct btrfs_fs_info *fs_info = root->fs_info; - /* No valid key type is 0, so all key should be larger than this key */ - struct btrfs_key prev_key = {0, 0, 0}; - struct btrfs_key key; - u32 nritems = btrfs_header_nritems(leaf); - int slot; - - /* - * Extent buffers from a relocation tree have a owner field that - * corresponds to the subvolume tree they are based on. So just from an - * extent buffer alone we can not find out what is the id of the - * corresponding subvolume tree, so we can not figure out if the extent - * buffer corresponds to the root of the relocation tree or not. So skip - * this check for relocation trees. - */ - if (nritems == 0 && !btrfs_header_flag(leaf, BTRFS_HEADER_FLAG_RELOC)) { - struct btrfs_root *check_root; - - key.objectid = btrfs_header_owner(leaf); - key.type = BTRFS_ROOT_ITEM_KEY; - key.offset = (u64)-1; - - check_root = btrfs_get_fs_root(fs_info, &key, false); - /* - * The only reason we also check NULL here is that during - * open_ctree() some roots has not yet been set up. - */ - if (!IS_ERR_OR_NULL(check_root)) { - struct extent_buffer *eb; - - eb = btrfs_root_node(check_root); - /* if leaf is the root, then it's fine */ - if (leaf != eb) { - CORRUPT("non-root leaf's nritems is 0", - leaf, check_root, 0); - free_extent_buffer(eb); - return -EUCLEAN; - } - free_extent_buffer(eb); - } - return 0; - } - - if (nritems == 0) - return 0; - - /* - * Check the following things to make sure this is a good leaf, and - * leaf users won't need to bother with similar sanity checks: - * - * 1) key order - * 2) item offset and size - * No overlap, no hole, all inside the leaf. - * 3) item content - * If possible, do comprehensive sanity check. - * NOTE: All checks must only rely on the item data itself. - */ - for (slot = 0; slot < nritems; slot++) { - u32 item_end_expected; - int ret; - - btrfs_item_key_to_cpu(leaf, &key, slot); - - /* Make sure the keys are in the right order */ - if (btrfs_comp_cpu_keys(&prev_key, &key) >= 0) { - CORRUPT("bad key order", leaf, root, slot); - return -EUCLEAN; - } - - /* - * Make sure the offset and ends are right, remember that the - * item data starts at the end of the leaf and grows towards the - * front. - */ - if (slot == 0) - item_end_expected = BTRFS_LEAF_DATA_SIZE(fs_info); - else - item_end_expected = btrfs_item_offset_nr(leaf, - slot - 1); - if (btrfs_item_end_nr(leaf, slot) != item_end_expected) { - CORRUPT("slot offset bad", leaf, root, slot); - return -EUCLEAN; - } - - /* - * Check to make sure that we don't point outside of the leaf, - * just in case all the items are consistent to each other, but - * all point outside of the leaf. - */ - if (btrfs_item_end_nr(leaf, slot) > - BTRFS_LEAF_DATA_SIZE(fs_info)) { - CORRUPT("slot end outside of leaf", leaf, root, slot); - return -EUCLEAN; - } - - /* Also check if the item pointer overlaps with btrfs item. */ - if (btrfs_item_nr_offset(slot) + sizeof(struct btrfs_item) > - btrfs_item_ptr_offset(leaf, slot)) { - CORRUPT("slot overlap with its data", leaf, root, slot); - return -EUCLEAN; - } - - /* Check if the item size and content meet other criteria */ - ret = check_leaf_item(root, leaf, &key, slot); - if (ret < 0) - return ret; - - prev_key.objectid = key.objectid; - prev_key.type = key.type; - prev_key.offset = key.offset; - } - - return 0; -} - -static int check_node(struct btrfs_root *root, struct extent_buffer *node) -{ - unsigned long nr = btrfs_header_nritems(node); - struct btrfs_key key, next_key; - int slot; - u64 bytenr; - int ret = 0; - - if (nr == 0 || nr > BTRFS_NODEPTRS_PER_BLOCK(root->fs_info)) { - btrfs_crit(root->fs_info, - "corrupt node: block %llu root %llu nritems %lu", - node->start, root->objectid, nr); - return -EIO; - } - - for (slot = 0; slot < nr - 1; slot++) { - bytenr = btrfs_node_blockptr(node, slot); - btrfs_node_key_to_cpu(node, &key, slot); - btrfs_node_key_to_cpu(node, &next_key, slot + 1); - - if (!bytenr) { - CORRUPT("invalid item slot", node, root, slot); - ret = -EIO; - goto out; - } - - if (btrfs_comp_cpu_keys(&key, &next_key) >= 0) { - CORRUPT("bad key order", node, root, slot); - ret = -EIO; - goto out; - } - } -out: - return ret; -} - static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio, u64 phy_offset, struct page *page, u64 start, u64 end, int mirror) @@ -886,12 +609,12 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio, * that we don't try and read the other copies of this block, just * return -EIO. */ - if (found_level == 0 && check_leaf(root, eb)) { + if (found_level == 0 && btrfs_check_leaf(root, eb)) { set_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags); ret = -EIO; } - if (found_level > 0 && check_node(root, eb)) + if (found_level > 0 && btrfs_check_node(root, eb)) ret = -EIO; if (!ret) @@ -4146,7 +3869,7 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf) buf->len, fs_info->dirty_metadata_batch); #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY - if (btrfs_header_level(buf) == 0 && check_leaf(root, buf)) { + if (btrfs_header_level(buf) == 0 && btrfs_check_leaf(root, buf)) { btrfs_print_leaf(buf); ASSERT(0); } diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c new file mode 100644 index 000000000000..56e25a630103 --- /dev/null +++ b/fs/btrfs/tree-checker.c @@ -0,0 +1,309 @@ +/* + * Copyright (C) Qu Wenruo 2017. 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. + * + * You should have received a copy of the GNU General Public + * License along with this program. + */ + +/* + * The module is used to catch unexpected/corrupted tree block data. + * Such behavior can be caused either by a fuzzed image or bugs. + * + * The objective is to do leaf/node validation checks when tree block is read + * from disk, and check *every* possible member, so other code won't + * need to checking them again. + * + * Due to the potential and unwanted damage, every checker needs to be + * carefully reviewed otherwise so it does not prevent mount of valid images. + */ + +#include "ctree.h" +#include "tree-checker.h" +#include "disk-io.h" +#include "compression.h" + +#define CORRUPT(reason, eb, root, slot) \ + btrfs_crit(root->fs_info, \ + "corrupt %s, %s: block=%llu, root=%llu, slot=%d", \ + btrfs_header_level(eb) == 0 ? "leaf" : "node", \ + reason, btrfs_header_bytenr(eb), root->objectid, slot) + +static int check_extent_data_item(struct btrfs_root *root, + struct extent_buffer *leaf, + struct btrfs_key *key, int slot) +{ + struct btrfs_file_extent_item *fi; + u32 sectorsize = root->fs_info->sectorsize; + u32 item_size = btrfs_item_size_nr(leaf, slot); + + if (!IS_ALIGNED(key->offset, sectorsize)) { + CORRUPT("unaligned key offset for file extent", + leaf, root, slot); + return -EUCLEAN; + } + + fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); + + if (btrfs_file_extent_type(leaf, fi) > BTRFS_FILE_EXTENT_TYPES) { + CORRUPT("invalid file extent type", leaf, root, slot); + return -EUCLEAN; + } + + /* + * Support for new compression/encrption must introduce incompat flag, + * and must be caught in open_ctree(). + */ + if (btrfs_file_extent_compression(leaf, fi) > BTRFS_COMPRESS_TYPES) { + CORRUPT("invalid file extent compression", leaf, root, slot); + return -EUCLEAN; + } + if (btrfs_file_extent_encryption(leaf, fi)) { + CORRUPT("invalid file extent encryption", leaf, root, slot); + return -EUCLEAN; + } + if (btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE) { + /* Inline extent must have 0 as key offset */ + if (key->offset) { + CORRUPT("inline extent has non-zero key offset", + leaf, root, slot); + return -EUCLEAN; + } + + /* Compressed inline extent has no on-disk size, skip it */ + if (btrfs_file_extent_compression(leaf, fi) != + BTRFS_COMPRESS_NONE) + return 0; + + /* Uncompressed inline extent size must match item size */ + if (item_size != BTRFS_FILE_EXTENT_INLINE_DATA_START + + btrfs_file_extent_ram_bytes(leaf, fi)) { + CORRUPT("plaintext inline extent has invalid size", + leaf, root, slot); + return -EUCLEAN; + } + return 0; + } + + /* Regular or preallocated extent has fixed item size */ + if (item_size != sizeof(*fi)) { + CORRUPT( + "regluar or preallocated extent data item size is invalid", + leaf, root, slot); + return -EUCLEAN; + } + if (!IS_ALIGNED(btrfs_file_extent_ram_bytes(leaf, fi), sectorsize) || + !IS_ALIGNED(btrfs_file_extent_disk_bytenr(leaf, fi), sectorsize) || + !IS_ALIGNED(btrfs_file_extent_disk_num_bytes(leaf, fi), sectorsize) || + !IS_ALIGNED(btrfs_file_extent_offset(leaf, fi), sectorsize) || + !IS_ALIGNED(btrfs_file_extent_num_bytes(leaf, fi), sectorsize)) { + CORRUPT( + "regular or preallocated extent data item has unaligned value", + leaf, root, slot); + return -EUCLEAN; + } + + return 0; +} + +static int check_csum_item(struct btrfs_root *root, struct extent_buffer *leaf, + struct btrfs_key *key, int slot) +{ + u32 sectorsize = root->fs_info->sectorsize; + u32 csumsize = btrfs_super_csum_size(root->fs_info->super_copy); + + if (key->objectid != BTRFS_EXTENT_CSUM_OBJECTID) { + CORRUPT("invalid objectid for csum item", leaf, root, slot); + return -EUCLEAN; + } + if (!IS_ALIGNED(key->offset, sectorsize)) { + CORRUPT("unaligned key offset for csum item", leaf, root, slot); + return -EUCLEAN; + } + if (!IS_ALIGNED(btrfs_item_size_nr(leaf, slot), csumsize)) { + CORRUPT("unaligned csum item size", leaf, root, slot); + return -EUCLEAN; + } + return 0; +} + +/* + * Common point to switch the item-specific validation. + */ +static int check_leaf_item(struct btrfs_root *root, + struct extent_buffer *leaf, + struct btrfs_key *key, int slot) +{ + int ret = 0; + + switch (key->type) { + case BTRFS_EXTENT_DATA_KEY: + ret = check_extent_data_item(root, leaf, key, slot); + break; + case BTRFS_EXTENT_CSUM_KEY: + ret = check_csum_item(root, leaf, key, slot); + break; + } + return ret; +} + +int btrfs_check_leaf(struct btrfs_root *root, struct extent_buffer *leaf) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + /* No valid key type is 0, so all key should be larger than this key */ + struct btrfs_key prev_key = {0, 0, 0}; + struct btrfs_key key; + u32 nritems = btrfs_header_nritems(leaf); + int slot; + + /* + * Extent buffers from a relocation tree have a owner field that + * corresponds to the subvolume tree they are based on. So just from an + * extent buffer alone we can not find out what is the id of the + * corresponding subvolume tree, so we can not figure out if the extent + * buffer corresponds to the root of the relocation tree or not. So + * skip this check for relocation trees. + */ + if (nritems == 0 && !btrfs_header_flag(leaf, BTRFS_HEADER_FLAG_RELOC)) { + struct btrfs_root *check_root; + + key.objectid = btrfs_header_owner(leaf); + key.type = BTRFS_ROOT_ITEM_KEY; + key.offset = (u64)-1; + + check_root = btrfs_get_fs_root(fs_info, &key, false); + /* + * The only reason we also check NULL here is that during + * open_ctree() some roots has not yet been set up. + */ + if (!IS_ERR_OR_NULL(check_root)) { + struct extent_buffer *eb; + + eb = btrfs_root_node(check_root); + /* if leaf is the root, then it's fine */ + if (leaf != eb) { + CORRUPT("non-root leaf's nritems is 0", + leaf, check_root, 0); + free_extent_buffer(eb); + return -EUCLEAN; + } + free_extent_buffer(eb); + } + return 0; + } + + if (nritems == 0) + return 0; + + /* + * Check the following things to make sure this is a good leaf, and + * leaf users won't need to bother with similar sanity checks: + * + * 1) key ordering + * 2) item offset and size + * No overlap, no hole, all inside the leaf. + * 3) item content + * If possible, do comprehensive sanity check. + * NOTE: All checks must only rely on the item data itself. + */ + for (slot = 0; slot < nritems; slot++) { + u32 item_end_expected; + int ret; + + btrfs_item_key_to_cpu(leaf, &key, slot); + + /* Make sure the keys are in the right order */ + if (btrfs_comp_cpu_keys(&prev_key, &key) >= 0) { + CORRUPT("bad key order", leaf, root, slot); + return -EUCLEAN; + } + + /* + * Make sure the offset and ends are right, remember that the + * item data starts at the end of the leaf and grows towards the + * front. + */ + if (slot == 0) + item_end_expected = BTRFS_LEAF_DATA_SIZE(fs_info); + else + item_end_expected = btrfs_item_offset_nr(leaf, + slot - 1); + if (btrfs_item_end_nr(leaf, slot) != item_end_expected) { + CORRUPT("slot offset bad", leaf, root, slot); + return -EUCLEAN; + } + + /* + * Check to make sure that we don't point outside of the leaf, + * just in case all the items are consistent to each other, but + * all point outside of the leaf. + */ + if (btrfs_item_end_nr(leaf, slot) > + BTRFS_LEAF_DATA_SIZE(fs_info)) { + CORRUPT("slot end outside of leaf", leaf, root, slot); + return -EUCLEAN; + } + + /* Also check if the item pointer overlaps with btrfs item. */ + if (btrfs_item_nr_offset(slot) + sizeof(struct btrfs_item) > + btrfs_item_ptr_offset(leaf, slot)) { + CORRUPT("slot overlap with its data", leaf, root, slot); + return -EUCLEAN; + } + + /* Check if the item size and content meet other criteria */ + ret = check_leaf_item(root, leaf, &key, slot); + if (ret < 0) + return ret; + + prev_key.objectid = key.objectid; + prev_key.type = key.type; + prev_key.offset = key.offset; + } + + return 0; +} + +int btrfs_check_node(struct btrfs_root *root, struct extent_buffer *node) +{ + unsigned long nr = btrfs_header_nritems(node); + struct btrfs_key key, next_key; + int slot; + u64 bytenr; + int ret = 0; + + if (nr == 0 || nr > BTRFS_NODEPTRS_PER_BLOCK(root->fs_info)) { + btrfs_crit(root->fs_info, + "corrupt node: block %llu root %llu nritems %lu", + node->start, root->objectid, nr); + return -EIO; + } + + for (slot = 0; slot < nr - 1; slot++) { + bytenr = btrfs_node_blockptr(node, slot); + btrfs_node_key_to_cpu(node, &key, slot); + btrfs_node_key_to_cpu(node, &next_key, slot + 1); + + if (!bytenr) { + CORRUPT("invalid item slot", node, root, slot); + ret = -EIO; + goto out; + } + + if (btrfs_comp_cpu_keys(&key, &next_key) >= 0) { + CORRUPT("bad key order", node, root, slot); + ret = -EIO; + goto out; + } + } +out: + return ret; +} diff --git a/fs/btrfs/tree-checker.h b/fs/btrfs/tree-checker.h new file mode 100644 index 000000000000..96c486e95d70 --- /dev/null +++ b/fs/btrfs/tree-checker.h @@ -0,0 +1,26 @@ +/* + * Copyright (C) Qu Wenruo 2017. 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. + * + * You should have received a copy of the GNU General Public + * License along with this program. + */ + +#ifndef __BTRFS_TREE_CHECKER__ +#define __BTRFS_TREE_CHECKER__ + +#include "ctree.h" +#include "extent_io.h" + +int btrfs_check_leaf(struct btrfs_root *root, struct extent_buffer *leaf); +int btrfs_check_node(struct btrfs_root *root, struct extent_buffer *node); + +#endif -- cgit v1.2.3 From f851689b5ae3eb8e1b4d397eaf2805ccfd88bf77 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Thu, 7 Sep 2017 11:22:20 -0600 Subject: Btrfs: remove nr_async_bios This was intended to congest higher layers to not send bios, but as 1) the congested bit has been taken by writeback Async bios come from buffered writes and DIO writes. For DIO writes, we want to submit them ASAP, while for buffered writes, writeback uses balance_dirty_pages() to throttle how much dirty pages we can have. 2) and no one is waiting for %nr_async_bios down to zero, Historically, it was introduced along with changes which let checksumming workload spread accross different cpus. And at that time, pdflush was used instead of per-bdi flushing, perhaps pdflush did not have the necessary information for writeback to do throttling. We can safely remove them now. Signed-off-by: Liu Bo [ additional explanation from mails, removed unused variable 'limit' ] Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 1 - fs/btrfs/disk-io.c | 1 - fs/btrfs/volumes.c | 17 ----------------- 3 files changed, 19 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 16b3537f31d4..bc1b6a033700 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -880,7 +880,6 @@ struct btrfs_fs_info { atomic_t nr_async_submits; atomic_t async_submit_draining; - atomic_t nr_async_bios; atomic_t async_delalloc_pages; atomic_t open_ioctl_trans; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 09407384e996..2b18b0c379c1 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2518,7 +2518,6 @@ int open_ctree(struct super_block *sb, atomic_set(&fs_info->nr_async_submits, 0); atomic_set(&fs_info->async_delalloc_pages, 0); atomic_set(&fs_info->async_submit_draining, 0); - atomic_set(&fs_info->nr_async_bios, 0); atomic_set(&fs_info->defrag_running, 0); atomic_set(&fs_info->qgroup_op_seq, 0); atomic_set(&fs_info->reada_works_cnt, 0); diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 9f26c576911f..763d04b4e5b7 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -360,7 +360,6 @@ static noinline void run_scheduled_bios(struct btrfs_device *device) int again = 0; unsigned long num_run; unsigned long batch_run = 0; - unsigned long limit; unsigned long last_waited = 0; int force_reg = 0; int sync_pending = 0; @@ -375,8 +374,6 @@ static noinline void run_scheduled_bios(struct btrfs_device *device) blk_start_plug(&plug); bdi = device->bdev->bd_bdi; - limit = btrfs_async_submit_limit(fs_info); - limit = limit * 2 / 3; loop: spin_lock(&device->io_lock); @@ -443,13 +440,6 @@ loop_lock: pending = pending->bi_next; cur->bi_next = NULL; - /* - * atomic_dec_return implies a barrier for waitqueue_active - */ - if (atomic_dec_return(&fs_info->nr_async_bios) < limit && - waitqueue_active(&fs_info->async_submit_wait)) - wake_up(&fs_info->async_submit_wait); - BUG_ON(atomic_read(&cur->__bi_cnt) == 0); /* @@ -6075,13 +6065,6 @@ static noinline void btrfs_schedule_bio(struct btrfs_device *device, return; } - /* - * nr_async_bios allows us to reliably return congestion to the - * higher layers. Otherwise, the async bio makes it appear we have - * made progress against dirty pages when we've really just put it - * on a queue for later - */ - atomic_inc(&fs_info->nr_async_bios); WARN_ON(bio->bi_next); bio->bi_next = NULL; -- cgit v1.2.3 From 736cd52e0c720103f52ab9da47b6cc3af6b083f6 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Thu, 7 Sep 2017 11:22:22 -0600 Subject: Btrfs: remove nr_async_submits and async_submit_draining Now that we have the combo of flushing twice, which can make sure IO have started since the second flush will wait for page lock which won't be unlocked unless setting page writeback and queuing ordered extents, we don't need %async_submit_draining, %async_delalloc_pages and %nr_async_submits to tell whether the IO has actually started. Moreover, all the flushers in use are followed by functions that wait for ordered extents to complete, so %nr_async_submits, which tracks whether bio's async submit has made progress, doesn't really make sense. However, %async_delalloc_pages is still required by shrink_delalloc() as that function doesn't flush twice in the normal case (just issues a writeback with WB_REASON_FS_FREE_SPACE). Signed-off-by: Liu Bo Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 2 -- fs/btrfs/disk-io.c | 24 ------------------------ fs/btrfs/inode.c | 28 ---------------------------- 3 files changed, 54 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index bc1b6a033700..7995666af959 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -878,8 +878,6 @@ struct btrfs_fs_info { rwlock_t tree_mod_log_lock; struct rb_root tree_mod_log; - atomic_t nr_async_submits; - atomic_t async_submit_draining; atomic_t async_delalloc_pages; atomic_t open_ioctl_trans; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 2b18b0c379c1..f3e6e8fa19b0 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -740,22 +740,9 @@ static void run_one_async_start(struct btrfs_work *work) static void run_one_async_done(struct btrfs_work *work) { - struct btrfs_fs_info *fs_info; struct async_submit_bio *async; - int limit; async = container_of(work, struct async_submit_bio, work); - fs_info = async->fs_info; - - limit = btrfs_async_submit_limit(fs_info); - limit = limit * 2 / 3; - - /* - * atomic_dec_return implies a barrier for waitqueue_active - */ - if (atomic_dec_return(&fs_info->nr_async_submits) < limit && - waitqueue_active(&fs_info->async_submit_wait)) - wake_up(&fs_info->async_submit_wait); /* If an error occurred we just want to clean up the bio and move on */ if (async->status) { @@ -803,19 +790,10 @@ blk_status_t btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio, async->status = 0; - atomic_inc(&fs_info->nr_async_submits); - if (op_is_sync(bio->bi_opf)) btrfs_set_work_high_priority(&async->work); btrfs_queue_work(fs_info->workers, &async->work); - - while (atomic_read(&fs_info->async_submit_draining) && - atomic_read(&fs_info->nr_async_submits)) { - wait_event(fs_info->async_submit_wait, - (atomic_read(&fs_info->nr_async_submits) == 0)); - } - return 0; } @@ -2515,9 +2493,7 @@ int open_ctree(struct super_block *sb, btrfs_init_block_rsv(&fs_info->empty_block_rsv, BTRFS_BLOCK_RSV_EMPTY); btrfs_init_block_rsv(&fs_info->delayed_block_rsv, BTRFS_BLOCK_RSV_DELOPS); - atomic_set(&fs_info->nr_async_submits, 0); atomic_set(&fs_info->async_delalloc_pages, 0); - atomic_set(&fs_info->async_submit_draining, 0); atomic_set(&fs_info->defrag_running, 0); atomic_set(&fs_info->qgroup_op_seq, 0); atomic_set(&fs_info->reada_works_cnt, 0); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index aaf36f78fa07..4ddb299af472 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1222,13 +1222,6 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page, btrfs_queue_work(fs_info->delalloc_workers, &async_cow->work); - while (atomic_read(&fs_info->async_submit_draining) && - atomic_read(&fs_info->async_delalloc_pages)) { - wait_event(fs_info->async_submit_wait, - (atomic_read(&fs_info->async_delalloc_pages) == - 0)); - } - *nr_written += nr_pages; start = cur_end + 1; } @@ -10332,19 +10325,6 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput) ret = __start_delalloc_inodes(root, delay_iput, -1); if (ret > 0) ret = 0; - /* - * the filemap_flush will queue IO into the worker threads, but - * we have to make sure the IO is actually started and that - * ordered extents get created before we return - */ - atomic_inc(&fs_info->async_submit_draining); - while (atomic_read(&fs_info->nr_async_submits) || - atomic_read(&fs_info->async_delalloc_pages)) { - wait_event(fs_info->async_submit_wait, - (atomic_read(&fs_info->nr_async_submits) == 0 && - atomic_read(&fs_info->async_delalloc_pages) == 0)); - } - atomic_dec(&fs_info->async_submit_draining); return ret; } @@ -10386,14 +10366,6 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput, spin_unlock(&fs_info->delalloc_root_lock); ret = 0; - atomic_inc(&fs_info->async_submit_draining); - while (atomic_read(&fs_info->nr_async_submits) || - atomic_read(&fs_info->async_delalloc_pages)) { - wait_event(fs_info->async_submit_wait, - (atomic_read(&fs_info->nr_async_submits) == 0 && - atomic_read(&fs_info->async_delalloc_pages) == 0)); - } - atomic_dec(&fs_info->async_submit_draining); out: if (!list_empty_careful(&splice)) { spin_lock(&fs_info->delalloc_root_lock); -- cgit v1.2.3 From fd708b81d972a0714b02a60eb4792fdbf15868c4 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 29 Sep 2017 15:43:50 -0400 Subject: Btrfs: add a extent ref verify tool We were having corruption issues that were tied back to problems with the extent tree. In order to track them down I built this tool to try and find the culprit, which was pretty successful. If you compile with this tool on it will live verify every ref update that the fs makes and make sure it is consistent and valid. I've run this through with xfstests and haven't gotten any false positives. Thanks, Signed-off-by: Josef Bacik Reviewed-by: David Sterba [ update error messages, add fixup from Dan Carpenter to handle errors of read_tree_block ] Signed-off-by: David Sterba --- fs/btrfs/Kconfig | 11 + fs/btrfs/Makefile | 1 + fs/btrfs/ctree.h | 5 + fs/btrfs/disk-io.c | 6 + fs/btrfs/extent-tree.c | 22 ++ fs/btrfs/ref-verify.c | 1031 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/btrfs/ref-verify.h | 62 +++ 7 files changed, 1138 insertions(+) create mode 100644 fs/btrfs/ref-verify.c create mode 100644 fs/btrfs/ref-verify.h (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig index a26c63b4ad68..2e558227931a 100644 --- a/fs/btrfs/Kconfig +++ b/fs/btrfs/Kconfig @@ -91,3 +91,14 @@ config BTRFS_ASSERT any of the assertions trip. This is meant for btrfs developers only. If unsure, say N. + +config BTRFS_FS_REF_VERIFY + bool "Btrfs with the ref verify tool compiled in" + depends on BTRFS_FS + default n + help + Enable run-time extent reference verification instrumentation. This + is meant to be used by btrfs developers for tracking down extent + reference problems or verifying they didn't break something. + + If unsure, say N. diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile index 88255e133ade..9216ff22d741 100644 --- a/fs/btrfs/Makefile +++ b/fs/btrfs/Makefile @@ -13,6 +13,7 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \ btrfs-$(CONFIG_BTRFS_FS_POSIX_ACL) += acl.o btrfs-$(CONFIG_BTRFS_FS_CHECK_INTEGRITY) += check-integrity.o +btrfs-$(CONFIG_BTRFS_FS_REF_VERIFY) += ref-verify.o btrfs-$(CONFIG_BTRFS_FS_RUN_SANITY_TESTS) += tests/free-space-tests.o \ tests/extent-buffer-tests.o tests/btrfs-tests.o \ diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 2fcc3c30d471..e2afe524e25e 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1097,6 +1097,11 @@ struct btrfs_fs_info { u32 nodesize; u32 sectorsize; u32 stripesize; + +#ifdef CONFIG_BTRFS_FS_REF_VERIFY + spinlock_t ref_verify_lock; + struct rb_root block_tree; +#endif }; static inline struct btrfs_fs_info *btrfs_sb(struct super_block *sb) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index f3e6e8fa19b0..f971d5680e6f 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -51,6 +51,7 @@ #include "qgroup.h" #include "compression.h" #include "tree-checker.h" +#include "ref-verify.h" #ifdef CONFIG_X86 #include @@ -2509,6 +2510,7 @@ int open_ctree(struct super_block *sb, /* readahead state */ INIT_RADIX_TREE(&fs_info->reada_tree, GFP_NOFS & ~__GFP_DIRECT_RECLAIM); spin_lock_init(&fs_info->reada_lock); + btrfs_init_ref_verify(fs_info); fs_info->thread_pool_size = min_t(unsigned long, num_online_cpus() + 2, 8); @@ -2920,6 +2922,9 @@ retry_root_backup: if (ret) goto fail_trans_kthread; + if (btrfs_build_ref_tree(fs_info)) + btrfs_err(fs_info, "couldn't build ref tree"); + /* do not make disk changes in broken FS or nologreplay is given */ if (btrfs_super_log_root(disk_super) != 0 && !btrfs_test_opt(fs_info, NOLOGREPLAY)) { @@ -3785,6 +3790,7 @@ void close_ctree(struct btrfs_fs_info *fs_info) cleanup_srcu_struct(&fs_info->subvol_srcu); btrfs_free_stripe_hash_table(fs_info); + btrfs_free_ref_cache(fs_info); __btrfs_free_block_rsv(root->orphan_block_rsv); root->orphan_block_rsv = NULL; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index a9787c3c0f7e..52d246994341 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -38,6 +38,7 @@ #include "math.h" #include "sysfs.h" #include "qgroup.h" +#include "ref-verify.h" #undef SCRAMBLE_DELAYED_REFS @@ -2188,6 +2189,9 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, BUG_ON(owner < BTRFS_FIRST_FREE_OBJECTID && root_objectid == BTRFS_TREE_LOG_OBJECTID); + btrfs_ref_tree_mod(root, bytenr, num_bytes, parent, root_objectid, + owner, offset, BTRFS_ADD_DELAYED_REF); + if (owner < BTRFS_FIRST_FREE_OBJECTID) { ret = btrfs_add_delayed_tree_ref(fs_info, trans, bytenr, num_bytes, parent, @@ -7280,6 +7284,10 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) { int old_ref_mod, new_ref_mod; + btrfs_ref_tree_mod(root, buf->start, buf->len, parent, + root->root_key.objectid, + btrfs_header_level(buf), 0, + BTRFS_DROP_DELAYED_REF); ret = btrfs_add_delayed_tree_ref(fs_info, trans, buf->start, buf->len, parent, root->root_key.objectid, @@ -7343,6 +7351,11 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, if (btrfs_is_testing(fs_info)) return 0; + if (root_objectid != BTRFS_TREE_LOG_OBJECTID) + btrfs_ref_tree_mod(root, bytenr, num_bytes, parent, + root_objectid, owner, offset, + BTRFS_DROP_DELAYED_REF); + /* * tree log blocks never actually go into the extent allocation * tree, just update pinning info and exit early. @@ -8318,6 +8331,10 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans, BUG_ON(root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID); + btrfs_ref_tree_mod(root, ins->objectid, ins->offset, 0, + root->root_key.objectid, owner, offset, + BTRFS_ADD_DELAYED_EXTENT); + ret = btrfs_add_delayed_data_ref(fs_info, trans, ins->objectid, ins->offset, 0, root->root_key.objectid, owner, @@ -8542,6 +8559,9 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, extent_op->is_data = false; extent_op->level = level; + btrfs_ref_tree_mod(root, ins.objectid, ins.offset, parent, + root_objectid, level, 0, + BTRFS_ADD_DELAYED_EXTENT); ret = btrfs_add_delayed_tree_ref(fs_info, trans, ins.objectid, ins.offset, parent, root_objectid, level, @@ -10391,6 +10411,8 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, * remove it. */ free_excluded_extents(fs_info, block_group); + btrfs_free_ref_tree_range(fs_info, block_group->key.objectid, + block_group->key.offset); memcpy(&key, &block_group->key, sizeof(key)); index = get_block_group_index(block_group); diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c new file mode 100644 index 000000000000..34878699d363 --- /dev/null +++ b/fs/btrfs/ref-verify.c @@ -0,0 +1,1031 @@ +/* + * Copyright (C) 2014 Facebook. 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. + * + * You should have received a copy of the GNU General Public + * License along with this program; if not, write to the + * Free Software Foundation, Inc., 59 Temple Place - Suite 330, + * Boston, MA 021110-1307, USA. + */ + +#include +#include +#include "ctree.h" +#include "disk-io.h" +#include "locking.h" +#include "delayed-ref.h" +#include "ref-verify.h" + +/* + * Used to keep track the roots and number of refs each root has for a given + * bytenr. This just tracks the number of direct references, no shared + * references. + */ +struct root_entry { + u64 root_objectid; + u64 num_refs; + struct rb_node node; +}; + +/* + * These are meant to represent what should exist in the extent tree, these can + * be used to verify the extent tree is consistent as these should all match + * what the extent tree says. + */ +struct ref_entry { + u64 root_objectid; + u64 parent; + u64 owner; + u64 offset; + u64 num_refs; + struct rb_node node; +}; + +#define MAX_TRACE 16 + +/* + * Whenever we add/remove a reference we record the action. The action maps + * back to the delayed ref action. We hold the ref we are changing in the + * action so we can account for the history properly, and we record the root we + * were called with since it could be different from ref_root. We also store + * stack traces because thats how I roll. + */ +struct ref_action { + int action; + u64 root; + struct ref_entry ref; + struct list_head list; + unsigned long trace[MAX_TRACE]; + unsigned int trace_len; +}; + +/* + * One of these for every block we reference, it holds the roots and references + * to it as well as all of the ref actions that have occured to it. We never + * free it until we unmount the file system in order to make sure re-allocations + * are happening properly. + */ +struct block_entry { + u64 bytenr; + u64 len; + u64 num_refs; + int metadata; + int from_disk; + struct rb_root roots; + struct rb_root refs; + struct rb_node node; + struct list_head actions; +}; + +static struct block_entry *insert_block_entry(struct rb_root *root, + struct block_entry *be) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent_node = NULL; + struct block_entry *entry; + + while (*p) { + parent_node = *p; + entry = rb_entry(parent_node, struct block_entry, node); + if (entry->bytenr > be->bytenr) + p = &(*p)->rb_left; + else if (entry->bytenr < be->bytenr) + p = &(*p)->rb_right; + else + return entry; + } + + rb_link_node(&be->node, parent_node, p); + rb_insert_color(&be->node, root); + return NULL; +} + +static struct block_entry *lookup_block_entry(struct rb_root *root, u64 bytenr) +{ + struct rb_node *n; + struct block_entry *entry = NULL; + + n = root->rb_node; + while (n) { + entry = rb_entry(n, struct block_entry, node); + if (entry->bytenr < bytenr) + n = n->rb_right; + else if (entry->bytenr > bytenr) + n = n->rb_left; + else + return entry; + } + return NULL; +} + +static struct root_entry *insert_root_entry(struct rb_root *root, + struct root_entry *re) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent_node = NULL; + struct root_entry *entry; + + while (*p) { + parent_node = *p; + entry = rb_entry(parent_node, struct root_entry, node); + if (entry->root_objectid > re->root_objectid) + p = &(*p)->rb_left; + else if (entry->root_objectid < re->root_objectid) + p = &(*p)->rb_right; + else + return entry; + } + + rb_link_node(&re->node, parent_node, p); + rb_insert_color(&re->node, root); + return NULL; + +} + +static int comp_refs(struct ref_entry *ref1, struct ref_entry *ref2) +{ + if (ref1->root_objectid < ref2->root_objectid) + return -1; + if (ref1->root_objectid > ref2->root_objectid) + return 1; + if (ref1->parent < ref2->parent) + return -1; + if (ref1->parent > ref2->parent) + return 1; + if (ref1->owner < ref2->owner) + return -1; + if (ref1->owner > ref2->owner) + return 1; + if (ref1->offset < ref2->offset) + return -1; + if (ref1->offset > ref2->offset) + return 1; + return 0; +} + +static struct ref_entry *insert_ref_entry(struct rb_root *root, + struct ref_entry *ref) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent_node = NULL; + struct ref_entry *entry; + int cmp; + + while (*p) { + parent_node = *p; + entry = rb_entry(parent_node, struct ref_entry, node); + cmp = comp_refs(entry, ref); + if (cmp > 0) + p = &(*p)->rb_left; + else if (cmp < 0) + p = &(*p)->rb_right; + else + return entry; + } + + rb_link_node(&ref->node, parent_node, p); + rb_insert_color(&ref->node, root); + return NULL; + +} + +static struct root_entry *lookup_root_entry(struct rb_root *root, u64 objectid) +{ + struct rb_node *n; + struct root_entry *entry = NULL; + + n = root->rb_node; + while (n) { + entry = rb_entry(n, struct root_entry, node); + if (entry->root_objectid < objectid) + n = n->rb_right; + else if (entry->root_objectid > objectid) + n = n->rb_left; + else + return entry; + } + return NULL; +} + +#ifdef CONFIG_STACKTRACE +static void __save_stack_trace(struct ref_action *ra) +{ + struct stack_trace stack_trace; + + stack_trace.max_entries = MAX_TRACE; + stack_trace.nr_entries = 0; + stack_trace.entries = ra->trace; + stack_trace.skip = 2; + save_stack_trace(&stack_trace); + ra->trace_len = stack_trace.nr_entries; +} + +static void __print_stack_trace(struct btrfs_fs_info *fs_info, + struct ref_action *ra) +{ + struct stack_trace trace; + + if (ra->trace_len == 0) { + btrfs_err(fs_info, " ref-verify: no stacktrace"); + return; + } + trace.nr_entries = ra->trace_len; + trace.entries = ra->trace; + print_stack_trace(&trace, 2); +} +#else +static void inline __save_stack_trace(struct ref_action *ra) +{ +} + +static void inline __print_stack_trace(struct btrfs_fs_info *fs_info, + struct ref_action *ra) +{ + btrfs_err(fs_info, " ref-verify: no stacktrace support"); +} +#endif + +static void free_block_entry(struct block_entry *be) +{ + struct root_entry *re; + struct ref_entry *ref; + struct ref_action *ra; + struct rb_node *n; + + while ((n = rb_first(&be->roots))) { + re = rb_entry(n, struct root_entry, node); + rb_erase(&re->node, &be->roots); + kfree(re); + } + + while((n = rb_first(&be->refs))) { + ref = rb_entry(n, struct ref_entry, node); + rb_erase(&ref->node, &be->refs); + kfree(ref); + } + + while (!list_empty(&be->actions)) { + ra = list_first_entry(&be->actions, struct ref_action, + list); + list_del(&ra->list); + kfree(ra); + } + kfree(be); +} + +static struct block_entry *add_block_entry(struct btrfs_fs_info *fs_info, + u64 bytenr, u64 len, + u64 root_objectid) +{ + struct block_entry *be = NULL, *exist; + struct root_entry *re = NULL; + + re = kzalloc(sizeof(struct root_entry), GFP_KERNEL); + be = kzalloc(sizeof(struct block_entry), GFP_KERNEL); + if (!be || !re) { + kfree(re); + kfree(be); + return ERR_PTR(-ENOMEM); + } + be->bytenr = bytenr; + be->len = len; + + re->root_objectid = root_objectid; + re->num_refs = 0; + + spin_lock(&fs_info->ref_verify_lock); + exist = insert_block_entry(&fs_info->block_tree, be); + if (exist) { + if (root_objectid) { + struct root_entry *exist_re; + + exist_re = insert_root_entry(&exist->roots, re); + if (exist_re) + kfree(re); + } + kfree(be); + return exist; + } + + be->num_refs = 0; + be->metadata = 0; + be->from_disk = 0; + be->roots = RB_ROOT; + be->refs = RB_ROOT; + INIT_LIST_HEAD(&be->actions); + if (root_objectid) + insert_root_entry(&be->roots, re); + else + kfree(re); + return be; +} + +static int add_tree_block(struct btrfs_fs_info *fs_info, u64 ref_root, + u64 parent, u64 bytenr, int level) +{ + struct block_entry *be; + struct root_entry *re; + struct ref_entry *ref = NULL, *exist; + + ref = kmalloc(sizeof(struct ref_entry), GFP_KERNEL); + if (!ref) + return -ENOMEM; + + if (parent) + ref->root_objectid = 0; + else + ref->root_objectid = ref_root; + ref->parent = parent; + ref->owner = level; + ref->offset = 0; + ref->num_refs = 1; + + be = add_block_entry(fs_info, bytenr, fs_info->nodesize, ref_root); + if (IS_ERR(be)) { + kfree(ref); + return PTR_ERR(be); + } + be->num_refs++; + be->from_disk = 1; + be->metadata = 1; + + if (!parent) { + ASSERT(ref_root); + re = lookup_root_entry(&be->roots, ref_root); + ASSERT(re); + re->num_refs++; + } + exist = insert_ref_entry(&be->refs, ref); + if (exist) { + exist->num_refs++; + kfree(ref); + } + spin_unlock(&fs_info->ref_verify_lock); + + return 0; +} + +static int add_shared_data_ref(struct btrfs_fs_info *fs_info, + u64 parent, u32 num_refs, u64 bytenr, + u64 num_bytes) +{ + struct block_entry *be; + struct ref_entry *ref; + + ref = kzalloc(sizeof(struct ref_entry), GFP_KERNEL); + if (!ref) + return -ENOMEM; + be = add_block_entry(fs_info, bytenr, num_bytes, 0); + if (IS_ERR(be)) { + kfree(ref); + return PTR_ERR(be); + } + be->num_refs += num_refs; + + ref->parent = parent; + ref->num_refs = num_refs; + if (insert_ref_entry(&be->refs, ref)) { + spin_unlock(&fs_info->ref_verify_lock); + btrfs_err(fs_info, "existing shared ref when reading from disk?"); + kfree(ref); + return -EINVAL; + } + spin_unlock(&fs_info->ref_verify_lock); + return 0; +} + +static int add_extent_data_ref(struct btrfs_fs_info *fs_info, + struct extent_buffer *leaf, + struct btrfs_extent_data_ref *dref, + u64 bytenr, u64 num_bytes) +{ + struct block_entry *be; + struct ref_entry *ref; + struct root_entry *re; + u64 ref_root = btrfs_extent_data_ref_root(leaf, dref); + u64 owner = btrfs_extent_data_ref_objectid(leaf, dref); + u64 offset = btrfs_extent_data_ref_offset(leaf, dref); + u32 num_refs = btrfs_extent_data_ref_count(leaf, dref); + + ref = kzalloc(sizeof(struct ref_entry), GFP_KERNEL); + if (!ref) + return -ENOMEM; + be = add_block_entry(fs_info, bytenr, num_bytes, ref_root); + if (IS_ERR(be)) { + kfree(ref); + return PTR_ERR(be); + } + be->num_refs += num_refs; + + ref->parent = 0; + ref->owner = owner; + ref->root_objectid = ref_root; + ref->offset = offset; + ref->num_refs = num_refs; + if (insert_ref_entry(&be->refs, ref)) { + spin_unlock(&fs_info->ref_verify_lock); + btrfs_err(fs_info, "existing ref when reading from disk?"); + kfree(ref); + return -EINVAL; + } + + re = lookup_root_entry(&be->roots, ref_root); + if (!re) { + spin_unlock(&fs_info->ref_verify_lock); + btrfs_err(fs_info, "missing root in new block entry?"); + return -EINVAL; + } + re->num_refs += num_refs; + spin_unlock(&fs_info->ref_verify_lock); + return 0; +} + +static int process_extent_item(struct btrfs_fs_info *fs_info, + struct btrfs_path *path, struct btrfs_key *key, + int slot, int *tree_block_level) +{ + struct btrfs_extent_item *ei; + struct btrfs_extent_inline_ref *iref; + struct btrfs_extent_data_ref *dref; + struct btrfs_shared_data_ref *sref; + struct extent_buffer *leaf = path->nodes[0]; + u32 item_size = btrfs_item_size_nr(leaf, slot); + unsigned long end, ptr; + u64 offset, flags, count; + int type, ret; + + ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item); + flags = btrfs_extent_flags(leaf, ei); + + if ((key->type == BTRFS_EXTENT_ITEM_KEY) && + flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) { + struct btrfs_tree_block_info *info; + + info = (struct btrfs_tree_block_info *)(ei + 1); + *tree_block_level = btrfs_tree_block_level(leaf, info); + iref = (struct btrfs_extent_inline_ref *)(info + 1); + } else { + if (key->type == BTRFS_METADATA_ITEM_KEY) + *tree_block_level = key->offset; + iref = (struct btrfs_extent_inline_ref *)(ei + 1); + } + + ptr = (unsigned long)iref; + end = (unsigned long)ei + item_size; + while (ptr < end) { + iref = (struct btrfs_extent_inline_ref *)ptr; + type = btrfs_extent_inline_ref_type(leaf, iref); + offset = btrfs_extent_inline_ref_offset(leaf, iref); + switch (type) { + case BTRFS_TREE_BLOCK_REF_KEY: + ret = add_tree_block(fs_info, offset, 0, key->objectid, + *tree_block_level); + break; + case BTRFS_SHARED_BLOCK_REF_KEY: + ret = add_tree_block(fs_info, 0, offset, key->objectid, + *tree_block_level); + break; + case BTRFS_EXTENT_DATA_REF_KEY: + dref = (struct btrfs_extent_data_ref *)(&iref->offset); + ret = add_extent_data_ref(fs_info, leaf, dref, + key->objectid, key->offset); + break; + case BTRFS_SHARED_DATA_REF_KEY: + sref = (struct btrfs_shared_data_ref *)(iref + 1); + count = btrfs_shared_data_ref_count(leaf, sref); + ret = add_shared_data_ref(fs_info, offset, count, + key->objectid, key->offset); + break; + default: + btrfs_err(fs_info, "invalid key type in iref"); + ret = -EINVAL; + break; + } + if (ret) + break; + ptr += btrfs_extent_inline_ref_size(type); + } + return ret; +} + +static int process_leaf(struct btrfs_root *root, + struct btrfs_path *path, u64 *bytenr, u64 *num_bytes) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + struct extent_buffer *leaf = path->nodes[0]; + struct btrfs_extent_data_ref *dref; + struct btrfs_shared_data_ref *sref; + u32 count; + int i = 0, tree_block_level = 0, ret; + struct btrfs_key key; + int nritems = btrfs_header_nritems(leaf); + + for (i = 0; i < nritems; i++) { + btrfs_item_key_to_cpu(leaf, &key, i); + switch (key.type) { + case BTRFS_EXTENT_ITEM_KEY: + *num_bytes = key.offset; + case BTRFS_METADATA_ITEM_KEY: + *bytenr = key.objectid; + ret = process_extent_item(fs_info, path, &key, i, + &tree_block_level); + break; + case BTRFS_TREE_BLOCK_REF_KEY: + ret = add_tree_block(fs_info, key.offset, 0, + key.objectid, tree_block_level); + break; + case BTRFS_SHARED_BLOCK_REF_KEY: + ret = add_tree_block(fs_info, 0, key.offset, + key.objectid, tree_block_level); + break; + case BTRFS_EXTENT_DATA_REF_KEY: + dref = btrfs_item_ptr(leaf, i, + struct btrfs_extent_data_ref); + ret = add_extent_data_ref(fs_info, leaf, dref, *bytenr, + *num_bytes); + break; + case BTRFS_SHARED_DATA_REF_KEY: + sref = btrfs_item_ptr(leaf, i, + struct btrfs_shared_data_ref); + count = btrfs_shared_data_ref_count(leaf, sref); + ret = add_shared_data_ref(fs_info, key.offset, count, + *bytenr, *num_bytes); + break; + default: + break; + } + if (ret) + break; + } + return ret; +} + +/* Walk down to the leaf from the given level */ +static int walk_down_tree(struct btrfs_root *root, struct btrfs_path *path, + int level, u64 *bytenr, u64 *num_bytes) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + struct extent_buffer *eb; + u64 block_bytenr, gen; + int ret = 0; + + while (level >= 0) { + if (level) { + block_bytenr = btrfs_node_blockptr(path->nodes[level], + path->slots[level]); + gen = btrfs_node_ptr_generation(path->nodes[level], + path->slots[level]); + eb = read_tree_block(fs_info, block_bytenr, gen); + if (IS_ERR(eb)) + return PTR_ERR(eb); + if (!extent_buffer_uptodate(eb)) { + free_extent_buffer(eb); + return -EIO; + } + btrfs_tree_read_lock(eb); + btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK); + path->nodes[level-1] = eb; + path->slots[level-1] = 0; + path->locks[level-1] = BTRFS_READ_LOCK_BLOCKING; + } else { + ret = process_leaf(root, path, bytenr, num_bytes); + if (ret) + break; + } + level--; + } + return ret; +} + +/* Walk up to the next node that needs to be processed */ +static int walk_up_tree(struct btrfs_root *root, struct btrfs_path *path, + int *level) +{ + int l; + + for (l = 0; l < BTRFS_MAX_LEVEL; l++) { + if (!path->nodes[l]) + continue; + if (l) { + path->slots[l]++; + if (path->slots[l] < + btrfs_header_nritems(path->nodes[l])) { + *level = l; + return 0; + } + } + btrfs_tree_unlock_rw(path->nodes[l], path->locks[l]); + free_extent_buffer(path->nodes[l]); + path->nodes[l] = NULL; + path->slots[l] = 0; + path->locks[l] = 0; + } + + return 1; +} + +static void dump_ref_action(struct btrfs_fs_info *fs_info, + struct ref_action *ra) +{ + btrfs_err(fs_info, +" Ref action %d, root %llu, ref_root %llu, parent %llu, owner %llu, offset %llu, num_refs %llu", + ra->action, ra->root, ra->ref.root_objectid, ra->ref.parent, + ra->ref.owner, ra->ref.offset, ra->ref.num_refs); + __print_stack_trace(fs_info, ra); +} + +/* + * Dumps all the information from the block entry to printk, it's going to be + * awesome. + */ +static void dump_block_entry(struct btrfs_fs_info *fs_info, + struct block_entry *be) +{ + struct ref_entry *ref; + struct root_entry *re; + struct ref_action *ra; + struct rb_node *n; + + btrfs_err(fs_info, +"dumping block entry [%llu %llu], num_refs %llu, metadata %d, from disk %d", + be->bytenr, be->len, be->num_refs, be->metadata, + be->from_disk); + + for (n = rb_first(&be->refs); n; n = rb_next(n)) { + ref = rb_entry(n, struct ref_entry, node); + btrfs_err(fs_info, +" ref root %llu, parent %llu, owner %llu, offset %llu, num_refs %llu", + ref->root_objectid, ref->parent, ref->owner, + ref->offset, ref->num_refs); + } + + for (n = rb_first(&be->roots); n; n = rb_next(n)) { + re = rb_entry(n, struct root_entry, node); + btrfs_err(fs_info, " root entry %llu, num_refs %llu", + re->root_objectid, re->num_refs); + } + + list_for_each_entry(ra, &be->actions, list) + dump_ref_action(fs_info, ra); +} + +/* + * btrfs_ref_tree_mod: called when we modify a ref for a bytenr + * @root: the root we are making this modification from. + * @bytenr: the bytenr we are modifying. + * @num_bytes: number of bytes. + * @parent: the parent bytenr. + * @ref_root: the original root owner of the bytenr. + * @owner: level in the case of metadata, inode in the case of data. + * @offset: 0 for metadata, file offset for data. + * @action: the action that we are doing, this is the same as the delayed ref + * action. + * + * This will add an action item to the given bytenr and do sanity checks to make + * sure we haven't messed something up. If we are making a new allocation and + * this block entry has history we will delete all previous actions as long as + * our sanity checks pass as they are no longer needed. + */ +int btrfs_ref_tree_mod(struct btrfs_root *root, u64 bytenr, u64 num_bytes, + u64 parent, u64 ref_root, u64 owner, u64 offset, + int action) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + struct ref_entry *ref = NULL, *exist; + struct ref_action *ra = NULL; + struct block_entry *be = NULL; + struct root_entry *re = NULL; + int ret = 0; + bool metadata = owner < BTRFS_FIRST_FREE_OBJECTID; + + if (!btrfs_test_opt(root->fs_info, REF_VERIFY)) + return 0; + + ref = kzalloc(sizeof(struct ref_entry), GFP_NOFS); + ra = kmalloc(sizeof(struct ref_action), GFP_NOFS); + if (!ra || !ref) { + kfree(ref); + kfree(ra); + ret = -ENOMEM; + goto out; + } + + if (parent) { + ref->parent = parent; + } else { + ref->root_objectid = ref_root; + ref->owner = owner; + ref->offset = offset; + } + ref->num_refs = (action == BTRFS_DROP_DELAYED_REF) ? -1 : 1; + + memcpy(&ra->ref, ref, sizeof(struct ref_entry)); + /* + * Save the extra info from the delayed ref in the ref action to make it + * easier to figure out what is happening. The real ref's we add to the + * ref tree need to reflect what we save on disk so it matches any + * on-disk refs we pre-loaded. + */ + ra->ref.owner = owner; + ra->ref.offset = offset; + ra->ref.root_objectid = ref_root; + __save_stack_trace(ra); + + INIT_LIST_HEAD(&ra->list); + ra->action = action; + ra->root = root->objectid; + + /* + * This is an allocation, preallocate the block_entry in case we haven't + * used it before. + */ + ret = -EINVAL; + if (action == BTRFS_ADD_DELAYED_EXTENT) { + /* + * For subvol_create we'll just pass in whatever the parent root + * is and the new root objectid, so let's not treat the passed + * in root as if it really has a ref for this bytenr. + */ + be = add_block_entry(root->fs_info, bytenr, num_bytes, ref_root); + if (IS_ERR(be)) { + kfree(ra); + ret = PTR_ERR(be); + goto out; + } + be->num_refs++; + if (metadata) + be->metadata = 1; + + if (be->num_refs != 1) { + btrfs_err(fs_info, + "re-allocated a block that still has references to it!"); + dump_block_entry(fs_info, be); + dump_ref_action(fs_info, ra); + goto out_unlock; + } + + while (!list_empty(&be->actions)) { + struct ref_action *tmp; + + tmp = list_first_entry(&be->actions, struct ref_action, + list); + list_del(&tmp->list); + kfree(tmp); + } + } else { + struct root_entry *tmp; + + if (!parent) { + re = kmalloc(sizeof(struct root_entry), GFP_NOFS); + if (!re) { + kfree(ref); + kfree(ra); + ret = -ENOMEM; + goto out; + } + /* + * This is the root that is modifying us, so it's the + * one we want to lookup below when we modify the + * re->num_refs. + */ + ref_root = root->objectid; + re->root_objectid = root->objectid; + re->num_refs = 0; + } + + spin_lock(&root->fs_info->ref_verify_lock); + be = lookup_block_entry(&root->fs_info->block_tree, bytenr); + if (!be) { + btrfs_err(fs_info, +"trying to do action %d to bytenr %llu num_bytes %llu but there is no existing entry!", + action, (unsigned long long)bytenr, + (unsigned long long)num_bytes); + dump_ref_action(fs_info, ra); + kfree(ref); + kfree(ra); + goto out_unlock; + } + + if (!parent) { + tmp = insert_root_entry(&be->roots, re); + if (tmp) { + kfree(re); + re = tmp; + } + } + } + + exist = insert_ref_entry(&be->refs, ref); + if (exist) { + if (action == BTRFS_DROP_DELAYED_REF) { + if (exist->num_refs == 0) { + btrfs_err(fs_info, +"dropping a ref for a existing root that doesn't have a ref on the block"); + dump_block_entry(fs_info, be); + dump_ref_action(fs_info, ra); + kfree(ra); + goto out_unlock; + } + exist->num_refs--; + if (exist->num_refs == 0) { + rb_erase(&exist->node, &be->refs); + kfree(exist); + } + } else if (!be->metadata) { + exist->num_refs++; + } else { + btrfs_err(fs_info, +"attempting to add another ref for an existing ref on a tree block"); + dump_block_entry(fs_info, be); + dump_ref_action(fs_info, ra); + kfree(ra); + goto out_unlock; + } + kfree(ref); + } else { + if (action == BTRFS_DROP_DELAYED_REF) { + btrfs_err(fs_info, +"dropping a ref for a root that doesn't have a ref on the block"); + dump_block_entry(fs_info, be); + dump_ref_action(fs_info, ra); + kfree(ra); + goto out_unlock; + } + } + + if (!parent && !re) { + re = lookup_root_entry(&be->roots, ref_root); + if (!re) { + /* + * This shouldn't happen because we will add our re + * above when we lookup the be with !parent, but just in + * case catch this case so we don't panic because I + * didn't thik of some other corner case. + */ + btrfs_err(fs_info, "failed to find root %llu for %llu", + root->objectid, be->bytenr); + dump_block_entry(fs_info, be); + dump_ref_action(fs_info, ra); + kfree(ra); + goto out_unlock; + } + } + if (action == BTRFS_DROP_DELAYED_REF) { + if (re) + re->num_refs--; + be->num_refs--; + } else if (action == BTRFS_ADD_DELAYED_REF) { + be->num_refs++; + if (re) + re->num_refs++; + } + list_add_tail(&ra->list, &be->actions); + ret = 0; +out_unlock: + spin_unlock(&root->fs_info->ref_verify_lock); +out: + if (ret) + btrfs_clear_opt(fs_info->mount_opt, REF_VERIFY); + return ret; +} + +/* Free up the ref cache */ +void btrfs_free_ref_cache(struct btrfs_fs_info *fs_info) +{ + struct block_entry *be; + struct rb_node *n; + + if (!btrfs_test_opt(fs_info, REF_VERIFY)) + return; + + spin_lock(&fs_info->ref_verify_lock); + while ((n = rb_first(&fs_info->block_tree))) { + be = rb_entry(n, struct block_entry, node); + rb_erase(&be->node, &fs_info->block_tree); + free_block_entry(be); + cond_resched_lock(&fs_info->ref_verify_lock); + } + spin_unlock(&fs_info->ref_verify_lock); +} + +void btrfs_free_ref_tree_range(struct btrfs_fs_info *fs_info, u64 start, + u64 len) +{ + struct block_entry *be = NULL, *entry; + struct rb_node *n; + + if (!btrfs_test_opt(fs_info, REF_VERIFY)) + return; + + spin_lock(&fs_info->ref_verify_lock); + n = fs_info->block_tree.rb_node; + while (n) { + entry = rb_entry(n, struct block_entry, node); + if (entry->bytenr < start) { + n = n->rb_right; + } else if (entry->bytenr > start) { + n = n->rb_left; + } else { + be = entry; + break; + } + /* We want to get as close to start as possible */ + if (be == NULL || + (entry->bytenr < start && be->bytenr > start) || + (entry->bytenr < start && entry->bytenr > be->bytenr)) + be = entry; + } + + /* + * Could have an empty block group, maybe have something to check for + * this case to verify we were actually empty? + */ + if (!be) { + spin_unlock(&fs_info->ref_verify_lock); + return; + } + + n = &be->node; + while (n) { + be = rb_entry(n, struct block_entry, node); + n = rb_next(n); + if (be->bytenr < start && be->bytenr + be->len > start) { + btrfs_err(fs_info, + "block entry overlaps a block group [%llu,%llu]!", + start, len); + dump_block_entry(fs_info, be); + continue; + } + if (be->bytenr < start) + continue; + if (be->bytenr >= start + len) + break; + if (be->bytenr + be->len > start + len) { + btrfs_err(fs_info, + "block entry overlaps a block group [%llu,%llu]!", + start, len); + dump_block_entry(fs_info, be); + } + rb_erase(&be->node, &fs_info->block_tree); + free_block_entry(be); + } + spin_unlock(&fs_info->ref_verify_lock); +} + +/* Walk down all roots and build the ref tree, meant to be called at mount */ +int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info) +{ + struct btrfs_path *path; + struct btrfs_root *root; + struct extent_buffer *eb; + u64 bytenr = 0, num_bytes = 0; + int ret, level; + + if (!btrfs_test_opt(fs_info, REF_VERIFY)) + return 0; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + eb = btrfs_read_lock_root_node(fs_info->extent_root); + btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK); + level = btrfs_header_level(eb); + path->nodes[level] = eb; + path->slots[level] = 0; + path->locks[level] = BTRFS_READ_LOCK_BLOCKING; + + while (1) { + /* + * We have to keep track of the bytenr/num_bytes we last hit + * because we could have run out of space for an inline ref, and + * would have had to added a ref key item which may appear on a + * different leaf from the original extent item. + */ + ret = walk_down_tree(fs_info->extent_root, path, level, + &bytenr, &num_bytes); + if (ret) + break; + ret = walk_up_tree(root, path, &level); + if (ret < 0) + break; + if (ret > 0) { + ret = 0; + break; + } + } + if (ret) { + btrfs_clear_opt(fs_info->mount_opt, REF_VERIFY); + btrfs_free_ref_cache(fs_info); + } + btrfs_free_path(path); + return ret; +} diff --git a/fs/btrfs/ref-verify.h b/fs/btrfs/ref-verify.h new file mode 100644 index 000000000000..3bf02ce0e1e2 --- /dev/null +++ b/fs/btrfs/ref-verify.h @@ -0,0 +1,62 @@ +/* + * Copyright (C) 2014 Facebook. 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. + * + * You should have received a copy of the GNU General Public + * License along with this program; if not, write to the + * Free Software Foundation, Inc., 59 Temple Place - Suite 330, + * Boston, MA 021110-1307, USA. + */ +#ifndef __REF_VERIFY__ +#define __REF_VERIFY__ + +#ifdef CONFIG_BTRFS_FS_REF_VERIFY +int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info); +void btrfs_free_ref_cache(struct btrfs_fs_info *fs_info); +int btrfs_ref_tree_mod(struct btrfs_root *root, u64 bytenr, u64 num_bytes, + u64 parent, u64 ref_root, u64 owner, u64 offset, + int action); +void btrfs_free_ref_tree_range(struct btrfs_fs_info *fs_info, u64 start, + u64 len); + +static inline void btrfs_init_ref_verify(struct btrfs_fs_info *fs_info) +{ + spin_lock_init(&fs_info->ref_verify_lock); + fs_info->block_tree = RB_ROOT; +} +#else +static inline int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info) +{ + return 0; +} + +static inline void btrfs_free_ref_cache(struct btrfs_fs_info *fs_info) +{ +} + +static inline int btrfs_ref_tree_mod(struct btrfs_root *root, u64 bytenr, + u64 num_bytes, u64 parent, u64 ref_root, + u64 owner, u64 offset, int action) +{ + return 0; +} + +static inline void btrfs_free_ref_tree_range(struct btrfs_fs_info *fs_info, + u64 start, u64 len) +{ +} + +static inline void btrfs_init_ref_verify(struct btrfs_fs_info *fs_info) +{ +} + +#endif /* CONFIG_BTRFS_FS_REF_VERIFY */ +#endif /* _REF_VERIFY__ */ -- cgit v1.2.3 From d278850eff3053ef166cf64c16f798dfe36278a2 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 29 Sep 2017 15:43:57 -0400 Subject: btrfs: remove delayed_ref_node from ref_head This is just excessive information in the ref_head, and makes the code complicated. It is a relic from when we had the heads and the refs in the same tree, which is no longer the case. With this removal I've cleaned up a bunch of the cruft around this old assumption as well. Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/backref.c | 4 +- fs/btrfs/delayed-ref.c | 126 +++++++++++++++++++------------------------ fs/btrfs/delayed-ref.h | 49 ++++++----------- fs/btrfs/disk-io.c | 12 ++--- fs/btrfs/extent-tree.c | 90 ++++++++++++------------------- include/trace/events/btrfs.h | 13 ++--- 6 files changed, 119 insertions(+), 175 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index b517ef1477ea..33cba1abf8b6 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -1178,7 +1178,7 @@ again: head = btrfs_find_delayed_ref_head(delayed_refs, bytenr); if (head) { if (!mutex_trylock(&head->mutex)) { - refcount_inc(&head->node.refs); + refcount_inc(&head->refs); spin_unlock(&delayed_refs->lock); btrfs_release_path(path); @@ -1189,7 +1189,7 @@ again: */ mutex_lock(&head->mutex); mutex_unlock(&head->mutex); - btrfs_put_delayed_ref(&head->node); + btrfs_put_delayed_ref_head(head); goto again; } spin_unlock(&delayed_refs->lock); diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 93ffa898df6d..b9b41c838da4 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -96,15 +96,15 @@ static struct btrfs_delayed_ref_head *htree_insert(struct rb_root *root, u64 bytenr; ins = rb_entry(node, struct btrfs_delayed_ref_head, href_node); - bytenr = ins->node.bytenr; + bytenr = ins->bytenr; while (*p) { parent_node = *p; entry = rb_entry(parent_node, struct btrfs_delayed_ref_head, href_node); - if (bytenr < entry->node.bytenr) + if (bytenr < entry->bytenr) p = &(*p)->rb_left; - else if (bytenr > entry->node.bytenr) + else if (bytenr > entry->bytenr) p = &(*p)->rb_right; else return entry; @@ -133,15 +133,15 @@ find_ref_head(struct rb_root *root, u64 bytenr, while (n) { entry = rb_entry(n, struct btrfs_delayed_ref_head, href_node); - if (bytenr < entry->node.bytenr) + if (bytenr < entry->bytenr) n = n->rb_left; - else if (bytenr > entry->node.bytenr) + else if (bytenr > entry->bytenr) n = n->rb_right; else return entry; } if (entry && return_bigger) { - if (bytenr > entry->node.bytenr) { + if (bytenr > entry->bytenr) { n = rb_next(&entry->href_node); if (!n) n = rb_first(root); @@ -164,17 +164,17 @@ int btrfs_delayed_ref_lock(struct btrfs_trans_handle *trans, if (mutex_trylock(&head->mutex)) return 0; - refcount_inc(&head->node.refs); + refcount_inc(&head->refs); spin_unlock(&delayed_refs->lock); mutex_lock(&head->mutex); spin_lock(&delayed_refs->lock); - if (!head->node.in_tree) { + if (RB_EMPTY_NODE(&head->href_node)) { mutex_unlock(&head->mutex); - btrfs_put_delayed_ref(&head->node); + btrfs_put_delayed_ref_head(head); return -EAGAIN; } - btrfs_put_delayed_ref(&head->node); + btrfs_put_delayed_ref_head(head); return 0; } @@ -183,15 +183,10 @@ static inline void drop_delayed_ref(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_head *head, struct btrfs_delayed_ref_node *ref) { - if (btrfs_delayed_ref_is_head(ref)) { - head = btrfs_delayed_node_to_head(ref); - rb_erase(&head->href_node, &delayed_refs->href_root); - } else { - assert_spin_locked(&head->lock); - list_del(&ref->list); - if (!list_empty(&ref->add_list)) - list_del(&ref->add_list); - } + assert_spin_locked(&head->lock); + list_del(&ref->list); + if (!list_empty(&ref->add_list)) + list_del(&ref->add_list); ref->in_tree = 0; btrfs_put_delayed_ref(ref); atomic_dec(&delayed_refs->num_entries); @@ -380,8 +375,8 @@ again: head->processing = 1; WARN_ON(delayed_refs->num_heads_ready == 0); delayed_refs->num_heads_ready--; - delayed_refs->run_delayed_start = head->node.bytenr + - head->node.num_bytes; + delayed_refs->run_delayed_start = head->bytenr + + head->num_bytes; return head; } @@ -469,20 +464,16 @@ add_tail: */ static noinline void update_existing_head_ref(struct btrfs_delayed_ref_root *delayed_refs, - struct btrfs_delayed_ref_node *existing, - struct btrfs_delayed_ref_node *update, + struct btrfs_delayed_ref_head *existing, + struct btrfs_delayed_ref_head *update, int *old_ref_mod_ret) { - struct btrfs_delayed_ref_head *existing_ref; - struct btrfs_delayed_ref_head *ref; int old_ref_mod; - existing_ref = btrfs_delayed_node_to_head(existing); - ref = btrfs_delayed_node_to_head(update); - BUG_ON(existing_ref->is_data != ref->is_data); + BUG_ON(existing->is_data != update->is_data); - spin_lock(&existing_ref->lock); - if (ref->must_insert_reserved) { + spin_lock(&existing->lock); + if (update->must_insert_reserved) { /* if the extent was freed and then * reallocated before the delayed ref * entries were processed, we can end up @@ -490,7 +481,7 @@ update_existing_head_ref(struct btrfs_delayed_ref_root *delayed_refs, * the must_insert_reserved flag set. * Set it again here */ - existing_ref->must_insert_reserved = ref->must_insert_reserved; + existing->must_insert_reserved = update->must_insert_reserved; /* * update the num_bytes so we make sure the accounting @@ -500,22 +491,22 @@ update_existing_head_ref(struct btrfs_delayed_ref_root *delayed_refs, } - if (ref->extent_op) { - if (!existing_ref->extent_op) { - existing_ref->extent_op = ref->extent_op; + if (update->extent_op) { + if (!existing->extent_op) { + existing->extent_op = update->extent_op; } else { - if (ref->extent_op->update_key) { - memcpy(&existing_ref->extent_op->key, - &ref->extent_op->key, - sizeof(ref->extent_op->key)); - existing_ref->extent_op->update_key = true; + if (update->extent_op->update_key) { + memcpy(&existing->extent_op->key, + &update->extent_op->key, + sizeof(update->extent_op->key)); + existing->extent_op->update_key = true; } - if (ref->extent_op->update_flags) { - existing_ref->extent_op->flags_to_set |= - ref->extent_op->flags_to_set; - existing_ref->extent_op->update_flags = true; + if (update->extent_op->update_flags) { + existing->extent_op->flags_to_set |= + update->extent_op->flags_to_set; + existing->extent_op->update_flags = true; } - btrfs_free_delayed_extent_op(ref->extent_op); + btrfs_free_delayed_extent_op(update->extent_op); } } /* @@ -523,23 +514,23 @@ update_existing_head_ref(struct btrfs_delayed_ref_root *delayed_refs, * only need the lock for this case cause we could be processing it * currently, for refs we just added we know we're a-ok. */ - old_ref_mod = existing_ref->total_ref_mod; + old_ref_mod = existing->total_ref_mod; if (old_ref_mod_ret) *old_ref_mod_ret = old_ref_mod; existing->ref_mod += update->ref_mod; - existing_ref->total_ref_mod += update->ref_mod; + existing->total_ref_mod += update->ref_mod; /* * If we are going to from a positive ref mod to a negative or vice * versa we need to make sure to adjust pending_csums accordingly. */ - if (existing_ref->is_data) { - if (existing_ref->total_ref_mod >= 0 && old_ref_mod < 0) + if (existing->is_data) { + if (existing->total_ref_mod >= 0 && old_ref_mod < 0) delayed_refs->pending_csums -= existing->num_bytes; - if (existing_ref->total_ref_mod < 0 && old_ref_mod >= 0) + if (existing->total_ref_mod < 0 && old_ref_mod >= 0) delayed_refs->pending_csums += existing->num_bytes; } - spin_unlock(&existing_ref->lock); + spin_unlock(&existing->lock); } /* @@ -550,14 +541,13 @@ update_existing_head_ref(struct btrfs_delayed_ref_root *delayed_refs, static noinline struct btrfs_delayed_ref_head * add_delayed_ref_head(struct btrfs_fs_info *fs_info, struct btrfs_trans_handle *trans, - struct btrfs_delayed_ref_node *ref, + struct btrfs_delayed_ref_head *head_ref, struct btrfs_qgroup_extent_record *qrecord, u64 bytenr, u64 num_bytes, u64 ref_root, u64 reserved, int action, int is_data, int *qrecord_inserted_ret, int *old_ref_mod, int *new_ref_mod) { struct btrfs_delayed_ref_head *existing; - struct btrfs_delayed_ref_head *head_ref = NULL; struct btrfs_delayed_ref_root *delayed_refs; int count_mod = 1; int must_insert_reserved = 0; @@ -593,26 +583,21 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info, delayed_refs = &trans->transaction->delayed_refs; - /* first set the basic ref node struct up */ - refcount_set(&ref->refs, 1); - ref->bytenr = bytenr; - ref->num_bytes = num_bytes; - ref->ref_mod = count_mod; - ref->type = 0; - ref->action = 0; - ref->is_head = 1; - ref->in_tree = 1; - ref->seq = 0; - - head_ref = btrfs_delayed_node_to_head(ref); + refcount_set(&head_ref->refs, 1); + head_ref->bytenr = bytenr; + head_ref->num_bytes = num_bytes; + head_ref->ref_mod = count_mod; head_ref->must_insert_reserved = must_insert_reserved; head_ref->is_data = is_data; INIT_LIST_HEAD(&head_ref->ref_list); INIT_LIST_HEAD(&head_ref->ref_add_list); + RB_CLEAR_NODE(&head_ref->href_node); head_ref->processing = 0; head_ref->total_ref_mod = count_mod; head_ref->qgroup_reserved = 0; head_ref->qgroup_ref_root = 0; + spin_lock_init(&head_ref->lock); + mutex_init(&head_ref->mutex); /* Record qgroup extent info if provided */ if (qrecord) { @@ -632,17 +617,14 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info, qrecord_inserted = 1; } - spin_lock_init(&head_ref->lock); - mutex_init(&head_ref->mutex); - - trace_add_delayed_ref_head(fs_info, ref, head_ref, action); + trace_add_delayed_ref_head(fs_info, head_ref, action); existing = htree_insert(&delayed_refs->href_root, &head_ref->href_node); if (existing) { WARN_ON(ref_root && reserved && existing->qgroup_ref_root && existing->qgroup_reserved); - update_existing_head_ref(delayed_refs, &existing->node, ref, + update_existing_head_ref(delayed_refs, existing, head_ref, old_ref_mod); /* * we've updated the existing ref, free the newly @@ -821,7 +803,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info, * insert both the head node and the new ref without dropping * the spin lock */ - head_ref = add_delayed_ref_head(fs_info, trans, &head_ref->node, record, + head_ref = add_delayed_ref_head(fs_info, trans, head_ref, record, bytenr, num_bytes, 0, 0, action, 0, &qrecord_inserted, old_ref_mod, new_ref_mod); @@ -888,7 +870,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info, * insert both the head node and the new ref without dropping * the spin lock */ - head_ref = add_delayed_ref_head(fs_info, trans, &head_ref->node, record, + head_ref = add_delayed_ref_head(fs_info, trans, head_ref, record, bytenr, num_bytes, ref_root, reserved, action, 1, &qrecord_inserted, old_ref_mod, new_ref_mod); @@ -920,7 +902,7 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info, delayed_refs = &trans->transaction->delayed_refs; spin_lock(&delayed_refs->lock); - add_delayed_ref_head(fs_info, trans, &head_ref->node, NULL, bytenr, + add_delayed_ref_head(fs_info, trans, head_ref, NULL, bytenr, num_bytes, 0, 0, BTRFS_UPDATE_DELAYED_HEAD, extent_op->is_data, NULL, NULL, NULL); diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index ce88e4ac5276..1ce11858d727 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -26,15 +26,6 @@ #define BTRFS_ADD_DELAYED_EXTENT 3 /* record a full extent allocation */ #define BTRFS_UPDATE_DELAYED_HEAD 4 /* not changing ref count on head ref */ -/* - * XXX: Qu: I really hate the design that ref_head and tree/data ref shares the - * same ref_node structure. - * Ref_head is in a higher logic level than tree/data ref, and duplicated - * bytenr/num_bytes in ref_node is really a waste or memory, they should be - * referred from ref_head. - * This gets more disgusting after we use list to store tree/data ref in - * ref_head. Must clean this mess up later. - */ struct btrfs_delayed_ref_node { /*data/tree ref use list, stored in ref_head->ref_list. */ struct list_head list; @@ -91,8 +82,9 @@ struct btrfs_delayed_extent_op { * reference count modifications we've queued up. */ struct btrfs_delayed_ref_head { - struct btrfs_delayed_ref_node node; - + u64 bytenr; + u64 num_bytes; + refcount_t refs; /* * the mutex is held while running the refs, and it is also * held when checking the sum of reference modifications. @@ -115,6 +107,14 @@ struct btrfs_delayed_ref_head { */ int total_ref_mod; + /* + * This is the current outstanding mod references for this bytenr. This + * is used with lookup_extent_info to get an accurate reference count + * for a bytenr, so it is adjusted as delayed refs are run so that any + * on disk reference count + ref_mod is accurate. + */ + int ref_mod; + /* * For qgroup reserved space freeing. * @@ -234,15 +234,18 @@ static inline void btrfs_put_delayed_ref(struct btrfs_delayed_ref_node *ref) case BTRFS_SHARED_DATA_REF_KEY: kmem_cache_free(btrfs_delayed_data_ref_cachep, ref); break; - case 0: - kmem_cache_free(btrfs_delayed_ref_head_cachep, ref); - break; default: BUG(); } } } +static inline void btrfs_put_delayed_ref_head(struct btrfs_delayed_ref_head *head) +{ + if (refcount_dec_and_test(&head->refs)) + kmem_cache_free(btrfs_delayed_ref_head_cachep, head); +} + int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info, struct btrfs_trans_handle *trans, u64 bytenr, u64 num_bytes, u64 parent, @@ -282,36 +285,18 @@ int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, struct btrfs_delayed_ref_root *delayed_refs, u64 seq); -/* - * a node might live in a head or a regular ref, this lets you - * test for the proper type to use. - */ -static int btrfs_delayed_ref_is_head(struct btrfs_delayed_ref_node *node) -{ - return node->is_head; -} - /* * helper functions to cast a node into its container */ static inline struct btrfs_delayed_tree_ref * btrfs_delayed_node_to_tree_ref(struct btrfs_delayed_ref_node *node) { - WARN_ON(btrfs_delayed_ref_is_head(node)); return container_of(node, struct btrfs_delayed_tree_ref, node); } static inline struct btrfs_delayed_data_ref * btrfs_delayed_node_to_data_ref(struct btrfs_delayed_ref_node *node) { - WARN_ON(btrfs_delayed_ref_is_head(node)); return container_of(node, struct btrfs_delayed_data_ref, node); } - -static inline struct btrfs_delayed_ref_head * -btrfs_delayed_node_to_head(struct btrfs_delayed_ref_node *node) -{ - WARN_ON(!btrfs_delayed_ref_is_head(node)); - return container_of(node, struct btrfs_delayed_ref_head, node); -} #endif diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index f971d5680e6f..69ce738c00d0 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -4121,12 +4121,12 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans, head = rb_entry(node, struct btrfs_delayed_ref_head, href_node); if (!mutex_trylock(&head->mutex)) { - refcount_inc(&head->node.refs); + refcount_inc(&head->refs); spin_unlock(&delayed_refs->lock); mutex_lock(&head->mutex); mutex_unlock(&head->mutex); - btrfs_put_delayed_ref(&head->node); + btrfs_put_delayed_ref_head(head); spin_lock(&delayed_refs->lock); continue; } @@ -4147,16 +4147,16 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans, if (head->processing == 0) delayed_refs->num_heads_ready--; atomic_dec(&delayed_refs->num_entries); - head->node.in_tree = 0; rb_erase(&head->href_node, &delayed_refs->href_root); + RB_CLEAR_NODE(&head->href_node); spin_unlock(&head->lock); spin_unlock(&delayed_refs->lock); mutex_unlock(&head->mutex); if (pin_bytes) - btrfs_pin_extent(fs_info, head->node.bytenr, - head->node.num_bytes, 1); - btrfs_put_delayed_ref(&head->node); + btrfs_pin_extent(fs_info, head->bytenr, + head->num_bytes, 1); + btrfs_put_delayed_ref_head(head); cond_resched(); spin_lock(&delayed_refs->lock); } diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 628ae71094f2..423d89145bac 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -912,7 +912,7 @@ search_again: head = btrfs_find_delayed_ref_head(delayed_refs, bytenr); if (head) { if (!mutex_trylock(&head->mutex)) { - refcount_inc(&head->node.refs); + refcount_inc(&head->refs); spin_unlock(&delayed_refs->lock); btrfs_release_path(path); @@ -923,7 +923,7 @@ search_again: */ mutex_lock(&head->mutex); mutex_unlock(&head->mutex); - btrfs_put_delayed_ref(&head->node); + btrfs_put_delayed_ref_head(head); goto search_again; } spin_lock(&head->lock); @@ -932,7 +932,7 @@ search_again: else BUG_ON(num_refs == 0); - num_refs += head->node.ref_mod; + num_refs += head->ref_mod; spin_unlock(&head->lock); mutex_unlock(&head->mutex); } @@ -2337,7 +2337,7 @@ static void __run_delayed_extent_op(struct btrfs_delayed_extent_op *extent_op, static int run_delayed_extent_op(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, - struct btrfs_delayed_ref_node *node, + struct btrfs_delayed_ref_head *head, struct btrfs_delayed_extent_op *extent_op) { struct btrfs_key key; @@ -2359,14 +2359,14 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans, if (!path) return -ENOMEM; - key.objectid = node->bytenr; + key.objectid = head->bytenr; if (metadata) { key.type = BTRFS_METADATA_ITEM_KEY; key.offset = extent_op->level; } else { key.type = BTRFS_EXTENT_ITEM_KEY; - key.offset = node->num_bytes; + key.offset = head->num_bytes; } again: @@ -2383,17 +2383,17 @@ again: path->slots[0]--; btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); - if (key.objectid == node->bytenr && + if (key.objectid == head->bytenr && key.type == BTRFS_EXTENT_ITEM_KEY && - key.offset == node->num_bytes) + key.offset == head->num_bytes) ret = 0; } if (ret > 0) { btrfs_release_path(path); metadata = 0; - key.objectid = node->bytenr; - key.offset = node->num_bytes; + key.objectid = head->bytenr; + key.offset = head->num_bytes; key.type = BTRFS_EXTENT_ITEM_KEY; goto again; } @@ -2562,7 +2562,7 @@ static int cleanup_extent_op(struct btrfs_trans_handle *trans, return 0; } spin_unlock(&head->lock); - ret = run_delayed_extent_op(trans, fs_info, &head->node, extent_op); + ret = run_delayed_extent_op(trans, fs_info, head, extent_op); btrfs_free_delayed_extent_op(extent_op); return ret ? ret : 1; } @@ -2597,39 +2597,37 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans, spin_unlock(&delayed_refs->lock); return 1; } - head->node.in_tree = 0; delayed_refs->num_heads--; rb_erase(&head->href_node, &delayed_refs->href_root); + RB_CLEAR_NODE(&head->href_node); spin_unlock(&delayed_refs->lock); spin_unlock(&head->lock); atomic_dec(&delayed_refs->num_entries); - trace_run_delayed_ref_head(fs_info, &head->node, head, - head->node.action); + trace_run_delayed_ref_head(fs_info, head, 0); if (head->total_ref_mod < 0) { struct btrfs_block_group_cache *cache; - cache = btrfs_lookup_block_group(fs_info, head->node.bytenr); + cache = btrfs_lookup_block_group(fs_info, head->bytenr); ASSERT(cache); percpu_counter_add(&cache->space_info->total_bytes_pinned, - -head->node.num_bytes); + -head->num_bytes); btrfs_put_block_group(cache); if (head->is_data) { spin_lock(&delayed_refs->lock); - delayed_refs->pending_csums -= head->node.num_bytes; + delayed_refs->pending_csums -= head->num_bytes; spin_unlock(&delayed_refs->lock); } } if (head->must_insert_reserved) { - btrfs_pin_extent(fs_info, head->node.bytenr, - head->node.num_bytes, 1); + btrfs_pin_extent(fs_info, head->bytenr, + head->num_bytes, 1); if (head->is_data) { - ret = btrfs_del_csums(trans, fs_info, - head->node.bytenr, - head->node.num_bytes); + ret = btrfs_del_csums(trans, fs_info, head->bytenr, + head->num_bytes); } } @@ -2637,7 +2635,7 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans, btrfs_qgroup_free_delayed_ref(fs_info, head->qgroup_ref_root, head->qgroup_reserved); btrfs_delayed_ref_unlock(head); - btrfs_put_delayed_ref(&head->node); + btrfs_put_delayed_ref_head(head); return 0; } @@ -2751,10 +2749,10 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, switch (ref->action) { case BTRFS_ADD_DELAYED_REF: case BTRFS_ADD_DELAYED_EXTENT: - locked_ref->node.ref_mod -= ref->ref_mod; + locked_ref->ref_mod -= ref->ref_mod; break; case BTRFS_DROP_DELAYED_REF: - locked_ref->node.ref_mod += ref->ref_mod; + locked_ref->ref_mod += ref->ref_mod; break; default: WARN_ON(1); @@ -3087,33 +3085,16 @@ again: spin_unlock(&delayed_refs->lock); goto out; } + head = rb_entry(node, struct btrfs_delayed_ref_head, + href_node); + refcount_inc(&head->refs); + spin_unlock(&delayed_refs->lock); - while (node) { - head = rb_entry(node, struct btrfs_delayed_ref_head, - href_node); - if (btrfs_delayed_ref_is_head(&head->node)) { - struct btrfs_delayed_ref_node *ref; - - ref = &head->node; - refcount_inc(&ref->refs); - - spin_unlock(&delayed_refs->lock); - /* - * Mutex was contended, block until it's - * released and try again - */ - mutex_lock(&head->mutex); - mutex_unlock(&head->mutex); + /* Mutex was contended, block until it's released and retry. */ + mutex_lock(&head->mutex); + mutex_unlock(&head->mutex); - btrfs_put_delayed_ref(ref); - cond_resched(); - goto again; - } else { - WARN_ON(1); - } - node = rb_next(node); - } - spin_unlock(&delayed_refs->lock); + btrfs_put_delayed_ref_head(head); cond_resched(); goto again; } @@ -3171,7 +3152,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, } if (!mutex_trylock(&head->mutex)) { - refcount_inc(&head->node.refs); + refcount_inc(&head->refs); spin_unlock(&delayed_refs->lock); btrfs_release_path(path); @@ -3182,7 +3163,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, */ mutex_lock(&head->mutex); mutex_unlock(&head->mutex); - btrfs_put_delayed_ref(&head->node); + btrfs_put_delayed_ref_head(head); return -EAGAIN; } spin_unlock(&delayed_refs->lock); @@ -7235,9 +7216,8 @@ static noinline int check_ref_cleanup(struct btrfs_trans_handle *trans, * at this point we have a head with no other entries. Go * ahead and process it. */ - head->node.in_tree = 0; rb_erase(&head->href_node, &delayed_refs->href_root); - + RB_CLEAR_NODE(&head->href_node); atomic_dec(&delayed_refs->num_entries); /* @@ -7256,7 +7236,7 @@ static noinline int check_ref_cleanup(struct btrfs_trans_handle *trans, ret = 1; mutex_unlock(&head->mutex); - btrfs_put_delayed_ref(&head->node); + btrfs_put_delayed_ref_head(head); return ret; out: spin_unlock(&head->lock); diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index 77437f545c63..bfe2f23b578c 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -798,11 +798,10 @@ DEFINE_EVENT(btrfs_delayed_data_ref, run_delayed_data_ref, DECLARE_EVENT_CLASS(btrfs_delayed_ref_head, TP_PROTO(const struct btrfs_fs_info *fs_info, - const struct btrfs_delayed_ref_node *ref, const struct btrfs_delayed_ref_head *head_ref, int action), - TP_ARGS(fs_info, ref, head_ref, action), + TP_ARGS(fs_info, head_ref, action), TP_STRUCT__entry_btrfs( __field( u64, bytenr ) @@ -812,8 +811,8 @@ DECLARE_EVENT_CLASS(btrfs_delayed_ref_head, ), TP_fast_assign_btrfs(fs_info, - __entry->bytenr = ref->bytenr; - __entry->num_bytes = ref->num_bytes; + __entry->bytenr = head_ref->bytenr; + __entry->num_bytes = head_ref->num_bytes; __entry->action = action; __entry->is_data = head_ref->is_data; ), @@ -828,21 +827,19 @@ DECLARE_EVENT_CLASS(btrfs_delayed_ref_head, DEFINE_EVENT(btrfs_delayed_ref_head, add_delayed_ref_head, TP_PROTO(const struct btrfs_fs_info *fs_info, - const struct btrfs_delayed_ref_node *ref, const struct btrfs_delayed_ref_head *head_ref, int action), - TP_ARGS(fs_info, ref, head_ref, action) + TP_ARGS(fs_info, head_ref, action) ); DEFINE_EVENT(btrfs_delayed_ref_head, run_delayed_ref_head, TP_PROTO(const struct btrfs_fs_info *fs_info, - const struct btrfs_delayed_ref_node *ref, const struct btrfs_delayed_ref_head *head_ref, int action), - TP_ARGS(fs_info, ref, head_ref, action) + TP_ARGS(fs_info, head_ref, action) ); #define show_chunk_type(type) \ -- cgit v1.2.3 From d4417e22551377c6e589c15ff2b931610e5230bc Mon Sep 17 00:00:00 2001 From: Nikolay Borisov Date: Mon, 16 Oct 2017 16:48:40 +0300 Subject: btrfs: Replace opencoded sizes with their symbolic constants Currently btrfs' code uses a mix of opencoded sizes and defines from sizes.h. Let's unifiy the code base to always use the symbolic constants. No functional changes Signed-off-by: Nikolay Borisov Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 2 +- fs/btrfs/disk-io.c | 2 +- fs/btrfs/super.c | 2 +- fs/btrfs/tests/inode-tests.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index e2afe524e25e..7bda8429e93f 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -523,7 +523,7 @@ struct btrfs_caching_control { }; /* Once caching_thread() finds this much free space, it will wake up waiters. */ -#define CACHING_CTL_WAKE_UP (1024 * 1024 * 2) +#define CACHING_CTL_WAKE_UP SZ_2M struct btrfs_io_ctl { void *cur, *orig; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 69ce738c00d0..484cf8fc952c 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2733,7 +2733,7 @@ int open_ctree(struct super_block *sb, sb->s_bdi->congested_fn = btrfs_congested_fn; sb->s_bdi->congested_data = fs_info; sb->s_bdi->capabilities |= BDI_CAP_CGROUP_WRITEBACK; - sb->s_bdi->ra_pages = VM_MAX_READAHEAD * 1024 / PAGE_SIZE; + sb->s_bdi->ra_pages = VM_MAX_READAHEAD * SZ_1K / PAGE_SIZE; sb->s_bdi->ra_pages *= btrfs_super_num_devices(disk_super); sb->s_bdi->ra_pages = max(sb->s_bdi->ra_pages, SZ_4M / PAGE_SIZE); diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 880ab4949f69..4fb7eefb80ae 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -2126,7 +2126,7 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf) * succeed even if the Avail is zero. But this is better than the other * way around. */ - thresh = 4 * 1024 * 1024; + thresh = SZ_4M; if (!mixed && total_free_meta - thresh < block_rsv->size) buf->f_bavail = 0; diff --git a/fs/btrfs/tests/inode-tests.c b/fs/btrfs/tests/inode-tests.c index 8c91d03cc82d..330815eb07b4 100644 --- a/fs/btrfs/tests/inode-tests.c +++ b/fs/btrfs/tests/inode-tests.c @@ -770,7 +770,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) offset = em->start + em->len; free_extent_map(em); - em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, 4096 * 1024, 0); + em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, SZ_4M, 0); if (IS_ERR(em)) { test_msg("Got an error when we shouldn't have\n"); goto out; -- cgit v1.2.3 From 69fe2d75dd91d0124ad2ab6e9fef07633bd730e0 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Thu, 19 Oct 2017 14:15:57 -0400 Subject: btrfs: make the delalloc block rsv per inode The way we handle delalloc metadata reservations has gotten progressively more complicated over the years. There is so much cruft and weirdness around keeping the reserved count and outstanding counters consistent and handling the error cases that it's impossible to understand. Fix this by making the delalloc block rsv per-inode. This way we can calculate the actual size of the outstanding metadata reservations every time we make a change, and then reserve the delta based on that amount. This greatly simplifies the code everywhere, and makes the error handling in btrfs_delalloc_reserve_metadata far less terrifying. Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/btrfs_inode.h | 26 ++-- fs/btrfs/ctree.h | 5 +- fs/btrfs/delayed-inode.c | 46 +------ fs/btrfs/disk-io.c | 18 ++- fs/btrfs/extent-tree.c | 320 ++++++++++++++++------------------------------- fs/btrfs/inode.c | 18 +-- 6 files changed, 141 insertions(+), 292 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 427c8738a3bd..63f0ccc92a71 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -36,14 +36,13 @@ #define BTRFS_INODE_ORPHAN_META_RESERVED 1 #define BTRFS_INODE_DUMMY 2 #define BTRFS_INODE_IN_DEFRAG 3 -#define BTRFS_INODE_DELALLOC_META_RESERVED 4 -#define BTRFS_INODE_HAS_ORPHAN_ITEM 5 -#define BTRFS_INODE_HAS_ASYNC_EXTENT 6 -#define BTRFS_INODE_NEEDS_FULL_SYNC 7 -#define BTRFS_INODE_COPY_EVERYTHING 8 -#define BTRFS_INODE_IN_DELALLOC_LIST 9 -#define BTRFS_INODE_READDIO_NEED_LOCK 10 -#define BTRFS_INODE_HAS_PROPS 11 +#define BTRFS_INODE_HAS_ORPHAN_ITEM 4 +#define BTRFS_INODE_HAS_ASYNC_EXTENT 5 +#define BTRFS_INODE_NEEDS_FULL_SYNC 6 +#define BTRFS_INODE_COPY_EVERYTHING 7 +#define BTRFS_INODE_IN_DELALLOC_LIST 8 +#define BTRFS_INODE_READDIO_NEED_LOCK 9 +#define BTRFS_INODE_HAS_PROPS 10 /* in memory btrfs inode */ struct btrfs_inode { @@ -176,7 +175,8 @@ struct btrfs_inode { * of extent items we've reserved metadata for. */ unsigned outstanding_extents; - unsigned reserved_extents; + + struct btrfs_block_rsv block_rsv; /* * Cached values of inode properties @@ -278,14 +278,6 @@ static inline void btrfs_mod_outstanding_extents(struct btrfs_inode *inode, mod); } -static inline void btrfs_mod_reserved_extents(struct btrfs_inode *inode, int mod) -{ - lockdep_assert_held(&inode->lock); - inode->reserved_extents += mod; - if (btrfs_is_free_space_inode(inode)) - return; -} - static inline int btrfs_inode_in_log(struct btrfs_inode *inode, u64 generation) { int ret = 0; diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 2ede3b6ceb68..f7df5536ab61 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -763,8 +763,6 @@ struct btrfs_fs_info { * delayed dir index item */ struct btrfs_block_rsv global_block_rsv; - /* block reservation for delay allocation */ - struct btrfs_block_rsv delalloc_block_rsv; /* block reservation for metadata operations */ struct btrfs_block_rsv trans_block_rsv; /* block reservation for chunk tree */ @@ -2757,6 +2755,9 @@ int btrfs_delalloc_reserve_space(struct inode *inode, void btrfs_init_block_rsv(struct btrfs_block_rsv *rsv, unsigned short type); struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_fs_info *fs_info, unsigned short type); +void btrfs_init_metadata_block_rsv(struct btrfs_fs_info *fs_info, + struct btrfs_block_rsv *rsv, + unsigned short type); void btrfs_free_block_rsv(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *rsv); void __btrfs_free_block_rsv(struct btrfs_block_rsv *rsv); diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 19e4ad2f3f2e..5d73f79ded8b 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -581,36 +581,12 @@ static int btrfs_delayed_inode_reserve_metadata( struct btrfs_block_rsv *dst_rsv; u64 num_bytes; int ret; - bool release = false; src_rsv = trans->block_rsv; dst_rsv = &fs_info->delayed_block_rsv; num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1); - /* - * If our block_rsv is the delalloc block reserve then check and see if - * we have our extra reservation for updating the inode. If not fall - * through and try to reserve space quickly. - * - * We used to try and steal from the delalloc block rsv or the global - * reserve, but we'd steal a full reservation, which isn't kind. We are - * here through delalloc which means we've likely just cowed down close - * to the leaf that contains the inode, so we would steal less just - * doing the fallback inode update, so if we do end up having to steal - * from the global block rsv we hopefully only steal one or two blocks - * worth which is less likely to hurt us. - */ - if (src_rsv && src_rsv->type == BTRFS_BLOCK_RSV_DELALLOC) { - spin_lock(&inode->lock); - if (test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED, - &inode->runtime_flags)) - release = true; - else - src_rsv = NULL; - spin_unlock(&inode->lock); - } - /* * btrfs_dirty_inode will update the inode under btrfs_join_transaction * which doesn't reserve space for speed. This is a problem since we @@ -618,7 +594,7 @@ static int btrfs_delayed_inode_reserve_metadata( * space. * * Now if src_rsv == delalloc_block_rsv we'll let it just steal since - * we're accounted for. + * we always reserve enough to update the inode item. */ if (!src_rsv || (!trans->bytes_reserved && src_rsv->type != BTRFS_BLOCK_RSV_DELALLOC)) { @@ -643,32 +619,12 @@ static int btrfs_delayed_inode_reserve_metadata( } ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1); - - /* - * Migrate only takes a reservation, it doesn't touch the size of the - * block_rsv. This is to simplify people who don't normally have things - * migrated from their block rsv. If they go to release their - * reservation, that will decrease the size as well, so if migrate - * reduced size we'd end up with a negative size. But for the - * delalloc_meta_reserved stuff we will only know to drop 1 reservation, - * but we could in fact do this reserve/migrate dance several times - * between the time we did the original reservation and we'd clean it - * up. So to take care of this, release the space for the meta - * reservation here. I think it may be time for a documentation page on - * how block rsvs. work. - */ if (!ret) { trace_btrfs_space_reservation(fs_info, "delayed_inode", btrfs_ino(inode), num_bytes, 1); node->bytes_reserved = num_bytes; } - if (release) { - trace_btrfs_space_reservation(fs_info, "delalloc", - btrfs_ino(inode), num_bytes, 0); - btrfs_block_rsv_release(fs_info, src_rsv, num_bytes); - } - return ret; } diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 484cf8fc952c..d1f396f72979 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2447,14 +2447,6 @@ int open_ctree(struct super_block *sb, goto fail_delalloc_bytes; } - fs_info->btree_inode = new_inode(sb); - if (!fs_info->btree_inode) { - err = -ENOMEM; - goto fail_bio_counter; - } - - mapping_set_gfp_mask(fs_info->btree_inode->i_mapping, GFP_NOFS); - INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC); INIT_RADIX_TREE(&fs_info->buffer_radix, GFP_ATOMIC); INIT_LIST_HEAD(&fs_info->trans_list); @@ -2487,8 +2479,6 @@ int open_ctree(struct super_block *sb, btrfs_mapping_init(&fs_info->mapping_tree); btrfs_init_block_rsv(&fs_info->global_block_rsv, BTRFS_BLOCK_RSV_GLOBAL); - btrfs_init_block_rsv(&fs_info->delalloc_block_rsv, - BTRFS_BLOCK_RSV_DELALLOC); btrfs_init_block_rsv(&fs_info->trans_block_rsv, BTRFS_BLOCK_RSV_TRANS); btrfs_init_block_rsv(&fs_info->chunk_block_rsv, BTRFS_BLOCK_RSV_CHUNK); btrfs_init_block_rsv(&fs_info->empty_block_rsv, BTRFS_BLOCK_RSV_EMPTY); @@ -2517,6 +2507,14 @@ int open_ctree(struct super_block *sb, INIT_LIST_HEAD(&fs_info->ordered_roots); spin_lock_init(&fs_info->ordered_root_lock); + + fs_info->btree_inode = new_inode(sb); + if (!fs_info->btree_inode) { + err = -ENOMEM; + goto fail_bio_counter; + } + mapping_set_gfp_mask(fs_info->btree_inode->i_mapping, GFP_NOFS); + fs_info->delayed_root = kmalloc(sizeof(struct btrfs_delayed_root), GFP_KERNEL); if (!fs_info->delayed_root) { diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index aaa346562df6..fc9720e28005 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -26,6 +26,7 @@ #include #include #include +#include #include "hash.h" #include "tree-log.h" #include "disk-io.h" @@ -4811,7 +4812,6 @@ static inline u64 calc_reclaim_items_nr(struct btrfs_fs_info *fs_info, static void shrink_delalloc(struct btrfs_fs_info *fs_info, u64 to_reclaim, u64 orig, bool wait_ordered) { - struct btrfs_block_rsv *block_rsv; struct btrfs_space_info *space_info; struct btrfs_trans_handle *trans; u64 delalloc_bytes; @@ -4827,8 +4827,7 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info, u64 to_reclaim, to_reclaim = items * EXTENT_SIZE_PER_ITEM; trans = (struct btrfs_trans_handle *)current->journal_info; - block_rsv = &fs_info->delalloc_block_rsv; - space_info = block_rsv->space_info; + space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); delalloc_bytes = percpu_counter_sum_positive( &fs_info->delalloc_bytes); @@ -5564,11 +5563,12 @@ again: } } -static void block_rsv_release_bytes(struct btrfs_fs_info *fs_info, +static u64 block_rsv_release_bytes(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *block_rsv, struct btrfs_block_rsv *dest, u64 num_bytes) { struct btrfs_space_info *space_info = block_rsv->space_info; + u64 ret; spin_lock(&block_rsv->lock); if (num_bytes == (u64)-1) @@ -5583,6 +5583,7 @@ static void block_rsv_release_bytes(struct btrfs_fs_info *fs_info, } spin_unlock(&block_rsv->lock); + ret = num_bytes; if (num_bytes > 0) { if (dest) { spin_lock(&dest->lock); @@ -5602,6 +5603,7 @@ static void block_rsv_release_bytes(struct btrfs_fs_info *fs_info, space_info_add_old_bytes(fs_info, space_info, num_bytes); } + return ret; } int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src, @@ -5625,6 +5627,15 @@ void btrfs_init_block_rsv(struct btrfs_block_rsv *rsv, unsigned short type) rsv->type = type; } +void btrfs_init_metadata_block_rsv(struct btrfs_fs_info *fs_info, + struct btrfs_block_rsv *rsv, + unsigned short type) +{ + btrfs_init_block_rsv(rsv, type); + rsv->space_info = __find_space_info(fs_info, + BTRFS_BLOCK_GROUP_METADATA); +} + struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_fs_info *fs_info, unsigned short type) { @@ -5634,9 +5645,7 @@ struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_fs_info *fs_info, if (!block_rsv) return NULL; - btrfs_init_block_rsv(block_rsv, type); - block_rsv->space_info = __find_space_info(fs_info, - BTRFS_BLOCK_GROUP_METADATA); + btrfs_init_metadata_block_rsv(fs_info, block_rsv, type); return block_rsv; } @@ -5719,6 +5728,66 @@ int btrfs_block_rsv_refill(struct btrfs_root *root, return ret; } +/** + * btrfs_inode_rsv_refill - refill the inode block rsv. + * @inode - the inode we are refilling. + * @flush - the flusing restriction. + * + * Essentially the same as btrfs_block_rsv_refill, except it uses the + * block_rsv->size as the minimum size. We'll either refill the missing amount + * or return if we already have enough space. This will also handle the resreve + * tracepoint for the reserved amount. + */ +int btrfs_inode_rsv_refill(struct btrfs_inode *inode, + enum btrfs_reserve_flush_enum flush) +{ + struct btrfs_root *root = inode->root; + struct btrfs_block_rsv *block_rsv = &inode->block_rsv; + u64 num_bytes = 0; + int ret = -ENOSPC; + + spin_lock(&block_rsv->lock); + if (block_rsv->reserved < block_rsv->size) + num_bytes = block_rsv->size - block_rsv->reserved; + spin_unlock(&block_rsv->lock); + + if (num_bytes == 0) + return 0; + + ret = reserve_metadata_bytes(root, block_rsv, num_bytes, flush); + if (!ret) { + block_rsv_add_bytes(block_rsv, num_bytes, 0); + trace_btrfs_space_reservation(root->fs_info, "delalloc", + btrfs_ino(inode), num_bytes, 1); + } + return ret; +} + +/** + * btrfs_inode_rsv_release - release any excessive reservation. + * @inode - the inode we need to release from. + * + * This is the same as btrfs_block_rsv_release, except that it handles the + * tracepoint for the reservation. + */ +void btrfs_inode_rsv_release(struct btrfs_inode *inode) +{ + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv; + struct btrfs_block_rsv *block_rsv = &inode->block_rsv; + u64 released = 0; + + /* + * Since we statically set the block_rsv->size we just want to say we + * are releasing 0 bytes, and then we'll just get the reservation over + * the size free'd. + */ + released = block_rsv_release_bytes(fs_info, block_rsv, global_rsv, 0); + if (released > 0) + trace_btrfs_space_reservation(fs_info, "delalloc", + btrfs_ino(inode), released, 0); +} + void btrfs_block_rsv_release(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *block_rsv, u64 num_bytes) @@ -5790,7 +5859,6 @@ static void init_global_block_rsv(struct btrfs_fs_info *fs_info) space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); fs_info->global_block_rsv.space_info = space_info; - fs_info->delalloc_block_rsv.space_info = space_info; fs_info->trans_block_rsv.space_info = space_info; fs_info->empty_block_rsv.space_info = space_info; fs_info->delayed_block_rsv.space_info = space_info; @@ -5810,8 +5878,6 @@ static void release_global_block_rsv(struct btrfs_fs_info *fs_info) { block_rsv_release_bytes(fs_info, &fs_info->global_block_rsv, NULL, (u64)-1); - WARN_ON(fs_info->delalloc_block_rsv.size > 0); - WARN_ON(fs_info->delalloc_block_rsv.reserved > 0); WARN_ON(fs_info->trans_block_rsv.size > 0); WARN_ON(fs_info->trans_block_rsv.reserved > 0); WARN_ON(fs_info->chunk_block_rsv.size > 0); @@ -5953,95 +6019,37 @@ void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info, btrfs_block_rsv_release(fs_info, rsv, (u64)-1); } -/** - * drop_over_reserved_extents - drop our extra extent reservations - * @inode: the inode we're dropping the extent for - * - * We reserve extents we may use, but they may have been merged with other - * extents and we may not need the extra reservation. - * - * We also call this when we've completed io to an extent or had an error and - * cleared the outstanding extent, in either case we no longer need our - * reservation and can drop the excess. - */ -static unsigned drop_over_reserved_extents(struct btrfs_inode *inode) -{ - unsigned num_extents = 0; - - if (inode->reserved_extents > inode->outstanding_extents) { - num_extents = inode->reserved_extents - - inode->outstanding_extents; - btrfs_mod_reserved_extents(inode, -num_extents); - } - - if (inode->outstanding_extents == 0 && - test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED, - &inode->runtime_flags)) - num_extents++; - return num_extents; -} - -/** - * calc_csum_metadata_size - return the amount of metadata space that must be - * reserved/freed for the given bytes. - * @inode: the inode we're manipulating - * @num_bytes: the number of bytes in question - * @reserve: 1 if we are reserving space, 0 if we are freeing space - * - * This adjusts the number of csum_bytes in the inode and then returns the - * correct amount of metadata that must either be reserved or freed. We - * calculate how many checksums we can fit into one leaf and then divide the - * number of bytes that will need to be checksumed by this value to figure out - * how many checksums will be required. If we are adding bytes then the number - * may go up and we will return the number of additional bytes that must be - * reserved. If it is going down we will return the number of bytes that must - * be freed. - * - * This must be called with BTRFS_I(inode)->lock held. - */ -static u64 calc_csum_metadata_size(struct btrfs_inode *inode, u64 num_bytes, - int reserve) +static void btrfs_calculate_inode_block_rsv_size(struct btrfs_fs_info *fs_info, + struct btrfs_inode *inode) { - struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); - u64 old_csums, num_csums; - - if (inode->flags & BTRFS_INODE_NODATASUM && inode->csum_bytes == 0) - return 0; - - old_csums = btrfs_csum_bytes_to_leaves(fs_info, inode->csum_bytes); - if (reserve) - inode->csum_bytes += num_bytes; - else - inode->csum_bytes -= num_bytes; - num_csums = btrfs_csum_bytes_to_leaves(fs_info, inode->csum_bytes); - - /* No change, no need to reserve more */ - if (old_csums == num_csums) - return 0; + struct btrfs_block_rsv *block_rsv = &inode->block_rsv; + u64 reserve_size = 0; + u64 csum_leaves; + unsigned outstanding_extents; - if (reserve) - return btrfs_calc_trans_metadata_size(fs_info, - num_csums - old_csums); + lockdep_assert_held(&inode->lock); + outstanding_extents = inode->outstanding_extents; + if (outstanding_extents) + reserve_size = btrfs_calc_trans_metadata_size(fs_info, + outstanding_extents + 1); + csum_leaves = btrfs_csum_bytes_to_leaves(fs_info, + inode->csum_bytes); + reserve_size += btrfs_calc_trans_metadata_size(fs_info, + csum_leaves); - return btrfs_calc_trans_metadata_size(fs_info, old_csums - num_csums); + spin_lock(&block_rsv->lock); + block_rsv->size = reserve_size; + spin_unlock(&block_rsv->lock); } int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); struct btrfs_root *root = inode->root; - struct btrfs_block_rsv *block_rsv = &fs_info->delalloc_block_rsv; - u64 to_reserve = 0; - u64 csum_bytes; - unsigned nr_extents, reserve_extents; + unsigned nr_extents; enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_FLUSH_ALL; int ret = 0; bool delalloc_lock = true; - u64 to_free = 0; - unsigned dropped; - bool release_extra = false; - bool underflow = false; - bool did_retry = false; /* If we are a free space inode we need to not flush since we will be in * the middle of a transaction commit. We also don't need the delalloc @@ -6066,33 +6074,13 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes) mutex_lock(&inode->delalloc_mutex); num_bytes = ALIGN(num_bytes, fs_info->sectorsize); -retry: + + /* Add our new extents and calculate the new rsv size. */ spin_lock(&inode->lock); - reserve_extents = nr_extents = count_max_extents(num_bytes); + nr_extents = count_max_extents(num_bytes); btrfs_mod_outstanding_extents(inode, nr_extents); - - /* - * Because we add an outstanding extent for ordered before we clear - * delalloc we will double count our outstanding extents slightly. This - * could mean that we transiently over-reserve, which could result in an - * early ENOSPC if our timing is unlucky. Keep track of the case that - * we had a reservation underflow so we can retry if we fail. - * - * Keep in mind we can legitimately have more outstanding extents than - * reserved because of fragmentation, so only allow a retry once. - */ - if (inode->outstanding_extents > - inode->reserved_extents + nr_extents) { - reserve_extents = inode->outstanding_extents - - inode->reserved_extents; - underflow = true; - } - - /* We always want to reserve a slot for updating the inode. */ - to_reserve = btrfs_calc_trans_metadata_size(fs_info, - reserve_extents + 1); - to_reserve += calc_csum_metadata_size(inode, num_bytes, 1); - csum_bytes = inode->csum_bytes; + inode->csum_bytes += num_bytes; + btrfs_calculate_inode_block_rsv_size(fs_info, inode); spin_unlock(&inode->lock); if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)) { @@ -6102,100 +6090,26 @@ retry: goto out_fail; } - ret = btrfs_block_rsv_add(root, block_rsv, to_reserve, flush); + ret = btrfs_inode_rsv_refill(inode, flush); if (unlikely(ret)) { btrfs_qgroup_free_meta(root, nr_extents * fs_info->nodesize); goto out_fail; } - spin_lock(&inode->lock); - if (test_and_set_bit(BTRFS_INODE_DELALLOC_META_RESERVED, - &inode->runtime_flags)) { - to_reserve -= btrfs_calc_trans_metadata_size(fs_info, 1); - release_extra = true; - } - btrfs_mod_reserved_extents(inode, reserve_extents); - spin_unlock(&inode->lock); - if (delalloc_lock) mutex_unlock(&inode->delalloc_mutex); - - if (to_reserve) - trace_btrfs_space_reservation(fs_info, "delalloc", - btrfs_ino(inode), to_reserve, 1); - if (release_extra) - btrfs_block_rsv_release(fs_info, block_rsv, - btrfs_calc_trans_metadata_size(fs_info, 1)); return 0; out_fail: spin_lock(&inode->lock); nr_extents = count_max_extents(num_bytes); btrfs_mod_outstanding_extents(inode, -nr_extents); - - dropped = drop_over_reserved_extents(inode); - /* - * If the inodes csum_bytes is the same as the original - * csum_bytes then we know we haven't raced with any free()ers - * so we can just reduce our inodes csum bytes and carry on. - */ - if (inode->csum_bytes == csum_bytes) { - calc_csum_metadata_size(inode, num_bytes, 0); - } else { - u64 orig_csum_bytes = inode->csum_bytes; - u64 bytes; - - /* - * This is tricky, but first we need to figure out how much we - * freed from any free-ers that occurred during this - * reservation, so we reset ->csum_bytes to the csum_bytes - * before we dropped our lock, and then call the free for the - * number of bytes that were freed while we were trying our - * reservation. - */ - bytes = csum_bytes - inode->csum_bytes; - inode->csum_bytes = csum_bytes; - to_free = calc_csum_metadata_size(inode, bytes, 0); - - - /* - * Now we need to see how much we would have freed had we not - * been making this reservation and our ->csum_bytes were not - * artificially inflated. - */ - inode->csum_bytes = csum_bytes - num_bytes; - bytes = csum_bytes - orig_csum_bytes; - bytes = calc_csum_metadata_size(inode, bytes, 0); - - /* - * Now reset ->csum_bytes to what it should be. If bytes is - * more than to_free then we would have freed more space had we - * not had an artificially high ->csum_bytes, so we need to free - * the remainder. If bytes is the same or less then we don't - * need to do anything, the other free-ers did the correct - * thing. - */ - inode->csum_bytes = orig_csum_bytes - num_bytes; - if (bytes > to_free) - to_free = bytes - to_free; - else - to_free = 0; - } + inode->csum_bytes -= num_bytes; + btrfs_calculate_inode_block_rsv_size(fs_info, inode); spin_unlock(&inode->lock); - if (dropped) - to_free += btrfs_calc_trans_metadata_size(fs_info, dropped); - if (to_free) { - btrfs_block_rsv_release(fs_info, block_rsv, to_free); - trace_btrfs_space_reservation(fs_info, "delalloc", - btrfs_ino(inode), to_free, 0); - } - if (underflow && !did_retry) { - did_retry = true; - underflow = false; - goto retry; - } + btrfs_inode_rsv_release(inode); if (delalloc_lock) mutex_unlock(&inode->delalloc_mutex); return ret; @@ -6213,25 +6127,17 @@ out_fail: void btrfs_delalloc_release_metadata(struct btrfs_inode *inode, u64 num_bytes) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); - u64 to_free = 0; - unsigned dropped; num_bytes = ALIGN(num_bytes, fs_info->sectorsize); spin_lock(&inode->lock); - dropped = drop_over_reserved_extents(inode); - if (num_bytes) - to_free = calc_csum_metadata_size(inode, num_bytes, 0); + inode->csum_bytes -= num_bytes; + btrfs_calculate_inode_block_rsv_size(fs_info, inode); spin_unlock(&inode->lock); - if (dropped > 0) - to_free += btrfs_calc_trans_metadata_size(fs_info, dropped); if (btrfs_is_testing(fs_info)) return; - trace_btrfs_space_reservation(fs_info, "delalloc", btrfs_ino(inode), - to_free, 0); - - btrfs_block_rsv_release(fs_info, &fs_info->delalloc_block_rsv, to_free); + btrfs_inode_rsv_release(inode); } /** @@ -6249,25 +6155,17 @@ void btrfs_delalloc_release_extents(struct btrfs_inode *inode, u64 num_bytes) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); unsigned num_extents; - u64 to_free; - unsigned dropped; spin_lock(&inode->lock); num_extents = count_max_extents(num_bytes); btrfs_mod_outstanding_extents(inode, -num_extents); - dropped = drop_over_reserved_extents(inode); + btrfs_calculate_inode_block_rsv_size(fs_info, inode); spin_unlock(&inode->lock); - if (!dropped) - return; - if (btrfs_is_testing(fs_info)) return; - to_free = btrfs_calc_trans_metadata_size(fs_info, dropped); - trace_btrfs_space_reservation(fs_info, "delalloc", btrfs_ino(inode), - to_free, 0); - btrfs_block_rsv_release(fs_info, &fs_info->delalloc_block_rsv, to_free); + btrfs_inode_rsv_release(inode); } /** diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 5b0de1e120d1..b71731ef28c4 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -42,6 +42,7 @@ #include #include #include +#include #include "ctree.h" #include "disk-io.h" #include "transaction.h" @@ -315,7 +316,7 @@ static noinline int cow_file_range_inline(struct btrfs_root *root, btrfs_free_path(path); return PTR_ERR(trans); } - trans->block_rsv = &fs_info->delalloc_block_rsv; + trans->block_rsv = &BTRFS_I(inode)->block_rsv; if (compressed_size && compressed_pages) extent_item_size = btrfs_file_extent_calc_inline_size( @@ -2954,7 +2955,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) trans = NULL; goto out; } - trans->block_rsv = &fs_info->delalloc_block_rsv; + trans->block_rsv = &BTRFS_I(inode)->block_rsv; ret = btrfs_update_inode_fallback(trans, root, inode); if (ret) /* -ENOMEM or corruption */ btrfs_abort_transaction(trans, ret); @@ -2990,7 +2991,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) goto out; } - trans->block_rsv = &fs_info->delalloc_block_rsv; + trans->block_rsv = &BTRFS_I(inode)->block_rsv; if (test_bit(BTRFS_ORDERED_COMPRESSED, &ordered_extent->flags)) compress_type = ordered_extent->compress_type; @@ -8845,7 +8846,6 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) if (iov_iter_rw(iter) == WRITE) { up_read(&BTRFS_I(inode)->dio_sem); current->journal_info = NULL; - btrfs_delalloc_release_extents(BTRFS_I(inode), count); if (ret < 0 && ret != -EIOCBQUEUED) { if (dio_data.reserve) btrfs_delalloc_release_space(inode, data_reserved, @@ -8866,6 +8866,7 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) } else if (ret >= 0 && (size_t)ret < count) btrfs_delalloc_release_space(inode, data_reserved, offset, count - (size_t)ret); + btrfs_delalloc_release_extents(BTRFS_I(inode), count); } out: if (wakeup) @@ -9430,6 +9431,7 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, struct inode *btrfs_alloc_inode(struct super_block *sb) { + struct btrfs_fs_info *fs_info = btrfs_sb(sb); struct btrfs_inode *ei; struct inode *inode; @@ -9456,8 +9458,9 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) spin_lock_init(&ei->lock); ei->outstanding_extents = 0; - ei->reserved_extents = 0; - + if (sb->s_magic != BTRFS_TEST_MAGIC) + btrfs_init_metadata_block_rsv(fs_info, &ei->block_rsv, + BTRFS_BLOCK_RSV_DELALLOC); ei->runtime_flags = 0; ei->prop_compress = BTRFS_COMPRESS_NONE; ei->defrag_compress = BTRFS_COMPRESS_NONE; @@ -9507,8 +9510,9 @@ void btrfs_destroy_inode(struct inode *inode) WARN_ON(!hlist_empty(&inode->i_dentry)); WARN_ON(inode->i_data.nrpages); + WARN_ON(BTRFS_I(inode)->block_rsv.reserved); + WARN_ON(BTRFS_I(inode)->block_rsv.size); WARN_ON(BTRFS_I(inode)->outstanding_extents); - WARN_ON(BTRFS_I(inode)->reserved_extents); WARN_ON(BTRFS_I(inode)->delalloc_bytes); WARN_ON(BTRFS_I(inode)->new_delalloc_bytes); WARN_ON(BTRFS_I(inode)->csum_bytes); -- cgit v1.2.3 From 0e0adbcfdc908684317c99a9bf5e13383f03b7ec Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Thu, 19 Oct 2017 14:16:00 -0400 Subject: btrfs: track refs in a rb_tree instead of a list If we get a significant amount of delayed refs for a single block (think modifying multiple snapshots) we can end up spending an ungodly amount of time looping through all of the entries trying to see if they can be merged. This is because we only add them to a list, so we have O(2n) for every ref head. This doesn't make any sense as we likely have refs for different roots, and so they cannot be merged. Tracking in a tree will allow us to break as soon as we hit an entry that doesn't match, making our worst case O(n). With this we can also merge entries more easily. Before we had to hope that matching refs were on the ends of our list, but with the tree we can search down to exact matches and merge them at insert time. Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/backref.c | 5 ++- fs/btrfs/delayed-ref.c | 108 +++++++++++++++++++++++++------------------------ fs/btrfs/delayed-ref.h | 5 +-- fs/btrfs/disk-io.c | 10 +++-- fs/btrfs/extent-tree.c | 21 ++++++---- 5 files changed, 82 insertions(+), 67 deletions(-) (limited to 'fs/btrfs/disk-io.c') diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 523d2dba7745..7d0dc100a09a 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -773,6 +773,7 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info, struct btrfs_key key; struct btrfs_key tmp_op_key; struct btrfs_key *op_key = NULL; + struct rb_node *n; int count; int ret = 0; @@ -782,7 +783,9 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info, } spin_lock(&head->lock); - list_for_each_entry(node, &head->ref_list, list) { + for (n = rb_first(&head->ref_tree); n; n = rb_next(n)) { + node = rb_entry(n, struct btrfs_delayed_ref_node, + ref_node); if (node->seq > seq) continue; diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 8c7d7db01f7a..83be8f9fd906 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -143,6 +143,34 @@ static struct btrfs_delayed_ref_head *htree_insert(struct rb_root *root, return NULL; } +static struct btrfs_delayed_ref_node* tree_insert(struct rb_root *root, + struct btrfs_delayed_ref_node *ins) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *node = &ins->ref_node; + struct rb_node *parent_node = NULL; + struct btrfs_delayed_ref_node *entry; + + while (*p) { + int comp; + + parent_node = *p; + entry = rb_entry(parent_node, struct btrfs_delayed_ref_node, + ref_node); + comp = comp_refs(ins, entry, true); + if (comp < 0) + p = &(*p)->rb_left; + else if (comp > 0) + p = &(*p)->rb_right; + else + return entry; + } + + rb_link_node(node, parent_node, p); + rb_insert_color(node, root); + return NULL; +} + /* * find an head entry based on bytenr. This returns the delayed ref * head if it was able to find one, or NULL if nothing was in that spot. @@ -212,7 +240,8 @@ static inline void drop_delayed_ref(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_node *ref) { assert_spin_locked(&head->lock); - list_del(&ref->list); + rb_erase(&ref->ref_node, &head->ref_tree); + RB_CLEAR_NODE(&ref->ref_node); if (!list_empty(&ref->add_list)) list_del(&ref->add_list); ref->in_tree = 0; @@ -229,24 +258,18 @@ static bool merge_ref(struct btrfs_trans_handle *trans, u64 seq) { struct btrfs_delayed_ref_node *next; + struct rb_node *node = rb_next(&ref->ref_node); bool done = false; - next = list_first_entry(&head->ref_list, struct btrfs_delayed_ref_node, - list); - while (!done && &next->list != &head->ref_list) { + while (!done && node) { int mod; - struct btrfs_delayed_ref_node *next2; - - next2 = list_next_entry(next, list); - - if (next == ref) - goto next; + next = rb_entry(node, struct btrfs_delayed_ref_node, ref_node); + node = rb_next(node); if (seq && next->seq >= seq) - goto next; - + break; if (comp_refs(ref, next, false)) - goto next; + break; if (ref->action == next->action) { mod = next->ref_mod; @@ -270,8 +293,6 @@ static bool merge_ref(struct btrfs_trans_handle *trans, WARN_ON(ref->type == BTRFS_TREE_BLOCK_REF_KEY || ref->type == BTRFS_SHARED_BLOCK_REF_KEY); } -next: - next = next2; } return done; @@ -283,11 +304,12 @@ void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_head *head) { struct btrfs_delayed_ref_node *ref; + struct rb_node *node; u64 seq = 0; assert_spin_locked(&head->lock); - if (list_empty(&head->ref_list)) + if (RB_EMPTY_ROOT(&head->ref_tree)) return; /* We don't have too many refs to merge for data. */ @@ -304,22 +326,13 @@ void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans, } spin_unlock(&fs_info->tree_mod_seq_lock); - ref = list_first_entry(&head->ref_list, struct btrfs_delayed_ref_node, - list); - while (&ref->list != &head->ref_list) { +again: + for (node = rb_first(&head->ref_tree); node; node = rb_next(node)) { + ref = rb_entry(node, struct btrfs_delayed_ref_node, ref_node); if (seq && ref->seq >= seq) - goto next; - - if (merge_ref(trans, delayed_refs, head, ref, seq)) { - if (list_empty(&head->ref_list)) - break; - ref = list_first_entry(&head->ref_list, - struct btrfs_delayed_ref_node, - list); continue; - } -next: - ref = list_next_entry(ref, list); + if (merge_ref(trans, delayed_refs, head, ref, seq)) + goto again; } } @@ -402,25 +415,19 @@ again: * Return 0 for insert. * Return >0 for merge. */ -static int -add_delayed_ref_tail_merge(struct btrfs_trans_handle *trans, - struct btrfs_delayed_ref_root *root, - struct btrfs_delayed_ref_head *href, - struct btrfs_delayed_ref_node *ref) +static int insert_delayed_ref(struct btrfs_trans_handle *trans, + struct btrfs_delayed_ref_root *root, + struct btrfs_delayed_ref_head *href, + struct btrfs_delayed_ref_node *ref) { struct btrfs_delayed_ref_node *exist; int mod; int ret = 0; spin_lock(&href->lock); - /* Check whether we can merge the tail node with ref */ - if (list_empty(&href->ref_list)) - goto add_tail; - exist = list_entry(href->ref_list.prev, struct btrfs_delayed_ref_node, - list); - /* No need to compare bytenr nor is_head */ - if (comp_refs(exist, ref, true)) - goto add_tail; + exist = tree_insert(&href->ref_tree, ref); + if (!exist) + goto inserted; /* Now we are sure we can merge */ ret = 1; @@ -451,9 +458,7 @@ add_delayed_ref_tail_merge(struct btrfs_trans_handle *trans, drop_delayed_ref(trans, root, href, exist); spin_unlock(&href->lock); return ret; - -add_tail: - list_add_tail(&ref->list, &href->ref_list); +inserted: if (ref->action == BTRFS_ADD_DELAYED_REF) list_add_tail(&ref->add_list, &href->ref_add_list); atomic_inc(&root->num_entries); @@ -593,7 +598,7 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info, head_ref->ref_mod = count_mod; head_ref->must_insert_reserved = must_insert_reserved; head_ref->is_data = is_data; - INIT_LIST_HEAD(&head_ref->ref_list); + head_ref->ref_tree = RB_ROOT; INIT_LIST_HEAD(&head_ref->ref_add_list); RB_CLEAR_NODE(&head_ref->href_node); head_ref->processing = 0; @@ -685,7 +690,7 @@ add_delayed_tree_ref(struct btrfs_fs_info *fs_info, ref->is_head = 0; ref->in_tree = 1; ref->seq = seq; - INIT_LIST_HEAD(&ref->list); + RB_CLEAR_NODE(&ref->ref_node); INIT_LIST_HEAD(&ref->add_list); full_ref = btrfs_delayed_node_to_tree_ref(ref); @@ -699,7 +704,7 @@ add_delayed_tree_ref(struct btrfs_fs_info *fs_info, trace_add_delayed_tree_ref(fs_info, ref, full_ref, action); - ret = add_delayed_ref_tail_merge(trans, delayed_refs, head_ref, ref); + ret = insert_delayed_ref(trans, delayed_refs, head_ref, ref); /* * XXX: memory should be freed at the same level allocated. @@ -742,7 +747,7 @@ add_delayed_data_ref(struct btrfs_fs_info *fs_info, ref->is_head = 0; ref->in_tree = 1; ref->seq = seq; - INIT_LIST_HEAD(&ref->list); + RB_CLEAR_NODE(&ref->ref_node); INIT_LIST_HEAD(&ref->add_list); full_ref = btrfs_delayed_node_to_data_ref(ref); @@ -758,8 +763,7 @@ add_delayed_data_ref(struct btrfs_fs_info *fs_info, trace_add_delayed_data_ref(fs_info, ref, full_ref, action); - ret = add_delayed_ref_tail_merge(trans, delayed_refs, head_ref, ref); - + ret = insert_delayed_ref(trans, delayed_refs, head_ref, ref); if (ret > 0) kmem_cache_free(btrfs_delayed_data_ref_cachep, full_ref); } diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 1ce11858d727..a43af432f859 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -27,8 +27,7 @@ #define BTRFS_UPDATE_DELAYED_HEAD 4 /* not changing ref count on head ref */ struct btrfs_delayed_ref_node { - /*data/tree ref use list, stored in ref_head->ref_list. */ - struct list_head list; + struct rb_node ref_node; /* * If action is BTRFS_ADD_DELAYED_REF, also link this node to * ref_head->ref_add_list, then we do not need to iterate the @@ -92,7 +91,7 @@ struct btrfs_delayed_ref_head { struct mutex mutex; spinlock_t lock; - struct list_head ref_list; + struct rb_root ref_tree; /* accumulate add BTRFS_ADD_DELAYED_REF nodes to this ref_add_list. */ struct list_head ref_add_list; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index d1f396f72979..efce9a2fa9be 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -4113,7 +4113,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans, while ((node = rb_first(&delayed_refs->href_root)) != NULL) { struct btrfs_delayed_ref_head *head; - struct btrfs_delayed_ref_node *tmp; + struct rb_node *n; bool pin_bytes = false; head = rb_entry(node, struct btrfs_delayed_ref_head, @@ -4129,10 +4129,12 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans, continue; } spin_lock(&head->lock); - list_for_each_entry_safe_reverse(ref, tmp, &head->ref_list, - list) { + while ((n = rb_first(&head->ref_tree)) != NULL) { + ref = rb_entry(n, struct btrfs_delayed_ref_node, + ref_node); ref->in_tree = 0; - list_del(&ref->list); + rb_erase(&ref->ref_node, &head->ref_tree); + RB_CLEAR_NODE(&ref->ref_node); if (!list_empty(&ref->add_list)) list_del(&ref->add_list); atomic_dec(&delayed_refs->num_entries); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index fc9720e28005..673ac4e01dd0 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -2519,7 +2519,7 @@ select_delayed_ref(struct btrfs_delayed_ref_head *head) { struct btrfs_delayed_ref_node *ref; - if (list_empty(&head->ref_list)) + if (RB_EMPTY_ROOT(&head->ref_tree)) return NULL; /* @@ -2532,8 +2532,8 @@ select_delayed_ref(struct btrfs_delayed_ref_head *head) return list_first_entry(&head->ref_add_list, struct btrfs_delayed_ref_node, add_list); - ref = list_first_entry(&head->ref_list, struct btrfs_delayed_ref_node, - list); + ref = rb_entry(rb_first(&head->ref_tree), + struct btrfs_delayed_ref_node, ref_node); ASSERT(list_empty(&ref->add_list)); return ref; } @@ -2593,7 +2593,7 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans, spin_unlock(&head->lock); spin_lock(&delayed_refs->lock); spin_lock(&head->lock); - if (!list_empty(&head->ref_list) || head->extent_op) { + if (!RB_EMPTY_ROOT(&head->ref_tree) || head->extent_op) { spin_unlock(&head->lock); spin_unlock(&delayed_refs->lock); return 1; @@ -2740,7 +2740,8 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, actual_count++; ref->in_tree = 0; - list_del(&ref->list); + rb_erase(&ref->ref_node, &locked_ref->ref_tree); + RB_CLEAR_NODE(&ref->ref_node); if (!list_empty(&ref->add_list)) list_del(&ref->add_list); /* @@ -3138,6 +3139,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, struct btrfs_delayed_data_ref *data_ref; struct btrfs_delayed_ref_root *delayed_refs; struct btrfs_transaction *cur_trans; + struct rb_node *node; int ret = 0; cur_trans = root->fs_info->running_transaction; @@ -3170,7 +3172,12 @@ static noinline int check_delayed_ref(struct btrfs_root *root, spin_unlock(&delayed_refs->lock); spin_lock(&head->lock); - list_for_each_entry(ref, &head->ref_list, list) { + /* + * XXX: We should replace this with a proper search function in the + * future. + */ + for (node = rb_first(&head->ref_tree); node; node = rb_next(node)) { + ref = rb_entry(node, struct btrfs_delayed_ref_node, ref_node); /* If it's a shared ref we know a cross reference exists */ if (ref->type != BTRFS_EXTENT_DATA_REF_KEY) { ret = 1; @@ -7141,7 +7148,7 @@ static noinline int check_ref_cleanup(struct btrfs_trans_handle *trans, goto out_delayed_unlock; spin_lock(&head->lock); - if (!list_empty(&head->ref_list)) + if (!RB_EMPTY_ROOT(&head->ref_tree)) goto out; if (head->extent_op) { -- cgit v1.2.3