Re: some free space cache corruptions
I am not sure I can remember a time when btrfs check did not print this "cache and super generation don't match, space cache will be invalidated" message, so I started ignoring it a long time ago because I never seemed to have problem with missing free space and never got any similar warnings/errors in the kernel log. On Mon, Dec 26, 2016 at 1:12 AM, Duncan <1i5t5.dun...@cox.net> wrote: > Christoph Anton Mitterer posted on Sun, 25 Dec 2016 23:00:34 +0100 as > excerpted: > >> # btrfs check /dev/mapper/data-a2 ; echo $? >> Checking filesystem on /dev/mapper/data-a2 > > [...] > >> checking free space cache >> block group 5431552376832 has wrong amount of free space >> failed to load free space cache for block group 5431552376832 > > [...] > >> 0 >> >> => same error again... >> >> Any ideas how to resolve? And is this some serious error that could have >> caused corruptions? > > By themselves, free-space cache warnings are minor and not a serious > issue at all -- the cache is just that, a cache, designed to speed > operation but not actually necessary, and btrfs can detect and route > around space-cache corruption on-the-fly so by itself it's not a big deal. > > These warnings are however hints that something out of the routine has > happened, and that you might wish to freshen your backups, run btrfs > check and scrub and see if anything else is wrong (if you get them at > mount, you got them /running/ btrfs check and nothing else out of the > ordinary was reported), etc. > > Three things to note: > > 1) Plain btrfs check, without options that trigger fixes, is read-only, > so you are likely to see anything unusual it reports again in repeated > runs, unless the filesystem itself, or a scrub, etc, has fixed things in > the mean time. (And as I said, the space-cache is only a cache, designed > to speed things up, cache corruption is fairly common and btrfs can and > does deal with it without issue. In fact btrfs has the nospace_cache > option to entirely disable it at mount.) > > 2) It recently came to the attention of the devs that the existing btrfs > mount-option method of clearing the free-space cache only clears it for > block-groups/chunks it encounters on-the-fly. It doesn't do a systematic > beginning-to-end clear. As such, in some instances it's possible to run > with the clear_cache mount option (see the btrfs (5) manpage for mount > option specifics, but to head off another question, it's recommended to > stay with v1 cache for now) and still see space-cache warnings you > thought should be cleared up, if btrfs didn't deal with those chunks in > the run where the cache was cleared. > > 3) As a result of #2, the devs only very recently added support in btrfs > check for a /full/ space-cache-v1 clear, using the new > --clear-space-cache option. But your btrfs-progs v4.7.3 is too old to > support it. I know it's in the v4.9 I just upgraded to... checking the > wiki it appears the option was added in btrfs-progs v4.8.3 (v4.8.4 for v2 > cache). > > So if you want you can try the clear_cache mount option, and if that > doesn't do it, upgrade to a current btrfs-progs and run it with the > --clear-space-cache option, but you're not endangering your filesystem or > anything by simply waiting until you get a btrfs-progs update from your > distro, if you decide to. The space-cache warnings aren't indicative of > a serious problem now and btrfs deals with them on its own, they are > simply hints that something, perhaps a crash with the btrfs mounted > writable, happened at some time in the past, and that it it might be wise > to investigate further for other damage, which you've already done, so > you're good. =:^) > > Tho if you haven't recently run a scrub, I'd do that as well (and in fact > recommend running it before check if you can successfully mount), since > the problems it detects and fixes are conceptually different than the > ones btrfs check deals with. Scrub deals with actual on-media > corruption, blocks not matching their checksum, while check deals with > filesystem logic errors, whether or not the blocks containing them match > the checksum. > > -- > Duncan - List replies preferred. No HTML msgs. > "Every nonfree program has a lord, a master -- > and if you use the program, he is your master." Richard Stallman > > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majord...@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 12/19] btrfs-progs: scrub: Introduce function to scrub one extent
Introduce a new function, scrub_one_extent(), as a wrapper to check one extent. It will accept a btrfs_path parameter @path, which must points to a META/EXTENT_ITEM. And @start, @len, which must be a subset of META/EXTENT_ITEM. Parameter @report will determine if we output error. Since the function will be reused by RAID56 code, we want it able to be silent. Signed-off-by: Qu Wenruo--- scrub.c | 79 + 1 file changed, 79 insertions(+) diff --git a/scrub.c b/scrub.c index 2563f407..f4ed0b78 100644 --- a/scrub.c +++ b/scrub.c @@ -272,3 +272,82 @@ out: return -EIO; return ret; } + +/* + * Check all copies of range @start, @len. + * Caller must ensure the range is covered by EXTENT_ITEM/METADATA_ITEM + * specified by leaf of @path. + * And @start, @len must be a subset of the EXTENT_ITEM/METADATA_ITEM. + * + * If @report is set, it will report if the range is recoverable or totally + * corrupted if it has corrupted mirror. + * This parameter is used for silent verification for RAID5/6 recovery code. + * + * Return 0 if the range is all OK or recoverable. + * Return <0 if the range can't be recoverable. + */ +static int scrub_one_extent(struct btrfs_fs_info *fs_info, + struct btrfs_scrub_progress *scrub_ctx, + struct btrfs_path *path, u64 start, u64 len, + int report) +{ + struct btrfs_key key; + struct btrfs_extent_item *ei; + struct extent_buffer *leaf = path->nodes[0]; + int slot = path->slots[0]; + int num_copies; + int corrupted = 0; + u64 extent_start; + u64 extent_len; + int metadata = 0; + int i; + int ret; + + btrfs_item_key_to_cpu(leaf, , slot); + if (key.type != BTRFS_METADATA_ITEM_KEY && + key.type != BTRFS_EXTENT_ITEM_KEY) + goto invalid_arg; + + extent_start = key.objectid; + if (key.type == BTRFS_METADATA_ITEM_KEY) { + extent_len = fs_info->tree_root->nodesize; + metadata = 1; + } else { + extent_len = key.offset; + ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item); + if (btrfs_extent_flags(leaf, ei) & BTRFS_EXTENT_FLAG_TREE_BLOCK) + metadata = 1; + } + if (start >= extent_start + extent_len || + start + len <= extent_start) + goto invalid_arg; + num_copies = btrfs_num_copies(_info->mapping_tree, start, len); + for (i = 1; i <= num_copies; i++) { + if (metadata) { + ret = scrub_tree_mirror(fs_info, scrub_ctx, + NULL, extent_start, i); + scrub_ctx->tree_extents_scrubbed++; + } else { + ret = scrub_data_mirror(fs_info, scrub_ctx, NULL, + start, len, i); + scrub_ctx->data_extents_scrubbed++; + } + if (ret < 0) + corrupted++; + } + + if (report) { + if (corrupted && corrupted < num_copies) + printf("bytenr %llu len %llu has corrupted mirror, but is recoverable\n", + start, len); + else if (corrupted >= num_copies) + error("bytenr %llu len %llu has corrupted mirror, can't be recovered", + start, len); + } + if (corrupted < num_copies) + return 0; + return -EIO; +invalid_arg: + error("invalid parameter for %s", __func__); + return -EINVAL; +} -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 05/19] btrfs-progs: Introduce wrapper to recover raid56 data
Introduce a wrapper to recover raid56 data. The logical is the same with kernel one, but with different interfaces, since kernel ones cares the performance while in btrfs we don't care that much. And the interface is more caller friendly inside btrfs-progs. Signed-off-by: Qu Wenruo--- kernel-lib/raid56.c | 77 + kernel-lib/raid56.h | 11 2 files changed, 88 insertions(+) diff --git a/kernel-lib/raid56.c b/kernel-lib/raid56.c index e078972b..e3a9339e 100644 --- a/kernel-lib/raid56.c +++ b/kernel-lib/raid56.c @@ -280,3 +280,80 @@ int raid6_recov_datap(int nr_devs, size_t stripe_len, int dest1, void **data) } return 0; } + +/* Original raid56 recovery wrapper */ +int raid56_recov(int nr_devs, size_t stripe_len, u64 profile, int dest1, +int dest2, void **data) +{ + int min_devs; + int ret; + + if (profile & BTRFS_BLOCK_GROUP_RAID5) + min_devs = 2; + else if (profile & BTRFS_BLOCK_GROUP_RAID6) + min_devs = 3; + else + return -EINVAL; + if (nr_devs < min_devs) + return -EINVAL; + + /* Nothing to recover */ + if (dest1 == -1 && dest2 == -1) + return 0; + + /* Reorder dest1/2, so only dest2 can be -1 */ + if (dest1 == -1) { + dest1 = dest2; + dest2 = -1; + } else if (dest2 != -1 && dest1 != -1) { + /* Reorder dest1/2, ensure dest2 > dest1 */ + if (dest1 > dest2) { + int tmp; + + tmp = dest2; + dest2 = dest1; + dest1 = tmp; + } + } + + if (profile & BTRFS_BLOCK_GROUP_RAID5) { + if (dest2 != -1) + return 1; + return raid5_gen_result(nr_devs, stripe_len, dest1, data); + } + + /* RAID6 one dev corrupted case*/ + if (dest2 == -1) { + /* Regenerate P/Q */ + if (dest1 == nr_devs - 1 || dest1 == nr_devs - 2) { + raid6_gen_syndrome(nr_devs, stripe_len, data); + return 0; + } + + /* Regerneate data from P */ + return raid5_gen_result(nr_devs - 1, stripe_len, dest1, data); + } + + /* P/Q bot corrupted */ + if (dest1 == nr_devs - 2 && dest2 == nr_devs - 1) { + raid6_gen_syndrome(nr_devs, stripe_len, data); + return 0; + } + + /* 2 Data corrupted */ + if (dest2 < nr_devs - 2) + return raid6_recov_data2(nr_devs, stripe_len, dest1, dest2, +data); + /* Data and P*/ + if (dest2 == nr_devs - 1) + return raid6_recov_datap(nr_devs, stripe_len, dest1, data); + + /* +* Final case, Data and Q, recover data first then regenerate Q +*/ + ret = raid5_gen_result(nr_devs - 1, stripe_len, dest1, data); + if (ret < 0) + return ret; + raid6_gen_syndrome(nr_devs, stripe_len, data); + return 0; +} diff --git a/kernel-lib/raid56.h b/kernel-lib/raid56.h index e088279b..9aee39aa 100644 --- a/kernel-lib/raid56.h +++ b/kernel-lib/raid56.h @@ -44,4 +44,15 @@ int raid6_recov_data2(int nr_devs, size_t stripe_len, int dest1, int dest2, void **data); /* Recover data and P */ int raid6_recov_datap(int nr_devs, size_t stripe_len, int dest1, void **data); + +/* + * Recover raid56 data + * @dest1/2 can be -1 to indicate correct data + * + * Return >0 for unrecoverable case. + * Return 0 for recoverable case, And recovered data will be stored into @data + * Return <0 for fatal error + */ +int raid56_recov(int nr_devs, size_t stripe_len, u64 profile, int dest1, +int dest2, void **data); #endif -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 09/19] btrfs-progs: scrub: Introduce structures to support fsck scrub for RAID56
Introuduce new local structures, scrub_full_stripe and scrub_stripe, for incoming offline RAID56 scrub support. For pure stripe/mirror based profiles, like raid0/1/10/dup/single, we will follow the original bytenr and mirror number based iteration, so they don't need any extra structures for these profiles. Signed-off-by: Qu Wenruo--- Makefile.in | 2 +- scrub.c | 120 2 files changed, 121 insertions(+), 1 deletion(-) create mode 100644 scrub.c diff --git a/Makefile.in b/Makefile.in index bb619bfe..41da7ab5 100644 --- a/Makefile.in +++ b/Makefile.in @@ -99,7 +99,7 @@ objects = ctree.o disk-io.o kernel-lib/radix-tree.o extent-tree.o print-tree.o \ qgroup.o free-space-cache.o kernel-lib/list_sort.o props.o \ ulist.o qgroup-verify.o backref.o string-table.o task-utils.o \ inode.o file.o find-root.o free-space-tree.o help.o send-dump.o \ - kernel-lib/tables.o kernel-lib/raid56.o csum.o + kernel-lib/tables.o kernel-lib/raid56.o csum.o scrub.o cmds_objects = cmds-subvolume.o cmds-filesystem.o cmds-device.o cmds-scrub.o \ cmds-inspect.o cmds-balance.o cmds-send.o cmds-receive.o \ cmds-quota.o cmds-qgroup.o cmds-replace.o cmds-check.o \ diff --git a/scrub.c b/scrub.c new file mode 100644 index ..c9ca817e --- /dev/null +++ b/scrub.c @@ -0,0 +1,120 @@ +/* + * Copyright (C) 2016 Fujitsu. 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. + */ + +/* + * Main part to implement offline(unmounted) btrfs scrub + */ + +#include +#include "ctree.h" +#include "volumes.h" +#include "disk-io.h" +#include "utils.h" + +/* + * For parity based profile(RAID56) + * Mirror/stripe based on won't need this. They are iterated by bytenr and + * mirror number. + */ +struct scrub_stripe { + /* For P/Q logical start will be BTRFS_RAID5/6_P/Q_STRIPE */ + u64 logical; + + /* Device is missing */ + unsigned int dev_missing:1; + + /* Any tree/data csum mismatches */ + unsigned int csum_mismatch:1; + + /* Some data doesn't have csum(nodatasum) */ + unsigned int csum_missing:1; + + char *data; +}; + +/* + * RAID56 full stripe(data stripes + P/Q) + */ +struct scrub_full_stripe { + u64 logical_start; + u64 logical_len; + u64 bg_type; + u32 nr_stripes; + u32 stripe_len; + + /* Read error stripes */ + u32 err_read_stripes; + + /* Missing devices */ + u32 err_missing_devs; + + /* Csum error data stripes */ + u32 err_csum_dstripes; + + /* Missing csum data stripes */ + u32 missing_csum_dstripes; + + /* currupted stripe index */ + int corrupted_index[2]; + + int nr_corrupted_stripes; + + /* Already recovered once? */ + unsigned int recovered:1; + + struct scrub_stripe stripes[]; +}; + +static void free_full_stripe(struct scrub_full_stripe *fstripe) +{ + int i; + + for (i = 0; i < fstripe->nr_stripes; i++) + free(fstripe->stripes[i].data); + free(fstripe); +} + +static struct scrub_full_stripe *alloc_full_stripe(int nr_stripes, + u32 stripe_len) +{ + struct scrub_full_stripe *ret; + int size = sizeof(*ret) + nr_stripes * sizeof(struct scrub_stripe); + int i; + + ret = malloc(size); + if (!ret) + return NULL; + + memset(ret, 0, size); + ret->nr_stripes = nr_stripes; + ret->stripe_len = stripe_len; + ret->corrupted_index[0] = -1; + ret->corrupted_index[1] = -1; + + /* Alloc data memory for each stripe */ + for (i = 0; i < nr_stripes; i++) { + struct scrub_stripe *stripe = >stripes[i]; + + stripe->data = malloc(stripe_len); + if (!stripe->data) { + free_full_stripe(ret); + return NULL; + } + } + return ret; +} -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 06/19] btrfs-progs: Introduce new btrfs_map_block function which returns more unified result.
Introduce a new function, __btrfs_map_block_v2(). Unlike old btrfs_map_block(), which needs different parameter to handle different RAID profile, this new function uses unified btrfs_map_block structure to handle all RAID profile in a more meaningful method: Return physical address along with logical address for each stripe. For RAID1/Single/DUP (none-stripped): result would be like: Map block: Logical 128M, Len 10M, Type RAID1, Stripe len 0, Nr_stripes 2 Stripe 0: Logical 128M, Physical X, Len: 10M Dev dev1 Stripe 1: Logical 128M, Physical Y, Len: 10M Dev dev2 Result will be as long as possible, since it's not stripped at all. For RAID0/10 (stripped without parity): Result will be aligned to full stripe size: Map block: Logical 64K, Len 128K, Type RAID10, Stripe len 64K, Nr_stripes 4 Stripe 0: Logical 64K, Physical X, Len 64K Dev dev1 Stripe 1: Logical 64K, Physical Y, Len 64K Dev dev2 Stripe 2: Logical 128K, Physical Z, Len 64K Dev dev3 Stripe 3: Logical 128K, Physical W, Len 64K Dev dev4 For RAID5/6 (stripped with parity and dev-rotation) Result will be aligned to full stripe size: Map block: Logical 64K, Len 128K, Type RAID6, Stripe len 64K, Nr_stripes 4 Stripe 0: Logical 64K, Physical X, Len 64K Dev dev1 Stripe 1: Logical 128K, Physical Y, Len 64K Dev dev2 Stripe 2: Logical RAID5_P, Physical Z, Len 64K Dev dev3 Stripe 3: Logical RAID6_Q, Physical W, Len 64K Dev dev4 The new unified layout should be very flex and can even handle things like N-way RAID1 (which old mirror_num basic one can't handle well). Signed-off-by: Qu Wenruo--- volumes.c | 181 ++ volumes.h | 49 + 2 files changed, 230 insertions(+) diff --git a/volumes.c b/volumes.c index f17bdeed..11d1f0e8 100644 --- a/volumes.c +++ b/volumes.c @@ -1593,6 +1593,187 @@ out: return 0; } +static inline struct btrfs_map_block *alloc_map_block(int num_stripes) +{ + struct btrfs_map_block *ret; + int size; + + size = sizeof(struct btrfs_map_stripe) * num_stripes + + sizeof(struct btrfs_map_block); + ret = malloc(size); + if (!ret) + return NULL; + memset(ret, 0, size); + return ret; +} + +static int fill_full_map_block(struct map_lookup *map, u64 start, u64 length, + struct btrfs_map_block *map_block) +{ + u64 profile = map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK; + u64 bg_start = map->ce.start; + u64 bg_end = bg_start + map->ce.size; + u64 bg_offset = start - bg_start; /* offset inside the block group */ + u64 fstripe_logical = 0;/* Full stripe start logical bytenr */ + u64 fstripe_size = 0; /* Full stripe logical size */ + u64 fstripe_phy_off = 0;/* Full stripe offset in each dev */ + u32 stripe_len = map->stripe_len; + int sub_stripes = map->sub_stripes; + int data_stripes = nr_data_stripes(map); + int dev_rotation; + int i; + + map_block->num_stripes = map->num_stripes; + map_block->type = profile; + + /* +* Common full stripe data for stripe based profiles +*/ + if (profile & (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID10 | + BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6)) { + fstripe_size = stripe_len * data_stripes; + if (sub_stripes) + fstripe_size /= sub_stripes; + fstripe_logical = bg_offset / fstripe_size * fstripe_size + + bg_start; + fstripe_phy_off = bg_offset / fstripe_size * stripe_len; + } + + switch (profile) { + case BTRFS_BLOCK_GROUP_DUP: + case BTRFS_BLOCK_GROUP_RAID1: + case 0: /* SINGLE */ + /* +* None-stripe mode,(Single, DUP and RAID1) +* Just use offset to fill map_block +*/ + map_block->stripe_len = 0; + map_block->start = start; + map_block->length = min(bg_end, start + length) - start; + for (i = 0; i < map->num_stripes; i++) { + struct btrfs_map_stripe *stripe; + + stripe = _block->stripes[i]; + + stripe->dev = map->stripes[i].dev; + stripe->logical = start; + stripe->physical = map->stripes[i].physical + bg_offset; + stripe->length = map_block->length; + } + break; + case BTRFS_BLOCK_GROUP_RAID10: + case BTRFS_BLOCK_GROUP_RAID0: + /* +* Stripe modes without parity(0 and 10) +* Return the whole full stripe +*/ + + map_block->start = fstripe_logical; + map_block->length = fstripe_size; + map_block->stripe_len =
[PATCH v2 18/19] btrfs-progs: scrub: Introduce function to check a whole block group
Introduce new function, scrub_one_block_group(), to scrub a block group. For Single/DUP/RAID0/RAID1/RAID10, we use old mirror number based map_block, and check extent by extent. For parity based profile (RAID5/6), we use new map_block_v2() and check full stripe by full stripe. Signed-off-by: Qu Wenruo--- scrub.c | 91 + 1 file changed, 91 insertions(+) diff --git a/scrub.c b/scrub.c index bb94fa9f..8f122012 100644 --- a/scrub.c +++ b/scrub.c @@ -862,3 +862,94 @@ out: free(map_block); return ret; } + +/* + * Scrub one block group. + * + * This function will handle all profiles current btrfs supports. + * Return 0 for scrubbing the block group. Found error will be recorded into + * scrub_ctx. + * Return <0 for fatal error preventing scrubing the block group. + */ +static int scrub_one_block_group(struct btrfs_fs_info *fs_info, +struct btrfs_scrub_progress *scrub_ctx, +struct btrfs_block_group_cache *bg_cache) +{ + struct btrfs_root *extent_root = fs_info->extent_root; + struct btrfs_path *path; + struct btrfs_key key; + u64 bg_start = bg_cache->key.objectid; + u64 bg_len = bg_cache->key.offset; + int ret; + + if (bg_cache->flags & + (BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6)) { + u64 cur = bg_start; + u64 next; + + while (cur < bg_start + bg_len) { + ret = scrub_one_full_stripe(fs_info, scrub_ctx, cur, + ); + /* Ignore any non-fatal error */ + if (ret < 0 && ret != -EIO) { + error("fatal error happens checking one full stripe at bytenr: %llu: %s", + cur, strerror(-ret)); + return ret; + } + cur = next; + } + /* Ignore any -EIO error, such error will be reported at last */ + return 0; + } + /* None parity based profile, check extent by extent */ + key.objectid = bg_start; + key.type = 0; + key.offset = 0; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + ret = btrfs_search_slot(NULL, extent_root, , path, 0, 0); + if (ret < 0) + goto out; + while (1) { + struct extent_buffer *eb = path->nodes[0]; + int slot = path->slots[0]; + u64 extent_start; + u64 extent_len; + + btrfs_item_key_to_cpu(eb, , slot); + if (key.objectid >= bg_start + bg_len) + break; + if (key.type != BTRFS_EXTENT_ITEM_KEY && + key.type != BTRFS_METADATA_ITEM_KEY) + goto next; + + extent_start = key.objectid; + if (key.type == BTRFS_METADATA_ITEM_KEY) + extent_len = extent_root->nodesize; + else + extent_len = key.offset; + + ret = scrub_one_extent(fs_info, scrub_ctx, path, extent_start, + extent_len, 1); + if (ret < 0 && ret != -EIO) { + error("fatal error checking extent bytenr %llu len %llu: %s", + extent_start, extent_len, strerror(-ret)); + goto out; + } + ret = 0; +next: + ret = btrfs_next_extent_item(extent_root, path, bg_start + +bg_len); + if (ret < 0) + goto out; + if (ret > 0) { + ret = 0; + break; + } + } +out: + btrfs_free_path(path); + return ret; +} -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 01/19] btrfs-progs: raid56: Introduce raid56 header for later recovery usage
Introduce a new header, kernel-lib/raid56.h, for later raid56 works. It contains 2 functions, from original btrfs-progs code: void raid6_gen_syndrome(int disks, size_t bytes, void **ptrs); int raid5_gen_result(int nr_devs, size_t stripe_len, int dest, void **data); Will be expanded later and some part of it(RAID6 recover part) may keep sync with kernel later. Signed-off-by: Qu Wenruo--- Makefile.in | 2 +- disk-io.h | 5 - kernel-lib/raid56.h | 28 volumes.c | 1 + 4 files changed, 30 insertions(+), 6 deletions(-) create mode 100644 kernel-lib/raid56.h diff --git a/Makefile.in b/Makefile.in index 0e3a0a0f..6e009bff 100644 --- a/Makefile.in +++ b/Makefile.in @@ -110,7 +110,7 @@ libbtrfs_objects = send-stream.o send-utils.o kernel-lib/rbtree.o btrfs-list.o \ uuid-tree.o utils-lib.o rbtree-utils.o libbtrfs_headers = send-stream.h send-utils.h send.h kernel-lib/rbtree.h btrfs-list.h \ kernel-lib/crc32c.h kernel-lib/list.h kerncompat.h \ - kernel-lib/radix-tree.h extent-cache.h \ + kernel-lib/radix-tree.h kernel-lib/raid56.h extent-cache.h \ extent_io.h ioctl.h ctree.h btrfsck.h version.h TESTS = fsck-tests.sh convert-tests.sh diff --git a/disk-io.h b/disk-io.h index 1c8387e7..4de9fef7 100644 --- a/disk-io.h +++ b/disk-io.h @@ -196,9 +196,4 @@ int write_tree_block(struct btrfs_trans_handle *trans, struct extent_buffer *eb); int write_and_map_eb(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct extent_buffer *eb); - -/* raid56.c */ -void raid6_gen_syndrome(int disks, size_t bytes, void **ptrs); -int raid5_gen_result(int nr_devs, size_t stripe_len, int dest, void **data); - #endif diff --git a/kernel-lib/raid56.h b/kernel-lib/raid56.h new file mode 100644 index ..7d4f4678 --- /dev/null +++ b/kernel-lib/raid56.h @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2016 Fujitsu. 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 _BTRFS_PROGS_RAID56_H +#define _BTRFS_PROGS_RAID56_H +/* + * Headers for RAID5/6 operations. + * Original headers from original RAID5/6 codes, not from kernel header. + */ + +void raid6_gen_syndrome(int disks, size_t bytes, void **ptrs); +int raid5_gen_result(int nr_devs, size_t stripe_len, int dest, void **data); +#endif diff --git a/volumes.c b/volumes.c index a0a85edd..f17bdeed 100644 --- a/volumes.c +++ b/volumes.c @@ -28,6 +28,7 @@ #include "print-tree.h" #include "volumes.h" #include "utils.h" +#include "kernel-lib/raid56.h" struct stripe { struct btrfs_device *dev; -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 13/19] btrfs-progs: scrub: Introduce function to scrub one data stripe
Introduce new function, scrub_one_data_stripe(), to check all data and tree blocks inside the data stripe. This function will not try to recovery any error, but only check if any data/tree blocks has mismatch csum. If data missing csum, which is completely valid for case like nodatasum, it will just record it, but not report as error. Signed-off-by: Qu Wenruo--- scrub.c | 129 1 file changed, 129 insertions(+) diff --git a/scrub.c b/scrub.c index f4ed0b78..15a1955c 100644 --- a/scrub.c +++ b/scrub.c @@ -351,3 +351,132 @@ invalid_arg: error("invalid parameter for %s", __func__); return -EINVAL; } + +/* + * Scrub one full data stripe of RAID5/6. + * This means it will check any data/metadata extent in the data stripe + * spcified by @stripe and @stripe_len + * + * This function will only *CHECK* if the data stripe has any corruption. + * Won't repair at this function. + * + * Return 0 if the full stripe is OK. + * Return <0 if any error is found. + * Note: Missing csum is not counted as error(NODATACSUM is valid) + */ +static int scrub_one_data_stripe(struct btrfs_fs_info *fs_info, +struct btrfs_scrub_progress *scrub_ctx, +struct scrub_stripe *stripe, u32 stripe_len) +{ + struct btrfs_path *path; + struct btrfs_root *extent_root = fs_info->extent_root; + struct btrfs_key key; + u64 extent_start; + u64 extent_len; + u64 orig_csum_discards; + int ret; + + if (!is_data_stripe(stripe)) + return -EINVAL; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + key.objectid = stripe->logical + stripe_len; + key.offset = 0; + key.type = 0; + + ret = btrfs_search_slot(NULL, extent_root, , path, 0, 0); + if (ret < 0) + goto out; + while (1) { + struct btrfs_extent_item *ei; + struct extent_buffer *eb; + char *data; + int slot; + int metadata = 0; + u64 check_start; + u64 check_len; + + ret = btrfs_previous_extent_item(extent_root, path, 0); + if (ret > 0) { + ret = 0; + goto out; + } + if (ret < 0) + goto out; + eb = path->nodes[0]; + slot = path->slots[0]; + btrfs_item_key_to_cpu(eb, , slot); + extent_start = key.objectid; + ei = btrfs_item_ptr(eb, slot, struct btrfs_extent_item); + + /* tree block scrub */ + if (key.type == BTRFS_METADATA_ITEM_KEY || + btrfs_extent_flags(eb, ei) & BTRFS_EXTENT_FLAG_TREE_BLOCK) { + extent_len = extent_root->nodesize; + metadata = 1; + } else { + extent_len = key.offset; + metadata = 0; + } + + /* Current extent is out of our range, loop comes to end */ + if (extent_start + extent_len <= stripe->logical) + break; + + if (metadata) { + /* +* Check crossing stripe first, which can't be scrubbed +*/ + if (check_crossing_stripes(fs_info, extent_start, + extent_root->nodesize)) { + error("tree block at %llu is crossing stripe boundary, unable to scrub", + extent_start); + ret = -EIO; + goto out; + } + data = stripe->data + extent_start - stripe->logical; + ret = scrub_tree_mirror(fs_info, scrub_ctx, + data, extent_start, 0); + /* Any csum/verify error means the stripe is screwed */ + if (ret < 0) { + stripe->csum_mismatch = 1; + ret = -EIO; + goto out; + } + ret = 0; + continue; + } + /* Restrict the extent range to fit stripe range */ + check_start = max(extent_start, stripe->logical); + check_len = min(extent_start + extent_len, stripe->logical + + stripe_len) - check_start; + + /* Record original csum_discards to detect missing csum case */ + orig_csum_discards = scrub_ctx->csum_discards; + + data = stripe->data + check_start - stripe->logical; + ret =
[PATCH v2 19/19] btrfs-progs: fsck: Introduce offline scrub function
Now, btrfs check has a kernel scrub equivalent. A new option, --scrub is added for "btrfs check". If --scrub is given, btrfs check will just act like kernel scrub, to check every copy of extent and do a report on corrupted data and if it's recoverable. The advantage compare to kernel scrub is: 1) No race Unlike kernel scrub, which is done in parallel, offline scrub is done by a single thread. Although it may be slower than kernel one, it's safer and no false alert. 2) Correctness Kernel has a known bug (fix submitted) which will recovery RAID5/6 data but screw up P/Q, due to the hardness coding in kernel. While in btrfs-progs, no page, (almost) no memory size limit, we're can focus on the scrub, and make things easier. Signed-off-by: Qu Wenruo--- Documentation/btrfs-check.asciidoc | 7 ++ cmds-check.c | 12 +- ctree.h| 3 +++ scrub.c| 49 ++ 4 files changed, 70 insertions(+), 1 deletion(-) diff --git a/Documentation/btrfs-check.asciidoc b/Documentation/btrfs-check.asciidoc index 633cbbf6..d421afa4 100644 --- a/Documentation/btrfs-check.asciidoc +++ b/Documentation/btrfs-check.asciidoc @@ -91,6 +91,13 @@ the entire free space cache. This option with 'v2' provides an alternative method of clearing the free space cache that doesn't require mounting the filesystem. +--scrub:: +kernel scrub equivalent. ++ +Off-line scrub has better reconstruction check than kernel. Won't cause +possible silent data corruption for RAID5 ++ +NOTE: Repair is not supported yet. DANGEROUS OPTIONS - diff --git a/cmds-check.c b/cmds-check.c index 1dba2985..3a16a1ff 100644 --- a/cmds-check.c +++ b/cmds-check.c @@ -12588,6 +12588,7 @@ int cmd_check(int argc, char **argv) int clear_space_cache = 0; int qgroup_report = 0; int qgroups_repaired = 0; + int scrub = 0; unsigned ctree_flags = OPEN_CTREE_EXCLUSIVE; while(1) { @@ -12595,7 +12596,8 @@ int cmd_check(int argc, char **argv) enum { GETOPT_VAL_REPAIR = 257, GETOPT_VAL_INIT_CSUM, GETOPT_VAL_INIT_EXTENT, GETOPT_VAL_CHECK_CSUM, GETOPT_VAL_READONLY, GETOPT_VAL_CHUNK_TREE, - GETOPT_VAL_MODE, GETOPT_VAL_CLEAR_SPACE_CACHE }; + GETOPT_VAL_MODE, GETOPT_VAL_CLEAR_SPACE_CACHE, + GETOPT_VAL_SCRUB }; static const struct option long_options[] = { { "super", required_argument, NULL, 's' }, { "repair", no_argument, NULL, GETOPT_VAL_REPAIR }, @@ -12617,6 +12619,7 @@ int cmd_check(int argc, char **argv) GETOPT_VAL_MODE }, { "clear-space-cache", required_argument, NULL, GETOPT_VAL_CLEAR_SPACE_CACHE}, + { "scrub", no_argument, NULL, GETOPT_VAL_SCRUB }, { NULL, 0, NULL, 0} }; @@ -12701,6 +12704,9 @@ int cmd_check(int argc, char **argv) } ctree_flags |= OPEN_CTREE_WRITES; break; + case GETOPT_VAL_SCRUB: + scrub = 1; + break; } } @@ -12755,6 +12761,10 @@ int cmd_check(int argc, char **argv) global_info = info; root = info->fs_root; + if (scrub) { + ret = btrfs_scrub(info, repair); + goto err_out; + } if (clear_space_cache == 1) { if (btrfs_fs_compat_ro(info, FREE_SPACE_TREE)) { error( diff --git a/ctree.h b/ctree.h index fe7c077e..8f669ee7 100644 --- a/ctree.h +++ b/ctree.h @@ -2803,4 +2803,7 @@ int btrfs_punch_hole(struct btrfs_trans_handle *trans, /* csum.c */ int btrfs_read_one_data_csum(struct btrfs_fs_info *fs_info, u64 bytenr, void *csum_ret); + +/* scrub.c */ +int btrfs_scrub(struct btrfs_fs_info *fs_info, int repair); #endif diff --git a/scrub.c b/scrub.c index 8f122012..00662da0 100644 --- a/scrub.c +++ b/scrub.c @@ -953,3 +953,52 @@ out: btrfs_free_path(path); return ret; } + +int btrfs_scrub(struct btrfs_fs_info *fs_info, int repair) +{ + struct btrfs_block_group_cache *bg_cache; + struct btrfs_scrub_progress scrub_ctx = {0}; + int ret = 0; + + /* +* TODO: To support repair, which should not be quite hard +*/ + if (repair) { + error("Read-write scrub is not supported yet"); + return 1; + } + + bg_cache = btrfs_lookup_first_block_group(fs_info, 0); + if (!bg_cache) { + error("no block group is found"); + return -ENOENT; + } + +
[PATCH v2 10/19] btrfs-progs: scrub: Introduce function to scrub mirror based tree block
Introduce a new function, scrub_tree_mirror(), to scrub mirror based tree blocks (Single/DUP/RAID0/1/10) This function can also be used on in-memory tree blocks using @data parameter. This is very handy for RAID5/6 case, either checking the data stripe tree block by @bytenr and 0 as @mirror, or using @data parameter for recovered in-memory data. Signed-off-by: Qu Wenruo--- disk-io.c | 4 ++-- disk-io.h | 2 ++ scrub.c | 72 +++ 3 files changed, 76 insertions(+), 2 deletions(-) diff --git a/disk-io.c b/disk-io.c index 9140a81b..d5011572 100644 --- a/disk-io.c +++ b/disk-io.c @@ -51,8 +51,8 @@ static u32 max_nritems(u8 level, u32 nodesize) sizeof(struct btrfs_key_ptr)); } -static int check_tree_block(struct btrfs_fs_info *fs_info, - struct extent_buffer *buf) +int check_tree_block(struct btrfs_fs_info *fs_info, +struct extent_buffer *buf) { struct btrfs_fs_devices *fs_devices; diff --git a/disk-io.h b/disk-io.h index 4de9fef7..db883d57 100644 --- a/disk-io.h +++ b/disk-io.h @@ -119,6 +119,8 @@ static inline struct extent_buffer* read_tree_block( parent_transid); } +int check_tree_block(struct btrfs_fs_info *fs_info, +struct extent_buffer *buf); int read_extent_data(struct btrfs_root *root, char *data, u64 logical, u64 *len, int mirror); void readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize, diff --git a/scrub.c b/scrub.c index c9ca817e..4cf678fb 100644 --- a/scrub.c +++ b/scrub.c @@ -118,3 +118,75 @@ static struct scrub_full_stripe *alloc_full_stripe(int nr_stripes, } return ret; } + +static inline int is_data_stripe(struct scrub_stripe *stripe) +{ + u64 bytenr = stripe->logical; + + if (bytenr == BTRFS_RAID5_P_STRIPE || bytenr == BTRFS_RAID6_Q_STRIPE) + return 0; + return 1; +} + +/* + * Scrub one tree mirror given by @bytenr and @mirror, or @data. + * If @data is not given(NULL), the function will try to read out tree block + * using @bytenr and @mirror. + * If @data is given, use data directly, won't try to read from disk. + * + * The extra @data prameter is handy for RAID5/6 recovery code to verify + * the recovered data. + * + * Return 0 if everything is OK. + * Return <0 something goes wrong, and @scrub_ctx accounting will be updated + * if it's a data corruption. + */ +static int scrub_tree_mirror(struct btrfs_fs_info *fs_info, +struct btrfs_scrub_progress *scrub_ctx, +char *data, u64 bytenr, int mirror) +{ + struct extent_buffer *eb; + u32 nodesize = fs_info->tree_root->nodesize; + int ret; + + if (!IS_ALIGNED(bytenr, fs_info->tree_root->sectorsize)) { + /* Such error will be reported by check_tree_block() */ + scrub_ctx->verify_errors++; + return -EIO; + } + + eb = btrfs_find_create_tree_block(fs_info, bytenr, nodesize); + if (!eb) + return -ENOMEM; + if (data) { + memcpy(eb->data, data, nodesize); + } else { + ret = read_whole_eb(fs_info, eb, mirror); + if (ret) { + scrub_ctx->read_errors++; + error("failed to read tree block %llu mirror %d", + bytenr, mirror); + goto out; + } + } + + scrub_ctx->tree_bytes_scrubbed += nodesize; + if (csum_tree_block(fs_info->tree_root, eb, 1)) { + error("tree block %llu mirror %d checksum mismatch", bytenr, + mirror); + scrub_ctx->csum_errors++; + ret = -EIO; + goto out; + } + ret = check_tree_block(fs_info, eb); + if (ret < 0) { + error("tree block %llu mirror %d is invalid", bytenr, mirror); + scrub_ctx->verify_errors++; + goto out; + } + + scrub_ctx->tree_extents_scrubbed++; +out: + free_extent_buffer(eb); + return ret; +} -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 11/19] btrfs-progs: scrub: Introduce function to scrub mirror based data blocks
Introduce a new function, scrub_data_mirror(), to check mirror based data blocks. It can also accept @data parameter to use in-memory data instead of reading them out of disk. This is a handy feature for RAID5/6 recovery verification code. Signed-off-by: Qu Wenruo--- scrub.c | 82 + 1 file changed, 82 insertions(+) diff --git a/scrub.c b/scrub.c index 4cf678fb..2563f407 100644 --- a/scrub.c +++ b/scrub.c @@ -190,3 +190,85 @@ out: free_extent_buffer(eb); return ret; } + +/* + * Scrub one data mirror given by @start @len and @mirror, or @data + * If @data is not given, try to read it from disk. + * This function will try to read out all the data then check sum. + * + * If @data is given, just use the data. + * This behavior is useful for RAID5/6 recovery code to verify recovered data. + * + * Return 0 if everything is OK. + * Return <0 if something goes wrong, and @scrub_ctx accounting will be updated + * if it's a data corruption. + */ +static int scrub_data_mirror(struct btrfs_fs_info *fs_info, +struct btrfs_scrub_progress *scrub_ctx, +char *data, u64 start, u64 len, int mirror) +{ + u64 cur = 0; + u32 csum; + u32 sectorsize = fs_info->tree_root->sectorsize; + char *buf = NULL; + int ret = 0; + int err = 0; + + if (!data) { + buf = malloc(len); + if (!buf) + return -ENOMEM; + /* Read out as much data as possible to speed up read */ + while (cur < len) { + u64 read_len = len - cur; + + ret = read_extent_data(fs_info->tree_root, buf + cur, + start + cur, _len, mirror); + if (ret < 0) { + error("failed to read out data at logical bytenr %llu mirror %d", + start + cur, mirror); + scrub_ctx->read_errors++; + goto out; + } + scrub_ctx->data_bytes_scrubbed += read_len; + cur += read_len; + } + } else { + buf = data; + } + + /* Check csum per-sectorsize */ + cur = 0; + while (cur < len) { + u32 data_csum = ~(u32)0; + + ret = btrfs_read_one_data_csum(fs_info, start + cur, ); + if (ret > 0) { + scrub_ctx->csum_discards++; + ret = 0; + + /* In case only some csum are missing */ + goto next; + } + data_csum = btrfs_csum_data(NULL, buf + cur, data_csum, + sectorsize); + btrfs_csum_final(data_csum, (u8 *)_csum); + if (data_csum != csum) { + error("data at bytenr %llu mirror %d csum mismatch, have %u expect %u", + start + cur, mirror, data_csum, csum); + err = 1; + scrub_ctx->csum_errors++; + cur += sectorsize; + continue; + } + scrub_ctx->data_bytes_scrubbed += sectorsize; +next: + cur += sectorsize; + } +out: + if (!data) + free(buf); + if (!ret && err) + return -EIO; + return ret; +} -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 17/19] btrfs-progs: scrub: Introduce a function to scrub one full stripe
Introduce a new function, scrub_one_full_stripe(), to check a full stripe. It handles the full stripe scrub in the following steps: 0) Check if we need to check full stripe If full stripe contains no extent, why waste our CPU and IO? 1) Read out full stripe Then we know how many devices are missing or have read error. If out of repair, then exit If have missing device or have read error, try recover here. 2) Check data stripe against csum We add data stripe with csum error as corrupted stripe, just like dev missing or read error. Then recheck if csum mismatch is still below tolerance. Finally we check the full stripe using 2 factors only: A) If the full stripe go through recover ever B) If the full stripe has csum error Combine factor A and B we get: 1) A && B: Recovered, csum mismatch Screwed up totally 2) A && !B: Recovered, csum match Recoverable, data corrupted but P/Q is good to recover 3) !A && B: Not recovered, csum mismatch Try to recover corrupted data stripes If recovered csum match, then recoverable Else, screwed up 4) !A && !B: Not recovered, no csum mismatch Best case, just check if P/Q matches. If P/Q matches, everything is good Else, just P/Q is screwed up, still recoverable. Signed-off-by: Qu Wenruo--- scrub.c | 262 1 file changed, 262 insertions(+) diff --git a/scrub.c b/scrub.c index 522802c2..bb94fa9f 100644 --- a/scrub.c +++ b/scrub.c @@ -600,3 +600,265 @@ static int recover_from_parities(struct btrfs_fs_info *fs_info, free(ptrs); return ret; } + +/* + * Return 0 if we still have chance to recover + * Return <0 if we have no more chance + */ +static int report_recoverablity(struct scrub_full_stripe *fstripe) +{ + int max_tolerance; + u64 start = fstripe->logical_start; + + if (fstripe->bg_type & BTRFS_BLOCK_GROUP_RAID5) + max_tolerance = 1; + else + max_tolerance = 2; + + if (fstripe->nr_corrupted_stripes > max_tolerance) { + error( + "full stripe %llu CORRUPTED: too many read error or corrupted devices", + start); + error( + "full stripe %llu: tolerance: %d, missing: %d, read error: %d, csum error: %d", + start, max_tolerance, fstripe->err_read_stripes, + fstripe->err_missing_devs, fstripe->err_csum_dstripes); + return -EIO; + } + return 0; +} + +static void clear_corrupted_stripe_record(struct scrub_full_stripe *fstripe) +{ + fstripe->corrupted_index[0] = -1; + fstripe->corrupted_index[1] = -1; + fstripe->nr_corrupted_stripes = 0; +} + +static void record_corrupted_stripe(struct scrub_full_stripe *fstripe, + int index) +{ + int i = 0; + + for (i = 0; i < 2; i++) { + if (fstripe->corrupted_index[i] == -1) { + fstripe->corrupted_index[i] = index; + break; + } + } + fstripe->nr_corrupted_stripes++; +} + +/* + * Scrub one full stripe. + * + * If everything matches, that's good. + * If data stripe corrupted badly, no mean to recovery, it will report it. + * If data stripe corrupted, try recovery first and recheck csum, to + * determine if it's recoverable or screwed up. + */ +static int scrub_one_full_stripe(struct btrfs_fs_info *fs_info, +struct btrfs_scrub_progress *scrub_ctx, +u64 start, u64 *next_ret) +{ + struct scrub_full_stripe *fstripe; + struct btrfs_map_block *map_block = NULL; + u32 stripe_len = BTRFS_STRIPE_LEN; + u64 bg_type; + u64 len; + int i; + int ret; + + if (!next_ret) { + error("invalid argument for %s", __func__); + return -EINVAL; + } + + ret = __btrfs_map_block_v2(fs_info, WRITE, start, stripe_len, + _block); + if (ret < 0) { + /* Let caller to skip the whole block group */ + *next_ret = (u64)-1; + return ret; + } + start = map_block->start; + len = map_block->length; + *next_ret = start + len; + + /* +* Step 0: Check if we need to scrub the full stripe +* +* If no extent lies in the full stripe, not need to check +*/ + ret = btrfs_check_extent_exists(fs_info, start, len); + if (ret < 0) { + free(map_block); + return ret; + } + /* No extents in range, no need to check */ + if (ret == 0) { + free(map_block); + return 0; + } + + bg_type = map_block->type & BTRFS_BLOCK_GROUP_PROFILE_MASK; + if (bg_type != BTRFS_BLOCK_GROUP_RAID5 && + bg_type != BTRFS_BLOCK_GROUP_RAID6) { +
[PATCH v2 03/19] btrfs-progs: raid56: Allow raid6 to recover 2 data stripes
Copied from kernel lib/raid6/recov.c raid6_2data_recov_intx1() function. With the following modification: - Rename to raid6_recov_data2() for shorter name - s/kfree/free/g modification Signed-off-by: Qu Wenruo--- Makefile.in | 4 +-- raid56.c => kernel-lib/raid56.c | 69 + kernel-lib/raid56.h | 5 +++ 3 files changed, 76 insertions(+), 2 deletions(-) rename raid56.c => kernel-lib/raid56.c (71%) diff --git a/Makefile.in b/Makefile.in index fecaaa6a..c3f0eeda 100644 --- a/Makefile.in +++ b/Makefile.in @@ -96,10 +96,10 @@ CHECKER_FLAGS := -include $(check_defs) -D__CHECKER__ \ objects = ctree.o disk-io.o kernel-lib/radix-tree.o extent-tree.o print-tree.o \ root-tree.o dir-item.o file-item.o inode-item.o inode-map.o \ extent-cache.o extent_io.o volumes.o utils.o repair.o \ - qgroup.o raid56.o free-space-cache.o kernel-lib/list_sort.o props.o \ + qgroup.o free-space-cache.o kernel-lib/list_sort.o props.o \ ulist.o qgroup-verify.o backref.o string-table.o task-utils.o \ inode.o file.o find-root.o free-space-tree.o help.o send-dump.o \ - kernel-lib/tables.o + kernel-lib/tables.o kernel-lib/raid56.o cmds_objects = cmds-subvolume.o cmds-filesystem.o cmds-device.o cmds-scrub.o \ cmds-inspect.o cmds-balance.o cmds-send.o cmds-receive.o \ cmds-quota.o cmds-qgroup.o cmds-replace.o cmds-check.o \ diff --git a/raid56.c b/kernel-lib/raid56.c similarity index 71% rename from raid56.c rename to kernel-lib/raid56.c index 8c79c456..dca8f8d4 100644 --- a/raid56.c +++ b/kernel-lib/raid56.c @@ -28,6 +28,7 @@ #include "disk-io.h" #include "volumes.h" #include "utils.h" +#include "kernel-lib/raid56.h" /* * This is the C data type to use @@ -170,3 +171,71 @@ int raid5_gen_result(int nr_devs, size_t stripe_len, int dest, void **data) } return 0; } + +/* + * Raid 6 recovery code copied from kernel lib/raid6/recov.c. + * With modifications: + * - rename from raid6_2data_recov_intx1 + * - kfree/free modification for btrfs-progs + */ +int raid6_recov_data2(int nr_devs, size_t stripe_len, int dest1, int dest2, + void **data) +{ + u8 *p, *q, *dp, *dq; + u8 px, qx, db; + const u8 *pbmul;/* P multiplier table for B data */ + const u8 *qmul; /* Q multiplier table (for both) */ + char *zero_mem1, *zero_mem2; + int ret = 0; + + /* Early check */ + if (dest1 < 0 || dest1 >= nr_devs - 2 || + dest2 < 0 || dest2 >= nr_devs - 2 || dest1 >= dest2) + return -EINVAL; + + zero_mem1 = calloc(1, stripe_len); + zero_mem2 = calloc(1, stripe_len); + if (!zero_mem1 || !zero_mem2) { + free(zero_mem1); + free(zero_mem2); + return -ENOMEM; + } + + p = (u8 *)data[nr_devs - 2]; + q = (u8 *)data[nr_devs - 1]; + + /* Compute syndrome with zero for the missing data pages + Use the dead data pages as temporary storage for + delta p and delta q */ + dp = (u8 *)data[dest1]; + data[dest1] = (void *)zero_mem1; + data[nr_devs - 2] = dp; + dq = (u8 *)data[dest2]; + data[dest2] = (void *)zero_mem2; + data[nr_devs - 1] = dq; + + raid6_gen_syndrome(nr_devs, stripe_len, data); + + /* Restore pointer table */ + data[dest1] = dp; + data[dest2] = dq; + data[nr_devs - 2] = p; + data[nr_devs - 1] = q; + + /* Now, pick the proper data tables */ + pbmul = raid6_gfmul[raid6_gfexi[dest2 - dest1]]; + qmul = raid6_gfmul[raid6_gfinv[raid6_gfexp[dest1]^raid6_gfexp[dest2]]]; + + /* Now do it... */ + while ( stripe_len-- ) { + px= *p ^ *dp; + qx= qmul[*q ^ *dq]; + *dq++ = db = pbmul[px] ^ qx; /* Reconstructed B */ + *dp++ = db ^ px; /* Reconstructed A */ + p++; q++; + } + + free(zero_mem1); + free(zero_mem2); + return ret; +} diff --git a/kernel-lib/raid56.h b/kernel-lib/raid56.h index 1bf2e01a..d397a23e 100644 --- a/kernel-lib/raid56.h +++ b/kernel-lib/raid56.h @@ -37,4 +37,9 @@ extern const u8 raid6_vgfmul[256][32] __attribute__((aligned(256))); extern const u8 raid6_gfexp[256] __attribute__((aligned(256))); extern const u8 raid6_gfinv[256] __attribute__((aligned(256))); extern const u8 raid6_gfexi[256] __attribute__((aligned(256))); + + +/* Recover raid6 with 2 data corrupted */ +int raid6_recov_data2(int nr_devs, size_t stripe_len, int dest1, int dest2, + void **data); #endif -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 00/19]
For any one who wants to try it, it can be get from my repo: https://github.com/adam900710/btrfs-progs/tree/offline_scrub Currently, I only tested it on SINGLE/DUP/RAID1/RAID5 filesystems, with mirror or parity or data corrupted. The tool are all able to detect them and give recoverbility report. Several reports on kernel scrub screwing up good data stripes are in ML for sometime. And since kernel scrub won't account P/Q corruption, it makes us quite to detect error like kernel screwing up P/Q when scrubbing. To get a comparable tool for kernel scrub, we need a user-space tool to act as benchmark to compare their different behaviors. So here is the patchset for user-space scrub. Which can do: 1) All mirror/backup check for non-parity based stripe Which means for RAID1/DUP/RAID10, we can really check all mirrors other than the 1st good mirror. Current "--check-data-csum" option will be finally replace by scrub. As it doesn't really check all mirrors, if it hits a good copy, then resting copies will just be ignored. 2) Comprehensive RAID5/6 full stripe check It will take full use of btrfs csum(both tree and data). It will only recover the full stripe if all recovered data matches with its csum. In fact, it can already expose several new btrfs kernel bug. As it's the main tool I'm using when developing the kernel fixes. For example, after screwing up a data stripe, kernel did repairs using parity, but recovered full stripe has wrong parity. Need to scrub again to fix it. And this patchset also introduced new map_block() function, which is more flex than current btrfs_map_block(), and has a unified interface for all profiles, not just an array of physical addresses. Check the 6th and 7th patch for details. They are already used in RAID5/6 scrub, but can also be used for other profiles too. The to-do list has been shortened, since RAID6 and new check logical is introduced. 1) Repair support In fact, current tool can already report recoverability, repair is not hard to implement. 2) Test cases Need to make the infrastructure able to handle multi-device first. 3) Make btrfsck able to handle RAID5 with missing device Now it doesn't even open RAID5 btrfs with missing device, even though scrub should be able to handle it. Changelog: V0.8 RFC: Initial RFC patchset v1: First formal patchset. RAID6 recovery support added, mainly copied from kernel radi6 lib. Cleaner recovery logical. v2: More comments in both code and commit message, suggested by David. File re-arrangement, no check/ dir, raid56.ch moved to kernel-lib, Suggested by David Qu Wenruo (19): btrfs-progs: raid56: Introduce raid56 header for later recovery usage btrfs-progs: raid56: Introduce tables for RAID6 recovery btrfs-progs: raid56: Allow raid6 to recover 2 data stripes btrfs-progs: raid56: Allow raid6 to recover data and p btrfs-progs: Introduce wrapper to recover raid56 data btrfs-progs: Introduce new btrfs_map_block function which returns more unified result. btrfs-progs: Allow __btrfs_map_block_v2 to remove unrelated stripes btrfs-progs: csum: Introduce function to read out one data csum btrfs-progs: scrub: Introduce structures to support fsck scrub for RAID56 btrfs-progs: scrub: Introduce function to scrub mirror based tree block btrfs-progs: scrub: Introduce function to scrub mirror based data blocks btrfs-progs: scrub: Introduce function to scrub one extent btrfs-progs: scrub: Introduce function to scrub one data stripe btrfs-progs: scrub: Introduce function to verify parities btrfs-progs: extent-tree: Introduce function to check if there is any extent in given range. btrfs-progs: scrub: Introduce function to recover data parity btrfs-progs: scrub: Introduce a function to scrub one full stripe btrfs-progs: scrub: Introduce function to check a whole block group btrfs-progs: fsck: Introduce offline scrub function .gitignore |2 + Documentation/btrfs-check.asciidoc |7 + Makefile.in| 19 +- cmds-check.c | 12 +- csum.c | 96 ctree.h|8 + disk-io.c |4 +- disk-io.h |7 +- extent-tree.c | 60 +++ kernel-lib/mktables.c | 148 ++ kernel-lib/raid56.c| 359 + kernel-lib/raid56.h| 58 +++ raid56.c | 172 -- scrub.c| 1004 volumes.c | 283 ++ volumes.h | 49 ++ 16 files changed, 2103 insertions(+), 185 deletions(-) create mode 100644 csum.c create mode 100644 kernel-lib/mktables.c create mode 100644 kernel-lib/raid56.c create mode 100644 kernel-lib/raid56.h delete mode
[PATCH v2 16/19] btrfs-progs: scrub: Introduce function to recover data parity
Introduce function, recover_from_parities(), to recover data stripes. It just wraps raid56_recov() with extra check functions to scrub_full_stripe structure. Signed-off-by: Qu Wenruo--- scrub.c | 51 +++ 1 file changed, 51 insertions(+) diff --git a/scrub.c b/scrub.c index 238feb3c..522802c2 100644 --- a/scrub.c +++ b/scrub.c @@ -549,3 +549,54 @@ out: free(ptrs); return ret; } + +/* + * Try to recover data stripe from P or Q stripe + * + * Return >0 if it can't be require any more. + * Return 0 for successful repair or no need to repair at all + * Return <0 for fatal error + */ +static int recover_from_parities(struct btrfs_fs_info *fs_info, + struct btrfs_scrub_progress *scrub_ctx, + struct scrub_full_stripe *fstripe) +{ + void **ptrs; + int nr_stripes = fstripe->nr_stripes; + int stripe_len = BTRFS_STRIPE_LEN; + int max_tolerance; + int i; + int ret; + + /* No need to recover */ + if (!fstripe->nr_corrupted_stripes) + return 0; + + /* Already recovered once, no more chance */ + if (fstripe->recovered) + return 1; + + if (fstripe->bg_type & BTRFS_BLOCK_GROUP_RAID5) + max_tolerance = 1; + else + max_tolerance = 2; + + /* Out of repair */ + if (fstripe->nr_corrupted_stripes > max_tolerance) + return 1; + + ptrs = malloc(sizeof(void *) * fstripe->nr_stripes); + if (!ptrs) + return -ENOMEM; + + /* Construct ptrs */ + for (i = 0; i < nr_stripes; i++) + ptrs[i] = fstripe->stripes[i].data; + + ret = raid56_recov(nr_stripes, stripe_len, fstripe->bg_type, + fstripe->corrupted_index[0], + fstripe->corrupted_index[1], ptrs); + fstripe->recovered = 1; + free(ptrs); + return ret; +} -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 14/19] btrfs-progs: scrub: Introduce function to verify parities
Introduce new function, verify_parities(), to check if parities matches for full stripe which all data stripes matches with their csum. Caller should fill the scrub_full_stripe structure properly before calling this function. Signed-off-by: Qu Wenruo--- scrub.c | 69 + 1 file changed, 69 insertions(+) diff --git a/scrub.c b/scrub.c index 15a1955c..238feb3c 100644 --- a/scrub.c +++ b/scrub.c @@ -25,6 +25,7 @@ #include "volumes.h" #include "disk-io.h" #include "utils.h" +#include "kernel-lib/raid56.h" /* * For parity based profile(RAID56) @@ -480,3 +481,71 @@ out: btrfs_free_path(path); return ret; } + +/* + * Verify parities for RAID56 + * Caller must fill @fstripe before calling this function + * + * Return 0 for parities matches. + * Return >0 for P or Q mismatch + * Return <0 for fatal error + */ +static int verify_parities(struct btrfs_fs_info *fs_info, + struct btrfs_scrub_progress *scrub_ctx, + struct scrub_full_stripe *fstripe) +{ + void **ptrs; + void *ondisk_p = NULL; + void *ondisk_q = NULL; + void *buf_p; + void *buf_q; + int nr_stripes = fstripe->nr_stripes; + int stripe_len = BTRFS_STRIPE_LEN; + int i; + int ret = 0; + + ptrs = malloc(sizeof(void *) * fstripe->nr_stripes); + buf_p = malloc(fstripe->stripe_len); + buf_q = malloc(fstripe->stripe_len); + if (!ptrs || !buf_p || !buf_q) { + ret = -ENOMEM; + goto out; + } + + for (i = 0; i < fstripe->nr_stripes; i++) { + struct scrub_stripe *stripe = >stripes[i]; + + if (stripe->logical == BTRFS_RAID5_P_STRIPE) { + ondisk_p = stripe->data; + ptrs[i] = buf_p; + continue; + } else if (stripe->logical == BTRFS_RAID6_Q_STRIPE) { + ondisk_q = stripe->data; + ptrs[i] = buf_q; + continue; + } else { + ptrs[i] = stripe->data; + continue; + } + } + /* RAID6 */ + if (ondisk_q) { + raid6_gen_syndrome(nr_stripes, stripe_len, ptrs); + + if (memcmp(ondisk_q, ptrs[nr_stripes - 1], stripe_len) != 0 || + memcmp(ondisk_p, ptrs[nr_stripes - 2], stripe_len)) + ret = 1; + } else { + ret = raid5_gen_result(nr_stripes, stripe_len, nr_stripes - 1, + ptrs); + if (ret < 0) + goto out; + if (memcmp(ondisk_p, ptrs[nr_stripes - 1], stripe_len) != 0) + ret = 1; + } +out: + free(buf_p); + free(buf_q); + free(ptrs); + return ret; +} -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 02/19] btrfs-progs: raid56: Introduce tables for RAID6 recovery
Use kernel RAID6 galois tables for later RAID6 recovery. Galois tables file, kernel-lib/tables.c is generated by user space program, mktable. Galois field tables declaration, in kernel-lib/raid56.h, is completely copied from kernel. The mktables.c is copied from kernel with minor header/macro modification, to ensure the generated tables.c works well in btrfs-progs. Signed-off-by: Qu Wenruo--- .gitignore| 2 + Makefile.in | 15 - kernel-lib/mktables.c | 148 ++ kernel-lib/raid56.h | 12 4 files changed, 174 insertions(+), 3 deletions(-) create mode 100644 kernel-lib/mktables.c diff --git a/.gitignore b/.gitignore index 98b3657b..554e8921 100644 --- a/.gitignore +++ b/.gitignore @@ -35,6 +35,8 @@ btrfs-select-super btrfs-calc-size btrfs-crc btrfstune +mktables +kernel-lib/tables.c libbtrfs.a libbtrfs.so libbtrfs.so.0 diff --git a/Makefile.in b/Makefile.in index 6e009bff..fecaaa6a 100644 --- a/Makefile.in +++ b/Makefile.in @@ -98,7 +98,8 @@ objects = ctree.o disk-io.o kernel-lib/radix-tree.o extent-tree.o print-tree.o \ extent-cache.o extent_io.o volumes.o utils.o repair.o \ qgroup.o raid56.o free-space-cache.o kernel-lib/list_sort.o props.o \ ulist.o qgroup-verify.o backref.o string-table.o task-utils.o \ - inode.o file.o find-root.o free-space-tree.o help.o send-dump.o + inode.o file.o find-root.o free-space-tree.o help.o send-dump.o \ + kernel-lib/tables.o cmds_objects = cmds-subvolume.o cmds-filesystem.o cmds-device.o cmds-scrub.o \ cmds-inspect.o cmds-balance.o cmds-send.o cmds-receive.o \ cmds-quota.o cmds-qgroup.o cmds-replace.o cmds-check.o \ @@ -318,6 +319,14 @@ version.h: version.sh version.h.in configure.ac @echo "[SH] $@" $(Q)bash ./config.status --silent $@ +mktables: kernel-lib/mktables.c + @echo "[CC] $@" + $(Q)$(CC) $(CFLAGS) $< -o $@ + +kernel-lib/tables.c: mktables + @echo "[TABLE] $@" + $(Q)./mktables > $@ || ($(RM) -f $@ && exit 1) + $(libs_shared): $(libbtrfs_objects) $(lib_links) send.h @echo "[LD] $@" $(Q)$(CC) $(CFLAGS) $(libbtrfs_objects) $(LDFLAGS) $(LIBBTRFS_LIBS) \ @@ -490,12 +499,12 @@ clean-all: clean clean-doc clean-gen clean: $(CLEANDIRS) @echo "Cleaning" $(Q)$(RM) -f -- $(progs) cscope.out *.o *.o.d \ - kernel-lib/*.o kernel-lib/*.o.d \ + kernel-lib/*.o kernel-lib/*.o.d kernel-lib/tables.c \ image/*.o image/*.o.d \ convert/*.o convert/*.o.d \ mkfs/*.o mkfs/*.o.d \ dir-test ioctl-test quick-test library-test library-test-static \ - btrfs.static mkfs.btrfs.static \ + mktables btrfs.static mkfs.btrfs.static \ $(check_defs) \ $(libs) $(lib_links) \ $(progs_static) $(progs_extra) diff --git a/kernel-lib/mktables.c b/kernel-lib/mktables.c new file mode 100644 index ..85f621fe --- /dev/null +++ b/kernel-lib/mktables.c @@ -0,0 +1,148 @@ +/* -*- linux-c -*- --- * + * + * Copyright 2002-2007 H. Peter Anvin - All Rights Reserved + * + * This file is part of the Linux kernel, and is made available under + * the terms of the GNU General Public License version 2 or (at your + * option) any later version; incorporated herein by reference. + * + * --- */ + +/* + * mktables.c + * + * Make RAID-6 tables. This is a host user space program to be run at + * compile time. + */ + +/* + * Btrfs-progs port, with following minor fixes: + * 1) Use "kerncompat.h" + * 2) Get rid of __KERNEL__ related macros + */ + +#include +#include +#include +#include +#include + +static uint8_t gfmul(uint8_t a, uint8_t b) +{ + uint8_t v = 0; + + while (b) { + if (b & 1) + v ^= a; + a = (a << 1) ^ (a & 0x80 ? 0x1d : 0); + b >>= 1; + } + + return v; +} + +static uint8_t gfpow(uint8_t a, int b) +{ + uint8_t v = 1; + + b %= 255; + if (b < 0) + b += 255; + + while (b) { + if (b & 1) + v = gfmul(v, a); + a = gfmul(a, a); + b >>= 1; + } + + return v; +} + +int main(int argc, char *argv[]) +{ + int i, j, k; + uint8_t v; + uint8_t exptbl[256], invtbl[256]; + + printf("#include \"kerncompat.h\"\n"); + + /* Compute multiplication table */ + printf("\nconst u8 __attribute__((aligned(256)))\n" + "raid6_gfmul[256][256] =\n" + "{\n"); + for (i = 0; i < 256; i++) { + printf("\t{\n"); + for (j = 0; j < 256; j += 8) { +
[PATCH v2 08/19] btrfs-progs: csum: Introduce function to read out one data csum
Introduce a new function: btrfs_read_one_data_csum(), to read out a csum for a sectorsize. This is quite useful for read out data csum so we don't need to do it using open code. Signed-off-by: Qu Wenruo--- Makefile.in | 2 +- csum.c | 96 + ctree.h | 3 ++ 3 files changed, 100 insertions(+), 1 deletion(-) create mode 100644 csum.c diff --git a/Makefile.in b/Makefile.in index c3f0eeda..bb619bfe 100644 --- a/Makefile.in +++ b/Makefile.in @@ -99,7 +99,7 @@ objects = ctree.o disk-io.o kernel-lib/radix-tree.o extent-tree.o print-tree.o \ qgroup.o free-space-cache.o kernel-lib/list_sort.o props.o \ ulist.o qgroup-verify.o backref.o string-table.o task-utils.o \ inode.o file.o find-root.o free-space-tree.o help.o send-dump.o \ - kernel-lib/tables.o kernel-lib/raid56.o + kernel-lib/tables.o kernel-lib/raid56.o csum.o cmds_objects = cmds-subvolume.o cmds-filesystem.o cmds-device.o cmds-scrub.o \ cmds-inspect.o cmds-balance.o cmds-send.o cmds-receive.o \ cmds-quota.o cmds-qgroup.o cmds-replace.o cmds-check.o \ diff --git a/csum.c b/csum.c new file mode 100644 index ..53195eaf --- /dev/null +++ b/csum.c @@ -0,0 +1,96 @@ +/* + * Copyright (C) 2016 Fujitsu. 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 "ctree.h" +#include "utils.h" +/* + * TODO: + * 1) Add write support for csum + *So we can write new data extents and add csum into csum tree + * 2) Add csum range search function + *So we don't need to search csum tree in a per-sectorsize loop. + */ + +int btrfs_read_one_data_csum(struct btrfs_fs_info *fs_info, u64 bytenr, +void *csum_ret) +{ + struct btrfs_path *path; + struct btrfs_key key; + struct btrfs_root *csum_root = fs_info->csum_root; + u32 item_offset; + u32 item_size; + u32 final_offset; + u32 sectorsize = fs_info->tree_root->sectorsize; + u16 csum_size = btrfs_super_csum_size(fs_info->super_copy); + int ret; + + if (!csum_ret) { + error("wrong parameter for %s", __func__); + return -EINVAL; + } + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + key.objectid = BTRFS_EXTENT_CSUM_OBJECTID; + key.type = BTRFS_EXTENT_CSUM_KEY; + key.offset = bytenr; + + ret = btrfs_search_slot(NULL, csum_root, , path, 0, 0); + if (ret < 0) + goto out; + if (ret == 0) { + btrfs_item_key_to_cpu(path->nodes[0], , path->slots[0]); + if (!IS_ALIGNED(key.offset, sectorsize)) { + error("csum item bytenr %llu is not aligned to %u", + key.offset, sectorsize); + ret = -EIO; + goto out; + } + u32 offset = btrfs_item_ptr_offset(path->nodes[0], + path->slots[0]); + + read_extent_buffer(path->nodes[0], csum_ret, offset, csum_size); + goto out; + } + ret = btrfs_previous_item(csum_root, path, BTRFS_EXTENT_CSUM_OBJECTID, + BTRFS_EXTENT_CSUM_KEY); + if (ret) + goto out; + btrfs_item_key_to_cpu(path->nodes[0], , path->slots[0]); + if (!IS_ALIGNED(key.offset, sectorsize)) { + error("csum item bytenr %llu is not aligned to %u", + key.offset, sectorsize); + ret = -EIO; + goto out; + } + item_offset = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]); + item_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]); + if (key.offset + item_size / csum_size * sectorsize <= bytenr) { + ret = 1; + goto out; + } + + final_offset = (bytenr - key.offset) / sectorsize * csum_size + + item_offset; + read_extent_buffer(path->nodes[0], csum_ret, final_offset, csum_size); + ret = 0; +out: + btrfs_free_path(path); + return ret; +}; diff --git a/ctree.h b/ctree.h index dd02ef86..506b107e 100644 --- a/ctree.h +++ b/ctree.h @@ -2798,4 +2798,7 @@ int
[PATCH v2 07/19] btrfs-progs: Allow __btrfs_map_block_v2 to remove unrelated stripes
For READ, caller normally hopes to get what they request, other than full stripe map. In this case, we should remove unrelated stripe map, just like the following case: 32K 96K |<-request range->| 0 64k 128K RAID0: |Data 1| Data 2| disk1 disk2 Before this patch, we return the full stripe: Stripe 0: Logical 0, Physical X, Len 64K, Dev disk1 Stripe 1: Logical 64k, Physical Y, Len 64K, Dev disk2 After this patch, we limit the stripe result to the request range: Stripe 0: Logical 32K, Physical X+32K, Len 32K, Dev disk1 Stripe 1: Logical 64k, Physical Y, Len 32K, Dev disk2 And if it's a RAID5/6 stripe, we just handle it like RAID0, ignoring parities. This should make caller easier to use. Signed-off-by: Qu Wenruo--- volumes.c | 103 +- 1 file changed, 102 insertions(+), 1 deletion(-) diff --git a/volumes.c b/volumes.c index 11d1f0e8..79b877c9 100644 --- a/volumes.c +++ b/volumes.c @@ -1733,6 +1733,107 @@ static int fill_full_map_block(struct map_lookup *map, u64 start, u64 length, return 0; } +static void del_one_stripe(struct btrfs_map_block *map_block, int i) +{ + int cur_nr = map_block->num_stripes; + int size_left = (cur_nr - 1 - i) * sizeof(struct btrfs_map_stripe); + + memmove(_block->stripes[i], _block->stripes[i + 1], size_left); + map_block->num_stripes--; +} + +static void remove_unrelated_stripes(struct map_lookup *map, +int rw, u64 start, u64 length, +struct btrfs_map_block *map_block) +{ + int i = 0; + /* +* RAID5/6 write must use full stripe. +* No need to do anything. +*/ + if (map->type & (BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6) && + rw == WRITE) + return; + + /* +* For RAID0/1/10/DUP, whatever read/write, we can remove unrelated +* stripes without causing anything wrong. +* RAID5/6 READ is just like RAID0, we don't care parity unless we need +* to recovery. +* For recovery, rw should be set to WRITE. +*/ + while (i < map_block->num_stripes) { + struct btrfs_map_stripe *stripe; + u64 orig_logical; /* Original stripe logical start */ + u64 orig_end; /* Original stripe logical end */ + + stripe = _block->stripes[i]; + + /* +* For READ, we don't really care parity +*/ + if (stripe->logical == BTRFS_RAID5_P_STRIPE || + stripe->logical == BTRFS_RAID6_Q_STRIPE) { + del_one_stripe(map_block, i); + continue; + } + /* Completely unrelated stripe */ + if (stripe->logical >= start + length || + stripe->logical + stripe->length <= start) { + del_one_stripe(map_block, i); + continue; + } + /* Covered stripe, modify its logical and physical */ + orig_logical = stripe->logical; + orig_end = stripe->logical + stripe->length; + if (start + length <= orig_end) { + /* +* |<--range-->| +* | stripe | +* Or +* || +* | stripe | +*/ + stripe->logical = max(orig_logical, start); + stripe->length = start + length; + stripe->physical += stripe->logical - orig_logical; + } else if (start >= orig_logical) { + /* +* |<-range--->| +* | stripe | +* Or +* || +* | stripe | +*/ + stripe->logical = start; + stripe->length = min(orig_end, start + length); + stripe->physical += stripe->logical - orig_logical; + } + /* +* Remaining case: +* | | +* | stripe | +* No need to do any modification +*/ + i++; + } + + /* Recaculate map_block size */ + map_block->start = 0; + map_block->length = 0; + for (i = 0; i < map_block->num_stripes; i++) { + struct btrfs_map_stripe *stripe; + + stripe = _block->stripes[i]; + if (stripe->logical > map_block->start) + map_block->start = stripe->logical; + if
[PATCH v2 15/19] btrfs-progs: extent-tree: Introduce function to check if there is any extent in given range.
Introduce a new function, btrfs_check_extent_exists(), to check if there is any extent in the range specified by user. The parameter can be a large range, and if any extent exists in the range, it will return >0 (in fact it will return 1). Or return 0 if no extent is found. Signed-off-by: Qu Wenruo--- ctree.h | 2 ++ extent-tree.c | 60 +++ 2 files changed, 62 insertions(+) diff --git a/ctree.h b/ctree.h index 506b107e..fe7c077e 100644 --- a/ctree.h +++ b/ctree.h @@ -2561,6 +2561,8 @@ int exclude_super_stripes(struct btrfs_root *root, u64 add_new_free_space(struct btrfs_block_group_cache *block_group, struct btrfs_fs_info *info, u64 start, u64 end); u64 hash_extent_data_ref(u64 root_objectid, u64 owner, u64 offset); +int btrfs_check_extent_exists(struct btrfs_fs_info *fs_info, u64 start, + u64 len); /* ctree.c */ int btrfs_comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2); diff --git a/extent-tree.c b/extent-tree.c index b2847ff9..92868395 100644 --- a/extent-tree.c +++ b/extent-tree.c @@ -4256,3 +4256,63 @@ u64 add_new_free_space(struct btrfs_block_group_cache *block_group, return total_added; } + +/* + * Check if there is any extent(both data and metadata) in the range + * [@start, @start + @len) + * + * Return 0 for no extent found. + * Return >0 for found extent. + * Return <0 for fatal error. + */ +int btrfs_check_extent_exists(struct btrfs_fs_info *fs_info, u64 start, + u64 len) +{ + struct btrfs_path *path; + struct btrfs_key key; + u64 extent_start; + u64 extent_len; + int ret; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + key.objectid = start + len; + key.type = 0; + key.offset = 0; + + ret = btrfs_search_slot(NULL, fs_info->extent_root, , path, 0, 0); + if (ret < 0) + goto out; + /* +* Now we're pointing at slot whose key.object >= end, skip to previous +* extent. +*/ + ret = btrfs_previous_extent_item(fs_info->extent_root, path, 0); + if (ret < 0) + goto out; + if (ret > 0) { + ret = 0; + goto out; + } + btrfs_item_key_to_cpu(path->nodes[0], , path->slots[0]); + extent_start = key.objectid; + if (key.type == BTRFS_METADATA_ITEM_KEY) + extent_len = fs_info->extent_root->nodesize; + else + extent_len = key.offset; + + /* +* search_slot() and previous_extent_item() has ensured that our +* extent_start < start + len, we only need to care extent end. +*/ + if (extent_start + extent_len <= start) + ret = 0; + else + ret = 1; + +out: + btrfs_free_path(path); + return ret; +} -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH v2 04/19] btrfs-progs: raid56: Allow raid6 to recover data and p
Copied from kernel lib/raid6/recov.c. Minor modifications includes: - Rename from raid6_datap_recov_intx() to raid5_recov_datap() - Rename parameter from faila to dest1 Signed-off-by: Qu Wenruo--- kernel-lib/raid56.c | 41 + kernel-lib/raid56.h | 2 ++ 2 files changed, 43 insertions(+) diff --git a/kernel-lib/raid56.c b/kernel-lib/raid56.c index dca8f8d4..e078972b 100644 --- a/kernel-lib/raid56.c +++ b/kernel-lib/raid56.c @@ -239,3 +239,44 @@ int raid6_recov_data2(int nr_devs, size_t stripe_len, int dest1, int dest2, free(zero_mem2); return ret; } + +/* + * Raid 6 recover code copied from kernel lib/raid6/recov.c + * - rename from raid6_datap_recov_intx1() + * - parameter changed from faila to dest1 + */ +int raid6_recov_datap(int nr_devs, size_t stripe_len, int dest1, void **data) +{ + u8 *p, *q, *dq; + const u8 *qmul; /* Q multiplier table */ + char *zero_mem; + + p = (u8 *)data[nr_devs - 2]; + q = (u8 *)data[nr_devs - 1]; + + zero_mem = calloc(1, stripe_len); + if (!zero_mem) + return -ENOMEM; + + /* Compute syndrome with zero for the missing data page + Use the dead data page as temporary storage for delta q */ + dq = (u8 *)data[dest1]; + data[dest1] = (void *)zero_mem; + data[nr_devs - 1] = dq; + + raid6_gen_syndrome(nr_devs, stripe_len, data); + + /* Restore pointer table */ + data[dest1] = dq; + data[nr_devs - 1] = q; + + /* Now, pick the proper data tables */ + qmul = raid6_gfmul[raid6_gfinv[raid6_gfexp[dest1]]]; + + /* Now do it... */ + while ( stripe_len-- ) { + *p++ ^= *dq = qmul[*q ^ *dq]; + q++; dq++; + } + return 0; +} diff --git a/kernel-lib/raid56.h b/kernel-lib/raid56.h index d397a23e..e088279b 100644 --- a/kernel-lib/raid56.h +++ b/kernel-lib/raid56.h @@ -42,4 +42,6 @@ extern const u8 raid6_gfexi[256] __attribute__((aligned(256))); /* Recover raid6 with 2 data corrupted */ int raid6_recov_data2(int nr_devs, size_t stripe_len, int dest1, int dest2, void **data); +/* Recover data and P */ +int raid6_recov_datap(int nr_devs, size_t stripe_len, int dest1, void **data); #endif -- 2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH] fstests: btrfs/132: Use better method to wait the writer to avoid EBUSY
The kill and wait method will only wait for the children process to exit, while the xfs_io can still run at background. This makes the test always fails on HDD backed physical machine. Use the "while ps aux | grep" method in btrfs/069 to truely wait the xfs_io to finish. Signed-off-by: Qu Wenruo--- tests/btrfs/132 | 7 --- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/tests/btrfs/132 b/tests/btrfs/132 index 0516177..db0dcb0 100755 --- a/tests/btrfs/132 +++ b/tests/btrfs/132 @@ -94,9 +94,10 @@ sleep $sleep_time kill $pids wait -# Sync the fs to avoid EBUSY while umount, which is quite common for btrfs -# compression -sync +# Wait all writers really exits +while ps aux | grep "$SCRATCH_MNT" | grep -qv grep; do + sleep 1 +done echo "Silence is golden" status=0 -- 2.7.4 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [CORRUPTION FILESYSTEM] Corrupted and unrecoverable file system during the snapshot receive
Xin Zhou posted on Mon, 26 Dec 2016 03:36:09 +0100 as excerpted: > One interesting thing to investigate might be the btrfs send / receive > result, under a disruptive network environment. If the connection breaks > in the middle of transfer (at different phase, maybe), see what could be > the file system status. Btrfs send, sends from a read-only snapshot, so the sending filesystem shouldn't be harmed no matter what happens to send. Btrfs receive does all its work in a new subvolume (basically a snapshot in an incremental send, tho I'm not sure it's a full snapshot in the technical sense), modifying the files therein using standard calls used in other contexts as well, so absent bugs that should appear in those other contexts too if they exist, the worst damage that a receive should be able to do is an unfaithful replay of the send stream, such that an appropriate copy of the sent snapshot doesn't appear on the receiver. Which means even in the case of error, cleanup is as simple as deleting the aborted/incompletely-received subvolume. Unless there are bugs that would show up in other situations as well (or an out-of-space condition is triggered that would likewise show up in other situations with a similar amount of data/metadata written), there should be no effects outside that received subvolume. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [CORRUPTION FILESYSTEM] Corrupted and unrecoverable file system during the snapshot receive
Hi, For free software with open source code, that is quite good. Most commercial product has a very robust error handling in transport, to guarantee no corruption due to transfer issues. One interesting thing to investigate might be the btrfs send / receive result, under a disruptive network environment. If the connection breaks in the middle of transfer (at different phase, maybe), see what could be the file system status. Thanks, Xin Sent: Sunday, December 25, 2016 at 2:57 PM From: Duncan <1i5t5.dun...@cox.net> To: linux-btrfs@vger.kernel.org Subject: Re: [CORRUPTION FILESYSTEM] Corrupted and unrecoverable file system during the snapshot receive Xin Zhou posted on Sat, 24 Dec 2016 21:15:40 +0100 as excerpted: > The code is relatively new to me, I did not see retry logic in stream > handling, please correct me if I am wrong about this. > So, I am not quite sure about the transfer behavior, if the system > subject to network issues in heavy workload, > in which packets missing or connect issues are not rare. As you likely know I'm neither a dev, just a list regular and btrfs user myself, and I'm not particularly familiar with send/receive as I don't use it myself, but... AFAIK, the send and receive sides are specifically designed to be separate and to work with STDOUT/STDIN, so it's possible with STDOUT redirection to "send" to a local file instead of directly to receive, and then to replay that file on the other end by cat-ing it to receive. As such, transfer behavior isn't really a factor at the btrfs layer, since handling problems in the transfer layer is the responsibility of whatever method the user is using to do that transfer, and the user is presumed to use a transfer method with whatever reliability guarantees they deem necessary. So network behavior isn't really a factor at the btrfs level as that's the transfer layer and btrfs isn't worrying about that, simply assuming it to have the necessary reliability. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH] fstests: btrfs: Test scrub and replace race for RAID56
At 12/24/2016 05:45 PM, Eryu Guan wrote: On Thu, Dec 22, 2016 at 10:02:51AM +0800, Qu Wenruo wrote: Although by design, btrfs scrub and replace share the same code path, so they are exclusive to each other. But the fact is, there is still some critical region not protected well, so we can have the following kernel panic, especially easy to trigger on RAID5/6 profiles. Could btrfs/069 reproduce the panic? It also races scrub and replace, but with fsstress running in background, raid5/6 profiles are part of the default test configs. Yes, but the possibility is very low and depends on the test machine. I never reproduced it with btrfs/069 on all my VMs, but can produce it on my physical machines. Nor btrfs/069 supports TIME_FACTOR to improve the possibility. Also, is there a known fix available in Linus tree or btrfs tree? If not, I'd push this new test after there's a known fix (if it's worth a new test). Not yet, still WIP. I'll CC you as the fix is submitted. Thanks, Qu Thanks, Eryu -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [CORRUPTION FILESYSTEM] Corrupted and unrecoverable file system during the snapshot receive
Xin Zhou posted on Sat, 24 Dec 2016 21:15:40 +0100 as excerpted: > The code is relatively new to me, I did not see retry logic in stream > handling, please correct me if I am wrong about this. > So, I am not quite sure about the transfer behavior, if the system > subject to network issues in heavy workload, > in which packets missing or connect issues are not rare. As you likely know I'm neither a dev, just a list regular and btrfs user myself, and I'm not particularly familiar with send/receive as I don't use it myself, but... AFAIK, the send and receive sides are specifically designed to be separate and to work with STDOUT/STDIN, so it's possible with STDOUT redirection to "send" to a local file instead of directly to receive, and then to replay that file on the other end by cat-ing it to receive. As such, transfer behavior isn't really a factor at the btrfs layer, since handling problems in the transfer layer is the responsibility of whatever method the user is using to do that transfer, and the user is presumed to use a transfer method with whatever reliability guarantees they deem necessary. So network behavior isn't really a factor at the btrfs level as that's the transfer layer and btrfs isn't worrying about that, simply assuming it to have the necessary reliability. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
some free space cache corruptions
Hey. Had the following on a Debian sid: Linux heisenberg 4.8.0-2-amd64 #1 SMP Debian 4.8.11-1 (2016-12-02) x86_64 GNU/Linux btrfs-progs v4.7.3 I was doing a btrfs check of a rather big btrfs (8TB device, nearly full), having many snapshots on it, all incrementally send from another 8TB device, which in turn functions as the master copy: # btrfs check /dev/mapper/data-a2 ; echo $? Checking filesystem on /dev/mapper/data-a2 UUID: f8acb432-7604-46ba-b3ad-0abe8e92c4db checking extents checking free space cache cache and super generation don't match, space cache will be invalidated checking fs roots checking csums checking root refs found 6805741969408 bytes used err is 0 total csum bytes: 6634558200 total tree bytes: 10292641792 total fs tree bytes: 2074869760 total extent tree bytes: 1100251136 btree space waste bytes: 885346193 file data blocks allocated: 6922343247872 referenced 7040929374208 0 => this already showed an unusual: cache and super generation don't match, space cache will be invalidated Where does it come from? Then I did some incremental send/receive (-p) from the other master 8TB master btrfs and another fsck afters wards: # btrfs check /dev/mapper/data-a2 ; echo $? Checking filesystem on /dev/mapper/data-a2 UUID: f8acb432-7604-46ba-b3ad-0abe8e92c4db checking extents checking free space cache checking fs roots checking csums checking root refs found 7467006156800 bytes used err is 0 total csum bytes: 7279407560 total tree bytes: 11069603840 total fs tree bytes: 2127314944 total extent tree bytes: 1141342208 btree space waste bytes: 922662895 file data blocks allocated: 7599280926720 referenced 7720960733184 0 => all fine... Afterwards I removed all ro-snapshots except the most recent one... and repeated the fsck: # btrfs check /dev/mapper/data-a2 ; echo $? Checking filesystem on /dev/mapper/data-a2 UUID: f8acb432-7604-46ba-b3ad-0abe8e92c4db checking extents checking free space cache block group 5431552376832 has wrong amount of free space failed to load free space cache for block group 5431552376832 checking fs roots checking csums checking root refs found 7427361222656 bytes used err is 0 total csum bytes: 7240763996 total tree bytes: 10998038528 total fs tree bytes: 2100297728 total extent tree bytes: 1137065984 btree space waste bytes: 992708933 file data blocks allocated: 7416363184128 referenced 7536754290688 0 => Isn't that some indication of a bug already? Nothing happened, just deletion of snapshots and there is apparently some free space cache corruption? Then I tried the usual recipe: mount /data/data-a/2/ -o clear_cache kernel said: Dec 25 22:14:17 heisenberg kernel: BTRFS info (device dm-2): force clearing of disk cache ...re-mounted,rw, deleted some regular files... repeated the fsck and again: # btrfs check /dev/mapper/data-a2 ; echo $? Checking filesystem on /dev/mapper/data-a2 UUID: f8acb432-7604-46ba-b3ad-0abe8e92c4db checking extents checking free space cache block group 5431552376832 has wrong amount of free space failed to load free space cache for block group 5431552376832 checking fs roots checking csums checking root refs found 7427284213760 bytes used err is 0 total csum bytes: 7240689688 total tree bytes: 10997907456 total fs tree bytes: 2100281344 total extent tree bytes: 1137049600 btree space waste bytes: 992679805 file data blocks allocated: 7416286306304 referenced 7536677412864 0 => same error again... Any ideas how to resolve? And is this some serious error that could have caused corruptions? Cheers, Chris. smime.p7s Description: S/MIME cryptographic signature