diff options
author | Heiko Schocher <hs@denx.de> | 2014-06-24 12:10:04 +0400 |
---|---|---|
committer | Tom Rini <trini@ti.com> | 2014-08-26 03:25:55 +0400 |
commit | ff94bc40af3481d47546595ba73c136de6af6929 (patch) | |
tree | 858119077e2ca8a992589185c36bd431e4a8c80e /drivers/mtd/ubi/wl.c | |
parent | 0c06db59836746c5caf397e642cd0f2bf1cc20a6 (diff) | |
download | u-boot-ff94bc40af3481d47546595ba73c136de6af6929.tar.xz |
mtd, ubi, ubifs: resync with Linux-3.14
resync ubi subsystem with linux:
commit 455c6fdbd219161bd09b1165f11699d6d73de11c
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date: Sun Mar 30 20:40:15 2014 -0700
Linux 3.14
A nice side effect of this, is we introduce UBI Fastmap support
to U-Boot.
Signed-off-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Sergey Lapin <slapin@ossfans.org>
Cc: Scott Wood <scottwood@freescale.com>
Cc: Joerg Krause <jkrause@posteo.de>
Diffstat (limited to 'drivers/mtd/ubi/wl.c')
-rw-r--r-- | drivers/mtd/ubi/wl.c | 1566 |
1 files changed, 1021 insertions, 545 deletions
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 1eaa88b36f..2987ffc093 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -7,97 +7,117 @@ */ /* - * UBI wear-leveling unit. + * UBI wear-leveling sub-system. * - * This unit is responsible for wear-leveling. It works in terms of physical - * eraseblocks and erase counters and knows nothing about logical eraseblocks, - * volumes, etc. From this unit's perspective all physical eraseblocks are of - * two types - used and free. Used physical eraseblocks are those that were - * "get" by the 'ubi_wl_get_peb()' function, and free physical eraseblocks are - * those that were put by the 'ubi_wl_put_peb()' function. + * This sub-system is responsible for wear-leveling. It works in terms of + * physical eraseblocks and erase counters and knows nothing about logical + * eraseblocks, volumes, etc. From this sub-system's perspective all physical + * eraseblocks are of two types - used and free. Used physical eraseblocks are + * those that were "get" by the 'ubi_wl_get_peb()' function, and free physical + * eraseblocks are those that were put by the 'ubi_wl_put_peb()' function. * * Physical eraseblocks returned by 'ubi_wl_get_peb()' have only erase counter - * header. The rest of the physical eraseblock contains only 0xFF bytes. + * header. The rest of the physical eraseblock contains only %0xFF bytes. * - * When physical eraseblocks are returned to the WL unit by means of the + * When physical eraseblocks are returned to the WL sub-system by means of the * 'ubi_wl_put_peb()' function, they are scheduled for erasure. The erasure is * done asynchronously in context of the per-UBI device background thread, - * which is also managed by the WL unit. + * which is also managed by the WL sub-system. * * The wear-leveling is ensured by means of moving the contents of used * physical eraseblocks with low erase counter to free physical eraseblocks * with high erase counter. * - * The 'ubi_wl_get_peb()' function accepts data type hints which help to pick - * an "optimal" physical eraseblock. For example, when it is known that the - * physical eraseblock will be "put" soon because it contains short-term data, - * the WL unit may pick a free physical eraseblock with low erase counter, and - * so forth. + * If the WL sub-system fails to erase a physical eraseblock, it marks it as + * bad. * - * If the WL unit fails to erase a physical eraseblock, it marks it as bad. + * This sub-system is also responsible for scrubbing. If a bit-flip is detected + * in a physical eraseblock, it has to be moved. Technically this is the same + * as moving it for wear-leveling reasons. * - * This unit is also responsible for scrubbing. If a bit-flip is detected in a - * physical eraseblock, it has to be moved. Technically this is the same as - * moving it for wear-leveling reasons. + * As it was said, for the UBI sub-system all physical eraseblocks are either + * "free" or "used". Free eraseblock are kept in the @wl->free RB-tree, while + * used eraseblocks are kept in @wl->used, @wl->erroneous, or @wl->scrub + * RB-trees, as well as (temporarily) in the @wl->pq queue. * - * As it was said, for the UBI unit all physical eraseblocks are either "free" - * or "used". Free eraseblock are kept in the @wl->free RB-tree, while used - * eraseblocks are kept in a set of different RB-trees: @wl->used, - * @wl->prot.pnum, @wl->prot.aec, and @wl->scrub. + * When the WL sub-system returns a physical eraseblock, the physical + * eraseblock is protected from being moved for some "time". For this reason, + * the physical eraseblock is not directly moved from the @wl->free tree to the + * @wl->used tree. There is a protection queue in between where this + * physical eraseblock is temporarily stored (@wl->pq). + * + * All this protection stuff is needed because: + * o we don't want to move physical eraseblocks just after we have given them + * to the user; instead, we first want to let users fill them up with data; + * + * o there is a chance that the user will put the physical eraseblock very + * soon, so it makes sense not to move it for some time, but wait. + * + * Physical eraseblocks stay protected only for limited time. But the "time" is + * measured in erase cycles in this case. This is implemented with help of the + * protection queue. Eraseblocks are put to the tail of this queue when they + * are returned by the 'ubi_wl_get_peb()', and eraseblocks are removed from the + * head of the queue on each erase operation (for any eraseblock). So the + * length of the queue defines how may (global) erase cycles PEBs are protected. + * + * To put it differently, each physical eraseblock has 2 main states: free and + * used. The former state corresponds to the @wl->free tree. The latter state + * is split up on several sub-states: + * o the WL movement is allowed (@wl->used tree); + * o the WL movement is disallowed (@wl->erroneous) because the PEB is + * erroneous - e.g., there was a read error; + * o the WL movement is temporarily prohibited (@wl->pq queue); + * o scrubbing is needed (@wl->scrub tree). + * + * Depending on the sub-state, wear-leveling entries of the used physical + * eraseblocks may be kept in one of those structures. * * Note, in this implementation, we keep a small in-RAM object for each physical * eraseblock. This is surely not a scalable solution. But it appears to be good * enough for moderately large flashes and it is simple. In future, one may - * re-work this unit and make it more scalable. + * re-work this sub-system and make it more scalable. * - * At the moment this unit does not utilize the sequence number, which was - * introduced relatively recently. But it would be wise to do this because the - * sequence number of a logical eraseblock characterizes how old is it. For + * At the moment this sub-system does not utilize the sequence number, which + * was introduced relatively recently. But it would be wise to do this because + * the sequence number of a logical eraseblock characterizes how old is it. For * example, when we move a PEB with low erase counter, and we need to pick the * target PEB, we pick a PEB with the highest EC if our PEB is "old" and we * pick target PEB with an average EC if our PEB is not very "old". This is a - * room for future re-works of the WL unit. - * - * FIXME: looks too complex, should be simplified (later). + * room for future re-works of the WL sub-system. */ -#ifdef UBI_LINUX +#define __UBOOT__ +#ifndef __UBOOT__ #include <linux/slab.h> #include <linux/crc32.h> #include <linux/freezer.h> #include <linux/kthread.h> +#else +#include <ubi_uboot.h> #endif -#include <ubi_uboot.h> #include "ubi.h" /* Number of physical eraseblocks reserved for wear-leveling purposes */ #define WL_RESERVED_PEBS 1 /* - * How many erase cycles are short term, unknown, and long term physical - * eraseblocks protected. - */ -#define ST_PROTECTION 16 -#define U_PROTECTION 10 -#define LT_PROTECTION 4 - -/* * Maximum difference between two erase counters. If this threshold is - * exceeded, the WL unit starts moving data from used physical eraseblocks with - * low erase counter to free physical eraseblocks with high erase counter. + * exceeded, the WL sub-system starts moving data from used physical + * eraseblocks with low erase counter to free physical eraseblocks with high + * erase counter. */ #define UBI_WL_THRESHOLD CONFIG_MTD_UBI_WL_THRESHOLD /* - * When a physical eraseblock is moved, the WL unit has to pick the target + * When a physical eraseblock is moved, the WL sub-system has to pick the target * physical eraseblock to move to. The simplest way would be just to pick the * one with the highest erase counter. But in certain workloads this could lead * to an unlimited wear of one or few physical eraseblock. Indeed, imagine a * situation when the picked physical eraseblock is constantly erased after the * data is written to it. So, we have a constant which limits the highest erase - * counter of the free physical eraseblock to pick. Namely, the WL unit does - * not pick eraseblocks with erase counter greater then the lowest erase + * counter of the free physical eraseblock to pick. Namely, the WL sub-system + * does not pick eraseblocks with erase counter greater than the lowest erase * counter plus %WL_FREE_MAX_DIFF. */ #define WL_FREE_MAX_DIFF (2*UBI_WL_THRESHOLD) @@ -108,89 +128,48 @@ */ #define WL_MAX_FAILURES 32 +static int self_check_ec(struct ubi_device *ubi, int pnum, int ec); +static int self_check_in_wl_tree(const struct ubi_device *ubi, + struct ubi_wl_entry *e, struct rb_root *root); +static int self_check_in_pq(const struct ubi_device *ubi, + struct ubi_wl_entry *e); + +#ifdef CONFIG_MTD_UBI_FASTMAP +#ifndef __UBOOT__ /** - * struct ubi_wl_prot_entry - PEB protection entry. - * @rb_pnum: link in the @wl->prot.pnum RB-tree - * @rb_aec: link in the @wl->prot.aec RB-tree - * @abs_ec: the absolute erase counter value when the protection ends - * @e: the wear-leveling entry of the physical eraseblock under protection - * - * When the WL unit returns a physical eraseblock, the physical eraseblock is - * protected from being moved for some "time". For this reason, the physical - * eraseblock is not directly moved from the @wl->free tree to the @wl->used - * tree. There is one more tree in between where this physical eraseblock is - * temporarily stored (@wl->prot). - * - * All this protection stuff is needed because: - * o we don't want to move physical eraseblocks just after we have given them - * to the user; instead, we first want to let users fill them up with data; - * - * o there is a chance that the user will put the physical eraseblock very - * soon, so it makes sense not to move it for some time, but wait; this is - * especially important in case of "short term" physical eraseblocks. - * - * Physical eraseblocks stay protected only for limited time. But the "time" is - * measured in erase cycles in this case. This is implemented with help of the - * absolute erase counter (@wl->abs_ec). When it reaches certain value, the - * physical eraseblocks are moved from the protection trees (@wl->prot.*) to - * the @wl->used tree. - * - * Protected physical eraseblocks are searched by physical eraseblock number - * (when they are put) and by the absolute erase counter (to check if it is - * time to move them to the @wl->used tree). So there are actually 2 RB-trees - * storing the protected physical eraseblocks: @wl->prot.pnum and - * @wl->prot.aec. They are referred to as the "protection" trees. The - * first one is indexed by the physical eraseblock number. The second one is - * indexed by the absolute erase counter. Both trees store - * &struct ubi_wl_prot_entry objects. - * - * Each physical eraseblock has 2 main states: free and used. The former state - * corresponds to the @wl->free tree. The latter state is split up on several - * sub-states: - * o the WL movement is allowed (@wl->used tree); - * o the WL movement is temporarily prohibited (@wl->prot.pnum and - * @wl->prot.aec trees); - * o scrubbing is needed (@wl->scrub tree). - * - * Depending on the sub-state, wear-leveling entries of the used physical - * eraseblocks may be kept in one of those trees. + * update_fastmap_work_fn - calls ubi_update_fastmap from a work queue + * @wrk: the work description object */ -struct ubi_wl_prot_entry { - struct rb_node rb_pnum; - struct rb_node rb_aec; - unsigned long long abs_ec; - struct ubi_wl_entry *e; -}; +static void update_fastmap_work_fn(struct work_struct *wrk) +{ + struct ubi_device *ubi = container_of(wrk, struct ubi_device, fm_work); + ubi_update_fastmap(ubi); +} +#endif /** - * struct ubi_work - UBI work description data structure. - * @list: a link in the list of pending works - * @func: worker function - * @priv: private data of the worker function - * - * @e: physical eraseblock to erase - * @torture: if the physical eraseblock has to be tortured - * - * The @func pointer points to the worker function. If the @cancel argument is - * not zero, the worker has to free the resources and exit immediately. The - * worker has to return zero in case of success and a negative error code in - * case of failure. + * ubi_ubi_is_fm_block - returns 1 if a PEB is currently used in a fastmap. + * @ubi: UBI device description object + * @pnum: the to be checked PEB */ -struct ubi_work { - struct list_head list; - int (*func)(struct ubi_device *ubi, struct ubi_work *wrk, int cancel); - /* The below fields are only relevant to erasure works */ - struct ubi_wl_entry *e; - int torture; -}; +static int ubi_is_fm_block(struct ubi_device *ubi, int pnum) +{ + int i; + + if (!ubi->fm) + return 0; + + for (i = 0; i < ubi->fm->used_blocks; i++) + if (ubi->fm->e[i]->pnum == pnum) + return 1; -#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID -static int paranoid_check_ec(struct ubi_device *ubi, int pnum, int ec); -static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e, - struct rb_root *root); + return 0; +} #else -#define paranoid_check_ec(ubi, pnum, ec) 0 -#define paranoid_check_in_wl_tree(e, root) +static int ubi_is_fm_block(struct ubi_device *ubi, int pnum) +{ + return 0; +} #endif /** @@ -210,7 +189,7 @@ static void wl_tree_add(struct ubi_wl_entry *e, struct rb_root *root) struct ubi_wl_entry *e1; parent = *p; - e1 = rb_entry(parent, struct ubi_wl_entry, rb); + e1 = rb_entry(parent, struct ubi_wl_entry, u.rb); if (e->ec < e1->ec) p = &(*p)->rb_left; @@ -225,8 +204,8 @@ static void wl_tree_add(struct ubi_wl_entry *e, struct rb_root *root) } } - rb_link_node(&e->rb, parent, p); - rb_insert_color(&e->rb, root); + rb_link_node(&e->u.rb, parent, p); + rb_insert_color(&e->u.rb, root); } /** @@ -289,18 +268,16 @@ static int produce_free_peb(struct ubi_device *ubi) { int err; - spin_lock(&ubi->wl_lock); while (!ubi->free.rb_node) { spin_unlock(&ubi->wl_lock); dbg_wl("do one work synchronously"); err = do_work(ubi); - if (err) - return err; spin_lock(&ubi->wl_lock); + if (err) + return err; } - spin_unlock(&ubi->wl_lock); return 0; } @@ -321,7 +298,7 @@ static int in_wl_tree(struct ubi_wl_entry *e, struct rb_root *root) while (p) { struct ubi_wl_entry *e1; - e1 = rb_entry(p, struct ubi_wl_entry, rb); + e1 = rb_entry(p, struct ubi_wl_entry, u.rb); if (e->pnum == e1->pnum) { ubi_assert(e == e1); @@ -345,223 +322,401 @@ static int in_wl_tree(struct ubi_wl_entry *e, struct rb_root *root) } /** - * prot_tree_add - add physical eraseblock to protection trees. + * prot_queue_add - add physical eraseblock to the protection queue. * @ubi: UBI device description object * @e: the physical eraseblock to add - * @pe: protection entry object to use - * @abs_ec: absolute erase counter value when this physical eraseblock has - * to be removed from the protection trees. * - * @wl->lock has to be locked. + * This function adds @e to the tail of the protection queue @ubi->pq, where + * @e will stay for %UBI_PROT_QUEUE_LEN erase operations and will be + * temporarily protected from the wear-leveling worker. Note, @wl->lock has to + * be locked. */ -static void prot_tree_add(struct ubi_device *ubi, struct ubi_wl_entry *e, - struct ubi_wl_prot_entry *pe, int abs_ec) +static void prot_queue_add(struct ubi_device *ubi, struct ubi_wl_entry *e) { - struct rb_node **p, *parent = NULL; - struct ubi_wl_prot_entry *pe1; - - pe->e = e; - pe->abs_ec = ubi->abs_ec + abs_ec; - - p = &ubi->prot.pnum.rb_node; - while (*p) { - parent = *p; - pe1 = rb_entry(parent, struct ubi_wl_prot_entry, rb_pnum); + int pq_tail = ubi->pq_head - 1; - if (e->pnum < pe1->e->pnum) - p = &(*p)->rb_left; - else - p = &(*p)->rb_right; - } - rb_link_node(&pe->rb_pnum, parent, p); - rb_insert_color(&pe->rb_pnum, &ubi->prot.pnum); - - p = &ubi->prot.aec.rb_node; - parent = NULL; - while (*p) { - parent = *p; - pe1 = rb_entry(parent, struct ubi_wl_prot_entry, rb_aec); - - if (pe->abs_ec < pe1->abs_ec) - p = &(*p)->rb_left; - else - p = &(*p)->rb_right; - } - rb_link_node(&pe->rb_aec, parent, p); - rb_insert_color(&pe->rb_aec, &ubi->prot.aec); + if (pq_tail < 0) + pq_tail = UBI_PROT_QUEUE_LEN - 1; + ubi_assert(pq_tail >= 0 && pq_tail < UBI_PROT_QUEUE_LEN); + list_add_tail(&e->u.list, &ubi->pq[pq_tail]); + dbg_wl("added PEB %d EC %d to the protection queue", e->pnum, e->ec); } /** * find_wl_entry - find wear-leveling entry closest to certain erase counter. + * @ubi: UBI device description object * @root: the RB-tree where to look for - * @max: highest possible erase counter + * @diff: maximum possible difference from the smallest erase counter * * This function looks for a wear leveling entry with erase counter closest to - * @max and less then @max. + * min + @diff, where min is the smallest erase counter. */ -static struct ubi_wl_entry *find_wl_entry(struct rb_root *root, int max) +static struct ubi_wl_entry *find_wl_entry(struct ubi_device *ubi, + struct rb_root *root, int diff) { struct rb_node *p; - struct ubi_wl_entry *e; + struct ubi_wl_entry *e, *prev_e = NULL; + int max; - e = rb_entry(rb_first(root), struct ubi_wl_entry, rb); - max += e->ec; + e = rb_entry(rb_first(root), struct ubi_wl_entry, u.rb); + max = e->ec + diff; p = root->rb_node; while (p) { struct ubi_wl_entry *e1; - e1 = rb_entry(p, struct ubi_wl_entry, rb); + e1 = rb_entry(p, struct ubi_wl_entry, u.rb); if (e1->ec >= max) p = p->rb_left; else { p = p->rb_right; + prev_e = e; e = e1; } } + /* If no fastmap has been written and this WL entry can be used + * as anchor PEB, hold it back and return the second best WL entry + * such that fastmap can use the anchor PEB later. */ + if (prev_e && !ubi->fm_disabled && + !ubi->fm && e->pnum < UBI_FM_MAX_START) + return prev_e; + return e; } /** - * ubi_wl_get_peb - get a physical eraseblock. + * find_mean_wl_entry - find wear-leveling entry with medium erase counter. * @ubi: UBI device description object - * @dtype: type of data which will be stored in this physical eraseblock + * @root: the RB-tree where to look for * - * This function returns a physical eraseblock in case of success and a - * negative error code in case of failure. Might sleep. + * This function looks for a wear leveling entry with medium erase counter, + * but not greater or equivalent than the lowest erase counter plus + * %WL_FREE_MAX_DIFF/2. */ -int ubi_wl_get_peb(struct ubi_device *ubi, int dtype) +static struct ubi_wl_entry *find_mean_wl_entry(struct ubi_device *ubi, + struct rb_root *root) { - int err, protect, medium_ec; struct ubi_wl_entry *e, *first, *last; - struct ubi_wl_prot_entry *pe; - ubi_assert(dtype == UBI_LONGTERM || dtype == UBI_SHORTTERM || - dtype == UBI_UNKNOWN); + first = rb_entry(rb_first(root), struct ubi_wl_entry, u.rb); + last = rb_entry(rb_last(root), struct ubi_wl_entry, u.rb); - pe = kmalloc(sizeof(struct ubi_wl_prot_entry), GFP_NOFS); - if (!pe) - return -ENOMEM; + if (last->ec - first->ec < WL_FREE_MAX_DIFF) { + e = rb_entry(root->rb_node, struct ubi_wl_entry, u.rb); + +#ifdef CONFIG_MTD_UBI_FASTMAP + /* If no fastmap has been written and this WL entry can be used + * as anchor PEB, hold it back and return the second best + * WL entry such that fastmap can use the anchor PEB later. */ + if (e && !ubi->fm_disabled && !ubi->fm && + e->pnum < UBI_FM_MAX_START) + e = rb_entry(rb_next(root->rb_node), + struct ubi_wl_entry, u.rb); +#endif + } else + e = find_wl_entry(ubi, root, WL_FREE_MAX_DIFF/2); + + return e; +} + +#ifdef CONFIG_MTD_UBI_FASTMAP +/** + * find_anchor_wl_entry - find wear-leveling entry to used as anchor PEB. + * @root: the RB-tree where to look for + */ +static struct ubi_wl_entry *find_anchor_wl_entry(struct rb_root *root) +{ + struct rb_node *p; + struct ubi_wl_entry *e, *victim = NULL; + int max_ec = UBI_MAX_ERASECOUNTER; + + ubi_rb_for_each_entry(p, e, root, u.rb) { + if (e->pnum < UBI_FM_MAX_START && e->ec < max_ec) { + victim = e; + max_ec = e->ec; + } + } + + return victim; +} + +static int anchor_pebs_avalible(struct rb_root *root) +{ + struct rb_node *p; + struct ubi_wl_entry *e; + + ubi_rb_for_each_entry(p, e, root, u.rb) + if (e->pnum < UBI_FM_MAX_START) + return 1; + + return 0; +} + +/** + * ubi_wl_get_fm_peb - find a physical erase block with a given maximal number. + * @ubi: UBI device description object + * @anchor: This PEB will be used as anchor PEB by fastmap + * + * The function returns a physical erase block with a given maximal number + * and removes it from the wl subsystem. + * Must be called with wl_lock held! + */ +struct ubi_wl_entry *ubi_wl_get_fm_peb(struct ubi_device *ubi, int anchor) +{ + struct ubi_wl_entry *e = NULL; + + if (!ubi->free.rb_node || (ubi->free_count - ubi->beb_rsvd_pebs < 1)) + goto out; + + if (anchor) + e = find_anchor_wl_entry(&ubi->free); + else + e = find_mean_wl_entry(ubi, &ubi->free); + + if (!e) + goto out; + + self_check_in_wl_tree(ubi, e, &ubi->free); + + /* remove it from the free list, + * the wl subsystem does no longer know this erase block */ + rb_erase(&e->u.rb, &ubi->free); + ubi->free_count--; +out: + return e; +} +#endif + +/** + * __wl_get_peb - get a physical eraseblock. + * @ubi: UBI device description object + * + * This function returns a physical eraseblock in case of success and a + * negative error code in case of failure. + */ +static int __wl_get_peb(struct ubi_device *ubi) +{ + int err; + struct ubi_wl_entry *e; retry: - spin_lock(&ubi->wl_lock); if (!ubi->free.rb_node) { if (ubi->works_count == 0) { - ubi_assert(list_empty(&ubi->works)); ubi_err("no free eraseblocks"); - spin_unlock(&ubi->wl_lock); - kfree(pe); + ubi_assert(list_empty(&ubi->works)); return -ENOSPC; } - spin_unlock(&ubi->wl_lock); err = produce_free_peb(ubi); - if (err < 0) { - kfree(pe); + if (err < 0) return err; - } goto retry; } - switch (dtype) { - case UBI_LONGTERM: - /* - * For long term data we pick a physical eraseblock - * with high erase counter. But the highest erase - * counter we can pick is bounded by the the lowest - * erase counter plus %WL_FREE_MAX_DIFF. - */ - e = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF); - protect = LT_PROTECTION; - break; - case UBI_UNKNOWN: - /* - * For unknown data we pick a physical eraseblock with - * medium erase counter. But we by no means can pick a - * physical eraseblock with erase counter greater or - * equivalent than the lowest erase counter plus - * %WL_FREE_MAX_DIFF. - */ - first = rb_entry(rb_first(&ubi->free), - struct ubi_wl_entry, rb); - last = rb_entry(rb_last(&ubi->free), - struct ubi_wl_entry, rb); - - if (last->ec - first->ec < WL_FREE_MAX_DIFF) - e = rb_entry(ubi->free.rb_node, - struct ubi_wl_entry, rb); - else { - medium_ec = (first->ec + WL_FREE_MAX_DIFF)/2; - e = find_wl_entry(&ubi->free, medium_ec); - } - protect = U_PROTECTION; - break; - case UBI_SHORTTERM: - /* - * For short term data we pick a physical eraseblock - * with the lowest erase counter as we expect it will - * be erased soon. - */ - e = rb_entry(rb_first(&ubi->free), - struct ubi_wl_entry, rb); - protect = ST_PROTECTION; - break; - default: - protect = 0; - e = NULL; - BUG(); + e = find_mean_wl_entry(ubi, &ubi->free); + if (!e) { + ubi_err("no free eraseblocks"); + return -ENOSPC; } + self_check_in_wl_tree(ubi, e, &ubi->free); + /* - * Move the physical eraseblock to the protection trees where it will + * Move the physical eraseblock to the protection queue where it will * be protected from being moved for some time. */ - paranoid_check_in_wl_tree(e, &ubi->free); - rb_erase(&e->rb, &ubi->free); - prot_tree_add(ubi, e, pe, protect); + rb_erase(&e->u.rb, &ubi->free); + ubi->free_count--; + dbg_wl("PEB %d EC %d", e->pnum, e->ec); +#ifndef CONFIG_MTD_UBI_FASTMAP + /* We have to enqueue e only if fastmap is disabled, + * is fastmap enabled prot_queue_add() will be called by + * ubi_wl_get_peb() after removing e from the pool. */ + prot_queue_add(ubi, e); +#endif + return e->pnum; +} - dbg_wl("PEB %d EC %d, protection %d", e->pnum, e->ec, protect); - spin_unlock(&ubi->wl_lock); +#ifdef CONFIG_MTD_UBI_FASTMAP +/** + * return_unused_pool_pebs - returns unused PEB to the free tree. + * @ubi: UBI device description object + * @pool: fastmap pool description object + */ +static void return_unused_pool_pebs(struct ubi_device *ubi, + struct ubi_fm_pool *pool) +{ + int i; + struct ubi_wl_entry *e; - return e->pnum; + for (i = pool->used; i < pool->size; i++) { + e = ubi->lookuptbl[pool->pebs[i]]; + wl_tree_add(e, &ubi->free); + ubi->free_count++; + } } /** - * prot_tree_del - remove a physical eraseblock from the protection trees + * refill_wl_pool - refills all the fastmap pool used by the + * WL sub-system. * @ubi: UBI device description object - * @pnum: the physical eraseblock to remove + */ +static void refill_wl_pool(struct ubi_device *ubi) +{ + struct ubi_wl_entry *e; + struct ubi_fm_pool *pool = &ubi->fm_wl_pool; + + return_unused_pool_pebs(ubi, pool); + + for (pool->size = 0; pool->size < pool->max_size; pool->size++) { + if (!ubi->free.rb_node || + (ubi->free_count - ubi->beb_rsvd_pebs < 5)) + break; + + e = find_wl_entry(ubi, &ubi->free, WL_FREE_MAX_DIFF); + self_check_in_wl_tree(ubi, e, &ubi->free); + rb_erase(&e->u.rb, &ubi->free); + ubi->free_count--; + + pool->pebs[pool->size] = e->pnum; + } + pool->used = 0; +} + +/** + * refill_wl_user_pool - refills all the fastmap pool used by ubi_wl_get_peb. + * @ubi: UBI device description object + */ +static void refill_wl_user_pool(struct ubi_device *ubi) +{ + struct ubi_fm_pool *pool = &ubi->fm_pool; + + return_unused_pool_pebs(ubi, pool); + + for (pool->size = 0; pool->size < pool->max_size; pool->size++) { + pool->pebs[pool->size] = __wl_get_peb(ubi); + if (pool->pebs[pool->size] < 0) + break; + } + pool->used = 0; +} + +/** + * ubi_refill_pools - refills all fastmap PEB pools. + * @ubi: UBI device description object + */ +void ubi_refill_pools(struct ubi_device *ubi) +{ + spin_lock(&ubi->wl_lock); + refill_wl_pool(ubi); + refill_wl_user_pool(ubi); + spin_unlock(&ubi->wl_lock); +} + +/* ubi_wl_get_peb - works exaclty like __wl_get_peb but keeps track of + * the fastmap pool. + */ +int ubi_wl_get_peb(struct ubi_device *ubi) +{ + int ret; + struct ubi_fm_pool *pool = &ubi->fm_pool; + struct ubi_fm_pool *wl_pool = &ubi->fm_wl_pool; + + if (!pool->size || !wl_pool->size || pool->used == pool->size || + wl_pool->used == wl_pool->size) + ubi_update_fastmap(ubi); + + /* we got not a single free PEB */ + if (!pool->size) + ret = -ENOSPC; + else { + spin_lock(&ubi->wl_lock); + ret = pool->pebs[pool->used++]; + prot_queue_add(ubi, ubi->lookuptbl[ret]); + spin_unlock(&ubi->wl_lock); + } + + return ret; +} + +/* get_peb_for_wl - returns a PEB to be used internally by the WL sub-system. * - * This function returns PEB @pnum from the protection trees and returns zero - * in case of success and %-ENODEV if the PEB was not found in the protection - * trees. + * @ubi: UBI device description object */ -static int prot_tree_del(struct ubi_device *ubi, int pnum) +static struct ubi_wl_entry *get_peb_for_wl(struct ubi_device *ubi) { - struct rb_node *p; - struct ubi_wl_prot_entry *pe = NULL; + struct ubi_fm_pool *pool = &ubi->fm_wl_pool; + int pnum; + + if (pool->used == pool->size || !pool->size) { + /* We cannot update the fastmap here because this + * function is called in atomic context. + * Let's fail here and refill/update it as soon as possible. */ +#ifndef __UBOOT__ + schedule_work(&ubi->fm_work); +#else + /* In U-Boot we must call this directly */ + ubi_update_fastmap(ubi); +#endif + return NULL; + } else { + pnum = pool->pebs[pool->used++]; + return ubi->lookuptbl[pnum]; + } +} +#else +static struct ubi_wl_entry *get_peb_for_wl(struct ubi_device *ubi) +{ + struct ubi_wl_entry *e; - p = ubi->prot.pnum.rb_node; - while (p) { + e = find_wl_entry(ubi, &ubi->free, WL_FREE_MAX_DIFF); + self_check_in_wl_tree(ubi, e, &ubi->free); + rb_erase(&e->u.rb, &ubi->free); - pe = rb_entry(p, struct ubi_wl_prot_entry, rb_pnum); + return e; +} - if (pnum == pe->e->pnum) - goto found; +int ubi_wl_get_peb(struct ubi_device *ubi) +{ + int peb, err; - if (pnum < pe->e->pnum) - p = p->rb_left; - else - p = p->rb_right; + spin_lock(&ubi->wl_lock); + peb = __wl_get_peb(ubi); + spin_unlock(&ubi->wl_lock); + + err = ubi_self_check_all_ff(ubi, peb, ubi->vid_hdr_aloffset, + ubi->peb_size - ubi->vid_hdr_aloffset); + if (err) { + ubi_err("new PEB %d does not contain all 0xFF bytes", peb); + return err; } - return -ENODEV; + return peb; +} +#endif -found: - ubi_assert(pe->e->pnum == pnum); - rb_erase(&pe->rb_aec, &ubi->prot.aec); - rb_erase(&pe->rb_pnum, &ubi->prot.pnum); - kfree(pe); +/** + * prot_queue_del - remove a physical eraseblock from the protection queue. + * @ubi: UBI device description object + * @pnum: the physical eraseblock to remove + * + * This function deletes PEB @pnum from the protection queue and returns zero + * in case of success and %-ENODEV if the PEB was not found. + */ +static int prot_queue_del(struct ubi_device *ubi, int pnum) +{ + struct ubi_wl_entry *e; + + e = ubi->lookuptbl[pnum]; + if (!e) + return -ENODEV; + + if (self_check_in_pq(ubi, e)) + return -ENODEV; + + list_del(&e->u.list); + dbg_wl("deleted PEB %d from the protection queue", e->pnum); return 0; } @@ -574,7 +729,8 @@ found: * This function returns zero in case of success and a negative error code in * case of failure. */ -static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, int torture) +static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, + int torture) { int err; struct ubi_ec_hdr *ec_hdr; @@ -582,8 +738,8 @@ static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, int tortur dbg_wl("erase PEB %d, old EC %llu", e->pnum, ec); - err = paranoid_check_ec(ubi, e->pnum, e->ec); - if (err > 0) + err = self_check_ec(ubi, e->pnum, e->ec); + if (err) return -EINVAL; ec_hdr = kzalloc(ubi->ec_hdr_alsize, GFP_NOFS); @@ -626,91 +782,124 @@ out_free: } /** - * check_protection_over - check if it is time to stop protecting some - * physical eraseblocks. + * serve_prot_queue - check if it is time to stop protecting PEBs. * @ubi: UBI device description object * - * This function is called after each erase operation, when the absolute erase - * counter is incremented, to check if some physical eraseblock have not to be - * protected any longer. These physical eraseblocks are moved from the - * protection trees to the used tree. + * This function is called after each erase operation and removes PEBs from the + * tail of the protection queue. These PEBs have been protected for long enough + * and should be moved to the used tree. */ -static void check_protection_over(struct ubi_device *ubi) +static void serve_prot_queue(struct ubi_device *ubi) { - struct ubi_wl_prot_entry *pe; + struct ubi_wl_entry *e, *tmp; + int count; /* * There may be several protected physical eraseblock to remove, * process them all. */ - while (1) { - spin_lock(&ubi->wl_lock); - if (!ubi->prot.aec.rb_node) { - spin_unlock(&ubi->wl_lock); - break; - } - - pe = rb_entry(rb_first(&ubi->prot.aec), - struct ubi_wl_prot_entry, rb_aec); +repeat: + count = 0; + spin_lock(&ubi->wl_lock); + list_for_each_entry_safe(e, tmp, &ubi->pq[ubi->pq_head], u.list) { + dbg_wl("PEB %d EC %d protection over, move to used tree", + e->pnum, e->ec); - if (pe->abs_ec > ubi->abs_ec) { + list_del(&e->u.list); + wl_tree_add(e, &ubi->used); + if (count++ > 32) { + /* + * Let's be nice and avoid holding the spinlock for + * too long. + */ spin_unlock(&ubi->wl_lock); - break; + cond_resched(); + goto repeat; } - - dbg_wl("PEB %d protection over, abs_ec %llu, PEB abs_ec %llu", - pe->e->pnum, ubi->abs_ec, pe->abs_ec); - rb_erase(&pe->rb_aec, &ubi->prot.aec); - rb_erase(&pe->rb_pnum, &ubi->prot.pnum); - wl_tree_add(pe->e, &ubi->used); - spin_unlock(&ubi->wl_lock); - - kfree(pe); - cond_resched(); } + + ubi->pq_head += 1; + if (ubi->pq_head == UBI_PROT_QUEUE_LEN) + ubi->pq_head = 0; + ubi_assert(ubi->pq_head >= 0 && ubi->pq_head < UBI_PROT_QUEUE_LEN); + spin_unlock(&ubi->wl_lock); } /** - * schedule_ubi_work - schedule a work. + * __schedule_ubi_work - schedule a work. * @ubi: UBI device description object * @wrk: the work to schedule * - * This function enqueues a work defined by @wrk to the tail of the pending - * works list. + * This function adds a work defined by @wrk to the tail of the pending works + * list. Can only be used of ubi->work_sem is already held in read mode! */ -static void schedule_ubi_work(struct ubi_device *ubi, struct ubi_work *wrk) +static void __schedule_ubi_work(struct ubi_device *ubi, struct ubi_work *wrk) { spin_lock(&ubi->wl_lock); list_add_tail(&wrk->list, &ubi->works); ubi_assert(ubi->works_count >= 0); ubi->works_count += 1; - +#ifndef __UBOOT__ + if (ubi->thread_enabled && !ubi_dbg_is_bgt_disabled(ubi)) + wake_up_process(ubi->bgt_thread); +#else /* * U-Boot special: We have no bgt_thread in U-Boot! * So just call do_work() here directly. */ do_work(ubi); - +#endif spin_unlock(&ubi->wl_lock); } +/** + * schedule_ubi_work - schedule a work. + * @ubi: UBI device description object + * @wrk: the work to schedule + * + * This function adds a work defined by @wrk to the tail of the pending works + * list. + */ +static void schedule_ubi_work(struct ubi_device *ubi, struct ubi_work *wrk) +{ + down_read(&ubi->work_sem); + __schedule_ubi_work(ubi, wrk); + up_read(&ubi->work_sem); +} + static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, int cancel); +#ifdef CONFIG_MTD_UBI_FASTMAP +/** + * ubi_is_erase_work - checks whether a work is erase work. + * @wrk: The work object to be checked + */ +int ubi_is_erase_work(struct ubi_work *wrk) +{ + return wrk->func == erase_worker; +} +#endif + /** * schedule_erase - schedule an erase work. * @ubi: UBI device description object * @e: the WL entry of the physical eraseblock to erase + * @vol_id: the volume ID that last used this PEB + * @lnum: the last used logical eraseblock number for the PEB * @torture: if the physical eraseblock has to be tortured * * This function returns zero in case of success and a %-ENOMEM in case of * failure. */ static int schedule_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, - int torture) + int vol_id, int lnum, int torture) { struct ubi_work *wl_wrk; + ubi_assert(e); + ubi_assert(!ubi_is_fm_block(ubi, e->pnum)); + dbg_wl("schedule erasure of PEB %d, EC %d, torture %d", e->pnum, e->ec, torture); @@ -720,6 +909,8 @@ static int schedule_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, wl_wrk->func = &erase_worker; wl_wrk->e = e; + wl_wrk->vol_id = vol_id; + wl_wrk->lnum = lnum; wl_wrk->torture = torture; schedule_ubi_work(ubi, wl_wrk); @@ -727,6 +918,79 @@ static int schedule_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, } /** + * do_sync_erase - run the erase worker synchronously. + * @ubi: UBI device description object + * @e: the WL entry of the physical eraseblock to erase + * @vol_id: the volume ID that last used this PEB + * @lnum: the last used logical eraseblock number for the PEB + * @torture: if the physical eraseblock has to be tortured + * + */ +static int do_sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, + int vol_id, int lnum, int torture) +{ + struct ubi_work *wl_wrk; + + dbg_wl("sync erase of PEB %i", e->pnum); + + wl_wrk = kmalloc(sizeof(struct ubi_work), GFP_NOFS); + if (!wl_wrk) + return -ENOMEM; + + wl_wrk->e = e; + wl_wrk->vol_id = vol_id; + wl_wrk->lnum = lnum; + wl_wrk->torture = torture; + + return erase_worker(ubi, wl_wrk, 0); +} + +#ifdef CONFIG_MTD_UBI_FASTMAP +/** + * ubi_wl_put_fm_peb - returns a PEB used in a fastmap to the wear-leveling + * sub-system. + * see: ubi_wl_put_peb() + * + * @ubi: UBI device description object + * @fm_e: physical eraseblock to return + * @lnum: the last used logical eraseblock number for the PEB + * @torture: if this physical eraseblock has to be tortured + */ +int ubi_wl_put_fm_peb(struct ubi_device *ubi, struct ubi_wl_entry *fm_e, + int lnum, int torture) +{ + struct ubi_wl_entry *e; + int vol_id, pnum = fm_e->pnum; + + dbg_wl("PEB %d", pnum); + + ubi_assert(pnum >= 0); + ubi_assert(pnum < ubi->peb_count); + + spin_lock(&ubi->wl_lock); + e = ubi->lookuptbl[pnum]; + + /* This can happen if we recovered from a fastmap the very + * first time and writing now a new one. In this case the wl system + * has never seen any PEB used by the original fastmap. + */ + if (!e) { + e = fm_e; + ubi_assert(e->ec >= 0); + ubi->lookuptbl[pnum] = e; + } else { + e->ec = fm_e->ec; + kfree(fm_e); + } + + spin_unlock(&ubi->wl_lock); + + vol_id = lnum ? UBI_FM_DATA_VOLUME_ID : UBI_FM_SB_VOLUME_ID; + return schedule_erase(ubi, e, vol_id, lnum, torture); +} +#endif + +/** * wear_leveling_worker - wear-leveling worker function. * @ubi: UBI device description object * @wrk: the work object @@ -739,13 +1003,15 @@ static int schedule_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk, int cancel) { - int err, put = 0, scrubbing = 0, protect = 0; - struct ubi_wl_prot_entry *uninitialized_var(pe); + int err, scrubbing = 0, torture = 0, protect = 0, erroneous = 0; + int vol_id = -1, uninitialized_var(lnum); +#ifdef CONFIG_MTD_UBI_FASTMAP + int anchor = wrk->anchor; +#endif struct ubi_wl_entry *e1, *e2; struct ubi_vid_hdr *vid_hdr; kfree(wrk); - if (cancel) return 0; @@ -775,36 +1041,61 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk, goto out_cancel; } +#ifdef CONFIG_MTD_UBI_FASTMAP + /* Check whether we need to produce an anchor PEB */ + if (!anchor) + anchor = !anchor_pebs_avalible(&ubi->free); + + if (anchor) { + e1 = find_anchor_wl_entry(&ubi->used); + if (!e1) + goto out_cancel; + e2 = get_peb_for_wl(ubi); + if (!e2) + goto out_cancel; + + self_check_in_wl_tree(ubi, e1, &ubi->used); + rb_erase(&e1->u.rb, &ubi->used); + dbg_wl("anchor-move PEB %d to PEB %d", e1->pnum, e2->pnum); + } else if (!ubi->scrub.rb_node) { +#else if (!ubi->scrub.rb_node) { +#endif /* * Now pick the least worn-out used physical eraseblock and a * highly worn-out free physical eraseblock. If the erase * counters differ much enough, start wear-leveling. */ - e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, rb); - e2 = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF); + e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, u.rb); + e2 = get_peb_for_wl(ubi); + if (!e2) + goto out_cancel; if (!(e2->ec - e1->ec >= UBI_WL_THRESHOLD)) { dbg_wl("no WL needed: min used EC %d, max free EC %d", e1->ec, e2->ec); + + /* Give the unused PEB back */ + wl_tree_add(e2, &ubi->free); goto out_cancel; } - paranoid_check_in_wl_tree(e1, &ubi->used); - rb_erase(&e1->rb, &ubi->used); + self_check_in_wl_tree(ubi, e1, &ubi->used); + rb_erase(&e1->u.rb, &ubi->used); dbg_wl("move PEB %d EC %d to PEB %d EC %d", e1->pnum, e1->ec, e2->pnum, e2->ec); } else { /* Perform scrubbing */ scrubbing = 1; - e1 = rb_entry(rb_first(&ubi->scrub), struct ubi_wl_entry, rb); - e2 = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF); - paranoid_check_in_wl_tree(e1, &ubi->scrub); - rb_erase(&e1->rb, &ubi->scrub); + e1 = rb_entry(rb_first(&ubi->scrub), struct ubi_wl_entry, u.rb); + e2 = get_peb_for_wl(ubi); + if (!e2) + goto out_cancel; + + self_check_in_wl_tree(ubi, e1, &ubi->scrub); + rb_erase(&e1->u.rb, &ubi->scrub); dbg_wl("scrub PEB %d to PEB %d", e1->pnum, e2->pnum); } - paranoid_check_in_wl_tree(e2, &ubi->free); - rb_erase(&e2->rb, &ubi->free); ubi->move_from = e1; ubi->move_to = e2; spin_unlock(&ubi->wl_lock); @@ -822,81 +1113,127 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk, err = ubi_io_read_vid_hdr(ubi, e1->pnum, vid_hdr, 0); if (err && err != UBI_IO_BITFLIPS) { - if (err == UBI_IO_PEB_FREE) { + if (err == UBI_IO_FF) { /* * We are trying to move PEB without a VID header. UBI * always write VID headers shortly after the PEB was - * given, so we have a situation when it did not have - * chance to write it down because it was preempted. - * Just re-schedule the work, so that next time it will - * likely have the VID header in place. + * given, so we have a situation when it has not yet + * had a chance to write it, because it was preempted. + * So add this PEB to the protection queue so far, + * because presumably more data will be written there + * (including the missing VID header), and then we'll + * move it. */ dbg_wl("PEB %d has no VID header", e1->pnum); + protect = 1; + goto out_not_moved; + } else if (err == UBI_IO_FF_BITFLIPS) { + /* + * The same situation as %UBI_IO_FF, but bit-flips were + * detected. It is better to schedule this PEB for + * scrubbing. + */ + dbg_wl("PEB %d has no VID header but has bit-flips", + e1->pnum); + scrubbing = 1; goto out_not_moved; } ubi_err("error %d while reading VID header from PEB %d", err, e1->pnum); - if (err > 0) - err = -EIO; goto out_error; } + vol_id = be32_to_cpu(vid_hdr->vol_id); + lnum = be32_to_cpu(vid_hdr->lnum); + err = ubi_eba_copy_leb(ubi, e1->pnum, e2->pnum, vid_hdr); if (err) { - - if (err < 0) - goto out_error; - if (err == 1) + if (err == MOVE_CANCEL_RACE) { + /* + * The LEB has not been moved because the volume is + * being deleted or the PEB has been put meanwhile. We + * should prevent this PEB from being selected for + * wear-leveling movement again, so put it to the + * protection queue. + */ + protect = 1; + goto out_not_moved; + } + if (err == MOVE_RETRY) { + scrubbing = 1; goto out_not_moved; + } + if (err == MOVE_TARGET_BITFLIPS || err == MOVE_TARGET_WR_ERR || + err == MOVE_TARGET_RD_ERR) { + /* + * Target PEB had bit-flips or write error - torture it. + */ + torture = 1; + goto out_not_moved; + } - /* - * For some reason the LEB was not moved - it might be because - * the volume is being deleted. We should prevent this PEB from - * being selected for wear-levelling movement for some "time", - * so put it to the protection tree. - */ + if (err == MOVE_SOURCE_RD_ERR) { + /* + * An error happened while reading the source PEB. Do + * not switch to R/O mode in this case, and give the + * upper layers a possibility to recover from this, + * e.g. by unmapping corresponding LEB. Instead, just + * put this PEB to the @ubi->erroneous list to prevent + * UBI from trying to move it over and over again. + */ + if (ubi->erroneous_peb_count > ubi->max_erroneous) { + ubi_err("too many erroneous eraseblocks (%d)", + ubi->erroneous_peb_count); + goto out_error; + } + erroneous = 1; + goto out_not_moved; + } - dbg_wl("cancelled moving PEB %d", e1->pnum); - pe = kmalloc(sizeof(struct ubi_wl_prot_entry), GFP_NOFS); - if (!pe) { - err = -ENOMEM; + if (err < 0) goto out_error; - } - protect = 1; + ubi_assert(0); } + /* The PEB has been successfully moved */ + if (scrubbing) + ubi_msg("scrubbed PEB %d (LEB %d:%d), data moved to PEB %d", + e1->pnum, vol_id, lnum, e2->pnum); ubi_free_vid_hdr(ubi, vid_hdr); + spin_lock(&ubi->wl_lock); - if (protect) - prot_tree_add(ubi, e1, pe, protect); - if (!ubi->move_to_put) + if (!ubi->move_to_put) { wl_tree_add(e2, &ubi->used); - else - put = 1; + e2 = NULL; + } ubi->move_from = ubi->move_to = NULL; ubi->move_to_put = ubi->wl_scheduled = 0; spin_unlock(&ubi->wl_lock); - if (put) { + err = do_sync_erase(ubi, e1, vol_id, lnum, 0); + if (err) { + kmem_cache_free(ubi_wl_entry_slab, e1); + if (e2) + kmem_cache_free(ubi_wl_entry_slab, e2); + goto out_ro; + } + + if (e2) { /* * Well, the target PEB was put meanwhile, schedule it for * erasure. */ - dbg_wl("PEB %d was put meanwhile, erase", e2->pnum); - err = schedule_erase(ubi, e2, 0); - if (err) - goto out_error; - } - - if (!protect) { - err = schedule_erase(ubi, e1, 0); - if (err) - goto out_error; + dbg_wl("PEB %d (LEB %d:%d) was put meanwhile, erase", + e2->pnum, vol_id, lnum); + err = do_sync_erase(ubi, e2, vol_id, lnum, 0); + if (err) { + kmem_cache_free(ubi_wl_entry_slab, e2); + goto out_ro; + } } - dbg_wl("done"); mutex_unlock(&ubi->move_mutex); return 0; @@ -904,42 +1241,60 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk, /* * For some reasons the LEB was not moved, might be an error, might be * something else. @e1 was not changed, so return it back. @e2 might - * be changed, schedule it for erasure. + * have been changed, schedule it for erasure. */ out_not_moved: - ubi_free_vid_hdr(ubi, vid_hdr); + if (vol_id != -1) + dbg_wl("cancel moving PEB %d (LEB %d:%d) to PEB %d (%d)", + e1->pnum, vol_id, lnum, e2->pnum, err); + else + dbg_wl("cancel moving PEB %d to PEB %d (%d)", + e1->pnum, e2->pnum, err); spin_lock(&ubi->wl_lock); - if (scrubbing) + if (protect) + prot_queue_add(ubi, e1); + else if (erroneous) { + wl_tree_add(e1, &ubi->erroneous); + ubi->erroneous_peb_count += 1; + } else if (scrubbing) wl_tree_add(e1, &ubi->scrub); else wl_tree_add(e1, &ubi->used); + ubi_assert(!ubi->move_to_put); ubi->move_from = ubi->move_to = NULL; - ubi->move_to_put = ubi->wl_scheduled = 0; + ubi->wl_scheduled = 0; spin_unlock(&ubi->wl_lock); - err = schedule_erase(ubi, e2, 0); - if (err) - goto out_error; - + ubi_free_vid_hdr(ubi, vid_hdr); + err = do_sync_erase(ubi, e2, vol_id, lnum, torture); + if (err) { + kmem_cache_free(ubi_wl_entry_slab, e2); + goto out_ro; + } mutex_unlock(&ubi->move_mutex); return 0; out_error: - ubi_err("error %d while moving PEB %d to PEB %d", - err, e1->pnum, e2->pnum); - - ubi_free_vid_hdr(ubi, vid_hdr); + if (vol_id != -1) + ubi_err("error %d while moving PEB %d to PEB %d", + err, e1->pnum, e2->pnum); + else + ubi_err("error %d while moving PEB %d (LEB %d:%d) to PEB %d", + err, e1->pnum, vol_id, lnum, e2->pnum); spin_lock(&ubi->wl_lock); ubi->move_from = ubi->move_to = NULL; ubi->move_to_put = ubi->wl_scheduled = 0; spin_unlock(&ubi->wl_lock); + ubi_free_vid_hdr(ubi, vid_hdr); kmem_cache_free(ubi_wl_entry_slab, e1); kmem_cache_free(ubi_wl_entry_slab, e2); - ubi_ro_mode(ubi); +out_ro: + ubi_ro_mode(ubi); mutex_unlock(&ubi->move_mutex); - return err; + ubi_assert(err != 0); + return err < 0 ? err : -EIO; out_cancel: ubi->wl_scheduled = 0; @@ -952,12 +1307,13 @@ out_cancel: /** * ensure_wear_leveling - schedule wear-leveling if it is needed. * @ubi: UBI device description object + * @nested: set to non-zero if this function is called from UBI worker * * This function checks if it is time to start wear-leveling and schedules it * if yes. This function returns zero in case of success and a negative error * code in case of failure. */ -static int ensure_wear_leveling(struct ubi_device *ubi) +static int ensure_wear_leveling(struct ubi_device *ubi, int nested) { int err = 0; struct ubi_wl_entry *e1; @@ -981,11 +1337,11 @@ static int ensure_wear_leveling(struct ubi_device *ubi) /* * We schedule wear-leveling only if the difference between the * lowest erase counter of used physical eraseblocks and a high - * erase counter of free physical eraseblocks is greater then + * erase counter of free physical eraseblocks is greater than * %UBI_WL_THRESHOLD. */ - e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, rb); - e2 = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF); + e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, u.rb); + e2 = find_wl_entry(ubi, &ubi->free, WL_FREE_MAX_DIFF); if (!(e2->ec - e1->ec >= UBI_WL_THRESHOLD)) goto out_unlock; @@ -1002,8 +1358,12 @@ static int ensure_wear_leveling(struct ubi_device *ubi) goto out_cancel; } + wrk->anchor = 0; wrk->func = &wear_leveling_worker; - schedule_ubi_work(ubi, wrk); + if (nested) + __schedule_ubi_work(ubi, wrk); + else + schedule_ubi_work(ubi, wrk); return err; out_cancel: @@ -1014,6 +1374,38 @@ out_unlock: return err; } +#ifdef CONFIG_MTD_UBI_FASTMAP +/** + * ubi_ensure_anchor_pebs - schedule wear-leveling to produce an anchor PEB. + * @ubi: UBI device description object + */ +int ubi_ensure_anchor_pebs(struct ubi_device *ubi) +{ + struct ubi_work *wrk; + + spin_lock(&ubi->wl_lock); + if (ubi->wl_scheduled) { + spin_unlock(&ubi->wl_lock); + return 0; + } + ubi->wl_scheduled = 1; + spin_unlock(&ubi->wl_lock); + + wrk = kmalloc(sizeof(struct ubi_work), GFP_NOFS); + if (!wrk) { + spin_lock(&ubi->wl_lock); + ubi->wl_scheduled = 0; + spin_unlock(&ubi->wl_lock); + return -ENOMEM; + } + + wrk->anchor = 1; + wrk->func = &wear_leveling_worker; + schedule_ubi_work(ubi, wrk); + return 0; +} +#endif + /** * erase_worker - physical eraseblock erase worker function. * @ubi: UBI device description object @@ -1029,7 +1421,10 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, int cancel) { struct ubi_wl_entry *e = wl_wrk->e; - int pnum = e->pnum, err, need; + int pnum = e->pnum; + int vol_id = wl_wrk->vol_id; + int lnum = wl_wrk->lnum; + int err, available_consumed = 0; if (cancel) { dbg_wl("cancel erasure of PEB %d EC %d", pnum, e->ec); @@ -1038,7 +1433,10 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, return 0; } - dbg_wl("erase PEB %d EC %d", pnum, e->ec); + dbg_wl("erase PEB %d EC %d LEB %d:%d", + pnum, e->ec, wl_wrk->vol_id, wl_wrk->lnum); + + ubi_assert(!ubi_is_fm_block(ubi, e->pnum)); err = sync_erase(ubi, e, wl_wrk->torture); if (!err) { @@ -1046,44 +1444,45 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, kfree(wl_wrk); spin_lock(&ubi->wl_lock); - ubi->abs_ec += 1; wl_tree_add(e, &ubi->free); + ubi->free_count++; spin_unlock(&ubi->wl_lock); /* - * One more erase operation has happened, take care about protected - * physical eraseblocks. + * One more erase operation has happened, take care about + * protected physical eraseblocks. */ - check_protection_over(ubi); + serve_prot_queue(ubi); /* And take care about wear-leveling */ - err = ensure_wear_leveling(ubi); + err = ensure_wear_leveling(ubi, 1); return err; } ubi_err("failed to erase PEB %d, error %d", pnum, err); kfree(wl_wrk); - kmem_cache_free(ubi_wl_entry_slab, e); if (err == -EINTR || err == -ENOMEM || err == -EAGAIN || err == -EBUSY) { int err1; /* Re-schedule the LEB for erasure */ - err1 = schedule_erase(ubi, e, 0); + err1 = schedule_erase(ubi, e, vol_id, lnum, 0); if (err1) { err = err1; goto out_ro; } return err; - } else if (err != -EIO) { + } + + kmem_cache_free(ubi_wl_entry_slab, e); + if (err != -EIO) /* * If this is not %-EIO, we have no idea what to do. Scheduling * this physical eraseblock for erasure again would cause - * errors again and again. Well, lets switch to RO mode. + * errors again and again. Well, lets switch to R/O mode. */ goto out_ro; - } /* It is %-EIO, the PEB went bad */ @@ -1093,48 +1492,62 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, } spin_lock(&ubi->volumes_lock); - need = ubi->beb_rsvd_level - ubi->beb_rsvd_pebs + 1; - if (need > 0) { - need = ubi->avail_pebs >= need ? need : ubi->avail_pebs; - ubi->avail_pebs -= need; - ubi->rsvd_pebs += need; - ubi->beb_rsvd_pebs += need; - if (need > 0) - ubi_msg("reserve more %d PEBs", need); - } - if (ubi->beb_rsvd_pebs == 0) { - spin_unlock(&ubi->volumes_lock); - ubi_err("no reserved physical eraseblocks"); - goto out_ro; + if (ubi->avail_pebs == 0) { + spin_unlock(&ubi->volumes_lock); + ubi_err("no reserved/available physical eraseblocks"); + goto out_ro; + } + ubi->avail_pebs -= 1; + available_consumed = 1; } - spin_unlock(&ubi->volumes_lock); - ubi_msg("mark PEB %d as bad", pnum); + ubi_msg("mark PEB %d as bad", pnum); err = ubi_io_mark_bad(ubi, pnum); if (err) goto out_ro; spin_lock(&ubi->volumes_lock); - ubi->beb_rsvd_pebs -= 1; + if (ubi->beb_rsvd_pebs > 0) { + if (available_consumed) { + /* + * The amount of reserved PEBs increased since we last + * checked. + */ + ubi->avail_pebs += 1; + available_consumed = 0; + } + ubi->beb_rsvd_pebs -= 1; + } ubi->bad_peb_count += 1; ubi->good_peb_count -= 1; ubi_calculate_reserved(ubi); - if (ubi->beb_rsvd_pebs == 0) - ubi_warn("last PEB from the reserved pool was used"); + if (available_consumed) + ubi_warn("no PEBs in the reserved pool, used an available PEB"); + else if (ubi->beb_rsvd_pebs) + ubi_msg("%d PEBs left in the reserve", ubi->beb_rsvd_pebs); + else + ubi_warn("last PEB from the reserve was used"); spin_unlock(&ubi->volumes_lock); return err; out_ro: + if (available_consumed) { + spin_lock(&ubi->volumes_lock); + ubi->avail_pebs += 1; + spin_unlock(&ubi->volumes_lock); + } ubi_ro_mode(ubi); return err; } /** - * ubi_wl_put_peb - return a physical eraseblock to the wear-leveling unit. + * ubi_wl_put_peb - return a PEB to the wear-leveling sub-system. * @ubi: UBI device description object + * @vol_id: the volume ID that last used this PEB + * @lnum: the last used logical eraseblock number for the PEB * @pnum: physical eraseblock to return * @torture: if this physical eraseblock has to be tortured * @@ -1143,7 +1556,8 @@ out_ro: * occurred to this @pnum and it has to be tested. This function returns zero * in case of success, and a negative error code in case of failure. */ -int ubi_wl_put_peb(struct ubi_device *ubi, int pnum, int torture) +int ubi_wl_put_peb(struct ubi_device *ubi, int vol_id, int lnum, + int pnum, int torture) { int err; struct ubi_wl_entry *e; @@ -1172,11 +1586,11 @@ retry: /* * User is putting the physical eraseblock which was selected * as the target the data is moved to. It may happen if the EBA - * unit already re-mapped the LEB in 'ubi_eba_copy_leb()' but - * the WL unit has not put the PEB to the "used" tree yet, but - * it is about to do this. So we just set a flag which will - * tell the WL worker that the PEB is not needed anymore and - * should be scheduled for erasure. + * sub-system already re-mapped the LEB in 'ubi_eba_copy_leb()' + * but the WL sub-system has not put the PEB to the "used" tree + * yet, but it is about to do this. So we just set a flag which + * will tell the WL worker that the PEB is not needed anymore + * and should be scheduled for erasure. */ dbg_wl("PEB %d is the target of data moving", pnum); ubi_assert(!ubi->move_to_put); @@ -1185,13 +1599,20 @@ retry: return 0; } else { if (in_wl_tree(e, &ubi->used)) { - paranoid_check_in_wl_tree(e, &ubi->used); - rb_erase(&e->rb, &ubi->used); + self_check_in_wl_tree(ubi, e, &ubi->used); + rb_erase(&e->u.rb, &ubi->used); } else if (in_wl_tree(e, &ubi->scrub)) { - paranoid_check_in_wl_tree(e, &ubi->scrub); - rb_erase(&e->rb, &ubi->scrub); + self_check_in_wl_tree(ubi, e, &ubi->scrub); + rb_erase(&e->u.rb, &ubi->scrub); + } else if (in_wl_tree(e, &ubi->erroneous)) { + self_check_in_wl_tree(ubi, e, &ubi->erroneous); + rb_erase(&e->u.rb, &ubi->erroneous); + ubi->erroneous_peb_count -= 1; + ubi_assert(ubi->erroneous_peb_count >= 0); + /* Erroneous PEBs should be tortured */ + torture = 1; } else { - err = prot_tree_del(ubi, e->pnum); + err = prot_queue_del(ubi, e->pnum); if (err) { ubi_err("PEB %d not found", pnum); ubi_ro_mode(ubi); @@ -1202,7 +1623,7 @@ retry: } spin_unlock(&ubi->wl_lock); - err = schedule_erase(ubi, e, torture); + err = schedule_erase(ubi, e, vol_id, lnum, torture); if (err) { spin_lock(&ubi->wl_lock); wl_tree_add(e, &ubi->used); @@ -1231,7 +1652,8 @@ int ubi_wl_scrub_peb(struct ubi_device *ubi, int pnum) retry: spin_lock(&ubi->wl_lock); e = ubi->lookuptbl[pnum]; - if (e == ubi->move_from || in_wl_tree(e, &ubi->scrub)) { + if (e == ubi->move_from || in_wl_tree(e, &ubi->scrub) || + in_wl_tree(e, &ubi->erroneous)) { spin_unlock(&ubi->wl_lock); return 0; } @@ -1250,12 +1672,12 @@ retry: } if (in_wl_tree(e, &ubi->used)) { - paranoid_check_in_wl_tree(e, &ubi->used); - rb_erase(&e->rb, &ubi->used); + self_check_in_wl_tree(ubi, e, &ubi->used); + rb_erase(&e->u.rb, &ubi->used); } else { int err; - err = prot_tree_del(ubi, e->pnum); + err = prot_queue_del(ubi, e->pnum); if (err) { ubi_err("PEB %d not found", pnum); ubi_ro_mode(ubi); @@ -1271,29 +1693,60 @@ retry: * Technically scrubbing is the same as wear-leveling, so it is done * by the WL worker. */ - return ensure_wear_leveling(ubi); + return ensure_wear_leveling(ubi, 0); } /** * ubi_wl_flush - flush all pending works. * @ubi: UBI device description object + * @vol_id: the volume id to flush for + * @lnum: the logical eraseblock number to flush for * - * This function returns zero in case of success and a negative error code in - * case of failure. + * This function executes all pending works for a particular volume id / + * logical eraseblock number pair. If either value is set to %UBI_ALL, then it + * acts as a wildcard for all of the corresponding volume numbers or logical + * eraseblock numbers. It returns zero in case of success and a negative error + * code in case of failure. */ -int ubi_wl_flush(struct ubi_device *ubi) +int ubi_wl_flush(struct ubi_device *ubi, int vol_id, int lnum) { - int err; + int err = 0; + int found = 1; /* - * Erase while the pending works queue is not empty, but not more then + * Erase while the pending works queue is not empty, but not more than * the number of currently pending works. */ - dbg_wl("flush (%d pending works)", ubi->works_count); - while (ubi->works_count) { - err = do_work(ubi); - if (err) - return err; + dbg_wl("flush pending work for LEB %d:%d (%d pending works)", + vol_id, lnum, ubi->works_count); + + while (found) { + struct ubi_work *wrk; + found = 0; + + down_read(&ubi->work_sem); + spin_lock(&ubi->wl_lock); + list_for_each_entry(wrk, &ubi->works, list) { + if ((vol_id == UBI_ALL || wrk->vol_id == vol_id) && + (lnum == UBI_ALL || wrk->lnum == lnum)) { + list_del(&wrk->list); + ubi->works_count -= 1; + ubi_assert(ubi->works_count >= 0); + spin_unlock(&ubi->wl_lock); + + err = wrk->func(ubi, wrk, 0); + if (err) { + up_read(&ubi->work_sem); + return err; + } + + spin_lock(&ubi->wl_lock); + found = 1; + break; + } + } + spin_unlock(&ubi->wl_lock); + up_read(&ubi->work_sem); } /* @@ -1303,18 +1756,7 @@ int ubi_wl_flush(struct ubi_device *ubi) down_write(&ubi->work_sem); up_write(&ubi->work_sem); - /* - * And in case last was the WL worker and it cancelled the LEB - * movement, flush again. - */ - while (ubi->works_count) { - dbg_wl("flush more (%d pending works)", ubi->works_count); - err = do_work(ubi); - if (err) - return err; - } - - return 0; + return err; } /** @@ -1333,11 +1775,11 @@ static void tree_destroy(struct rb_root *root) else if (rb->rb_right) rb = rb->rb_right; else { - e = rb_entry(rb, struct ubi_wl_entry, rb); + e = rb_entry(rb, struct ubi_wl_entry, u.rb); rb = rb_parent(rb); if (rb) { - if (rb->rb_left == &e->rb) + if (rb->rb_left == &e->u.rb) rb->rb_left = NULL; else rb->rb_right = NULL; @@ -1372,7 +1814,7 @@ int ubi_thread(void *u) spin_lock(&ubi->wl_lock); if (list_empty(&ubi->works) || ubi->ro_mode || - !ubi->thread_enabled) { + !ubi->thread_enabled || ubi_dbg_is_bgt_disabled(ubi)) { set_current_state(TASK_INTERRUPTIBLE); spin_unlock(&ubi->wl_lock); schedule(); @@ -1392,7 +1834,8 @@ int ubi_thread(void *u) ubi_msg("%s: %d consecutive failures", ubi->bgt_name, WL_MAX_FAILURES); ubi_ro_mode(ubi); - break; + ubi->thread_enabled = 0; + continue; } } else failures = 0; @@ -1422,30 +1865,32 @@ static void cancel_pending(struct ubi_device *ubi) } /** - * ubi_wl_init_scan - initialize the wear-leveling unit using scanning - * information. + * ubi_wl_init - initialize the WL sub-system using attaching information. * @ubi: UBI device description object - * @si: scanning information + * @ai: attaching information * * This function returns zero in case of success, and a negative error code in * case of failure. */ -int ubi_wl_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si) +int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai) { - int err; + int err, i, reserved_pebs, found_pebs = 0; struct rb_node *rb1, *rb2; - struct ubi_scan_volume *sv; - struct ubi_scan_leb *seb, *tmp; + struct ubi_ainf_volume *av; + struct ubi_ainf_peb *aeb, *tmp; struct ubi_wl_entry *e; - - ubi->used = ubi->free = ubi->scrub = RB_ROOT; - ubi->prot.pnum = ubi->prot.aec = RB_ROOT; + ubi->used = ubi->erroneous = ubi->free = ubi->scrub = RB_ROOT; spin_lock_init(&ubi->wl_lock); mutex_init(&ubi->move_mutex); init_rwsem(&ubi->work_sem); - ubi->max_ec = si->max_ec; + ubi->max_ec = ai->max_ec; INIT_LIST_HEAD(&ubi->works); +#ifndef __UBOOT__ +#ifdef CONFIG_MTD_UBI_FASTMAP + INIT_WORK(&ubi->fm_work, update_fastmap_work_fn); +#endif +#endif sprintf(ubi->bgt_name, UBI_BGT_NAME_PATTERN, ubi->ubi_num); @@ -1454,64 +1899,63 @@ int ubi_wl_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si) if (!ubi->lookuptbl) return err; - list_for_each_entry_safe(seb, tmp, &si->erase, u.list) { + for (i = 0; i < UBI_PROT_QUEUE_LEN; i++) + INIT_LIST_HEAD(&ubi->pq[i]); + ubi->pq_head = 0; + + list_for_each_entry_safe(aeb, tmp, &ai->erase, u.list) { cond_resched(); e = kmem_cache_alloc(ubi_wl_entry_slab, GFP_KERNEL); if (!e) goto out_free; - e->pnum = seb->pnum; - e->ec = seb->ec; + e->pnum = aeb->pnum; + e->ec = aeb->ec; + ubi_assert(!ubi_is_fm_block(ubi, e->pnum)); ubi->lookuptbl[e->pnum] = e; - if (schedule_erase(ubi, e, 0)) { + if (schedule_erase(ubi, e, aeb->vol_id, aeb->lnum, 0)) { kmem_cache_free(ubi_wl_entry_slab, e); goto out_free; } + + found_pebs++; } - list_for_each_entry(seb, &si->free, u.list) { + ubi->free_count = 0; + list_for_each_entry(aeb, &ai->free, u.list) { cond_resched(); e = kmem_cache_alloc(ubi_wl_entry_slab, GFP_KERNEL); if (!e) goto out_free; - e->pnum = seb->pnum; - e->ec = seb->ec; + e->pnum = aeb->pnum; + e->ec = aeb->ec; ubi_assert(e->ec >= 0); - wl_tree_add(e, &ubi->free); - ubi->lookuptbl[e->pnum] = e; - } - - list_for_each_entry(seb, &si->corr, u.list) { - cond_resched(); + ubi_assert(!ubi_is_fm_block(ubi, e->pnum)); - e = kmem_cache_alloc(ubi_wl_entry_slab, GFP_KERNEL); - if (!e) - goto out_free; + wl_tree_add(e, &ubi->free); + ubi->free_count++; - e->pnum = seb->pnum; - e->ec = seb->ec; ubi->lookuptbl[e->pnum] = e; - if (schedule_erase(ubi, e, 0)) { - kmem_cache_free(ubi_wl_entry_slab, e); - goto out_free; - } + + found_pebs++; } - ubi_rb_for_each_entry(rb1, sv, &si->volumes, rb) { - ubi_rb_for_each_entry(rb2, seb, &sv->root, u.rb) { + ubi_rb_for_each_entry(rb1, av, &ai->volumes, rb) { + ubi_rb_for_each_entry(rb2, aeb, &av->root, u.rb) { cond_resched(); e = kmem_cache_alloc(ubi_wl_entry_slab, GFP_KERNEL); if (!e) goto out_free; - e->pnum = seb->pnum; - e->ec = seb->ec; + e->pnum = aeb->pnum; + e->ec = aeb->ec; ubi->lookuptbl[e->pnum] = e; - if (!seb->scrub) { + + if (!aeb->scrub) { dbg_wl("add PEB %d EC %d to the used tree", e->pnum, e->ec); wl_tree_add(e, &ubi->used); @@ -1520,20 +1964,38 @@ int ubi_wl_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si) e->pnum, e->ec); wl_tree_add(e, &ubi->scrub); } + + found_pebs++; } } - if (ubi->avail_pebs < WL_RESERVED_PEBS) { + dbg_wl("found %i PEBs", found_pebs); + + if (ubi->fm) + ubi_assert(ubi->good_peb_count == \ + found_pebs + ubi->fm->used_blocks); + else + ubi_assert(ubi->good_peb_count == found_pebs); + + reserved_pebs = WL_RESERVED_PEBS; +#ifdef CONFIG_MTD_UBI_FASTMAP + /* Reserve enough LEBs to store two fastmaps. */ + reserved_pebs += (ubi->fm_size / ubi->leb_size) * 2; +#endif + + if (ubi->avail_pebs < reserved_pebs) { ubi_err("no enough physical eraseblocks (%d, need %d)", - ubi->avail_pebs, WL_RESERVED_PEBS); - err = -ENOSPC; + ubi->avail_pebs, reserved_pebs); + if (ubi->corr_peb_count) + ubi_err("%d PEBs are corrupted and not used", + ubi->corr_peb_count); goto out_free; } - ubi->avail_pebs -= WL_RESERVED_PEBS; - ubi->rsvd_pebs += WL_RESERVED_PEBS; + ubi->avail_pebs -= reserved_pebs; + ubi->rsvd_pebs += reserved_pebs; /* Schedule wear-leveling if needed */ - err = ensure_wear_leveling(ubi); + err = ensure_wear_leveling(ubi, 0); if (err) goto out_free; @@ -1549,72 +2011,57 @@ out_free: } /** - * protection_trees_destroy - destroy the protection RB-trees. + * protection_queue_destroy - destroy the protection queue. * @ubi: UBI device description object */ -static void protection_trees_destroy(struct ubi_device *ubi) +static void protection_queue_destroy(struct ubi_device *ubi) { - struct rb_node *rb; - struct ubi_wl_prot_entry *pe; + int i; + struct ubi_wl_entry *e, *tmp; - rb = ubi->prot.aec.rb_node; - while (rb) { - if (rb->rb_left) - rb = rb->rb_left; - else if (rb->rb_right) - rb = rb->rb_right; - else { - pe = rb_entry(rb, struct ubi_wl_prot_entry, rb_aec); - - rb = rb_parent(rb); - if (rb) { - if (rb->rb_left == &pe->rb_aec) - rb->rb_left = NULL; - else - rb->rb_right = NULL; - } - - kmem_cache_free(ubi_wl_entry_slab, pe->e); - kfree(pe); + for (i = 0; i < UBI_PROT_QUEUE_LEN; ++i) { + list_for_each_entry_safe(e, tmp, &ubi->pq[i], u.list) { + list_del(&e->u.list); + kmem_cache_free(ubi_wl_entry_slab, e); } } } /** - * ubi_wl_close - close the wear-leveling unit. + * ubi_wl_close - close the wear-leveling sub-system. * @ubi: UBI device description object */ void ubi_wl_close(struct ubi_device *ubi) { - dbg_wl("close the UBI wear-leveling unit"); - + dbg_wl("close the WL sub-system"); cancel_pending(ubi); - protection_trees_destroy(ubi); + protection_queue_destroy(ubi); tree_destroy(&ubi->used); + tree_destroy(&ubi->erroneous); tree_destroy(&ubi->free); tree_destroy(&ubi->scrub); kfree(ubi->lookuptbl); } -#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID - /** - * paranoid_check_ec - make sure that the erase counter of a physical eraseblock - * is correct. + * self_check_ec - make sure that the erase counter of a PEB is correct. * @ubi: UBI device description object * @pnum: the physical eraseblock number to check * @ec: the erase counter to check * * This function returns zero if the erase counter of physical eraseblock @pnum - * is equivalent to @ec, %1 if not, and a negative error code if an error + * is equivalent to @ec, and a negative error code if not or if an error * occurred. */ -static int paranoid_check_ec(struct ubi_device *ubi, int pnum, int ec) +static int self_check_ec(struct ubi_device *ubi, int pnum, int ec) { int err; long long read_ec; struct ubi_ec_hdr *ec_hdr; + if (!ubi_dbg_chk_gen(ubi)) + return 0; + ec_hdr = kzalloc(ubi->ec_hdr_alsize, GFP_NOFS); if (!ec_hdr) return -ENOMEM; @@ -1627,10 +2074,10 @@ static int paranoid_check_ec(struct ubi_device *ubi, int pnum, int ec) } read_ec = be64_to_cpu(ec_hdr->ec); - if (ec != read_ec) { - ubi_err("paranoid check failed for PEB %d", pnum); + if (ec != read_ec && read_ec - ec > 1) { + ubi_err("self-check failed for PEB %d", pnum); ubi_err("read EC is %lld, should be %d", read_ec, ec); - ubi_dbg_dump_stack(); + dump_stack(); err = 1; } else err = 0; @@ -1641,24 +2088,53 @@ out_free: } /** - * paranoid_check_in_wl_tree - make sure that a wear-leveling entry is present - * in a WL RB-tree. + * self_check_in_wl_tree - check that wear-leveling entry is in WL RB-tree. + * @ubi: UBI device description object * @e: the wear-leveling entry to check * @root: the root of the tree * - * This function returns zero if @e is in the @root RB-tree and %1 if it + * This function returns zero if @e is in the @root RB-tree and %-EINVAL if it * is not. */ -static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e, - struct rb_root *root) +static int self_check_in_wl_tree(const struct ubi_device *ubi, + struct ubi_wl_entry *e, struct rb_root *root) { + if (!ubi_dbg_chk_gen(ubi)) + return 0; + if (in_wl_tree(e, root)) return 0; - ubi_err("paranoid check failed for PEB %d, EC %d, RB-tree %p ", + ubi_err("self-check failed for PEB %d, EC %d, RB-tree %p ", e->pnum, e->ec, root); - ubi_dbg_dump_stack(); - return 1; + dump_stack(); + return -EINVAL; } -#endif /* CONFIG_MTD_UBI_DEBUG_PARANOID */ +/** + * self_check_in_pq - check if wear-leveling entry is in the protection + * queue. + * @ubi: UBI device description object + * @e: the wear-leveling entry to check + * + * This function returns zero if @e is in @ubi->pq and %-EINVAL if it is not. + */ +static int self_check_in_pq(const struct ubi_device *ubi, + struct ubi_wl_entry *e) +{ + struct ubi_wl_entry *p; + int i; + + if (!ubi_dbg_chk_gen(ubi)) + return 0; + + for (i = 0; i < UBI_PROT_QUEUE_LEN; ++i) + list_for_each_entry(p, &ubi->pq[i], u.list) + if (p == e) + return 0; + + ubi_err("self-check failed for PEB %d, EC %d, Protect queue", + e->pnum, e->ec); + dump_stack(); + return -EINVAL; +} |