[RFC] ubihealthd

[RFC] ubihealthd

[Richard Weinberger]

ubihealthd is a tiny C program which takes care of your NAND.
It will trigger re-reads and scrubbing such that read-disturb and
data retention will be addressed before data is lost.
Currently the policy is rather trivial. It re-reads every PEB within
a given time frame, same for scrubbing and if a PEB's read counter exceeds
a given threshold it will also trigger a re-read.

At ELCE some people asked why this is done in userspace.
The reason is that this is a classical example of kernel offers mechanism
and userspace the policy. Also ubihealthd is not mandatory.
Depending on your NAND it can help you increasing its lifetime.
But you won't lose data immediately if it does not run for a while.
It is something like smartd is for hard disks.
I did this also in kernel space and it was messy.

[PATCH 1/4] Add kernel style linked lists
[PATCH 2/4] Include new ioctls and struct in ubi-user.h
[PATCH 3/4] Initial implementation for ubihealthd.
[PATCH 4/4] Documentation for ubihealthd

[PATCH 1/4] Add kernel style linked lists

[Richard Weinberger]

From: Daniel Walter <dwalter@sigma-star.at>

Signed-off-by: Daniel Walter <dwalter@sigma-star.at>
Signed-off-by: Richard Weinberger <richard@nod.at>
---
 include/list.h | 611 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 611 insertions(+)
 create mode 100644 include/list.h

diff --git a/include/list.h b/include/list.h
new file mode 100644
index 0000000..e8d58eb
--- /dev/null
+++ b/include/list.h
@@ -0,0 +1,611 @@
+/*
+ * 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 _LINUX_LIST_H
+#define _LINUX_LIST_H
+
+struct list_head {
+	struct list_head *next, *prev;
+};
+
+#define LIST_POISON1  ((struct list_head *) 0x00100100)
+#define LIST_POISON2  ((struct list_head *) 0x00200200)
+
+#define _offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
+
+#define container_of(ptr, type, member) ({                      \
+        const typeof( ((type *)0)->member ) *__mptr = (ptr);    \
+               (type *)( (char *)__mptr - _offsetof(type,member) );})
+
+/*
+ * Simple doubly linked list implementation.
+ *
+ * Some of the internal functions ("__xxx") are useful when
+ * manipulating whole lists rather than single entries, as
+ * sometimes we already know the next/prev entries and we can
+ * generate better code by using them directly rather than
+ * using the generic single-entry routines.
+ */
+
+#define LIST_HEAD_INIT(name) { &(name), &(name) }
+
+#define LIST_HEAD(name) \
+	struct list_head name = LIST_HEAD_INIT(name)
+
+static inline void INIT_LIST_HEAD(struct list_head *list)
+{
+	list->next = list;
+	list->prev = list;
+}
+
+/*
+ * Insert a new entry between two known consecutive entries.
+ *
+ * This is only for internal list manipulation where we know
+ * the prev/next entries already!
+ */
+#ifndef CONFIG_DEBUG_LIST
+static inline void __list_add(struct list_head *new,
+			      struct list_head *prev,
+			      struct list_head *next)
+{
+	next->prev = new;
+	new->next = next;
+	new->prev = prev;
+	prev->next = new;
+}
+#else
+extern void __list_add(struct list_head *new,
+			      struct list_head *prev,
+			      struct list_head *next);
+#endif
+
+/**
+ * list_add - add a new entry
+ * @new: new entry to be added
+ * @head: list head to add it after
+ *
+ * Insert a new entry after the specified head.
+ * This is good for implementing stacks.
+ */
+static inline void list_add(struct list_head *new, struct list_head *head)
+{
+	__list_add(new, head, head->next);
+}
+
+
+/**
+ * list_add_tail - add a new entry
+ * @new: new entry to be added
+ * @head: list head to add it before
+ *
+ * Insert a new entry before the specified head.
+ * This is useful for implementing queues.
+ */
+static inline void list_add_tail(struct list_head *new, struct list_head *head)
+{
+	__list_add(new, head->prev, head);
+}
+
+/*
+ * Delete a list entry by making the prev/next entries
+ * point to each other.
+ *
+ * This is only for internal list manipulation where we know
+ * the prev/next entries already!
+ */
+static inline void __list_del(struct list_head * prev, struct list_head * next)
+{
+	next->prev = prev;
+	prev->next = next;
+}
+
+/**
+ * list_del - deletes entry from list.
+ * @entry: the element to delete from the list.
+ * Note: list_empty() on entry does not return true after this, the entry is
+ * in an undefined state.
+ */
+#ifndef CONFIG_DEBUG_LIST
+static inline void __list_del_entry(struct list_head *entry)
+{
+	__list_del(entry->prev, entry->next);
+}
+
+static inline void list_del(struct list_head *entry)
+{
+	__list_del(entry->prev, entry->next);
+	entry->next = LIST_POISON1;
+	entry->prev = LIST_POISON2;
+}
+#else
+extern void __list_del_entry(struct list_head *entry);
+extern void list_del(struct list_head *entry);
+#endif
+
+/**
+ * list_replace - replace old entry by new one
+ * @old : the element to be replaced
+ * @new : the new element to insert
+ *
+ * If @old was empty, it will be overwritten.
+ */
+static inline void list_replace(struct list_head *old,
+				struct list_head *new)
+{
+	new->next = old->next;
+	new->next->prev = new;
+	new->prev = old->prev;
+	new->prev->next = new;
+}
+
+static inline void list_replace_init(struct list_head *old,
+					struct list_head *new)
+{
+	list_replace(old, new);
+	INIT_LIST_HEAD(old);
+}
+
+/**
+ * list_del_init - deletes entry from list and reinitialize it.
+ * @entry: the element to delete from the list.
+ */
+static inline void list_del_init(struct list_head *entry)
+{
+	__list_del_entry(entry);
+	INIT_LIST_HEAD(entry);
+}
+
+/**
+ * list_move - delete from one list and add as another's head
+ * @list: the entry to move
+ * @head: the head that will precede our entry
+ */
+static inline void list_move(struct list_head *list, struct list_head *head)
+{
+	__list_del_entry(list);
+	list_add(list, head);
+}
+
+/**
+ * list_move_tail - delete from one list and add as another's tail
+ * @list: the entry to move
+ * @head: the head that will follow our entry
+ */
+static inline void list_move_tail(struct list_head *list,
+				  struct list_head *head)
+{
+	__list_del_entry(list);
+	list_add_tail(list, head);
+}
+
+/**
+ * list_is_last - tests whether @list is the last entry in list @head
+ * @list: the entry to test
+ * @head: the head of the list
+ */
+static inline int list_is_last(const struct list_head *list,
+				const struct list_head *head)
+{
+	return list->next == head;
+}
+
+/**
+ * list_empty - tests whether a list is empty
+ * @head: the list to test.
+ */
+static inline int list_empty(const struct list_head *head)
+{
+	return head->next == head;
+}
+
+/**
+ * list_empty_careful - tests whether a list is empty and not being modified
+ * @head: the list to test
+ *
+ * Description:
+ * tests whether a list is empty _and_ checks that no other CPU might be
+ * in the process of modifying either member (next or prev)
+ *
+ * NOTE: using list_empty_careful() without synchronization
+ * can only be safe if the only activity that can happen
+ * to the list entry is list_del_init(). Eg. it cannot be used
+ * if another CPU could re-list_add() it.
+ */
+static inline int list_empty_careful(const struct list_head *head)
+{
+	struct list_head *next = head->next;
+	return (next == head) && (next == head->prev);
+}
+
+/**
+ * list_rotate_left - rotate the list to the left
+ * @head: the head of the list
+ */
+static inline void list_rotate_left(struct list_head *head)
+{
+	struct list_head *first;
+
+	if (!list_empty(head)) {
+		first = head->next;
+		list_move_tail(first, head);
+	}
+}
+
+/**
+ * list_is_singular - tests whether a list has just one entry.
+ * @head: the list to test.
+ */
+static inline int list_is_singular(const struct list_head *head)
+{
+	return !list_empty(head) && (head->next == head->prev);
+}
+
+static inline void __list_cut_position(struct list_head *list,
+		struct list_head *head, struct list_head *entry)
+{
+	struct list_head *new_first = entry->next;
+	list->next = head->next;
+	list->next->prev = list;
+	list->prev = entry;
+	entry->next = list;
+	head->next = new_first;
+	new_first->prev = head;
+}
+
+/**
+ * list_cut_position - cut a list into two
+ * @list: a new list to add all removed entries
+ * @head: a list with entries
+ * @entry: an entry within head, could be the head itself
+ *	and if so we won't cut the list
+ *
+ * This helper moves the initial part of @head, up to and
+ * including @entry, from @head to @list. You should
+ * pass on @entry an element you know is on @head. @list
+ * should be an empty list or a list you do not care about
+ * losing its data.
+ *
+ */
+static inline void list_cut_position(struct list_head *list,
+		struct list_head *head, struct list_head *entry)
+{
+	if (list_empty(head))
+		return;
+	if (list_is_singular(head) &&
+		(head->next != entry && head != entry))
+		return;
+	if (entry == head)
+		INIT_LIST_HEAD(list);
+	else
+		__list_cut_position(list, head, entry);
+}
+
+static inline void __list_splice(const struct list_head *list,
+				 struct list_head *prev,
+				 struct list_head *next)
+{
+	struct list_head *first = list->next;
+	struct list_head *last = list->prev;
+
+	first->prev = prev;
+	prev->next = first;
+
+	last->next = next;
+	next->prev = last;
+}
+
+/**
+ * list_splice - join two lists, this is designed for stacks
+ * @list: the new list to add.
+ * @head: the place to add it in the first list.
+ */
+static inline void list_splice(const struct list_head *list,
+				struct list_head *head)
+{
+	if (!list_empty(list))
+		__list_splice(list, head, head->next);
+}
+
+/**
+ * list_splice_tail - join two lists, each list being a queue
+ * @list: the new list to add.
+ * @head: the place to add it in the first list.
+ */
+static inline void list_splice_tail(struct list_head *list,
+				struct list_head *head)
+{
+	if (!list_empty(list))
+		__list_splice(list, head->prev, head);
+}
+
+/**
+ * list_splice_init - join two lists and reinitialise the emptied list.
+ * @list: the new list to add.
+ * @head: the place to add it in the first list.
+ *
+ * The list at @list is reinitialised
+ */
+static inline void list_splice_init(struct list_head *list,
+				    struct list_head *head)
+{
+	if (!list_empty(list)) {
+		__list_splice(list, head, head->next);
+		INIT_LIST_HEAD(list);
+	}
+}
+
+/**
+ * list_splice_tail_init - join two lists and reinitialise the emptied list
+ * @list: the new list to add.
+ * @head: the place to add it in the first list.
+ *
+ * Each of the lists is a queue.
+ * The list at @list is reinitialised
+ */
+static inline void list_splice_tail_init(struct list_head *list,
+					 struct list_head *head)
+{
+	if (!list_empty(list)) {
+		__list_splice(list, head->prev, head);
+		INIT_LIST_HEAD(list);
+	}
+}
+
+/**
+ * list_entry - get the struct for this entry
+ * @ptr:	the &struct list_head pointer.
+ * @type:	the type of the struct this is embedded in.
+ * @member:	the name of the list_head within the struct.
+ */
+#define list_entry(ptr, type, member) \
+	container_of(ptr, type, member)
+
+/**
+ * list_first_entry - get the first element from a list
+ * @ptr:	the list head to take the element from.
+ * @type:	the type of the struct this is embedded in.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Note, that list is expected to be not empty.
+ */
+#define list_first_entry(ptr, type, member) \
+	list_entry((ptr)->next, type, member)
+
+/**
+ * list_last_entry - get the last element from a list
+ * @ptr:	the list head to take the element from.
+ * @type:	the type of the struct this is embedded in.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Note, that list is expected to be not empty.
+ */
+#define list_last_entry(ptr, type, member) \
+	list_entry((ptr)->prev, type, member)
+
+/**
+ * list_first_entry_or_null - get the first element from a list
+ * @ptr:	the list head to take the element from.
+ * @type:	the type of the struct this is embedded in.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Note that if the list is empty, it returns NULL.
+ */
+#define list_first_entry_or_null(ptr, type, member) \
+	(!list_empty(ptr) ? list_first_entry(ptr, type, member) : NULL)
+
+/**
+ * list_next_entry - get the next element in list
+ * @pos:	the type * to cursor
+ * @member:	the name of the list_head within the struct.
+ */
+#define list_next_entry(pos, member) \
+	list_entry((pos)->member.next, typeof(*(pos)), member)
+
+/**
+ * list_prev_entry - get the prev element in list
+ * @pos:	the type * to cursor
+ * @member:	the name of the list_head within the struct.
+ */
+#define list_prev_entry(pos, member) \
+	list_entry((pos)->member.prev, typeof(*(pos)), member)
+
+/**
+ * list_for_each	-	iterate over a list
+ * @pos:	the &struct list_head to use as a loop cursor.
+ * @head:	the head for your list.
+ */
+#define list_for_each(pos, head) \
+	for (pos = (head)->next; pos != (head); pos = pos->next)
+
+/**
+ * list_for_each_prev	-	iterate over a list backwards
+ * @pos:	the &struct list_head to use as a loop cursor.
+ * @head:	the head for your list.
+ */
+#define list_for_each_prev(pos, head) \
+	for (pos = (head)->prev; pos != (head); pos = pos->prev)
+
+/**
+ * list_for_each_safe - iterate over a list safe against removal of list entry
+ * @pos:	the &struct list_head to use as a loop cursor.
+ * @n:		another &struct list_head to use as temporary storage
+ * @head:	the head for your list.
+ */
+#define list_for_each_safe(pos, n, head) \
+	for (pos = (head)->next, n = pos->next; pos != (head); \
+		pos = n, n = pos->next)
+
+/**
+ * list_for_each_prev_safe - iterate over a list backwards safe against removal of list entry
+ * @pos:	the &struct list_head to use as a loop cursor.
+ * @n:		another &struct list_head to use as temporary storage
+ * @head:	the head for your list.
+ */
+#define list_for_each_prev_safe(pos, n, head) \
+	for (pos = (head)->prev, n = pos->prev; \
+	     pos != (head); \
+	     pos = n, n = pos->prev)
+
+/**
+ * list_for_each_entry	-	iterate over list of given type
+ * @pos:	the type * to use as a loop cursor.
+ * @head:	the head for your list.
+ * @member:	the name of the list_head within the struct.
+ */
+#define list_for_each_entry(pos, head, member)				\
+	for (pos = list_first_entry(head, typeof(*pos), member);	\
+	     &pos->member != (head);					\
+	     pos = list_next_entry(pos, member))
+
+/**
+ * list_for_each_entry_reverse - iterate backwards over list of given type.
+ * @pos:	the type * to use as a loop cursor.
+ * @head:	the head for your list.
+ * @member:	the name of the list_head within the struct.
+ */
+#define list_for_each_entry_reverse(pos, head, member)			\
+	for (pos = list_last_entry(head, typeof(*pos), member);		\
+	     &pos->member != (head); 					\
+	     pos = list_prev_entry(pos, member))
+
+/**
+ * list_prepare_entry - prepare a pos entry for use in list_for_each_entry_continue()
+ * @pos:	the type * to use as a start point
+ * @head:	the head of the list
+ * @member:	the name of the list_head within the struct.
+ *
+ * Prepares a pos entry for use as a start point in list_for_each_entry_continue().
+ */
+#define list_prepare_entry(pos, head, member) \
+	((pos) ? : list_entry(head, typeof(*pos), member))
+
+/**
+ * list_for_each_entry_continue - continue iteration over list of given type
+ * @pos:	the type * to use as a loop cursor.
+ * @head:	the head for your list.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Continue to iterate over list of given type, continuing after
+ * the current position.
+ */
+#define list_for_each_entry_continue(pos, head, member) 		\
+	for (pos = list_next_entry(pos, member);			\
+	     &pos->member != (head);					\
+	     pos = list_next_entry(pos, member))
+
+/**
+ * list_for_each_entry_continue_reverse - iterate backwards from the given point
+ * @pos:	the type * to use as a loop cursor.
+ * @head:	the head for your list.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Start to iterate over list of given type backwards, continuing after
+ * the current position.
+ */
+#define list_for_each_entry_continue_reverse(pos, head, member)		\
+	for (pos = list_prev_entry(pos, member);			\
+	     &pos->member != (head);					\
+	     pos = list_prev_entry(pos, member))
+
+/**
+ * list_for_each_entry_from - iterate over list of given type from the current point
+ * @pos:	the type * to use as a loop cursor.
+ * @head:	the head for your list.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Iterate over list of given type, continuing from current position.
+ */
+#define list_for_each_entry_from(pos, head, member) 			\
+	for (; &pos->member != (head);					\
+	     pos = list_next_entry(pos, member))
+
+/**
+ * list_for_each_entry_safe - iterate over list of given type safe against removal of list entry
+ * @pos:	the type * to use as a loop cursor.
+ * @n:		another type * to use as temporary storage
+ * @head:	the head for your list.
+ * @member:	the name of the list_head within the struct.
+ */
+#define list_for_each_entry_safe(pos, n, head, member)			\
+	for (pos = list_first_entry(head, typeof(*pos), member),	\
+		n = list_next_entry(pos, member);			\
+	     &pos->member != (head); 					\
+	     pos = n, n = list_next_entry(n, member))
+
+/**
+ * list_for_each_entry_safe_continue - continue list iteration safe against removal
+ * @pos:	the type * to use as a loop cursor.
+ * @n:		another type * to use as temporary storage
+ * @head:	the head for your list.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Iterate over list of given type, continuing after current point,
+ * safe against removal of list entry.
+ */
+#define list_for_each_entry_safe_continue(pos, n, head, member) 		\
+	for (pos = list_next_entry(pos, member), 				\
+		n = list_next_entry(pos, member);				\
+	     &pos->member != (head);						\
+	     pos = n, n = list_next_entry(n, member))
+
+/**
+ * list_for_each_entry_safe_from - iterate over list from current point safe against removal
+ * @pos:	the type * to use as a loop cursor.
+ * @n:		another type * to use as temporary storage
+ * @head:	the head for your list.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Iterate over list of given type from current point, safe against
+ * removal of list entry.
+ */
+#define list_for_each_entry_safe_from(pos, n, head, member) 			\
+	for (n = list_next_entry(pos, member);					\
+	     &pos->member != (head);						\
+	     pos = n, n = list_next_entry(n, member))
+
+/**
+ * list_for_each_entry_safe_reverse - iterate backwards over list safe against removal
+ * @pos:	the type * to use as a loop cursor.
+ * @n:		another type * to use as temporary storage
+ * @head:	the head for your list.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Iterate backwards over list of given type, safe against removal
+ * of list entry.
+ */
+#define list_for_each_entry_safe_reverse(pos, n, head, member)		\
+	for (pos = list_last_entry(head, typeof(*pos), member),		\
+		n = list_prev_entry(pos, member);			\
+	     &pos->member != (head); 					\
+	     pos = n, n = list_prev_entry(n, member))
+
+/**
+ * list_safe_reset_next - reset a stale list_for_each_entry_safe loop
+ * @pos:	the loop cursor used in the list_for_each_entry_safe loop
+ * @n:		temporary storage used in list_for_each_entry_safe
+ * @member:	the name of the list_head within the struct.
+ *
+ * list_safe_reset_next is not safe to use in general if the list may be
+ * modified concurrently (eg. the lock is dropped in the loop body). An
+ * exception to this is if the cursor element (pos) is pinned in the list,
+ * and list_safe_reset_next is called after re-taking the lock and before
+ * completing the current iteration of the loop body.
+ */
+#define list_safe_reset_next(pos, n, member)				\
+	n = list_next_entry(pos, member)
+
+#endif
-- 
2.5.0

[PATCH 2/4] Include new ioctls and struct in ubi-user.h

[Richard Weinberger]

From: Daniel Walter <dwalter@sigma-star.at>

Add ioctls and struct definitions for ubi
statistic interface and PEB read/scrub calls

Signed-off-by: Daniel Walter <dwalter@sigma-star.at>
Signed-off-by: Richard Weinberger <richard@nod.at>
---
 include/mtd/ubi-user.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/include/mtd/ubi-user.h b/include/mtd/ubi-user.h
index 2b50dad..f805d89 100644
--- a/include/mtd/ubi-user.h
+++ b/include/mtd/ubi-user.h
@@ -168,6 +168,10 @@
 /* Re-name volumes */
 #define UBI_IOCRNVOL _IOW(UBI_IOC_MAGIC, 3, struct ubi_rnvol_req)
 
+#define UBI_IOCRPEB _IOW(UBI_IOC_MAGIC, 4, int32_t)
+#define UBI_IOCSPEB _IOW(UBI_IOC_MAGIC, 5, int32_t)
+#define UBI_IOCSTATS _IOW(UBI_IOC_MAGIC, 6, struct ubi_stats_req)
+
 /* ioctl commands of the UBI control character device */
 
 #define UBI_CTRL_IOC_MAGIC 'o'
@@ -437,4 +441,18 @@ struct ubi_blkcreate_req {
 	int8_t  padding[128];
 }  __attribute__((packed));
 
+struct ubi_stats_entry {
+       int32_t pnum;
+       int32_t ec;
+       int32_t rc;
+       int32_t padding;
+} __attribute__((packed));
+
+struct ubi_stats_req {
+       int32_t req_len;
+       int32_t req_pnum;
+       int32_t padding[2];
+       struct ubi_stats_entry stats[0];
+} __attribute__((packed));
+
 #endif /* __UBI_USER_H__ */
-- 
2.5.0

[PATCH 3/4] Initial implementation for ubihealthd.

[Richard Weinberger]

From: Daniel Walter <dwalter@sigma-star.at>

ubihealthd will read / scrub all PEBs
of a given ubi device over a given amount
of time. This should detect and fix common
errors on various mtds.

Signed-off-by: Daniel Walter <dwalter@sigma-star.at>
Signed-off-by: Richard Weinberger <richard@nod.at>
---
 Makefile               |   3 +-
 ubi-utils/.gitignore   |   1 +
 ubi-utils/ubihealthd.c | 686 +++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 689 insertions(+), 1 deletion(-)
 create mode 100644 ubi-utils/ubihealthd.c

diff --git a/Makefile b/Makefile
index 3ce8587..9556da9 100644
--- a/Makefile
+++ b/Makefile
@@ -28,7 +28,8 @@ MTD_BINS = \
 	sumtool jffs2reader
 UBI_BINS = \
 	ubiupdatevol ubimkvol ubirmvol ubicrc32 ubinfo ubiattach \
-	ubidetach ubinize ubiformat ubirename mtdinfo ubirsvol ubiblock
+	ubidetach ubinize ubiformat ubirename mtdinfo ubirsvol ubiblock \
+	ubihealthd
 
 BINS = $(MTD_BINS)
 BINS += mkfs.ubifs/mkfs.ubifs
diff --git a/ubi-utils/.gitignore b/ubi-utils/.gitignore
index 19653a8..80edc23 100644
--- a/ubi-utils/.gitignore
+++ b/ubi-utils/.gitignore
@@ -11,3 +11,4 @@
 /ubirsvol
 /ubiblock
 /mtdinfo
+/ubihealthd
diff --git a/ubi-utils/ubihealthd.c b/ubi-utils/ubihealthd.c
new file mode 100644
index 0000000..97ffbac
--- /dev/null
+++ b/ubi-utils/ubihealthd.c
@@ -0,0 +1,686 @@
+#include <stdio.h>
+#include <stdarg.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <stdint.h>
+#include <sys/ioctl.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <string.h>
+#include <errno.h>
+#include <time.h>
+#include <mtd/ubi-user.h>
+#include <sys/signalfd.h>
+#include <signal.h>
+#include <poll.h>
+#include <sys/timerfd.h>
+#include <inttypes.h>
+
+#include <getopt.h>
+#include <libubi.h>
+
+#include "list.h"
+
+#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+#ifndef NDEBUG
+#define _log(lvl, M, ...) __log(lvl, "[%s:%d] " M, __FILE__, __LINE__, ##__VA_ARGS__);
+#else
+#define _log(lvl, M, ...) __log(lvl, M, ##__VA_ARGS__);
+#endif
+#define log(M, ...) _log(2, M, ##__VA_ARGS__);
+#define log_fatal(M, ...) _log(0, "[FATAL]" M, ##__VA_ARGS__);
+#define log_err(M, ...) _log(1, "[ERR]" M, ##__VA_ARGS__);
+#define log_warn(M, ...) _log(2, "[WARN]" M, ##__VA_ARGS__);
+#define log_info(M, ...) _log(3, "[INFO]" M, ##__VA_ARGS__);
+#define log_debug(M, ...) _log(4, "[DEBUG]" M, ##__VA_ARGS__);
+
+
+int log_level;
+
+static void __log(int level, const char *fmt, ...)
+{
+	va_list ap;
+	if (level > log_level)
+		return;
+	va_start(ap, fmt);
+	vfprintf(stderr, fmt, ap);
+	va_end(ap);
+	fprintf(stderr, "\n");
+}
+
+static const uint64_t UBIHEALTHD_MAGIC_VERSION = 0x00000001LLU;
+
+/*
+ * Basic algorithm:
+ *  - get number of PEBs and identify sleep times between scheduling
+ *  - read stats to identify hotspots (schedule full block read if identified as such)
+ *  - read PEB and remove from list (move to tail ?)
+ */
+
+static const char opt_string[] = "d:f:hr:s:x:v:";
+static const struct option options[] = {
+	{
+		.name = "device",
+		.has_arg = required_argument,
+		.flag = NULL,
+		.val = 'd'
+	},
+	{
+		.name = "file",
+		.has_arg = required_argument,
+		.flag = NULL,
+		.val = 'f'
+	},
+	{
+		.name = "read_complete",
+		.has_arg = required_argument,
+		.flag = NULL,
+		.val = 'r'
+	},
+	{
+		.name = "scrub_complete",
+		.has_arg = required_argument,
+		.flag = NULL,
+		.val = 's'
+	},
+	{
+		.name = "read_threshold",
+		.has_arg = required_argument,
+		.flag = NULL,
+		.val = 'x'
+	},
+	{
+		.name = "help",
+		.has_arg = no_argument,
+		.flag = NULL,
+		.val = 'h'
+	},
+	{
+		.name = "verbosity",
+		.has_arg = required_argument,
+		.flag = NULL,
+		.val = 'v'
+	}
+};
+
+struct peb_info {
+	int64_t peb_num;
+	uint64_t err_cnt;
+	uint64_t read_cnt;
+	uint64_t prev_read_cnt;
+	time_t last_stat_update;
+	time_t last_read;
+	time_t last_err;
+} __attribute__((packed));
+
+typedef enum {
+	SCHED_READ,
+	SCHED_SCRUB
+} sched_type;
+
+struct sched_peb {
+	struct peb_info *peb;
+	sched_type type;
+	struct list_head list;
+};
+
+struct peb_list {
+	struct peb_info *peb;
+	struct list_head list;
+};
+
+static const char *help_str = \
+"[OPTIONS]\n" \
+"  -h, --help\t\tShow this message and exit\n" \
+"  -d, --device\t\tDevice to be monitored (default: /dev/ubi0)\n" \
+"  -f, --file\t\tPath to statistics save file\n" \
+"  -r, --read_complete\tTimeframe for reading all PEBs in seconds\n" \
+"  -s, --scrub_complete\tTimeframe for scrubbing all PEBs in seconds\n" \
+"  -x, --read_threshold\tNumber of reads between two stats updates\n" \
+"                      \twhich will trigger a PEB read\n" \
+"  -v, --verbosity\t\tlog level (0-4)\n";
+
+static void usage(const char* progname)
+{
+	printf("usage: %s [OPTIONS]", progname);
+	printf("\n%s\n", help_str);
+	_exit(1);
+}
+
+static int64_t get_num_pebs(const char *ubi_dev)
+{
+	libubi_t libubi = libubi_open();
+	struct ubi_dev_info dev_info;
+	int err;
+	err = ubi_get_dev_info(libubi, ubi_dev, &dev_info);
+	if (err) {
+		log_err("Could not get ubi info for device %s", ubi_dev);
+		return -1;
+	}
+	libubi_close(libubi);
+	return dev_info.total_lebs;
+}
+
+static int write_stats_file(const char *filename, struct peb_list *peb_head, struct sched_peb *sched_read_head, struct sched_peb *sched_scrub_head, int pnum)
+{
+	int64_t next_read_peb = 0;
+	int64_t next_scrub_peb = 0;
+	struct peb_info *peb = NULL;
+	struct peb_list *tmpp = NULL;
+	struct sched_peb *p = NULL;
+	FILE *file = fopen(filename, "wb");
+	if (file == NULL)
+		return -1;
+	p = list_first_entry_or_null(&sched_read_head->list, struct sched_peb, list);
+	if (p)
+		next_read_peb = p->peb->peb_num;
+	p = list_first_entry_or_null(&sched_scrub_head->list, struct sched_peb, list);
+	if (p)
+		next_scrub_peb = p->peb->peb_num;
+	fwrite(&UBIHEALTHD_MAGIC_VERSION, sizeof(UBIHEALTHD_MAGIC_VERSION), 1, file);
+	fwrite(&pnum, sizeof(pnum), 1, file);
+	fwrite(&next_read_peb, sizeof(next_read_peb), 1, file);
+	fwrite(&next_scrub_peb, sizeof(next_scrub_peb), 1, file);
+	list_for_each_entry(tmpp, &peb_head->list, list) {
+		peb = tmpp->peb;
+		fwrite(peb, sizeof(struct peb_info), 1, file);
+	}
+	fclose(file);
+	return 0;
+}
+
+
+static int init_stats(int fd, struct list_head *head, int pnum)
+{
+	int i, err = 0;
+	size_t req_size = pnum * sizeof(struct ubi_stats_entry);
+	struct ubi_stats_req *req = malloc(sizeof(struct ubi_stats_req) + req_size);
+	if (!req) {
+		log_err("Could not alloc ubi_stats_req: %s", strerror(errno));
+		return -1;
+	}
+	req->req_len = req_size + sizeof(struct ubi_stats_req);
+	req->req_pnum = -1;
+	err = ioctl(fd, UBI_IOCSTATS, req);
+	if (err < 0) {
+		log_err("Could not init stats via ioctl: %s", strerror(errno));
+		free(req);
+		return -1;
+	}
+	log_info("Kernel reported stats for %d PEBs", err);
+	struct peb_info *peb = NULL;
+	struct peb_list *p = NULL;
+	time_t now = time(NULL);
+	for (i = 0; i < err; i++) {
+		struct ubi_stats_entry *s = &req->stats[i];
+		peb = malloc(sizeof(struct peb_info));
+		if (!peb) {
+			log_err("Could not alloc peb_info");
+			free(req);
+			return -1;
+		}
+		peb->peb_num = s->pnum;
+		peb->err_cnt = s->ec;
+		peb->read_cnt = s->rc;
+		peb->prev_read_cnt = s->rc;
+		peb->last_stat_update = now;
+		p = malloc(sizeof(struct peb_list));
+		if (!p) {
+			log_err("Could not alloc peb_list element");
+			free(req);
+			return -1;
+		}
+		p->peb = peb;
+		list_add_tail(&p->list, head);
+	}
+	free(req);
+	return 0;
+}
+
+static void free_list(struct peb_list *head)
+{
+	if (list_empty(&head->list))
+		return;
+	struct peb_list *p = NULL;
+	struct peb_list *tmp = NULL;
+	list_for_each_entry_safe(p, tmp, &head->list, list) {
+		list_del(&p->list);
+		free(p->peb);
+		free(p);
+	}
+}
+
+static int update_stats(int fd, struct peb_list *head, int pnum)
+{
+	if (list_empty(&head->list)) {
+		log_fatal("PEB list not initialized");
+		return -1;
+	}
+	int i, err = 0;
+	size_t req_size = pnum * sizeof(struct ubi_stats_entry);
+	struct ubi_stats_req *req = malloc(sizeof(struct ubi_stats_req) + req_size);
+	if (!req) {
+		log_err("Could not alloc ubi_stats_req: %s", strerror(errno));
+		return -1;
+	}
+	memset(req, 0, sizeof(struct ubi_stats_req));
+	req->req_len = req_size + sizeof(struct ubi_stats_req);
+	req->req_pnum = -1;
+	log_debug("req_len: %d, req_pnum: %d", req->req_len, req->req_pnum);
+	err = ioctl(fd, UBI_IOCSTATS, req);
+	if (err < 0) {
+		log_err("Could not get stats for PEBs, [%d] %s", errno, strerror(errno));
+		free(req);
+		return -1;
+	}
+	log_debug("Kernel reported stats for %d PEBs", err);
+	time_t now = time(NULL);
+	for (i = 0; i < err; i++) {
+		struct ubi_stats_entry *s = &req->stats[i];
+		struct peb_list *p = NULL;
+		struct peb_info *peb = NULL;
+		list_for_each_entry(p, &head->list, list) {
+			if (p->peb && (p->peb->peb_num == s->pnum)) {
+				peb = p->peb;
+				break;
+			}
+		}
+		if (!peb) {
+			log_warn("Could not get stats for PEB %d", i);
+			continue;
+		}
+		/* TODO(sahne): check for overflow ! */
+		peb->err_cnt = s->ec;
+		peb->prev_read_cnt = peb->read_cnt;
+		peb->read_cnt = s->rc;
+		/* check if peb was erased (read_cnt would be reset to 0 if it was) */
+		if (peb->read_cnt < peb->prev_read_cnt)
+			peb->prev_read_cnt = peb->read_cnt;
+		peb->last_stat_update = now;
+	}
+	free(req);
+	return 0;
+}
+
+static int read_peb(int fd, struct peb_info *peb)
+{
+	time_t now = time(NULL);
+	log_debug("Reading PEB %"PRIu64 , peb->peb_num);
+	int err = ioctl(fd, UBI_IOCRPEB, &peb->peb_num);
+	if (err < 0) {
+		log_err("Error while reading PEB %" PRIu64, peb->peb_num);
+		return -1;
+	}
+	peb->last_read = now;
+	return 0;
+}
+
+static int scrub_peb(int fd, struct peb_info *peb)
+{
+	time_t now = time(NULL);
+	log_debug("Scrubbing PEB %"PRIu64, peb->peb_num);
+	int err = ioctl (fd, UBI_IOCSPEB, &peb->peb_num);
+	if (err < 0) {
+		log_err("Error while scrubbing PEB %" PRIu64, peb->peb_num);
+		return -1;
+	}
+	peb->last_read = now;
+	return 0;
+}
+
+static int schedule_peb(struct list_head *sched_list, struct peb_info *peb, sched_type type)
+{
+	struct sched_peb *s = malloc(sizeof(struct sched_peb));
+	if (!s) {
+		log_err("Could not allocate memory");
+		return -1;
+	}
+	s->peb = peb;
+	s->type = type;
+	list_add_tail(&s->list, sched_list);
+	return 0;
+}
+
+static int work(struct sched_peb *sched_list, int fd)
+{
+	if (list_empty(&sched_list->list))
+		return 0;
+	struct sched_peb *sched = list_first_entry(&sched_list->list, struct sched_peb, list);
+	struct peb_info *peb = sched->peb;
+	if (peb == NULL) {
+		log_warn("invalid peb");
+		return -1;
+	}
+	/* delete entry from list, we will add it if needed */
+	list_del(&sched->list);
+	switch(sched->type) {
+	case SCHED_READ:
+		read_peb(fd, peb);
+		break;
+	case SCHED_SCRUB:
+		scrub_peb(fd, peb);
+		break;
+	default:
+		log_warn("Unknown work type: %d", sched->type);
+		free(sched);
+		return -1;
+	}
+	/* reschedule PEB */
+	/* TODO(sahne): check error read/scrub in case PEB went bad (so we don't reschedule it) */
+	schedule_peb(&sched_list->list, peb, sched->type);
+	free(sched);
+	return 1;
+}
+
+static int create_and_arm_timer(int seconds)
+{
+	int tfd = timerfd_create(CLOCK_MONOTONIC, TFD_NONBLOCK | TFD_CLOEXEC);
+	if (tfd < 0) {
+		log_err("Could not create timer");
+		return -1;
+	}
+	struct itimerspec tspec = {
+		.it_interval = {
+			.tv_sec = seconds,
+			.tv_nsec = 0,
+		},
+		.it_value = {
+			.tv_sec = 0,
+			.tv_nsec = 1,
+		},
+	};
+	if (timerfd_settime(tfd, 0, &tspec, NULL) < 0) {
+		log_err("Could not arm timer");
+		close(tfd);
+		return -1;
+	}
+
+	return tfd;
+}
+
+static int read_stats_file(const char *filename, struct peb_list *peb_head, struct sched_peb *sched_read_head, struct sched_peb *sched_scrub_head)
+{
+	int num_pebs = 0;
+	int64_t next_read_peb;
+	int64_t next_scrub_peb;
+	uint64_t magic_version;
+	FILE *file = fopen(filename, "rb");
+	ssize_t i;
+	if (file == NULL)
+		return -1;
+	fread(&magic_version, sizeof(magic_version), 1, file);
+	if (magic_version != UBIHEALTHD_MAGIC_VERSION) {
+		log_warn("Magic mismatching, aborting reading from stats file");
+		fclose(file);
+		return -1;
+	}
+	fread(&num_pebs, sizeof(num_pebs), 1, file);
+	fread(&next_read_peb, sizeof(next_read_peb), 1, file);
+	fread(&next_scrub_peb, sizeof(next_scrub_peb), 1, file);
+	for (i = 0; i < num_pebs; i++) {
+		struct peb_info *peb = malloc(sizeof(struct peb_info));
+		if (!peb) {
+			log_err("Could not allocate peb_info");
+			return -1;
+		}
+		struct peb_list *p = NULL;
+		fread(peb, sizeof(struct peb_info), 1, file);
+		list_for_each_entry(p, &peb_head->list, list) {
+			if (p->peb && (p->peb->peb_num == peb->peb_num)) {
+				free(p->peb);
+				p->peb = peb;
+			}
+		}
+	}
+	/* init read and scrub lists */
+	struct peb_list *p = NULL;
+	list_for_each_entry(p, &peb_head->list, list) {
+		if (p->peb->peb_num >= next_read_peb)
+			schedule_peb(&sched_read_head->list, p->peb, SCHED_READ);
+		if (p->peb->peb_num >= next_scrub_peb)
+			schedule_peb(&sched_scrub_head->list, p->peb, SCHED_SCRUB);
+	}
+	p = NULL;
+	list_for_each_entry(p, &peb_head->list, list) {
+		if (p->peb->peb_num < next_read_peb)
+			schedule_peb(&sched_read_head->list, p->peb, SCHED_READ);
+		if (p->peb->peb_num < next_scrub_peb)
+			schedule_peb(&sched_scrub_head->list, p->peb, SCHED_SCRUB);
+	}
+
+	return 0;
+}
+
+static int init_sigfd()
+{
+	int sigfd;
+	sigset_t mask;
+	sigemptyset(&mask);
+	sigaddset(&mask, SIGINT);
+	sigaddset(&mask, SIGQUIT);
+	sigaddset(&mask, SIGHUP);
+	sigaddset(&mask, SIGTERM);
+	sigaddset(&mask, SIGUSR1);
+	if (sigprocmask(SIG_BLOCK, &mask, NULL) == -1) {
+		log_warn("Could not init sigprocmask");
+		return -1;
+	}
+	sigfd = signalfd(-1, &mask, SFD_CLOEXEC | SFD_NONBLOCK);
+	if (sigfd < 0) {
+		log_warn("Could not init signal handling");
+		return -1;
+	}
+	return sigfd;
+}
+
+int main(int argc, char **argv)
+{
+	int c, i;
+	int64_t num_pebs;
+	time_t read_completion = 100000;
+	time_t scrub_completion = 1000000;
+	uint64_t read_threshold = 10000;
+	struct sched_peb *sched_read_head;
+	struct sched_peb *sched_scrub_head;
+	struct peb_list *peb_head;
+	const char *stats_file = "/tmp/ubihealth_stats";
+	const char *ubi_dev = "/dev/ubi0";
+	log_level = 4;
+
+	while ((c = getopt_long(argc, argv, opt_string, options, &i)) != -1) {
+		switch(c) {
+		case 'd':
+			ubi_dev = optarg;
+			break;
+		case 'h':
+			usage(argv[0]);
+			break;
+		case 'f':
+			stats_file = optarg;
+			break;
+		case 'r':
+			read_completion = atoi(optarg);
+			break;
+		case 's':
+			scrub_completion = atoi(optarg);
+			break;
+		case 'x':
+			read_threshold = atoi(optarg);
+			break;
+		case 'v':
+			log_level = atoi(optarg);
+			if (log_level < 0)
+				log_level = 0;
+			else if (log_level > 4)
+				log_level = 4;
+		case '?':
+		default:
+			break;
+
+		}
+	}
+	/* signal handling */
+	struct signalfd_siginfo fdsi;
+	int sigfd = init_sigfd();
+	if (sigfd < 0) {
+		log_fatal("Could not init signal handling, aborting");
+		_exit(EXIT_FAILURE);
+	}
+
+	/* init sched_list */
+	peb_head = malloc(sizeof(struct peb_list));
+	if (!peb_head) {
+		log_fatal("Could not allocate peb_list");
+		_exit(EXIT_FAILURE);
+	}
+	peb_head->peb = NULL;
+	INIT_LIST_HEAD(&peb_head->list);
+	sched_read_head = malloc(sizeof(struct sched_peb));
+	if (!sched_read_head) {
+		log_fatal("Could not allocate read scheduler");
+		_exit(EXIT_FAILURE);
+	}
+	INIT_LIST_HEAD(&sched_read_head->list);
+	sched_read_head->peb = NULL;
+	sched_scrub_head = malloc(sizeof(struct sched_peb));
+	if (!sched_read_head) {
+		log_fatal("Could not allocate scrub scheduler");
+		_exit(EXIT_FAILURE);
+	}
+	INIT_LIST_HEAD(&sched_scrub_head->list);
+	sched_scrub_head->peb = NULL;
+	int fd = open(ubi_dev, O_RDONLY);
+	if (fd < 0) {
+		log_fatal("Could not open device %s", ubi_dev);
+		return 1;
+	}
+
+	/* get peb info */
+	num_pebs = get_num_pebs(ubi_dev);
+	if (num_pebs < 1) {
+		log_err("Invalid number of PEBs");
+		return 1;
+	}
+	if (init_stats(fd, &peb_head->list, num_pebs) < 0) {
+		log_fatal("Could not init statistics, aborting");
+		_exit(EXIT_FAILURE);
+	}
+	/* init peb list */
+	log_debug("Number of PEBs: %" PRIu64, num_pebs);
+
+
+	if (read_stats_file(stats_file, peb_head, sched_read_head, sched_scrub_head) < 0) {
+		log_warn("Could not init stats from file %s", stats_file);
+		/* init read and scrub lists */
+		struct peb_list *p = NULL;
+		list_for_each_entry(p, &peb_head->list, list) {
+			schedule_peb(&sched_read_head->list, p->peb, SCHED_READ);
+			schedule_peb(&sched_scrub_head->list, p->peb, SCHED_SCRUB);
+		}
+	}
+
+	int shutdown = 0;
+	int stats_timer = create_and_arm_timer(60);
+	if (stats_timer < 0) {
+		log_fatal("Could not create stats timer, aborting");
+		_exit(1);
+	}
+	int read_peb_timer = create_and_arm_timer(read_completion / num_pebs);
+	if (read_peb_timer < 0) {
+		log_fatal("Could not create read timer, aborting");
+		_exit(1);
+	}
+	int scrub_peb_timer = create_and_arm_timer(scrub_completion / num_pebs);
+	if (scrub_peb_timer < 0) {
+		log_fatal("Could not create scrubbing timer, aborting");
+		_exit(1);
+	}
+	struct pollfd pfd[4];
+	pfd[0].fd = sigfd;
+	pfd[0].events = POLLIN;
+	pfd[1].fd = stats_timer;
+	pfd[1].events = POLLIN;
+	pfd[2].fd = read_peb_timer;
+	pfd[2].events = POLLIN;
+	pfd[3].fd = scrub_peb_timer;
+	pfd[3].events = POLLIN;
+	while (!shutdown) {
+		int n = poll(pfd, ARRAY_SIZE(pfd), -1);
+		if (n == -1) {
+			log_err("poll error: %s", strerror(errno));
+			shutdown = 1;
+		}
+		if (n == 0) {
+			continue;
+		}
+		/* signalfd */
+		if (pfd[0].revents & POLLIN) {
+			ssize_t s = read(sigfd, &fdsi, sizeof(fdsi));
+			if (s != sizeof(fdsi)) {
+				log_warn("Could not read from signal fd");
+				continue;
+			}
+			switch(fdsi.ssi_signo) {
+			case SIGUSR1:
+				/* write back stats to disk */
+				write_stats_file(stats_file, peb_head, sched_read_head, sched_scrub_head, num_pebs);
+				break;
+			default:
+				shutdown = 1;
+				break;
+			}
+		}
+		/* stats timer */
+		if (pfd[1].revents & POLLIN) {
+			uint64_t tmp;
+			read(stats_timer, &tmp, sizeof(tmp));
+			/* update stats */
+			if (update_stats(fd, peb_head, num_pebs) < 0) {
+				log_warn("Could not update stats");
+				continue;
+			}
+
+			struct peb_list *p = NULL;
+			/* check if we need to act on any block */
+			list_for_each_entry(p, &peb_head->list, list) {
+				struct peb_info *peb = p->peb;
+				if (!peb)
+					continue;
+				uint64_t read_stats = peb->read_cnt - peb->prev_read_cnt;
+				/* read whole PEB if number of reads since last check is above threshold */
+				if (read_stats >= read_threshold) {
+					log_info("Too many reads for PEB %" PRIu64 " between stats updates, scheduling READ", peb->peb_num);
+					read_peb(fd, peb);
+				}
+			}
+		}
+
+		/* read_peb_timer */
+		if (pfd[2].revents & POLLIN) {
+			uint64_t tmp;
+			read(pfd[2].fd, &tmp, sizeof(tmp));
+			/* do next peb read */
+			if (work(sched_read_head, fd) < 0) {
+				log_err("Error while reading PEB");
+			}
+		}
+
+		/* scrub pebs */
+		if (pfd[3].revents & POLLIN) {
+			uint64_t tmp;
+			read(pfd[3].fd, &tmp, sizeof(tmp));
+			/* do next peb scrub */
+			if (work(sched_scrub_head, fd) < 0) {
+				log_err("Error while scrubbing PEB");
+			}
+		}
+
+	}
+	log_info("Shutting down");
+	write_stats_file(stats_file, peb_head, sched_read_head, sched_scrub_head, num_pebs);
+	close(fd);
+	free_list(peb_head);
+
+	return 0;
+}
-- 
2.5.0

Re: [PATCH 3/4] Initial implementation for ubihealthd.

[Sascha Hauer]

On Fri, Nov 06, 2015 at 12:00:02AM +0100, Richard Weinberger wrote:

> +
> +	int shutdown = 0;
> +	int stats_timer = create_and_arm_timer(60);
> +	if (stats_timer < 0) {
> +		log_fatal("Could not create stats timer, aborting");
> +		_exit(1);
> +	}
> +	int read_peb_timer = create_and_arm_timer(read_completion / num_pebs);
> +	if (read_peb_timer < 0) {
> +		log_fatal("Could not create read timer, aborting");
> +		_exit(1);
> +	}
> +	int scrub_peb_timer = create_and_arm_timer(scrub_completion / num_pebs);
> +	if (scrub_peb_timer < 0) {
> +		log_fatal("Could not create scrubbing timer, aborting");
> +		_exit(1);
> +	}

I tried to configure a "do as fast as you can" and realized that when
read_/scrub_completion is configured lower than the number of pebs and
create_and_arm_timer is called with 0 as argument then the program no
longer works. This should probably be catched.

Sascha

-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

[PATCH 4/4] Documentation for ubihealthd

[Richard Weinberger]

From: Daniel Walter <dwalter@sigma-star.at>

Add documentation for ubihealthd

Signed-off-by: Daniel Walter <dwalter@sigma-star.at>
Signed-off-by: Richard Weinberger <richard@nod.at>
---
 ubi-utils/README.ubihealthd | 39 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 39 insertions(+)
 create mode 100644 ubi-utils/README.ubihealthd

diff --git a/ubi-utils/README.ubihealthd b/ubi-utils/README.ubihealthd
new file mode 100644
index 0000000..da4501a
--- /dev/null
+++ b/ubi-utils/README.ubihealthd
@@ -0,0 +1,39 @@
+# ubihealthd
+
+ubihealthd is a small daemon which monitors ubi devices.
+The main purpose of this daemon is to read and scrub
+all PEBs of a given device over a specified amount of time.
+Additionally if a PEBs was read above a given threshold
+the complete PEB will be read to detect and fix read-disturbs.
+
+## Basic Algorithm
+
+ubihealthd currently uses 2 lists to keep track of the PEB
+which should be read/scrubbed periodically.
+Additionally after each statistics update (default: every 60 seconds)
+a PEB read is performed if the number of reads since the last time
+passes a given threshold.
+In order to allow the daemon to operate after a reboot or restart of
+the daemon, the statistics are written to a given file on shutdown or
+as soon as a SIGUSR1 is received.
+
+## Statistics File format
+The statistics file is a binary format and written as follows.
+    * number of PEBs (int64_t)
+    * next scheduled PEB for reads (int64_t)
+    * next scheduled PEB for scrubbing (int64_t)
+    * dump of all PEBs (struct peb_info)
+
+## Usage:
+ubihealthd [OPTIONS]
+
+OPTIONS
+  -h, --help		Show this message
+  -d, --device		Device to be monitored (default: /dev/ubi0)
+  -f, --file		Path to statistics save file
+  -r, --read_complete	Timeframe for reading all PEBs in seconds
+  -s, --scrub_complete	Timeframe for scrubbing all PEBs in seconds
+  -x, --read_threshold	Number of reads between two stats updates
+                      	which will trigger a PEB read
+  -v, --verbosity	Set log level (0-4)
+
-- 
2.5.0

Re: [RFC] ubihealthd

[Sascha Hauer]

Hi Richard, Daniel,

On Thu, Nov 05, 2015 at 11:59:59PM +0100, Richard Weinberger wrote:
> ubihealthd is a tiny C program which takes care of your NAND.
> It will trigger re-reads and scrubbing such that read-disturb and
> data retention will be addressed before data is lost.
> Currently the policy is rather trivial. It re-reads every PEB within
> a given time frame, same for scrubbing and if a PEB's read counter exceeds
> a given threshold it will also trigger a re-read.
> 
> At ELCE some people asked why this is done in userspace.
> The reason is that this is a classical example of kernel offers mechanism
> and userspace the policy. Also ubihealthd is not mandatory.
> Depending on your NAND it can help you increasing its lifetime.
> But you won't lose data immediately if it does not run for a while.
> It is something like smartd is for hard disks.
> I did this also in kernel space and it was messy.

I gave ubihealthd a try and it basically works as expected. I let it run
on a UBI device with a ton of (artificial) bitflips and the demon crawls
over them moving the data away.

Do you have plans to further work on this and to integrate it into the
kernel and mtd-utils?

One thing I noticed is that ubihealthd always scrubs blocks, even when
there are no bitflips in that block. Why is that done? I would assume
that rewriting a block when there are more bitflips than we can accept
is enough, no?

Sascha

-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

Re: [RFC] ubihealthd

[Boris Brezillon]

On Fri, 15 Apr 2016 08:26:04 +0200
Sascha Hauer <s.hauer@pengutronix.de> wrote:

> Hi Richard, Daniel,
> 
> On Thu, Nov 05, 2015 at 11:59:59PM +0100, Richard Weinberger wrote:
> > ubihealthd is a tiny C program which takes care of your NAND.
> > It will trigger re-reads and scrubbing such that read-disturb and
> > data retention will be addressed before data is lost.
> > Currently the policy is rather trivial. It re-reads every PEB within
> > a given time frame, same for scrubbing and if a PEB's read counter exceeds
> > a given threshold it will also trigger a re-read.
> > 
> > At ELCE some people asked why this is done in userspace.
> > The reason is that this is a classical example of kernel offers mechanism
> > and userspace the policy. Also ubihealthd is not mandatory.
> > Depending on your NAND it can help you increasing its lifetime.
> > But you won't lose data immediately if it does not run for a while.
> > It is something like smartd is for hard disks.
> > I did this also in kernel space and it was messy.
> 
> I gave ubihealthd a try and it basically works as expected. I let it run
> on a UBI device with a ton of (artificial) bitflips and the demon crawls
> over them moving the data away.
> 
> Do you have plans to further work on this and to integrate it into the
> kernel and mtd-utils?
> 
> One thing I noticed is that ubihealthd always scrubs blocks, even when
> there are no bitflips in that block. Why is that done? I would assume
> that rewriting a block when there are more bitflips than we can accept
> is enough, no?

Yep, that's my opinion too: we should not scrub the block if we're
below the bitflip_threshold. If one wants to be conservative, and
scrub as soon as there's a single bitflip, he can always manually set
bitflips_threshold to something really low.


-- 
Boris Brezillon, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

Re: [RFC] ubihealthd

[Daniel Walter]

On 04/15/2016 08:26 AM, Sascha Hauer wrote:
> Hi Richard, Daniel,
> 
> On Thu, Nov 05, 2015 at 11:59:59PM +0100, Richard Weinberger wrote:
>> ubihealthd is a tiny C program which takes care of your NAND.
>> It will trigger re-reads and scrubbing such that read-disturb and
>> data retention will be addressed before data is lost.
>> Currently the policy is rather trivial. It re-reads every PEB within
>> a given time frame, same for scrubbing and if a PEB's read counter exceeds
>> a given threshold it will also trigger a re-read.
>>
>> At ELCE some people asked why this is done in userspace.
>> The reason is that this is a classical example of kernel offers mechanism
>> and userspace the policy. Also ubihealthd is not mandatory.
>> Depending on your NAND it can help you increasing its lifetime.
>> But you won't lose data immediately if it does not run for a while.
>> It is something like smartd is for hard disks.
>> I did this also in kernel space and it was messy.
> 
> I gave ubihealthd a try and it basically works as expected. I let it run
> on a UBI device with a ton of (artificial) bitflips and the demon crawls
> over them moving the data away.
> 
> Do you have plans to further work on this and to integrate it into the
> kernel and mtd-utils?
> 
> One thing I noticed is that ubihealthd always scrubs blocks, even when
> there are no bitflips in that block. Why is that done? I would assume
> that rewriting a block when there are more bitflips than we can accept
> is enough, no?
> 
> Sascha
> 

Hi Sascha,

sorry for the late reply.

I've picked up working on ubihealthd again and after your comments and
the comments from Brian, I came to the conclusion, that we can indeed
skip the scrubbing, since it will be done by the kernel anyways as soon
as a read requests produces bitfilps.

I assume that I'll finish the work for the next version for ubihealthd
within the next few days and will send an updated RFC to the list.

daniel

Re: [RFC] ubihealthd

[Richard Weinberger]

Bean,

Am 27.06.2017 um 15:35 schrieb Bean Huo (beanhuo):
> Hi, Richard
> I see ubihealthd tool doesn't merge in to mainline MTD-utils, also the kernel 
> " UBI statistics and bitrot interface".  
> http://lists.infradead.org/pipermail/linux-mtd/2015-November/063122.html
> I have a little bit confusion.  What is it difference with your patch series " UBI: Bitrot checking "?
> http://lists.infradead.org/pipermail/linux-mtd/2015-November/063122.html

this is two times the same link.
The most recent series is the latest approach. I sent multiple approaches.

> the ubihealthd is a user space tool and can trigger re-reads and scrubbing.
> I see " UBI: Bitrot checking " also has this action. They are separate or working together?
> Can you please provide more information about both?

The planned mode was that all logic happens in userspace, therefore ubihealthd.

Thanks,
//richard

Re: [RFC] ubihealthd

[Richard Weinberger]

Bean,

Am 27.06.2017 um 16:25 schrieb Bean Huo (beanhuo):
> I mean these two links:
> http://lists.infradead.org/pipermail/linux-mtd/2015-November/063122.html
> http://lists.infradead.org/pipermail/linux-mtd/2015-March/058519.html
> so, you are preferring to implement latest approach in user space by ubihealthd.
> But so far, you don't have plan to merge into mainline kernel and mtd-utils. Right?

I have plans but no time.

> do you have a roughly estimated data for the CPU and memory usage for the ubihealthd?

42.

Really, read the source. :-)

Thanks,
//richard