When the WL worker is moving an LEB, the volume might go away
occasionally. UBI does not handle these situations correctly.
This patch introduces a new mutex which serializes wear-levelling
worker and the the 'ubi_wl_put_peb()' function. Now, if one puts
an LEB, and its PEB is being moved, it will wait on the mutex.
And because we unmap all LEBs when removing volumes, this will make
the volume remove function to wait while the LEB movement
finishes.
Below is an example of an oops which should be fixed by this patch:
Pid: 9167, comm: io_paral Not tainted (2.6.24-rc5-ubi-2.6.git #2)
EIP: 0060:[<f884a379>] EFLAGS: 00010246 CPU: 0
EIP is at prot_tree_del+0x2a/0x63 [ubi]
EAX: f39a90e0 EBX: 00000000 ECX: 00000000 EDX: 00000134
ESI: f39a90e0 EDI: f39a90e0 EBP: f2d55ddc ESP: f2d55dd4
DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
Process io_paral (pid: 9167, ti=f2d54000 task=f72a8030 task.ti=f2d54000)
Stack: f39a95f8 ef6aae50 f2d55e08 f884a511 f88538e1 f884ecea 00000134 00000000
f39a9604 f39a95f0 efea8280 00000000 f39a90e0 f2d55e40 f8847261 f8850c3c
f884eaad 00000001 000000b9 00000134 00000172 000000b9 00000134 00000001
Call Trace:
[<c0105227>] show_trace_log_lvl+0x1a/0x30
[<c01052e2>] show_stack_log_lvl+0xa5/0xca
[<c01053d6>] show_registers+0xcf/0x21b
[<c0105648>] die+0x126/0x224
[<c0119a62>] do_page_fault+0x27f/0x60d
[<c037dd62>] error_code+0x72/0x78
[<f884a511>] ubi_wl_put_peb+0xf0/0x191 [ubi]
[<f8847261>] ubi_eba_unmap_leb+0xaf/0xcc [ubi]
[<f8843c21>] ubi_remove_volume+0x102/0x1e8 [ubi]
[<f8846077>] ubi_cdev_ioctl+0x22a/0x383 [ubi]
[<c017d768>] do_ioctl+0x68/0x71
[<c017d7c6>] vfs_ioctl+0x55/0x271
[<c017da15>] sys_ioctl+0x33/0x52
[<c0104152>] sysenter_past_esp+0x5f/0xa5
=======================
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Add ref_count field to UBI volumes and remove weired "vol->removed"
field. This way things are better understandable and we do not have
to do whold show_attr operation under spinlock.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Pass volume description object to the EBA function which makes
more sense, and EBA function do not have to find the volume
description object by volume ID.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Similarly to ltree_entry_slab, it makes more sense to create
and destroy ubi_wl_entry slab on module initialization/exit.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Since the ltree_entry slab cache is a global entity, which is
used by all UBI devices, it is more logical to create it on
module initialization time and destro on module exit time.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
When the UBI device is nearly full, i.e. all LEBs are mapped, we have
only one spare LEB left - the one we reserved for WL purposes. Well,
I do not count the LEBs which were reserved for bad PEB handling -
suppose NOR flash for simplicity. If an "atomic LEB change operation"
is run, and the WL unit is moving a LEB, we have no spare LEBs to
finish the operation and fail, which is not good. Moreover, if there
are 2 or more simultanious "atomic LEB change" requests, only one of
them has chances to succeed, the other will fail with -ENOSPC. Not
good either.
This patch does 2 things:
1. Reserves one PEB for the "atomic LEB change" operation.
2. Serealize the operations so that only on of them may run
at a time (by means of a mutex).
Pointed-to-by: Brijesh Singh <brijesh.s.singh@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Similar reason as in case of the previous patch: it causes
deadlocks if a filesystem with writeback support works on top
of UBI. So pre-allocate needed buffers when attaching MTD device.
We also need mutexes to protect the buffers, but they do not
cause much contantion because they are used in recovery, torture,
and WL copy routines, which are called seldom.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Use GFP_NOFS flag when allocating memory on I/O path, because otherwise
we may deadlock the filesystem which works on top of us. We observed
the deadlocks with UBIFS. Example:
VFS->FS lock a lock->UBI->kmalloc()->VFS writeback->FS locks the same
lock again.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
UBI allocates temporary buffers of PEB size, which may be 256KiB.
Use vmalloc instead of kmalloc for such big temporary buffers.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
In case of static volumes, make emulated MTD device size to
be equivalent to data size, rather then volume size.
Reported-by: John Smith <john@arrows.demon.co.uk>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
UBI (Latin: "where?") manages multiple logical volumes on a single
flash device, specifically supporting NAND flash devices. UBI provides
a flexible partitioning concept which still allows for wear-levelling
across the whole flash device.
In a sense, UBI may be compared to the Logical Volume Manager
(LVM). Whereas LVM maps logical sector numbers to physical HDD sector
numbers, UBI maps logical eraseblocks to physical eraseblocks.
More information may be found at
http://www.linux-mtd.infradead.org/doc/ubi.html
Partitioning/Re-partitioning
An UBI volume occupies a certain number of erase blocks. This is
limited by a configured maximum volume size, which could also be
viewed as the partition size. Each individual UBI volume's size can
be changed independently of the other UBI volumes, provided that the
sum of all volume sizes doesn't exceed a certain limit.
UBI supports dynamic volumes and static volumes. Static volumes are
read-only and their contents are protected by CRC check sums.
Bad eraseblocks handling
UBI transparently handles bad eraseblocks. When a physical
eraseblock becomes bad, it is substituted by a good physical
eraseblock, and the user does not even notice this.
Scrubbing
On a NAND flash bit flips can occur on any write operation,
sometimes also on read. If bit flips persist on the device, at first
they can still be corrected by ECC, but once they accumulate,
correction will become impossible. Thus it is best to actively scrub
the affected eraseblock, by first copying it to a free eraseblock
and then erasing the original. The UBI layer performs this type of
scrubbing under the covers, transparently to the UBI volume users.
Erase Counts
UBI maintains an erase count header per eraseblock. This frees
higher-level layers (like file systems) from doing this and allows
for centralized erase count management instead. The erase counts are
used by the wear-levelling algorithm in the UBI layer. The algorithm
itself is exchangeable.
Booting from NAND
For booting directly from NAND flash the hardware must at least be
capable of fetching and executing a small portion of the NAND
flash. Some NAND flash controllers have this kind of support. They
usually limit the window to a few kilobytes in erase block 0. This
"initial program loader" (IPL) must then contain sufficient logic to
load and execute the next boot phase.
Due to bad eraseblocks, which may be randomly scattered over the
flash device, it is problematic to store the "secondary program
loader" (SPL) statically. Also, due to bit-flips it may become
corrupted over time. UBI allows to solve this problem gracefully by
storing the SPL in a small static UBI volume.
UBI volumes vs. static partitions
UBI volumes are still very similar to static MTD partitions:
* both consist of eraseblocks (logical eraseblocks in case of UBI
volumes, and physical eraseblocks in case of static partitions;
* both support three basic operations - read, write, erase.
But UBI volumes have the following advantages over traditional
static MTD partitions:
* there are no eraseblock wear-leveling constraints in case of UBI
volumes, so the user should not care about this;
* there are no bit-flips and bad eraseblocks in case of UBI volumes.
So, UBI volumes may be considered as flash devices with relaxed
restrictions.
Where can it be found?
Documentation, kernel code and applications can be found in the MTD
gits.
What are the applications for?
The applications help to create binary flash images for two purposes: pfi
files (partial flash images) for in-system update of UBI volumes, and plain
binary images, with or without OOB data in case of NAND, for a manufacturing
step. Furthermore some tools are/and will be created that allow flash content
analysis after a system has crashed..
Who did UBI?
The original ideas, where UBI is based on, were developed by Andreas
Arnez, Frank Haverkamp and Thomas Gleixner. Josh W. Boyer and some others
were involved too. The implementation of the kernel layer was done by Artem
B. Bityutskiy. The user-space applications and tools were written by Oliver
Lohmann with contributions from Frank Haverkamp, Andreas Arnez, and Artem.
Joern Engel contributed a patch which modifies JFFS2 so that it can be run on
a UBI volume. Thomas Gleixner did modifications to the NAND layer. Alexander
Schmidt made some testing work as well as core functionality improvements.
Signed-off-by: Artem B. Bityutskiy <dedekind@linutronix.de>
Signed-off-by: Frank Haverkamp <haver@vnet.ibm.com>
Artem Bityutskiy