This patch enables NAND subpage read functionality.
If upper layer drivers are requesting to read non page aligned data NAND
subpage-read functionality reads the only whose ECC regions which include
requested data when original code reads whole page.
This significantly improves performance in many cases.
Here are some digits :
UBI volume mount time
No subpage reads: 5.75 seconds
Subpage read patch: 2.42 seconds
Open/stat time for files on JFFS2 volume:
No subpage read 0m 5.36s
Subpage read 0m 2.88s
Signed-off-by Alexey Korolev <akorolev@infradead.org>
Acked-by: Artem Bityutskiy <dedekind@infradead.org>
Acked-by: Jörn Engel <joern@logfs.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Add support for CPU frequency scalling to the S3C24XX NAND driver.
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The unlock_addr rework in kernel 2.6.25 breaks 16-bit SST chips. SST
39LF160 and SST 39VF1601 are both 16-bit only chip (do not have BYTE#
pin) and new uaddr value is not correct for them. Add
MTD_UADDR_0xAAAA_0x5555 for those chips. Tested with SST 39VF1601
chip.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
When detecting a partition beyond the end of the device, skip most of
the initialisation, in particular those bits causing a division by zero.
Signed-off-by: Jörn Engel <joern@logfs.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Mostly simplifying the loops. Now everything fits into 80 columns,
is easier to read and the finer details have extra comments.
Signed-off-by: Jörn Engel <joern@logfs.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Remaining are 12 warnings about long lines and 1 about braces that
could be argued about.
Signed-off-by: Jörn Engel <joern@logfs.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
add_mtd_partition was a 150+ line monster consisting mostly of a single
loop. Seperate the loop from most of the body. Now it should be
obvious which variables are carried around from iteration to iteration.
Signed-off-by: Jörn Engel <joern@logfs.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
A nice side effect of this patch is that the return value of
physmap_flash_suspend in the error path is the value of the first failing
suspend callback and not the bitwise OR of all of them.
Signed-off-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Don't call suspend/resume functions if they have not been
defined.
Signed-off-by: Robert Jarzmik <rjarzmik@free.fr>
Acked-By: Jörn Engel <joern@logfs.org>
Signed-off-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Existing CFI driver has problems with excessive writes during erase.
If CFI driver does many writes during one erase cycle we may face the
messages with -ETIMEO error on erase operation. It may cause the
following data corruption and kernel panics.
The reason of the issue is related to specifics of suspend operation:
if we write to flash during erase, suspend operation will cost some time
to erase procedure (for P30 it could be significant). In current version of
cfi driver the problem of many suspends is partially workarounded by adding
some time reserv to any operation (8xerase_time) but if we have many writes
during one erase the problem appears.
This patch detects the suspend and resets timer if suspend occured. It
has been well verified on different chips. No problems were found.
Could you please include the patch as it is simple and fixes bad issue.
Signed-off-by: Alexey Korolev <akorolev@infradead.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
With CONFIG_MTD_OF_PARTS=y I'm getting this new section mismatch in reference
from the function fsl_elbc_chip_probe() to the function
.devinit.text:of_mtd_parse_partitions()
This patch fixes the mismatch by providing __devinit annotation to the
fsl_elbc_chip_probe() function.
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Acked-By: Scott Wood <scottwood@freescale.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
This fix only affects UBI debugging.
If the the background thread is disabled for debugging purposes,
start it anyway, because otherwise we see tonns of kernel debugging
complaints like this:
INFO: task ubi_bgt0d:26857 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
ubi_bgt0d D dd37bf94 0 26857 2
dd37bfcc 00000086 f8e17cea dd37bf94 00000046 00000000 00000000 f5c62430
f5c62430 f5c62590 c2a09c80 f6cbd498 dd8e9cbc 00000296 dd37bfb0 00000296
dd8e9cb8 dd8e9cbc dd37bfcc c0119774 00000000 00000000 c0132e89 f6961560
Call Trace:
[<f8e17cea>] ? ubi_thread+0x0/0x127 [ubi]
[<c0119774>] ? complete+0x43/0x4b
[<c0132e89>] ? kthread+0x0/0x5b
[<f8e17cea>] ? ubi_thread+0x0/0x127 [ubi]
[<c0132eae>] kthread+0x25/0x5b
[<c0132e89>] ? kthread+0x0/0x5b
[<c0104953>] kernel_thread_helper+0x7/0x14
=======================
So start it, and go sleep inside it, instead of creating it and never
start.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
On 32-bit architectures PAGE_ALIGN() truncates 64-bit values to the 32-bit
boundary. For example:
u64 val = PAGE_ALIGN(size);
always returns a value < 4GB even if size is greater than 4GB.
The problem resides in PAGE_MASK definition (from include/asm-x86/page.h for
example):
#define PAGE_SHIFT 12
#define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT)
#define PAGE_MASK (~(PAGE_SIZE-1))
...
#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK)
The "~" is performed on a 32-bit value, so everything in "and" with
PAGE_MASK greater than 4GB will be truncated to the 32-bit boundary.
Using the ALIGN() macro seems to be the right way, because it uses
typeof(addr) for the mask.
Also move the PAGE_ALIGN() definitions out of include/asm-*/page.h in
include/linux/mm.h.
See also lkml discussion: http://lkml.org/lkml/2008/6/11/237
[akpm@linux-foundation.org: fix drivers/media/video/uvc/uvc_queue.c]
[akpm@linux-foundation.org: fix v850]
[akpm@linux-foundation.org: fix powerpc]
[akpm@linux-foundation.org: fix arm]
[akpm@linux-foundation.org: fix mips]
[akpm@linux-foundation.org: fix drivers/media/video/pvrusb2/pvrusb2-dvb.c]
[akpm@linux-foundation.org: fix drivers/mtd/maps/uclinux.c]
[akpm@linux-foundation.org: fix powerpc]
Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix the following warning:
drivers/mtd/ubi/vmt.c: In function 'ubi_rename_volumes':
drivers/mtd/ubi/vmt.c:642: warning: statement with no effect
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Before UBI got into mainline, there was a slight flash format
change - we did not have sequence number support, then added it.
We have carried full support of those ancient images till this
moment. Now the support is removed, well, not fully removed.
Now UBI will support only _clean_ old images, which were cleanly
detached last time (just before kernel upgrade). This is most
likely the case.
But we will not support unclean ancient images. Surprisingly,
this allows us to remove a big chunk of legacy code.
And the same should be true for downgrading: clean images should
downgrade fine, but unclean ones will not.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
No functional changes, just tweak comments to make kernel-doc
work fine and stop complaining.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Just out or curiousity ran checkpatch.pl for whole UBI,
and discovered there are quite a few of stylistic issues.
Fix them.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
This is probably a copy-paste bug - we torture the old PEB
in the atomic LEB change function, but we should not do this.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
If bit-flips happen often, UBI prints to many messages. Lessen
the amount by only printing the messages when the PEB has been
scrubbed. Also, print torturing messages.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Quite useful ioctl which allows to make atomic system upgrades.
The idea belongs to Richard Titmuss <richard_titmuss@logitech.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Hch asked not to use "unit" for sub-systems, let it be so.
Also some other commentaries modifications.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Check that volume name is not shorter than 'name_len'.
No need to copy the trailing zero byte because whole array
was zeroed earlier.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
leb_read_unlock() may be called simultaniously by several tasks.
The would race at the following code:
up_read(&le->mutex);
if (free)
kfree(le);
And it is possible that one task frees 'le' before the other tasks
do 'up_read()'. Fix this by doing up_read and free inside the
'ubi->ltree' lock. Below it the oops we had because of this:
BUG: spinlock bad magic on CPU#0, integck/7504
BUG: unable to handle kernel paging request at 6b6b6c4f
IP: [<c0211221>] spin_bug+0x5c/0xdb
*pde = 00000000 Oops: 0000 [#1] PREEMPT SMP Modules linked in: ubifs ubi nandsim nand nand_ids nand_ecc video output
Pid: 7504, comm: integck Not tainted (2.6.26-rc3ubifs26 #8)
EIP: 0060:[<c0211221>] EFLAGS: 00010002 CPU: 0
EIP is at spin_bug+0x5c/0xdb
EAX: 00000032 EBX: 6b6b6b6b ECX: 6b6b6b6b EDX: f7f7ce30
ESI: f76491dc EDI: c044f51f EBP: e8a736cc ESP: e8a736a8
DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
Process integck (pid: 7504, ti=e8a72000 task=f7f7ce30 task.ti=e8a72000)
Stack: c044f754 c044f51f 00000000 f7f7d024 00001d50 00000001 f76491dc 00000296 f6df50e0 e8a736d8 c02112f0 f76491dc e8a736e8 c039157a f7d9e830 f76491d8 e8a7370c c020b975 f76491dc 00000296 f76491f8 00000000 f76491d8 00000000 Call Trace:
[<c02112f0>] ? _raw_spin_unlock+0x50/0x7c
[<c039157a>] ? _spin_unlock_irqrestore+0x20/0x58
[<c020b975>] ? rwsem_wake+0x4b/0x122
[<c0390e0a>] ? call_rwsem_wake+0xa/0xc
[<c0139ee7>] ? up_read+0x28/0x31
[<f8873b3c>] ? leb_read_unlock+0x73/0x7b [ubi]
[<f88742a3>] ? ubi_eba_read_leb+0x195/0x2b0 [ubi]
[<f8872a04>] ? ubi_leb_read+0xaf/0xf8 [ubi]
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Normally UBI volumes are freed in the release function of
the struct device object. However, on error path they may
have to be freed before the struct device objects have been
initialized.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
UBI forgets to free internal volumes when detaching MTD device.
Fix this.
Pointed-out-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
ubi_free_volume() function sets ubi->volumes[] to NULL, so
ubi_eba_close() is useless, it does not free what has to be freed.
So zap it and free vol->eba_tbl at the volume release function.
Pointed-out-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
UBI already checks that @min io size is the power of 2 at io_init.
It is save to use bit operations then.
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Instead of correctly pad the buffer wich we are writing to the
eraseblock during update, we used weird construct:
memset(buf + len, 0xFF, len - len);
Fix this.
Signed-off-by: Kyungmin Park <kmpark@infradead.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
It is not clear why we schedule PEB for scrubbing in case of
-EBADMSG. Elaborate.
Requested-by: Kyungmin Park <kmpark@infradead.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
This changes the MTD core to handle pci_name() now returning a constant
string.
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
device_create() is race-prone, so use the race-free
device_create_drvdata() instead as device_create() is going away.
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Fix const to non-const pointer assignment in the MTD command line partitioning
driver.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Use pr_debug(...) instead of printk(KERN_DEBUG ...) so that the message
is only printed when debugging is enabled.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Tested-by: John stoffel <john@stoffel.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Such a hardcoded address can cause a checkstop or machine check if
the driver is in the kernel but the address is not acknowledged.
Both drivers allow an address to be specified as either a module
parameter or config option. Any future powerpc board should either
use one of these methods or find the address in the device tree.
Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Remove the Simtec BAST flash driver as this has been replaced by using
the platform flash driver.
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
This patch deletes oobavail assignments, they're calculated by the nand
core code in nand_scan_tail, plus current oobavail values are wrong for
the LP NANDs.
Also remove mtd->ecclayout and mtd->oobavail assignments, mtd core
handles this all by itself.
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
This patch implements support for flash-based BBT for chips working
through ELBC NAND controller, so that NAND core will not have to re-scan
for bad blocks on every boot.
Because ELBC controller may provide HW-generated ECCs we should adjust
bbt pattern and bbt version positions in the OOB free area.
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Acked-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
For large page chips, nand_bbt is looking into OOB area, and checking
for "0xff 0xff" pattern at OOB offset 0. That is, two bytes should be
reserved for bbt means.
But ELBC driver is specifying ecclayout so that oobfree area starts at
offset 1, so only one byte left for the bbt purposes.
This causes problems with any OOB users, namely JFFS2: after first mount
JFFS2 will fill all OOBs with "erased marker", so OOBs will contain:
OOB Data: ff 19 85 20 03 00 ff ff ff 00 00 08 ff ff ff ff
OOB Data: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
OOB Data: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
OOB Data: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
And on the next boot, NAND core will rescan for bad blocks, then will
see "0xff 0x19" pattern, and will mark all blocks as bad ones.
To fix the issue we should implement our own bad block pattern: just one
byte at OOB start. Though, this will work only for x8 chips. For x16
chips two bytes must be checked. Since ELBC driver does not support x16
NANDs (yet), we're safe for now.
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Acked-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
There's no reason to prevent the Atmel NAND driver from
building as a module.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The ALE signal isn't correctly wired up to the ECC controller on the
AP7000, so it starts calculating ECC during the address cycles.
Work around this by resetting the ECC controller between the address and
data cycles.
Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
This uses __raw_{read,write}s{b,w}() primitives to access data on NAND
chips for more efficient I/O.
On an arm926 with memory clocked at 100 MHz, this reduced the elapsed time
for a 64 MiB read by 16%. ("dd" /dev/mtd0 to /dev/null, with an 8-bit
NAND using hardware ECC and 128KiB blocksize.)
Also some minor section tweaks:
- Use platform_driver_probe() so no pointer to probe() lingers
after that code has been removed at run-time.
- Use __exit and __exit_p so the remove() code will normally be
removed by the linker.
Since these buffer read/write calls are new, this increases the runtime
code footprint (by 88 bytes on my build, after the section tweaks).
[haavard.skinnemoen@atmel.com: rebase onto atmel_nand rename]
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Use einfo, oinfo for the inner erase_info and otp_info structs used in
individual case statements.
drivers/mtd/mtdchar.c:582:26: warning: symbol 'info' shadows an earlier one
drivers/mtd/mtdchar.c:380:23: originally declared here
drivers/mtd/mtdchar.c:596:26: warning: symbol 'info' shadows an earlier one
drivers/mtd/mtdchar.c:380:23: originally declared here
drivers/mtd/mtdchar.c:704:19: warning: symbol 'info' shadows an earlier one
drivers/mtd/mtdchar.c:380:23: originally declared here
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The copy_to_user was casting away the address space to get the offset of
the length member. Use offsetof() instead and add it to the void __user
*argp.
drivers/mtd/mtdchar.c:527:23: warning: cast removes address space of expression
drivers/mtd/mtdchar.c:527:23: warning: incorrect type in argument 1 (different address spaces)
drivers/mtd/mtdchar.c:527:23: expected void [noderef] <asn:1>*to
drivers/mtd/mtdchar.c:527:23: got unsigned int *<noident>
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Alexey Korolev