Thomas Gleixner [Mon, 29 May 2006 12:56:39 +0000 (14:56 +0200)]
[MTD] NAND Signal that a bitflip was corrected by ECC
Return -EUCLEAN on read when a bitflip was detected and corrected, so the
clients can react and eventually copy the affected block to a spare one.
Make all in kernel users aware of the change.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Mon, 29 May 2006 01:26:58 +0000 (03:26 +0200)]
[MTD] Rework the out of band handling completely
Hopefully the last iteration on this!
The handling of out of band data on NAND was accompanied by tons of fruitless
discussions and halfarsed patches to make it work for a particular
problem. Sufficiently annoyed by I all those "I know it better" mails and the
resonable amount of discarded "it solves my problem" patches, I finally decided
to go for the big rework. After removing the _ecc variants of mtd read/write
functions the solution to satisfy the various requirements was to refactor the
read/write _oob functions in mtd.
The major change is that read/write_oob now takes a pointer to an operation
descriptor structure "struct mtd_oob_ops".instead of having a function with at
least seven arguments.
read/write_oob which should probably renamed to a more descriptive name, can do
the following tasks:
- read/write out of band data
- read/write data content and out of band data
- read/write raw data content and out of band data (ecc disabled)
struct mtd_oob_ops has a mode field, which determines the oob handling mode.
Aside of the MTD_OOB_RAW mode, which is intended to be especially for
diagnostic purposes and some internal functions e.g. bad block table creation,
the other two modes are for mtd clients:
MTD_OOB_PLACE puts/gets the given oob data exactly to/from the place which is
described by the ooboffs and ooblen fields of the mtd_oob_ops strcuture. It's
up to the caller to make sure that the byte positions are not used by the ECC
placement algorithms.
MTD_OOB_AUTO puts/gets the given oob data automaticaly to/from the places in
the out of band area which are described by the oobfree tuples in the ecclayout
data structre which is associated to the devicee.
The decision whether data plus oob or oob only handling is done depends on the
setting of the datbuf member of the data structure. When datbuf == NULL then
the internal read/write_oob functions are selected, otherwise the read/write
data routines are invoked.
Tested on a few platforms with all variants. Please be aware of possible
regressions for your particular device / application scenario
Disclaimer: Any whining will be ignored from those who just contributed "hot
air blurb" and never sat down to tackle the underlying problem of the mess in
the NAND driver grown over time and the big chunk of work to fix up the
existing users. The problem was not the holiness of the existing MTD
interfaces. The problems was the lack of time to go for the big overhaul. It's
easy to add more mess to the existing one, but it takes alot of effort to go
for a real solution.
Improvements and bugfixes are welcome!
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Sun, 28 May 2006 09:01:53 +0000 (11:01 +0200)]
[MTD] Remove silly MTD_WRITE/READ macros
Most of those macros are unused and the used ones just obfuscate
the code. Remove them and fixup all users.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Sat, 27 May 2006 20:16:10 +0000 (22:16 +0200)]
[MTD] NAND Replace oobinfo by ecclayout
The nand_oobinfo structure is not fitting the newer error correction
demands anymore. Replace it by struct nand_ecclayout and fixup the users
all over the place. Keep the nand_oobinfo based ioctl for user space
compability reasons.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Sat, 27 May 2006 18:36:12 +0000 (20:36 +0200)]
[MTD] NAND Consolidate oobinfo handling
The info structure for out of band data was copied into
the mtd structure. Make it a pointer and remove the ability
to set it from userspace. The position of ecc bytes is
defined by the hardware and should not be changed by software.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Sat, 27 May 2006 18:05:26 +0000 (20:05 +0200)]
[MTD] NAND Fix platform structure and NDFC driver
The platform structure was lacking an oobinfo field.
The NDFC driver had some remains from another tree.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Andrew Morton [Mon, 29 May 2006 10:33:33 +0000 (11:33 +0100)]
[MTD] Fix debug printk format warning in m25p80 SPI chip driver, again.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Mon, 29 May 2006 10:24:29 +0000 (11:24 +0100)]
[MTD] AMD Geode NAND support can depend on X86_32; we won't see it on x86_64
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 28 May 2006 23:41:11 +0000 (00:41 +0100)]
[JFFS2] Preallocate raw_node_refs in a couple of missing places in scan
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 28 May 2006 21:13:25 +0000 (22:13 +0100)]
[JFFS2] Fix oops when marking space dirty in scan, but no previous node exists.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sat, 27 May 2006 12:15:16 +0000 (13:15 +0100)]
[JFFS2] Fix wbuf recovery of f->metadata->raw node.
A data node might not be in the fraglist; it could be f->metadata.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Thomas Gleixner [Fri, 26 May 2006 23:02:13 +0000 (01:02 +0200)]
[MTD] NAND Consolidate references and add back default name setting
We have a type pointer. Make use of it instead of the error prone nand_ids[i]
reference.
The NAND driver used to set default name settings from the chip ID
string for the device. The feature got lost during the rework. Add it back.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Fri, 26 May 2006 22:47:18 +0000 (00:47 +0200)]
[MTD] NAND simplify nand_chip_select
nCE setting can be done when the first command is issued to the device.
We keep the deselect functionality as it makes sense to deassert nCE
when the device becomes idle.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
David Woodhouse [Fri, 26 May 2006 22:05:44 +0000 (23:05 +0100)]
[MTD] Fix thinko in nand_write_page_hwecc()
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Fri, 26 May 2006 20:19:05 +0000 (21:19 +0100)]
[JFFS2] Switch to using an array of jffs2_raw_node_refs instead of a list.
This allows us to drop another pointer from the struct jffs2_raw_node_ref,
shrinking it to 8 bytes on 32-bit machines (if the TEST_TOTLEN) paranoia
check is turned off, which will be committed soon).
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Thomas Gleixner [Fri, 26 May 2006 16:52:08 +0000 (18:52 +0200)]
[MTD] NAND modularize write function
Modularize the write function and reorganaize the internal buffer
management. Remove obsolete chip options and fixup all affected
users.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
David Woodhouse [Fri, 26 May 2006 01:06:27 +0000 (02:06 +0100)]
[MTD] Remove PCI dependency for Geode CS553[56] NAND controller.
PCI is faked on these devices by SMM traps. Don't depend on that --
check for the chipset directly instead.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Thu, 25 May 2006 12:30:24 +0000 (13:30 +0100)]
[JFFS2] Fix 64-bit size_t problems in XATTR code.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Thu, 25 May 2006 12:25:17 +0000 (13:25 +0100)]
[MTD] Fix NAND_VERIFY_WRITE case to build with tglx's recent changes
Bad tglx. No biscuit.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Thomas Gleixner [Thu, 25 May 2006 11:38:45 +0000 (13:38 +0200)]
[MTD] Remove the only useless readv implementation
Removing readv from struct mtd_info broke block2mtd. Remove the
reference and the useless default implementation.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Thu, 25 May 2006 08:07:16 +0000 (10:07 +0200)]
[MTD] NAND Modularize read function
Split the core of the read function out and implement
seperate handling functions for software and hardware
ECC.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Thu, 25 May 2006 08:04:31 +0000 (10:04 +0200)]
[MTD] NAND Add read/write function pointers to struct nand_ecc_ctrl
Add read/write function pointers to struct nand_ecc_ctrl to
prepare the modulaization of nand_read/write functions. The
current implementation handles every type of ecc mode
software/hardware and all kinds of strange ecc placement
schemes in one switch/if construct. Thats too complex to
maintain and too inflexible to expand. Modularization will
also shorten the code pathes of the read/write functions.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Thu, 25 May 2006 07:57:31 +0000 (09:57 +0200)]
[MTD] Add ECC statistics to struct mtd_info
FLASH - especially NAND FLASH - will become less reliable
and bit flips more likely. Add an ECC statistics struct
to struct mtd_info to keep track of this.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Thu, 25 May 2006 07:51:54 +0000 (09:51 +0200)]
[MTD] NAND Cleanup oob functions
Cleanup the code in the oob related functions and
make use of the new NO_READRDY flag.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Thu, 25 May 2006 07:50:16 +0000 (09:50 +0200)]
[MTD] NAND Introduce NAND_NO_READRDY option
The nand driver has a superflous read ready / command
delay in the read functions. This was added to handle
chips which have an automatic read forward. Newer
chips do not have this functionality anymore. Add this
option to avoid the delay / I/O operation. Mark all
large page chips with the new option flag.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Thu, 25 May 2006 07:45:29 +0000 (09:45 +0200)]
[MTD] NAND Initialize controller lock and wq only once
The lock simplifying patch did not move the lock and waitqueue
initialization into the controller allocation patch.
This reinitializes waitqueue and spinlocks also for driver
supplied controller stuctures. Move it into the allocation path.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
David Woodhouse [Thu, 25 May 2006 00:53:09 +0000 (01:53 +0100)]
[JFFS2] Fix and improve debugging output during scan.
Print wasted_size in scanned eraseblocks, print range correctly for
summary dirent and inode entries.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Thu, 25 May 2006 00:50:35 +0000 (01:50 +0100)]
[JFFS2] Add 'jeb' argument to jffs2_prealloc_raw_node_refs()
Preallocation of refs is shortly going to be a per-eraseblock thing,
rather than per-filesystem. Add the required argument to the function.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Thu, 25 May 2006 00:42:40 +0000 (01:42 +0100)]
[JFFS2] Correctly handle wasted space before summary node.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Thu, 25 May 2006 00:38:27 +0000 (01:38 +0100)]
[JFFS2] jffs2_free_all_node_refs() doesn't free them all. Rename it.
... to jffs2_free_jeb_node_refs() since that's what it does.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Thu, 25 May 2006 00:37:28 +0000 (01:37 +0100)]
[JFFS2] Allocate node_ref for wasted space when skipping to page boundary
One more place where we were changing the accounting info without
actually allocating a ref for the lost space...
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Thomas Gleixner [Wed, 24 May 2006 20:57:09 +0000 (22:57 +0200)]
[MTD] NAND fix cmd_ctrl breakage
The cmd_ctrl rework lacks some state transition flags.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Wed, 24 May 2006 10:07:37 +0000 (12:07 +0200)]
[MTD] NAND coding style and namespace cleanup
Cleanup the functions which are not going to change in the
next steps.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
David Woodhouse [Wed, 24 May 2006 13:24:02 +0000 (14:24 +0100)]
[JFFS2] Revert Artem's Bunkage in debug messages.
Random unthinking 'cleanup' caused debug messages like this:
Obsoleting node at 0x0006daf4 of len 0x3a4: <7>Dirtying
If messages are continuation of an existing line, they don't need
to be prefixed with KERN_DEBUG.
THINK. Or you will be replaced by a small shell script.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Wed, 24 May 2006 09:01:43 +0000 (10:01 +0100)]
Merge branch 'master' of git://git.infradead.org/~gleixner/mtd-nand-2.6.git
David Woodhouse [Wed, 24 May 2006 08:04:17 +0000 (09:04 +0100)]
[JFFS2] Introduce ref_next() macro for finding next physical node
Another part of the preparation for switching to an array...
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Wed, 24 May 2006 01:04:45 +0000 (02:04 +0100)]
[JFFS2] Reduce visibility of raw_node_ref to upper layers of JFFS2 code.
As the first step towards eliminating the ref->next_phys member and saving
memory by using an _array_ of struct jffs2_raw_node_ref per eraseblock,
stop the write functions from allocating their own refs; have them just
_reserve_ the appropriate number instead. Then jffs2_link_node_ref() can
just fill them in.
Use a linked list of pre-allocated refs in the superblock, for now. Once
we switch to an array, it'll just be a case of extending that array.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Thomas Gleixner [Tue, 23 May 2006 21:48:57 +0000 (23:48 +0200)]
[MTD] NAND LED support cleanup
Move the define out of the middle of the code and add an
appropriate comment.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 21:28:48 +0000 (23:28 +0200)]
[MTD] NAND remove write_byte/word function from nand_chip
The previous change of the command / hardware control allows to
remove the write_byte/word functions completely, as their only
user were nand_command and nand_command_lp.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 21:25:53 +0000 (23:25 +0200)]
[MTD] Refactor NAND hwcontrol to cmd_ctrl
The hwcontrol function enforced a step by step state machine
for any kind of hardware chip access. Let the hardware driver
know which control bits are set and inform it about a change
of the control lines. Let the hardware driver write out the
command and address bytes directly. This gives a peformance
advantage for address bus controlled chips and simplifies the
quirks in the hardware drivers.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 20:33:52 +0000 (22:33 +0200)]
[MTD] Export nand_write_raw
The previous _ecc removal / cleanup broke (i)nftl module usage.
Export the missing symbol.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 20:03:39 +0000 (22:03 +0200)]
[MTD] Mark NAND drivers TOTO and PPChameleon broken
Both drivers can not be fixed and compiled due to missing header files.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 15:21:03 +0000 (17:21 +0200)]
[MTD] Remove read/write _ecc variants
MTD clients are agnostic of FLASH which needs ECC suppport.
Remove the functions and fixup the callers.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 14:10:00 +0000 (16:10 +0200)]
[MTD] Remove readv/readv_ecc
These functions were never implemented and added only bloat to
partition and concat code.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 14:06:03 +0000 (16:06 +0200)]
[MTD] Remove nand writev support
NAND writev(_ecc) support is not longer necessary. Remove it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 13:59:58 +0000 (15:59 +0200)]
[MTD] ECC rework broke diskonchip
Fix the diskonchip ecc setup.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 13:58:23 +0000 (15:58 +0200)]
[MTD] NAND ECC hwctl function has no return value
Fix the broken prototype
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 10:37:31 +0000 (12:37 +0200)]
Merge branch 'master' of /home/tglx/work/kernel/git/mtd-2.6/
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 10:00:46 +0000 (12:00 +0200)]
[MTD] NAND modularize ECC
First step of modularizing ECC support.
- Move ECC related functionality into a seperate embedded data structure
- Get rid of the hardware dependend constants to simplify new ECC models
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 09:54:38 +0000 (11:54 +0200)]
[MTD] NAND cleanup nand_scan
Seperate functionality out of nand_scan so the code is more
readable. No functional change. First step of simplifying
the nand driver.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 09:52:35 +0000 (11:52 +0200)]
[MTD] NAND consolidate data types
The NAND driver used a mix of unsigned char, u_char amd uint8_t
data types. Consolidate to uint8_t usage
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 09:50:56 +0000 (11:50 +0200)]
[MTD] NAND whitespace and formatting cleanup
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 09:49:14 +0000 (11:49 +0200)]
[JFFS2] Simplify writebuffer handling
The writev based write buffer implementation was far to complex as
in most use cases the write buffer had to be handled anyway.
Simplify the write buffer handling and use mtd->write instead.
From extensive testing no performance impact has been noted.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 09:43:28 +0000 (11:43 +0200)]
[MTD] Add support for NDFC NAND controller
NDFC NAND Flash controller is embedded in PPC EP44x SoCs.
Add platform driver based support.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 09:38:59 +0000 (11:38 +0200)]
[MTD] Add platform support for NAND
Add the data structures necessary to provide platform device support
for NAND
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 09:37:03 +0000 (11:37 +0200)]
[MTD] Simplify NAND locking
Replace the chip lock by a the controller lock. For simple drivers a
dummy controller structure is created by the scan code.
This simplifies the locking algorithm in nand_get/release_chip().
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Thomas Gleixner [Tue, 23 May 2006 09:32:45 +0000 (11:32 +0200)]
[MTD] Improve software ECC calculation
Unrolling the loops produces denser and much faster code.
Add a config switch which allows to select the byte order of the
resulting ecc code. The current Linux implementation has a byte
swap versus the SmartMedia specification
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
David Woodhouse [Mon, 22 May 2006 23:38:06 +0000 (00:38 +0100)]
[JFFS2] Remove flash offset argument from various functions.
We don't need the upper layers to deal with the physical offset. It's
_always_ c->nextblock->offset + c->sector_size - c->nextblock->free_size
so we might as well just let the actual write functions deal with that.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Joern Engel [Mon, 22 May 2006 21:18:29 +0000 (23:18 +0200)]
[MTD] Introduce MTD_BIT_WRITEABLE
o Add a flag MTD_BIT_WRITEABLE for devices that allow single bits to be
cleared.
o Replace MTD_PROGRAM_REGIONS with a cleared MTD_BIT_WRITEABLE flag for
STMicro and Intel Sibley flashes with internal ECC. Those flashes
disallow clearing of single bits, unlike regular NOR flashes, so the
new flag models their behaviour better.
o Remove MTD_ECC. After the STMicro/Sibley merge, this flag is only set
and never checked.
Signed-off-by: Joern Engel <joern@wh.fh-wedel.de>
Joern Engel [Mon, 22 May 2006 21:18:12 +0000 (23:18 +0200)]
[MTD] Merge STMicro NOR_ECC code with Intel Sibley code
In 2002, STMicro started producing NOR flashes with internal ECC protection
for small blocks (8 or 16 bytes). Support for those flashes was added by me.
In 2005, Intel Sibley flashes copied this strategy and Nico added support for
those. Merge the code for both.
Signed-off-by: Joern Engel <joern@wh.fh-wedel.de>
Joern Engel [Mon, 22 May 2006 21:18:05 +0000 (23:18 +0200)]
[MTD] Introduce writesize
At least two flashes exists that have the concept of a minimum write unit,
similar to NAND pages, but no other NAND characteristics. Therefore, rename
the minimum write unit to "writesize" for all flashes, including NAND.
Signed-off-by: Joern Engel <joern@wh.fh-wedel.de>
Joern Engel [Mon, 22 May 2006 21:17:23 +0000 (23:17 +0200)]
[MTD] Use single flag to mark writeable devices.
Two flags exist to decide whether a device is writeable or not. None of
those two flags is checked for independently, so they are clearly redundant,
if not an invitation to bugs. This patch removed both of them, replacing
them with a single new flag.
Signed-off-by: Joern Engel <joern@wh.fh-wedel.de>
David Woodhouse [Mon, 22 May 2006 15:32:05 +0000 (16:32 +0100)]
[JFFS2] Put list of nodes in common part of ic/x_ref/x_datum structure
We'll be using a proper list of nodes in the jffs2_xattr_datum and
jffs2_xattr_ref structures, because the existing code to overwrite
them is just broken. Put it in the common part at the front of the
structure which is shared with the jffs2_inode_cache, so that the
jffs2_link_node_ref() function can do the right thing.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Mon, 22 May 2006 15:29:23 +0000 (16:29 +0100)]
[JFFS2] Add some preemptive BUG checks for XATTR code
In a couple of places, we assume that what's at the end of the
->next_in_ino list is a struct jffs2_inode_cache. Let's check
for that, since we expect it to change soon.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Mon, 22 May 2006 14:23:10 +0000 (15:23 +0100)]
[JFFS2] Extend jffs2_link_node_ref() to link into per-inode list too.
Let's avoid the potential for forgetting to set ref->next_in_ino, by doing
it within jffs2_link_node_ref() instead.
This highlights the ugliness of what we're currently doing with
xattr_datum and xattr_ref structures -- we should find a nicer way of
dealing with that.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Mon, 22 May 2006 12:55:46 +0000 (13:55 +0100)]
[JFFS2] Initialise ref->next_in_ino when marking dirty space in wbuf flush
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Mon, 22 May 2006 11:15:47 +0000 (12:15 +0100)]
[JFFS2] Fix accounting error in jffs2_link_node_ref()
When filing REF_OBSOLETE nodes, we'd add their size to the global
'dirty_size' count, but then to the eraseblock's 'used_size' count.
That's not clever.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Mon, 22 May 2006 10:27:14 +0000 (11:27 +0100)]
[JFFS2] Fix dummy jffs2_sum_scan_sumnode() macro for !SUMMARY case.
I added an argument to the real function...
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 18:05:55 +0000 (19:05 +0100)]
Merge git://git.infradead.org/jffs2-devel-2.6
David Woodhouse [Sun, 21 May 2006 18:03:21 +0000 (19:03 +0100)]
[MTD] Account for MODULE_SYMBOL_PREFIX when requesting NOR chip driver
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 17:38:51 +0000 (18:38 +0100)]
[MTD] Use symbol_request() in old DiskOnChip probe code to find actual driver
The previous code wouldn't work correctly on architectures which have a
non-empty MODULE_SYMBOL_PREFIX, and this version is neater if slightly
less optimal in the built-in case.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Jonathan McDowell [Sun, 21 May 2006 17:11:55 +0000 (18:11 +0100)]
[MTD] Add Amstrad Delta NAND support
The patch below adds support for the NAND device on the Amstrad Delta.
This is a 32MiB 8bit Toshiba device, with the data bus connected to the
OMAP MPUIO pins and ALE, CLE, NCE, NRE, NWE and NWP all connected to the
Delta's latch2 16bit latch.
Signed-Off-By: Jonathan McDowell <noodles@earth.li>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 12:29:11 +0000 (13:29 +0100)]
[JFFS2] Finally eliminate __totlen field from struct jffs2_raw_node_ref
Well, almost. We'll actually keep a 'TEST_TOTLEN' macro set for now, and keep
doing some paranoia checks to make sure it's all working correctly. But if
TEST_TOTLEN is unset, the size of struct jffs2_raw_node_ref drops from 16
bytes to 12 on 32-bit machines. That's a saving of about half a megabyte of
memory on the OLPC prototype board, with 125K or so nodes in its 512MiB of
flash.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 12:15:59 +0000 (13:15 +0100)]
[JFFS2] Locking issues in summary write code.
We can't use jffs2_scan_dirty_space() because it doesn't do any locking; it's
only for use at scan time -- hence the 'scan' in the name.
Also, don't allocate refs while we have c->erase_completion_lock held.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 12:13:45 +0000 (13:13 +0100)]
[JFFS2] Remove stray kfree of summary info in XATTR code.
We don't allocate this locally any more -- it's given to us and owner by
our caller. Also improve the debug messages a little.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 12:00:54 +0000 (13:00 +0100)]
[JFFS2] File node reference for wasted space when flushing wbuf
Next step in ongoing campaign to file a struct jffs2_raw_node_ref for every
piece of dirty space in the system, so that __totlen can be killed off....
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 03:36:45 +0000 (04:36 +0100)]
[JFFS2] Add length argument to jffs2_add_physical_node_ref()
If __totlen is going away, we need to pass the length in separately.
Also stop callers from needlessly setting ref->next_phys to NULL,
since that's done for them... and since that'll also be going away soon.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 03:00:01 +0000 (04:00 +0100)]
[JFFS2] Mark gaps in summary list as dirty space
Make sure we allocate a ref for any dirty space which exists between nodes
which we find in an eraseblock summary.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 02:57:56 +0000 (03:57 +0100)]
[JFFS2] Discard remaining free space when filing a dirty block in scan.
The incoming ref_totlen() calculation is going to rely on the existence
of nodes which cover all dirty space. We can't just tweak the accounting
data any more; we have to call jffs2_scan_dirty_space() to do it.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 02:46:05 +0000 (03:46 +0100)]
[JFFS2] Introduce jffs2_scan_dirty_space() function.
To eliminate the __totlen field from struct jffs2_raw_node_ref, we need
to allocate nodes for dirty space instead of just tweaking the accounting
data. Introduce jffs2_scan_dirty_space() in preparation for that.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 02:45:27 +0000 (03:45 +0100)]
[JFFS2] Fix summary handling of unknown but compatible nodes.
For RWCOMPAT and ROCOMPAT nodes, we should still allow the mount to
succeed. Just abandon the summary and fall through to the full scan.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sun, 21 May 2006 00:28:05 +0000 (01:28 +0100)]
[JFFS2] Fix memory leak in scan code; improve comments.
If we had to allocate extra space for the summary node, we weren't
correctly freeing it when jffs2_sum_scan_sumnode() returned nonzero --
which is both the success and the failure case. Only when it returned
zero, which means fall through to the full scan, were we correctly freeing
the buffer.
Document the meaning of those return codes while we're at it.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sat, 20 May 2006 23:02:06 +0000 (00:02 +0100)]
[JFFS2] Correct handling of JFFS2_FEATURE_RWCOMPAT_COPY nodes.
We should preserve these when we come to garbage collect them, not let
them get erased. Use jffs2_garbage_collect_pristine() for this, and make
sure the summary code copes -- just refrain from writing a summary for any
block which contains a node we don't understand.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sat, 20 May 2006 19:08:42 +0000 (20:08 +0100)]
[JFFS2] Correct accounting of erroneous cleanmarkers and failed summaries.
It should all be counted as dirty space, not wasted and _definitely_ not
unchecked.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sat, 20 May 2006 18:45:26 +0000 (19:45 +0100)]
[JFFS2] Introduce jffs2_link_node_ref() function to reduce code duplication
The same sequence of code was repeated in many places, to add a new
struct jffs2_raw_node_ref to an eraseblock and adjust the space accounting
accordingly. Move it out-of-line.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sat, 20 May 2006 16:27:32 +0000 (17:27 +0100)]
Merge git://git.infradead.org/jffs2-xattr-2.6
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sat, 20 May 2006 15:20:19 +0000 (16:20 +0100)]
[JFFS2] Reduce calls to ref_totlen() in jffs2_mark_node_obsolete()
We were calling ref_totlen() 18 times. Even before that becomes a real
function rather than just a dereference, apparently some compilers still
suck anyway. It'll _certainly_ suck after ref_totlen() becomes more
complicated, so calculate it once and don't rely on CSE.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sat, 20 May 2006 15:13:34 +0000 (16:13 +0100)]
[JFFS2] Optimise reading of eraseblock summary nodes
This improves the time to mount 512MiB of NAND flash on my OLPC prototype
by about 4%. We used to read the last page of the eraseblock twice -- once
to find the offset of the summary node, and again to actually _read_ the
summary node. Now we read the last page only once, and read more only if
we need to.
We also don't allocate a new buffer just for the summary code -- we use
the buffer which was already allocated for the scan. Better still, if the
'buffer' for the scan is actually just a pointer directly into NOR flash,
we use that too, avoiding the memcpy() which we used to do.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Andrew Morton [Sat, 20 May 2006 09:17:21 +0000 (10:17 +0100)]
[MTD] Avoid 64-bit division in mtdconcat
WARNING: "__moddi3" [drivers/mtd/mtdconcat.ko] undefined!
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Sat, 20 May 2006 01:41:34 +0000 (02:41 +0100)]
[MTD] Use __symbol_get() instead of symbol_get() in NOR chip probe
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Andrew Morton [Sat, 20 May 2006 01:06:34 +0000 (18:06 -0700)]
git-mtd: symbol_get() fix
drivers/mtd/devices/docprobe.c: In function `DoC_Probe':
drivers/mtd/devices/docprobe.c:338: warning: assignment from incompatible pointer type
drivers/mtd/devices/docprobe.c:341: warning: assignment from incompatible pointer type
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Ferenc Havasi [Fri, 19 May 2006 20:00:36 +0000 (21:00 +0100)]
[JFFS2] Remove forgotten summary code
Remove forgotten lines from jffs2_scan_eraseblock() which
were unnecessary and may cause problem in some environments.
Thanks to Alexander Belyakov <alexander.belyakov@intel.com>.
Signed-off-by: Ferenc Havasi <havasi@inf.u-szeged.hu>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Thu, 18 May 2006 23:28:49 +0000 (00:28 +0100)]
[JFFS2] Support new device nodes
Device node major/minor numbers are just stored in the payload of a single
data node. Just extend that to 4 bytes and use new_encode_dev() for it.
We only use the 4-byte format if we _need_ to, if !old_valid_dev(foo).
This preserves backwards compatibility with older code as much as
possible. If we do make devices with major or minor numbers above 255, and
then mount the file system with the old code, it'll just read the first
two bytes and get the numbers wrong. If it comes to garbage-collect it,
it'll then write back those wrong numbers. But that's about the best we
can expect.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
KaiGai Kohei [Thu, 18 May 2006 15:43:53 +0000 (00:43 +0900)]
Merge git://git.infradead.org/mtd-2.6
David Woodhouse [Wed, 17 May 2006 21:03:10 +0000 (22:03 +0100)]
[MTD] Fix printk format error in gen_probe.c
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
David Woodhouse [Wed, 17 May 2006 20:53:58 +0000 (21:53 +0100)]
[MTD] Fix mtdconcat build. We didn't introduce mtd->writesize yet.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Alexander Belyakov [Wed, 17 May 2006 15:11:16 +0000 (19:11 +0400)]
MTD: mtdconcat NAND/Sibley support (rev.2)
There is a second revision of "mtdconcat NAND/Sibley" patch. I hope
the patch will not get damaged as I'm posting it from gmail account,
thanks to Jorn.
The patch adds previously missing concat_writev(),
concat_writev_ecc(), concat_block_isbad(), concat_block_markbad()
functions to make concatenation layer compatible with Sibley and NAND
chips.
Patch has been cleared from whitespaces, fixed some lines of code as
requested. Also I have added code for alignment check that should
support Jorn's "writesize" patch.
Signed-off-by: Alexander Belyakov <alexander.belyakov@intel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Martin Michlmayr [Mon, 20 Mar 2006 04:40:46 +0000 (04:40 +0000)]
LASAT depends on MTD_CFI
The following difference was found between the mainline and linux-mips
kernel. LASAT depends on MTD_CFI.
Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Martin Michlmayr [Mon, 20 Mar 2006 04:40:26 +0000 (04:40 +0000)]
Re-add module description for ms02-nv to Kconfig
In an unrelated MTD commit, a description about the ms02-nv module
got removed from Kconfig. While I personally agree with this
removal, the module maintainer (Maciej W. Rozycki) would like to
see it added back. In the absense of any consistency regarding
Kconfig descriptions his wish should be followed.
Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Acked-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Sean Young [Wed, 17 May 2006 11:45:34 +0000 (12:45 +0100)]
[MTD] RFD FTL: Be noisier, and don't assume block without RFD magic are erased
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>