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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
'oobavail' parameter of mtd_info structure is now propagated to the MTD
partitions
Signed-off-by: Vitaly Wool <vwool@ru.mvista.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
This allows for much better abstraction and separation of the XIP and
non-XIP cases with their own specific implementations. This fixes the
case where a timeout was tripped on in the XIP case by the code that
was meant for the non-XIP case only.
This also makes for a nice code reduction.
Signed-off-by: Nicolas Pitre <nico@cam.org>
CC: "Alexey, Korolev" <alexey.korolev@intel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
The patch below fixes a potential starvation issue that can arise when
there is contention on a chip during a period when a process is
currently writing to it. The starvation is avoided by conditionally
rescheduling when the chip is left in a state usable by other processes.
Signed-off-by: Josh Boyer <jdub@us.ibm.com>
Signed-off-by: Tom Gall <tom_gall@vnet.ibm.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
We have to pack at least the jint16_t structure, because otherwise it'll
be four bytes in size. Thankfully, we can do that and _not_ pack the
actual node structures, and the compiler still doesn't emit stupid code.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
On AMD Au1550 the static bus controller fails to keep -CE asserted during
chip ready delay on read commands and the NAND chip being used requires this.
So, the current driver allows nand_base.c to drive -CE manually during the
entire sector read. When the PCMCIA driver is enabled however, occasionally
the ECC errors occur on NAND reads. This happens because the PCMCIA driver
polls sockets periodically and reads one of the board's control/status regs
(BCSRs) which are on the same static bus as the NAND flash, and just use
another chip select (and the NOR flash also resides on that bus), so as the
NAND driver forces NAND chip select asserted and the -RE signal is shared, a
contention occurs on the static bus when BCSR or NOR flash is read while we're
reading from NAND.
So, we either can't keep interrupts enabled during the whole NAND sector
read (which is hardly acceptable), or have to implement some interlocking
scheme between multiple drivers (which is painful, and makes me shudder :-).
There's a third way which has proven to work: to force -CE asserted only
while we're waiting for a NAND chip to become ready after a read command,
disabling interrupts for a maximum of 25 microseconds (according to Toshiba
TC58DVM92A1FT00 datasheet -- this chip is mentioned in the board schematics);
for Samsung NAND chip which seems to be actually used this delay is even less,
12 us.
Signed-off-by: Konstantin Baydarov <kbaidarov@ru.mvista.com>
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Fix build warnings from drivers/mtd/redboot.c due to
use of `unsigned long` in `struct fis_image_desc` for
fields being passed to swab32s() which expects __u32 *
Change the entries to uint32_t to make them compatible
with the swab32s() function
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
During the last cleanup of the AMD Au1550 NAND driver the old buglet was
reintroduced: as the MEM_STNDCTL register is write-only and seem to always
read as 0x31, read-modify-write to it done in au1xxx_nand_init() will have the
side effect of enabling -RCS0/1 pin override (via bits 4/5 of this reg.), thus
possibly causing a contention on the static bus when the NOR flash (using
-RCS0) or board control status registers (using -RCS2) are read. Luckily, this
goes away with a first NAND access, since au1550_hwcontrol() doesn't try to
read this register before writing anymore.
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
