We only need to call sunxi_nfc_wait_cmd_fifo_empty() if we want to send
a new command. Move the sunxi_nfc_wait_cmd_fifo_empty() call to right
place to avoid extra register reads.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Some operations, like read/write an entire page of data with the ECC
engine enabled, are known to take a lot of time. Use the interrupt-based
waiting mode in these situation.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
wait_for_completion_timeout() returns 0 if a timeout occurred, 1
otherwise. Fix the sunxi_nfc_wait_events() accordingly.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
The driver is incorrectly assuming that the ECC block size is always 1k
which is not always true.
Also take the other cases into account.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
The NAND framework is now able to select the best NAND timings for us.
All we have to do is implement a ->setup_data_interface() function to
apply those timings and remove the timing selection code from the sunxi
driver.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
The code to initialize a struct nand_hw_control is duplicated across
several drivers. Factorize it using an inline function.
Signed-off-by: Marc Gonzalez <marc_gonzalez@sigmadesigns.com>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
I moved the sanity check on ecc->size before the allocation so that we
don't leak memory on error.
Fixes: 05af074a4b73 ('mtd: nand: sunxi: check ecc->size values')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
The NAND controller on some sun8i chips needs its reset line to be
deasserted before they can enter working state.
Signed-off-by: Icenowy Zheng <icenowy@aosc.xyz>
Acked-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Implement ecc->write_subpage() to prevent core code from assigning this
hook to nand_write_subpage_hwecc(). This default implementation tries
to call ecc->hwctl() which in our case is NULL, thus leading to a NULL
pointer dereference.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Verify that the ecc->size value is either 512 or 1024 bytes.
This should always be the case if this field was assigned to the
nand->ecc_step_size_ds value, but can be wrong when the user overloaded
this value with the nand-ecc-step-size DT property.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Switching to 1k ECC blocks when possible provides better resistance against
concentrated bitflips. Say you have those two configurations:
1/ 16bits/512bytes
2/ 32bits/1024bytes
Both of them require the same amount of ECC bytes (only true for this
specific engine), but the second config allows you to correct the case
where most of your bitflips are concentrated in a single 512bytes portion.
This fact makes the 1k ECC block size more advantageous than the 512bytes
one.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
The column address passed to the RNDOUT operation was missing the page
size offset.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Fixes: 614049a8d9 ("mtd: nand: sunxi: add support for DMA assisted operations")
In case of error, the function dmaengine_prep_slave_sg() returns NULL
pointer not ERR_PTR(). The IS_ERR() test in the return value check
should be replaced with NULL test.
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Fixes: 614049a8d9 ("mtd: nand: sunxi: add support for DMA assisted operations")
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
The sunxi NAND controller is able to pipeline ECC operations only when
operated in DMA mode, which improves a lot NAND throughput while keeping
CPU usage low.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
This value should not be part of nand_ecc_modes_t as it specifies
algorithm not a mode. We successfully managed to introduce new "algo"
field which is respected now.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Implementing the mtd_ooblayout_ops interface is the new way of exposing
ECC/OOB layout to MTD users.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Allow for NULL cur_offs values when the caller does not know where the
NAND page register pointer points to.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
sunxi_nfc_hw_ecc_read_chunk() always retrieves the ECC and protected free
bytes, no matter if the user really asked for it or not. This can take a
non negligible amount of time, especially on NAND chips exposing large OOB
areas (> 1KB). Make it optional.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
In order to support DMA operations in a clean way we need to extract some
of the logic coded in sunxi_nfc_hw_ecc_read/write_page() into their own
function.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Some NAND operations are so fast that it doesn't make any sense to use
interrupt based waits (the scheduling overhead is not worth it).
Rename sunxi_nfc_wait_int() into sunxi_nfc_wait_events() and add a
parameter to specify whether polling should be used or not.
Note that all sunxi_nfc_wait_int() are moved to the polling approach now,
but this should change as soon as we have more information about the
approximate time we are about to wait (can be extracted from the NAND
timings, and the type of operation).
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
->dev_ready() is not supposed to wait for busy to ready solution (this is
the role of ->waitfunc()).
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
When the NAND controller operates in DMA mode it can pipeline ECC
operations which improves the throughput.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
NFC_ECC_ERR_CNT() is not taking into account the case when the NAND chip
contains more than 4 ECC blocks (NANDs with 4kB+ pages).
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
We don't need to manually toggle the CE line since the controller handles
it for us. Moreover, keeping the CE line low when interacting with a DDR
NAND can be problematic (data loss in some corner cases).
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Try to pack address and command cycles into a single NAND controller
command to avoid polling the status register for each single change
on the NAND bus.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Being able to read subpages can greatly improve read performances if the
MTD user is only interested in a small section of a NAND page.
This is particularly true with large pages (>= 8k).
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Allwinner's ECC engine is capable of protecting a few bytes of the OOB
area. Implement specific OOB functions to benefit from this capability.
Also, when in raw mode, the randomizer is disabled, which means you'll
only be able to retrieve randomized data, which is not really useful
for most applications.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Adapt the NAND controller clk rate to the tWB, tADL, tWHR and tRHW
timings instead of returning an error when the maximum clk divisor is
not big enough to provide an appropriate timing.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
The ONFI spec says that EDO should be enabled if the host drives tRC less
than 30ns, but the code just tests for the tRC_min value extracted from
the timings exposed by the NAND chip not the timings actually configured
in the NAND controller.
Fix that by first rounding down the requested clk_rate with
clk_round_rate() and then checking if tRC is actually smaller than 30ns.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Unlike what is specified in the Allwinner datasheets, the NAND clock rate
is not equal to 2/T but 1/T. Fix the clock rate selection accordingly.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
sunxi_nand_chip_set_timings() is extracting a pointer to the nfc from the
nand->controller field, but this field is initialized after
sunxi_nand_chip_set_timings() call.
Reorder the calls to avoid any problem.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Some drivers are including linux/of_mtd.h even if they don't use any of
the of_get_nand_xxx() helpers.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Acked-by: Harvey Hunt <harvey.hunt@imgtec.com>
mtd->priv is no longer pointing to the struct nand_chip it is attached to.
Replace those accesses by mtd_to_nand() calls.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Fixes: 4be4e03efc ("mtd: nand: sunxi: add randomizer support")
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Suggested-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Add support for the randomizer engine available in Allwinner's NFC IP.
Randomization is useful to support modern NAND chips which are sensitive to
repeated patterns. On such NANDs you might experience an unexpectedly high
number of bitflips when you repeat the same pattern all over a given NAND
block.
Randomizing input data mitigate this problem by avoiding such repeated
patterns.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
mtd_to_nand() now uses the container_of() approach to transform an
mtd_info pointer into a nand_chip one. Drop useless mtd->priv
assignments from NAND controller drivers.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
struct nand_chip now embeds an mtd device. Make use of this mtd instance.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
mtd_to_nand() was recently introduced to avoid direct accesses to the
mtd->priv field. Update all NAND drivers to use it.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
for_each_child_of_node performs an of_node_get on each iteration, so
a break out of the loop requires an of_node_put.
A simplified version of the semantic patch that fixes this problem is as
follows (http://coccinelle.lip6.fr):
// <smpl>
@@
expression root,e;
local idexpression child;
@@
for_each_child_of_node(root, child) {
... when != of_node_put(child)
when != e = child
(
return child;
|
+ of_node_put(child);
? return ...;
)
...
}
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Acked-by: Chen-Yu Tsai <wens@csie.org>
Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
All of these drivers set up a parser data struct just to communicate DT
partition data. This field has been deprecated and is instead supported
by telling nand_scan_ident() about the 'flash_node'.
This patch:
* sets chip->flash_node for those drivers that didn't already (but used
OF partitioning)
* drops the parser data
* switches to the simpler mtd_device_register() where possible, now
that we've eliminated one of the auxiliary parameters
Now that we've assigned chip->flash_node for these drivers, we can
probably rely on nand_dt_init() to do more of the DT parsing for us, but
for now, I don't want to fiddle with each of these drivers. The parsing
is done in duplicate for now on some drivers. I don't think this should
break things. (Famous last words.)
(Rolled in some changes by Boris Brezillon)
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
The in-band data are copied twice: before ECC correction and after the
ECC engine has fixed all the fixable bitflips.
Drop the useless memcpy_fromio operation by passing a NULL pointer when
calling sunxi_nfc_read_buf().
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
The sunxi_nfc_hw_ecc_read/write_chunk() functions try to avoid changing
the column address if unnecessary, but the logic to determine whether it's
necessary or not is currently wrong: it adds the ecc->bytes value to the
current offset where it should actually add ecc->size.
Fixes: 913821bdd2 ("mtd: nand: sunxi: introduce sunxi_nfc_hw_ecc_read/write_chunk()")
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
The ->read_xxx() methods are all passed the page number the NAND controller
is supposed to read, but ->write_xxx() do not have such a parameter.
This is a problem if we want to properly implement data
scrambling/randomization in order to mitigate MLC sensibility to repeated
pattern: to prevent bitflips in adjacent pages in the same block we need
to avoid repeating the same pattern at the same offset in those pages,
hence the randomizer/scrambler engine need to be passed the page value
in order to adapt its seed accordingly.
Moreover, adding the page parameter to the ->write_xxx() methods add some
consistency to the current API.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
CC: Josh Wu <josh.wu@atmel.com>
CC: Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
CC: Maxime Ripard <maxime.ripard@free-electrons.com>
CC: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CC: Huang Shijie <shijie.huang@arm.com>
CC: Stefan Agner <stefan@agner.ch>
CC: devel@driverdev.osuosl.org
CC: linux-arm-kernel@lists.infradead.org
CC: linux-kernel@vger.kernel.org
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Owner is automatically set by mtdcore. Make use of that.
Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Use the nand_check_erased_ecc_chunk() function to test if the ECC error
was triggered by an erased page containing a few bitflips.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Boris Brezillon