* Raw NAND changes
The use of for_each_child_of_node_scoped() has been spread into the
subsystem drivers. Aside from that, a couple of exit path have been
fixed (mtk, denali), the TI GPMC bindings have been enhanced to comply
with up-to-date partition descriptions and as always there is a load of
small and misc fixes.
* SPI-NAND changes
The most impacting series this cycle is bringing support for continuous
reads in the SPI-NAND subsystem. This is a feature already merged in the
raw NAND subsystem which allows optimizing the internal fetch times in
the chip while reading sequential pages within an eraseblock. For now
only Macronix NANDs benefit from this feature. While we are talking
about Macronix, some of their chip need an explicit action for selecting
a different plane, and support for it has also been brought.
The bitflip threshold has also been set to the same arbitrary level as
in the raw NAND subsystem to optimize wear leveling decisions, and
finally support for a new Winbond chip has been added.
SPI NOR changes for 6.12
Notable changes:
- Add Write Protect support for N25Q064A.
- New flash support for Zetta ZD25Q128C and Spansion S28HS256T.
- Fix a NULL dereference in probe path for flashes without a name. The
probe path tries to access the name without checking its existence
first. S28HS256T is the first flash to define its entry without a
name, uncovering this issue.
Writing to the Flash in `sst_nor_write()` is a 3-step process:
first an optional one-byte write to get 2-byte-aligned, then the
bulk of the data is written out in vendor-specific 2-byte writes.
Finally, if there's a byte left over, another one-byte write.
This was implemented 3 times in the body of `sst_nor_write()`.
To reduce code duplication, factor out these sub-steps to their
own function.
Signed-off-by: Csókás, Bence <csokas.bence@prolan.hu>
Reviewed-by: Pratyush Yadav <pratyush@kernel.org>
[pratyush@kernel.org: fixup whitespace, use %zu instead of %i in WARN()]
Signed-off-by: Pratyush Yadav <pratyush@kernel.org>
Link: https://lore.kernel.org/r/20240710091401.1282824-1-csokas.bence@prolan.hu
Add two flags for inserting the Plane Select bit into the column
address during the write_to_cache and the read_from_cache operation.
Add the SPINAND_HAS_PROG_PLANE_SELECT_BIT flag for serial NAND flash
that require inserting the Plane Select bit into the column address
during the write_to_cache operation.
Add the SPINAND_HAS_READ_PLANE_SELECT_BIT flag for serial NAND flash
that require inserting the Plane Select bit into the column address
during the read_from_cache operation.
Signed-off-by: Cheng Ming Lin <chengminglin@mxic.com.tw>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20240909092643.2434479-2-linchengming884@gmail.com
Reviewing a series converting the for_each_chil_of_node() loops into
their _scoped variants made me realize there was no cleanup of the
already registered NAND devices upon error which may leak memory on
systems with more than a chip when this error occurs. We should call the
_nand_chips_cleanup() function when this happens.
Fixes: 1d6b1e4649 ("mtd: mediatek: driver for MTK Smart Device")
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Pratyush Yadav <pratyush@kernel.org>
Reviewed-by: Matthias Brugger <matthias.bgg@kernel.org>
Link: https://lore.kernel.org/linux-mtd/20240826153019.67106-2-miquel.raynal@bootlin.com
There is a reason why sometime we write "NAND chip" with an 's'. It
usually means several chips can be managed by the same controller. So
when initializing a single chip at a time, the wording "chip" must be
used, otherwise when talking about all the chips managed by the
controller, we want to use "chips". Fix the function name to clarify the
meson_nfc_nand_chip*s*_cleanup() helper intend.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Pratyush Yadav <pratyush@kernel.org>
Link: https://lore.kernel.org/linux-mtd/20240826153158.67334-1-miquel.raynal@bootlin.com
Enabling continuous read support implies several changes which must be
done atomically in order to keep the code base consistent and
bisectable.
1/ Retrieving bitflips differently
Improve the helper retrieving the number of bitflips to support the case
where many pages have been read instead of just one. In this case, if
there is one page with bitflips, we cannot know the detail and just get
the information of the maximum number of bitflips corrected in the most
corrupted chunk. Compatible Macronix flashes return:
- the ECC status for the last page read (bits 0-3),
- the amount of bitflips for the whole read operation (bits 4-7).
Hence, when reading two consecutive pages, if there was 2 bits corrected
at most in one chunk, we return this amount times (arbitrary) the number
read pages. It is probably a very pessimistic calculation in most cases,
but still less pessimistic than if we multiplied this amount by the
number of chunks. Anyway, this is just for statistics, the important
data is the maximum amount of bitflips, which leads to wear leveling.
2/ Configuring, enabling and disabling the feature
Create an init function for allocating a vendor structure. Use this
vendor structure to cache the internal continuous read state. The state
is being used to discriminate between the two bitflips retrieval
methods. Finally, helpers for enabling and disabling sequential reads
are also created.
3/ Fill the chips table
Flag all the chips supporting the feature with the ->set_cont_read()
helper.
In order to validate the changes, I modified the mtd-utils test suite
with extended versions of nandbiterrs, nanddump and flash_speed in order
to support, test and benchmark continuous reads. I also ran all the UBI
tests successfully.
The nandbiterrs tool allows to track the ECC efficiency and
feedback. Here is its default output (stripped):
Successfully corrected 0 bit errors per subpage
Read reported 1 corrected bit errors
Successfully corrected 1 bit errors per subpage
Read reported 2 corrected bit errors
Successfully corrected 2 bit errors per subpage
Read reported 3 corrected bit errors
Successfully corrected 3 bit errors per subpage
Read reported 4 corrected bit errors
Successfully corrected 4 bit errors per subpage
Read reported 5 corrected bit errors
Successfully corrected 5 bit errors per subpage
Read reported 6 corrected bit errors
Successfully corrected 6 bit errors per subpage
Read reported 7 corrected bit errors
Successfully corrected 7 bit errors per subpage
Read reported 8 corrected bit errors
Successfully corrected 8 bit errors per subpage
Failed to recover 1 bitflips
Read error after 9 bit errors per page
The output using the continuous option over two pages (the second page
is kept intact):
Successfully corrected 0 bit errors per subpage
Read reported 2 corrected bit errors
Successfully corrected 1 bit errors per subpage
Read reported 4 corrected bit errors
Successfully corrected 2 bit errors per subpage
Read reported 6 corrected bit errors
Successfully corrected 3 bit errors per subpage
Read reported 8 corrected bit errors
Successfully corrected 4 bit errors per subpage
Read reported 10 corrected bit errors
Successfully corrected 5 bit errors per subpage
Read reported 12 corrected bit errors
Successfully corrected 6 bit errors per subpage
Read reported 14 corrected bit errors
Successfully corrected 7 bit errors per subpage
Read reported 16 corrected bit errors
Successfully corrected 8 bit errors per subpage
Failed to recover 1 bitflips
Read error after 9 bit errors per page
Regarding the throughput improvements, tests have been conducted in
1-1-1 and 1-1-4 modes, reading a full block X pages at a
time, X ranging from 1 to 64 (size of a block with the tested device).
The percent value on the right is the comparison of the same test
conducted without the continuous read feature, ie. reading X pages in
one single user request, which got naturally split by the core whit the
continuous read optimization disabled into single-page reads.
* 1-1-1 result:
1 page read speed is 2634 KiB/s
2 page read speed is 2704 KiB/s (+3%)
3 page read speed is 2747 KiB/s (+5%)
4 page read speed is 2804 KiB/s (+7%)
5 page read speed is 2782 KiB/s
6 page read speed is 2826 KiB/s
7 page read speed is 2834 KiB/s
8 page read speed is 2821 KiB/s
9 page read speed is 2846 KiB/s
10 page read speed is 2819 KiB/s
11 page read speed is 2871 KiB/s (+10%)
12 page read speed is 2823 KiB/s
13 page read speed is 2880 KiB/s
14 page read speed is 2842 KiB/s
15 page read speed is 2862 KiB/s
16 page read speed is 2837 KiB/s
32 page read speed is 2879 KiB/s
64 page read speed is 2842 KiB/s
* 1-1-4 result:
1 page read speed is 7562 KiB/s
2 page read speed is 8904 KiB/s (+15%)
3 page read speed is 9655 KiB/s (+25%)
4 page read speed is 10118 KiB/s (+30%)
5 page read speed is 10084 KiB/s
6 page read speed is 10300 KiB/s
7 page read speed is 10434 KiB/s (+35%)
8 page read speed is 10406 KiB/s
9 page read speed is 10769 KiB/s (+40%)
10 page read speed is 10666 KiB/s
11 page read speed is 10757 KiB/s
12 page read speed is 10835 KiB/s
13 page read speed is 10976 KiB/s
14 page read speed is 11200 KiB/s
15 page read speed is 11009 KiB/s
16 page read speed is 11082 KiB/s
32 page read speed is 11352 KiB/s (+45%)
64 page read speed is 11403 KiB/s
This work has received support and could be achieved thanks to
Alvin Zhou <alvinzhou@mxic.com.tw>.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20240826101412.20644-10-miquel.raynal@bootlin.com
Macronix SPI-NANDs encode the ECC status into two bits. There are three
standard situations (no bitflip, bitflips, error), and an additional
possible situation which is only triggered when configuring the 0x10
configuration register, allowing to know, if there have been bitflips,
whether the maximum amount of bitflips was above a configurable
threshold or not. In all cases, for now, s this configuration register
is unset, it means the same as "there are bitflips".
This value is maybe standard, maybe not. For now, let's define it in the
Macronix driver, we can safely move it to a shared place later if that
is relevant.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20240826101412.20644-9-miquel.raynal@bootlin.com
With GET_STATUS commands, SPI-NAND devices can tell the status of the
last read operation, in particular if there was:
- no bitflips
- corrected bitflips
- uncorrectable bitflips
The next step then to read an ECC status register and retrieve the
amount of bitflips, when relevant, if possible. The logic used here
works well for now, but will no longer apply to continuous reads. In
order to prepare the introduction of continuous reads, let's factorize
out the code that is specific to single-page reads.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20240826101412.20644-8-miquel.raynal@bootlin.com
A regular page read consist in:
- Asking one page of content from the NAND array to be loaded in the
chip's SRAM,
- Waiting for the operation to be done,
- Retrieving the data (I/O phase) from the chip's SRAM.
When reading several sequential pages, the above operation is repeated
over and over. There is however a way to optimize these accesses, by
enabling continuous reads. The feature requires the NAND chip to have a
second internal SRAM area plus a bit of additional internal logic to
trigger another internal transfer between the NAND array and the second
SRAM area while the I/O phase is ongoing. Once the first I/O phase is
done, the host can continue reading more data, continuously, as the chip
will automatically switch to the second SRAM content (which has already
been loaded) and in turns trigger the next load into the first SRAM area
again.
From an instruction perspective, the command op-codes are different, but
the same cycles are required. The only difference is that after a
continuous read (which is stopped by a CS deassert), the host must
observe a delay of tRST. However, because there is no guarantee in Linux
regarding the actual state of the CS pin after a transfer (in order to
speed-up the next transfer if targeting the same device), it was
necessary to manually end the continuous read with a configuration
register write operation.
Continuous reads have two main drawbacks:
* They only work on full pages (column address ignored)
* Only the main data area is pulled, out-of-band bytes are not
accessible. Said otherwise, the feature can only be useful with on-die
ECC engines.
Performance wise, measures have been performed on a Zynq platform using
Macronix SPI-NAND controller with a Macronix chip (based on the
flash_speed tool modified for testing sequential reads):
- 1-1-1 mode: performances improved from +3% (2-pages) up to +10% after
a dozen pages.
- 1-1-4 mode: performances improved from +15% (2-pages) up to +40% after
a dozen pages.
This series is based on a previous work from Macronix engineer Jaime
Liao.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Pratyush Yadav <pratyush@kernel.org>
Link: https://lore.kernel.org/linux-mtd/20240826101412.20644-5-miquel.raynal@bootlin.com
There is currently only a single path for performing page reads as
requested by the MTD layer. Soon there will be two:
- a "regular" page read
- a continuous page read
Let's extract the page read logic in a dedicated helper, so the
introduction of continuous page reads will be as easy as checking whether
continuous reads shall/can be used and calling one helper or the other.
There is not behavioral change intended.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20240826101412.20644-4-miquel.raynal@bootlin.com
Use of_property_read_bool() to read boolean properties rather than
of_get_property(). This is part of a larger effort to remove callers
of of_get_property() and similar functions. of_get_property() leaks
the DT property data pointer which is a problem for dynamically
allocated nodes which may be freed.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20240731191312.1710417-22-robh@kernel.org
Pull UBI and UBIFS updates from Richard Weinberger:
- Many fixes for power-cut issues by Zhihao Cheng
- Another ubiblock error path fix
- ubiblock section mismatch fix
- Misc fixes all over the place
* tag 'ubifs-for-linus-6.11-rc1-take2' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/ubifs:
ubi: Fix ubi_init() ubiblock_exit() section mismatch
ubifs: add check for crypto_shash_tfm_digest
ubifs: Fix inconsistent inode size when powercut happens during appendant writing
ubi: block: fix null-pointer-dereference in ubiblock_create()
ubifs: fix kernel-doc warnings
ubifs: correct UBIFS_DFS_DIR_LEN macro definition and improve code clarity
mtd: ubi: Restore missing cleanup on ubi_init() failure path
ubifs: dbg_orphan_check: Fix missed key type checking
ubifs: Fix unattached inode when powercut happens in creating
ubifs: Fix space leak when powercut happens in linking tmpfile
ubifs: Move ui->data initialization after initializing security
ubifs: Fix adding orphan entry twice for the same inode
ubifs: Remove insert_dead_orphan from replaying orphan process
Revert "ubifs: ubifs_symlink: Fix memleak of inode->i_link in error path"
ubifs: Don't add xattr inode into orphan area
ubifs: Fix unattached xattr inode if powercut happens after deleting
mtd: ubi: avoid expensive do_div() on 32-bit machines
mtd: ubi: make ubi_class constant
ubi: eba: properly rollback inside self_check_eba
Pull MTD updates from Miquel Raynal:
"Nothing stands out for this merge window, mostly minor fixes, such as
module descriptions, the use of debug macros and Makefile
improvements.
Raw NAND changes;
- The Freescale MXC driver has been converted to the newer
'->exec_op()' interface
- The meson driver now supports handling the boot ROM area with very
specific ECC needs
- Support for the iMX8QXP has been added to the GPMI driver
- The lpx32xx driver now can get the DMA channels using DT entries
- The Qcom binding has been improved to be more future proof by Rob
- And then there is the usual load of misc and minor changes
SPI-NAND changes:
- The Macronix vendor driver has been improved to support an extended
ID to avoid conflicting with older devices after an ID reuse issue
SPI NOR changes:
- Drop support for Xilinx S3AN flashes. These flashes are for the
very old Xilinx Spartan 3 FPGAs and they need some awkward code in
the core to support.
Drop support for these flashes, along with the special handling we
needed for them in the core like non-power-of-2 page size handling
and the .setup() callback.
- Fix regression for old w25q128 flashes without SFDP tables.
Commit 83e824a4a5 ("mtd: spi-nor: Correct flags for Winbond
w25q128") dropped support for such devices under the assumption
that they aren't being used anymore. Users have now surfaced [0] so
fix the regression by supporting both kind of devices.
- Core cleanups including removal of SPI_NOR_NO_FR flag and
simplification of spi_nor_get_flash_info()"
Link: https://lore.kernel.org/r/CALxbwRo_-9CaJmt7r7ELgu+vOcgk=xZcGHobnKf=oT2=u4d4aA@mail.gmail.com/ [0]
* tag 'mtd/for-6.11' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux: (28 commits)
mtd: rawnand: lpx32xx: Fix dma_request_chan() error checks
mtd: spinand: macronix: Add support for serial NAND flash
mtd: spinand: macronix: Add support for reading Device ID 2
mtd: rawnand: lpx32xx: Request DMA channels using DT entries
dt-bindings: mtd: qcom,nandc: Define properties at top-level
mtd: rawnand: intel: use 'time_left' variable with wait_for_completion_timeout()
mtd: rawnand: mxc: use 'time_left' variable with wait_for_completion_timeout()
mtd: rawnand: gpmi: add iMX8QXP support.
mtd: rawnand: gpmi: add 'support_edo_timing' in gpmi_devdata
mtd: cmdlinepart: Replace `dbg()` macro with `pr_debug()`
mtd: add missing MODULE_DESCRIPTION() macros
mtd: make mtd_test.c a separate module
dt-bindings: mtd: gpmi-nand: Add 'fsl,imx8qxp-gpmi-nand' compatible string
mtd: rawnand: cadence: remove unused struct 'ecc_info'
mtd: rawnand: mxc: support software ECC
mtd: rawnand: mxc: implement exec_op
mtd: rawnand: mxc: separate page read from ecc calc
mtd: spi-nor: winbond: fix w25q128 regression
mtd: spi-nor: simplify spi_nor_get_flash_info()
mtd: spi-nor: get rid of SPI_NOR_NO_FR
...
Raw NAND changes;
The Freescale MXC driver has been converted to the newer ->exec_op()
interface. The meson driver now supports handling the boot ROM area with
very specific ECC needs. Support for the iMX8QXP has been added to the
GPMI driver. The lpx32xx driver now can get the DMA channels using DT
entries. The Qcom binding has been improved to be more future proof by
Rob. And then there is the usual load of misc and minor changes.
SPI-NAND changes:
The Macronix vendor driver has been improved to support an extended ID
to avoid conflicting with older devices after an ID reuse issue.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
SPI NOR changes for 6.11
Notable changes:
- Drop support for Xilinx S3AN flashes. These flashes are for the very
old Xilinx Spartan 3 FPGAs and they need some awkward code in the core
to support. Drop support for these flashes, along with the special
handling we needed for them in the core like non-power-of-2 page size
handling and the .setup() callback.
- Fix regression for old w25q128 flashes without SFDP tables. Commit
83e824a4a5 ("mtd: spi-nor: Correct flags for Winbond w25q128")
dropped support for such devices under the assumption that they aren't
being used anymore. Users have now surfaced [0] so fix the regression
by supporting both kind of devices.
- Core cleanups including removal of SPI_NOR_NO_FR flag and
simplification of spi_nor_get_flash_info().
[0] https://lore.kernel.org/r/CALxbwRo_-9CaJmt7r7ELgu+vOcgk=xZcGHobnKf=oT2=u4d4aA@mail.gmail.com/
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
There is a confusing pattern in the kernel to use a variable named 'timeout' to
store the result of wait_for_completion_timeout() causing patterns like:
timeout = wait_for_completion_timeout(...)
if (!timeout) return -ETIMEDOUT;
with all kinds of permutations. Use 'time_left' as a variable to make the code
self explaining.
Fix to the proper variable type 'unsigned long' while here.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20240604212919.5038-6-wsa+renesas@sang-engineering.com
