Group NOR flags initialization. Introduce a dedicated function for
setting the fixup_flags and emphasise when those flash_info flags
should be set: when the SNOR_F_4B_OPCODES/SNOR_F_IO_MODE_EN_VOLATILE
setttings can not be discovered by SFDP for this particular flash
because the SFDP table that indicates this support is not defined
in the flash.
In case the table for his support is defined but has wrong values,
one should instead use a post_sfdp() hook to set the SNOR_F equivalent
flag.
No functional change intended in this patch.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211207140254.87681-9-tudor.ambarus@microchip.com
Used to initialize the NOR flags for settings that are not defined
in the JESD216 SFDP standard, thus can not be retrieved when parsing
SFDP.
This moves the setting of SNOR_F_READY_XSR_RDY and SNOR_F_HAS_LOCK
late in the init call, without any functional change expected.
The rest of the flags were already set after the spi_nor_init_params().
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211207140254.87681-8-tudor.ambarus@microchip.com
Clarify for what the flash_info flags are used for. Split them in
four categories and a bool:
1/ FLAGS: flags that indicate support that is not defined by the JESD216
standard in its SFDP tables.
2/ NO_SFDP_FLAGS: these flags are used when the flash does not define the
SFDP tables. These flags indicate support that can be discovered via
SFDP. Used together with SPI_NOR_SKIP_SFDP flag.
3/ FIXUP_FLAGS: flags that indicate support that can be discovered
via SFDP ideally, but can not be discovered for this particular flash
because the SFDP table that indicates this support is not defined by
the flash. In case the table for this support is defined but has wrong
values, one should instead use a post_sfdp() hook to set the SNOR_F
equivalent flag.
4/ MFR_FLAGS: manufacturer private flags. Used in the manufacturer
fixup hooks to differentiate support between flashes of the same
manufacturer.
5/ PARSE_SFDP: sets info->parse_sfdp to true. All flash_info entries
that support SFDP should be converted to set info->parse_sfdp to true.
SPI NOR flashes that statically declare one of the
SPI_NOR_{DUAL, QUAD, OCTAL, OCTAL_DTR}_READ flags and do not support
the RDSFDP command are gratuiously receiving the RDSFDP command
in the attempt of parsing the SFDP tables. It is not desirable to issue
commands that are not supported, so introduce PARSE_SFDP to help on this
situation.
New flash additions/updates should be declared/updated to use either
PARSE_SFDP or SPI_NOR_SKIP_SFDP. Once all the flash_info entries are
converted to use SPI_NOR_SKIP_SFDP or PARSE_SFDP, we can get rid of the
SPI_NOR_SKIP_SFDP flag and use just the bool nor->info->parse_sfdp to
determine whether to parse SFDP or not. SPI_NOR_SKIP_SFDP flag is kept
just as a way to differentiate whether a flash is converted to the new
flags logic or not.
Support that can be discovered when parsing SFDP should not be duplicated
by explicit flags at flash declaration. All the flash parameters will be
discovered when parsing SFDP. Sometimes manufacturers wrongly define some
fields in the SFDP tables. If that's the case, SFDP data can be amended
with the fixups() hooks. It is not common, but if the SFDP tables are
entirely wrong, and it does not worth the hassle to tweak the SFDP
parameters by using the fixups hooks, or if the flash does not define the
SFDP tables at all, then statically init the flash with the
SPI_NOR_SKIP_SFDP flag and specify the rest of flash capabilities with
the flash info flags.
With time, we want to convert all flashes to use PARSE_SFDP and
stop triggering the SFDP parsing with the
SPI_NOR_{DUAL, QUAD, OCTAL*}_READ flags. Getting rid of the
SPI_NOR_{OCTAL, OCTAL_DTR}_READ trigger is easily achievable,
the rest are a long term goal.
Manufacturer specific flags like USE_CLSR, USE_FSR, SPI_NOR_XSR_RDY,
will be removed in a future series.
No functional changes intended in this patch.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211207140254.87681-7-tudor.ambarus@microchip.com
Used in the manufacturer fixup hooks to differentiate support
between flashes of the same manufacturer. Not used in the
SPI NOR core.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211207140254.87681-6-tudor.ambarus@microchip.com
spi_nor_post_sfdp_fixups() was called even when there were no SFDP
tables defined. late_init() should be instead used for flashes that
do not define SFDP tables.
Use spi_nor_post_sfdp_fixups() just to fix SFDP data. post_sfdp()
hook is as of now used just by s28hs512t, mt35xu512aba, and both
support SFDP, there's no functional change with this patch.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211207140254.87681-5-tudor.ambarus@microchip.com
Used to init all the mtd_info fields. Move the mtd_info init
the last thing in the spi_nor_scan(), so that we avoid superfluous
initialization of the mtd_info fields in case of errors.
While here use common naming scheme for functions that are setting
mtd_info fields:
s/spi_nor_register_locking_ops/spi_nor_set_mtd_locking_ops
s/spi_nor_otp_init/spi_nor_set_mtd_otp_ops
The functions names are self explanatory, get rid of the comment
for the OTP function.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211207140254.87681-4-tudor.ambarus@microchip.com
Use NOR parameters in the probe's sequence of calls, thus
nor->params->size instead of nor->mtd.size and let the mtd_info
fields be used by the mtd calls (mtd->_erase, mtd->_read, mtd->_write).
mtd_info fields should not be used during probe because we haven't
registered mtd yet. It's safe to drop xilinx's setting of nor->mtd.size,
now that we use nor->params->size in spi_nor_set_addr_width().
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211207140254.87681-3-tudor.ambarus@microchip.com
As it was before the blamed commit, s3an_nor_scan() was called
after mtd size was set with params->size, and it overwrote the mtd
size value with '8 * nor->page_size * nor->info->n_sectors' when
XSR_PAGESIZE was set. With the introduction of
s3an_post_sfdp_fixups(), we missed to update the mtd size for the
s3an flashes. Fix the mtd size by updating both nor->params->size,
(which will update the mtd_info size later on) and nor->mtd.size
(which is used in spi_nor_set_addr_width()).
Fixes: 641edddb4f ("mtd: spi-nor: Add s3an_post_sfdp_fixups()")
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211207140254.87681-2-tudor.ambarus@microchip.com
spansion_post_sfdp_fixups() was called regardless if the flash defined
SFDP tables or not. A better place for this kind of parameters init is
in manufacturer's late_init() hook. post_sfdp() should be called only
when SFDP is defined. No functional change in this patch.
Instead of doing the 4b opcodes settings at manufacturer level, thus
also for every flash that will be introduced, this should be done
just where it is needed, per flash. I'll let this for other patch.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211029172633.886453-12-tudor.ambarus@microchip.com
Setting the correct nor->mtd._write in a fixup hook was misleading,
since this is not a fixup, just a specific setting for SST, that differs
from the SPI NOR core default init.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211029172633.886453-11-tudor.ambarus@microchip.com
post_sfdp was misleading in this case, as SFDP is not supported by
xilinx. Plus, there's no fixup here, just setting the correct
setup method, as required by xilinx parts.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211029172633.886453-10-tudor.ambarus@microchip.com
OTP is not described in the JESD216 SFDP standard, place the
OTP ops init in late_init().
We can't get rid of the default_init() hook for winbond, as the
4byte_addr_mode is SFDP specific and will require to have all
flashes at hand, in order to check which has the SFDP tables defined,
in which case there's nothing to do if the SFDP tables are corect,
and which of the flashes do not define the SFDP tables in which case
each flash should declare a late_init() fixup.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211029172633.886453-9-tudor.ambarus@microchip.com
Locking is not described in JESD216 SFDP standard, place the
locking init in late_init().
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211029172633.886453-8-tudor.ambarus@microchip.com
Locking is not described in JESD216 SFDP standard, place the
locking init in late_init().
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211029172633.886453-7-tudor.ambarus@microchip.com
Flash parameters init is done in a spaghetti way right now.
There is the init based on the flash_info data, then there is the
default_init() hook, then SFDP init, an intermediary post_bft(),
then post_sfdp() and a spi_nor_late_init_params(). Each method can
overwrite previuosly initialized parameters.
We want to separate what is SFDP and non-SFDP specific. late_init()
will replace the default_init() hook and will be used only to initialize
flash parameters that are not declared in the JESD216 SFDP standard, or
where SFDP tables are not defined at all.
We cut a member in the chain of initializing parameters by getting rid
of the default_init() hook, and we make it clear that everything that is
in late_init() is not covered by the SFDP tables defined by the flash.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211029172633.886453-6-tudor.ambarus@microchip.com
nor->page_size duplicated what nor->params->page_size indicates
for no good reason. page_size is a flash parameter of fixed value
and it is better suited to be found in nor->params->page_size.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Link: https://lore.kernel.org/r/20211029172633.886453-5-tudor.ambarus@microchip.com
"struct mtd_info mtd" is member of "struct spi_nor", there's no need
to use "mtd->priv". Get the pointer to the containing struct spi_nor
by using container_of. While here, make the function inline and
get rid of the __maybe_unused.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20211029172633.886453-3-tudor.ambarus@microchip.com
Update the description of the otp member of the
struct spi_nor_flash_parameter.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: Pratyush Yadav <p.yadav@ti.com>
Reviewed-by: Michael Walle <michael@walle.cc>
Link: https://lore.kernel.org/r/20211029172633.886453-2-tudor.ambarus@microchip.com
-----BEGIN PGP SIGNATURE-----
iQJEBAABCAAuFiEEwPw5LcreJtl1+l5K99NY+ylx4KYFAmGKqOcQHGF4Ym9lQGtl
cm5lbC5kawAKCRD301j7KXHgpr3yEADD9Cx8oNk3KzWV3c3JlIR4JQtvpczS3dho
KkGU0D5fOh1sViXbLBNr6VxypcEIKQoHxDQQ6qid1kOu/B3mCNM1duLsVjyj3Qa0
7nbm2dVUsD/EVDuXedRmMvcfCUx6Z23DbpI182wXtIPaCsEEmsANzHnZNg38OV44
25SYG0QUvb9ViSz1Y1GORu0ttEJNF2GhZfiBpb0WveRnY7eTSL/PnHNDzHsSeFv4
zD0W205g7jKbt0+57kgNElTz7DbdM3p8XVex+aXPlFaHz2qx4ZoJJIsaMv/P8tT5
14b50cB41xnPvlGTvqr1WfZZfJocDNq2rG+fh6N5D1sO86ogWpj7psiiADfa0pb6
ZWoJqhk3BvEUMPQ5N/BJ/8j3FWGIYWtKQf4QcyxrJYpqDwtwbBfMlzKkc7JMPFYk
JAi6uq1uF5SbA4x99G90tK85LvxsbkseyIYXgBJ/GIyW5doIPkD9TPDEzJMCdHOe
laynHS5PMHzuhPLuEDDn9sTVXpZWAMBnoy4j1L4wGmBjiogYWLTSJVobODzCAqHY
1Va2oP6SXfCdVRkCysFbcrdsjJuoIWlMKrdE40tNvkmU0v7sEX0Zd+GLHiaWdIZa
fgxC9fmZtDDOowCp+Iw0VaAqPeeptmyUrof06ZktJleOAscX7kSwbxPdmr1FM0jy
dbnLDyaq/A==
=QaFI
-----END PGP SIGNATURE-----
Merge tag 'for-5.16/drivers-2021-11-09' of git://git.kernel.dk/linux-block
Pull more block driver updates from Jens Axboe:
- Last series adding error handling support for add_disk() in drivers.
After this one, and once the SCSI side has been merged, we can
finally annotate add_disk() as must_check. (Luis)
- bcache fixes (Coly)
- zram fixes (Ming)
- ataflop locking fix (Tetsuo)
- nbd fixes (Ye, Yu)
- MD merge via Song
- Cleanup (Yang)
- sysfs fix (Guoqing)
- Misc fixes (Geert, Wu, luo)
* tag 'for-5.16/drivers-2021-11-09' of git://git.kernel.dk/linux-block: (34 commits)
bcache: Revert "bcache: use bvec_virt"
ataflop: Add missing semicolon to return statement
floppy: address add_disk() error handling on probe
ataflop: address add_disk() error handling on probe
block: update __register_blkdev() probe documentation
ataflop: remove ataflop_probe_lock mutex
mtd/ubi/block: add error handling support for add_disk()
block/sunvdc: add error handling support for add_disk()
z2ram: add error handling support for add_disk()
nvdimm/pmem: use add_disk() error handling
nvdimm/pmem: cleanup the disk if pmem_release_disk() is yet assigned
nvdimm/blk: add error handling support for add_disk()
nvdimm/blk: avoid calling del_gendisk() on early failures
nvdimm/btt: add error handling support for add_disk()
nvdimm/btt: use goto error labels on btt_blk_init()
loop: Remove duplicate assignments
drbd: Fix double free problem in drbd_create_device
nvdimm/btt: do not call del_gendisk() if not needed
bcache: fix use-after-free problem in bcache_device_free()
zram: replace fsync_bdev with sync_blockdev
...
* Remove obsolete macros only used by the old nand_ecclayout struct
* Don't remove debugfs directory if device is in use
* MAINTAINERS:
- Add entry for Qualcomm NAND controller driver
- Update the devicetree documentation path of hyperbus
MTD devices:
* block2mtd:
- Add support for an optional custom MTD label
- Minor refactor to avoid hard coded constant
* mtdswap: Remove redundant assignment of pointer eb
CFI:
* Fixup CFI on ixp4xx
Raw NAND controller drivers:
* Arasan:
- Prevent an unsupported configuration
* Xway, Socrates: plat_nand, Pasemi, Orion, mpc5121, GPIO, Au1550nd, AMS-Delta:
- Keep the driver compatible with on-die ECC engines
* cs553x, lpc32xx_slc, ndfc, sharpsl, tmio, txx9ndfmc:
- Revert the commits: "Fix external use of SW Hamming ECC helper"
- And let callers use the bare Hamming helpers
* Fsmc: Fix use of SM ORDER
* Intel:
- Fix potential buffer overflow in probe
* xway, vf610, txx9ndfm, tegra, stm32, plat_nand, oxnas, omap, mtk, hisi504,
gpmi, gpio, denali, bcm6368, atmel:
- Make use of the helper function devm_platform_ioremap_resource{,byname}()
Onenand drivers:
* Samsung: Drop Exynos4 and describe driver in KConfig
Raw NAND chip drivers:
* Hynix: Add support for H27UCG8T2ETR-BC MLC NAND
SPI NOR core:
* Add spi-nor device tree binding under SPI NOR maintainers
SPI NOR manufacturer drivers:
* Enable locking for n25q128a13
SPI NOR controller drivers:
* Use devm_platform_ioremap_resource_byname()
-----BEGIN PGP SIGNATURE-----
iQEzBAABCgAdFiEE9HuaYnbmDhq/XIDIJWrqGEe9VoQFAmGIAdEACgkQJWrqGEe9
VoQHEAgAkFvFxIOyPSwAJfyeEklWmGSh+cm+X9EqWxZ5d6iwFE3FZH4X4XiqdyVY
+rzy37o6Tp3pQVYDqxmPM8GGoHy8wYPR5h/U1IozbBaNeCJma1HyK4Sjb/cuX1eQ
23EM1pWT8N9CbRG2S9mz5C9uHcP9mImtyvU4WhTLKOEc9XGbpuj+k1cacSuEEbeF
rZklOa7tvUiOFIDfr7Kf+DfxbpYabgB2dJgWPEtZmKmefWMiODHgvBJo0MT6MKr7
kOlIkgq5zYIFRNaxApWLI3JmH9+lsw6MHBe+cCqemV6q2OZlMV/5VGFElEywJt+4
L3VmlJQpMiCEMo2Cgc1jigHKryk4Xw==
=h6o3
-----END PGP SIGNATURE-----
Merge tag 'mtd/for-5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux
Pull mtd updates from Miquel Raynal:
"Core:
- Remove obsolete macros only used by the old nand_ecclayout struct
- Don't remove debugfs directory if device is in use
- MAINTAINERS:
- Add entry for Qualcomm NAND controller driver
- Update the devicetree documentation path of hyperbus
MTD devices:
- block2mtd:
- Add support for an optional custom MTD label
- Minor refactor to avoid hard coded constant
- mtdswap: Remove redundant assignment of pointer eb
CFI:
- Fixup CFI on ixp4xx
Raw NAND controller drivers:
- Arasan:
- Prevent an unsupported configuration
- Xway, Socrates: plat_nand, Pasemi, Orion, mpc5121, GPIO, Au1550nd,
AMS-Delta:
- Keep the driver compatible with on-die ECC engines
- cs553x, lpc32xx_slc, ndfc, sharpsl, tmio, txx9ndfmc:
- Revert the commits: "Fix external use of SW Hamming ECC helper"
- And let callers use the bare Hamming helpers
- Fsmc: Fix use of SM ORDER
- Intel:
- Fix potential buffer overflow in probe
- xway, vf610, txx9ndfm, tegra, stm32, plat_nand, oxnas, omap, mtk,
hisi504, gpmi, gpio, denali, bcm6368, atmel:
- Make use of the helper function devm_platform_ioremap_resource{,byname}()
Onenand drivers:
- Samsung: Drop Exynos4 and describe driver in KConfig
Raw NAND chip drivers:
- Hynix: Add support for H27UCG8T2ETR-BC MLC NAND
SPI NOR core:
- Add spi-nor device tree binding under SPI NOR maintainers
SPI NOR manufacturer drivers:
- Enable locking for n25q128a13
SPI NOR controller drivers:
- Use devm_platform_ioremap_resource_byname()"
* tag 'mtd/for-5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux: (50 commits)
mtd: core: don't remove debugfs directory if device is in use
MAINTAINERS: Update the devicetree documentation path of hyperbus
mtd: block2mtd: add support for an optional custom MTD label
mtd: block2mtd: minor refactor to avoid hard coded constant
mtd: fixup CFI on ixp4xx
mtd: rawnand: arasan: Prevent an unsupported configuration
MAINTAINERS: Add entry for Qualcomm NAND controller driver
mtd: rawnand: hynix: Add support for H27UCG8T2ETR-BC MLC NAND
mtd: rawnand: xway: Keep the driver compatible with on-die ECC engines
mtd: rawnand: socrates: Keep the driver compatible with on-die ECC engines
mtd: rawnand: plat_nand: Keep the driver compatible with on-die ECC engines
mtd: rawnand: pasemi: Keep the driver compatible with on-die ECC engines
mtd: rawnand: orion: Keep the driver compatible with on-die ECC engines
mtd: rawnand: mpc5121: Keep the driver compatible with on-die ECC engines
mtd: rawnand: gpio: Keep the driver compatible with on-die ECC engines
mtd: rawnand: au1550nd: Keep the driver compatible with on-die ECC engines
mtd: rawnand: ams-delta: Keep the driver compatible with on-die ECC engines
Revert "mtd: rawnand: cs553x: Fix external use of SW Hamming ECC helper"
Revert "mtd: rawnand: lpc32xx_slc: Fix external use of SW Hamming ECC helper"
Revert "mtd: rawnand: ndfc: Fix external use of SW Hamming ECC helper"
...
- Add spi-nor device tree binding under SPI NOR maintainers
SPI NOR manufacturer drivers changes:
- Enable locking for n25q128a13
SPI NOR controller drivers changes:
- Use devm_platform_ioremap_resource_byname()
-----BEGIN PGP SIGNATURE-----
iQEzBAABCgAdFiEEHUIqys8OyG1eHf7fS1VPR6WNFOkFAmGA9wgACgkQS1VPR6WN
FOnkZAf9E0iin+gx05X6V68NuuQ2iG+gJoHTqCL1WMvBaDbT2vr9gYgclqKAM4FO
2QL66b+zrIhmNP3qPy1HobjxpdCJo8b5PDDShR8aQ1c1RaeOfRNyK9XXsfrI8t1s
mJwIvIcqEAZLskyCotlJVzvHmyTx30RfFbtJD1hjArFUKi1vbdROhx2G1tnAMdVj
tnbRTVL6WTsrSTvgpydtJIfdZ3lS/9RnewCUjQAM9V1PlfJ3ZF4cnRsGPRArGuSb
5f9iHqK6uQtW+K+zuRke6Vf+Xrh6Qan5BZGbn5QhijSaCKgL6aAr6uQ3RDMwu4AU
G0tnqjfeaS6GVV6NV7Fay0M4CVJChA==
=Q/zo
-----END PGP SIGNATURE-----
Merge tag 'spi-nor/for-5.16' into mtd/next
SPI NOR core changes:
- Add spi-nor device tree binding under SPI NOR maintainers
SPI NOR manufacturer drivers changes:
- Enable locking for n25q128a13
SPI NOR controller drivers changes:
- Use devm_platform_ioremap_resource_byname()
* Remove obsolete macros only used by the old nand_ecclayout struct
* MAINTAINERS: Add entry for Qualcomm NAND controller driver
Raw NAND controller drivers:
* Arasan:
- Prevent an unsupported configuration
* Xway, Socrates: plat_nand, Pasemi, Orion, mpc5121, GPIO, Au1550nd, AMS-Delta:
- Keep the driver compatible with on-die ECC engines
* cs553x, lpc32xx_slc, ndfc, sharpsl, tmio, txx9ndfmc:
- Revert the commits: "Fix external use of SW Hamming ECC helper"
- And let callers use the bare Hamming helpers
* Fsmc: Fix use of SM ORDER
* Intel:
- Fix potential buffer overflow in probe
* xway, vf610, txx9ndfm, tegra, stm32, plat_nand, oxnas, omap, mtk, hisi504,
gpmi, gpio, denali, bcm6368, atmel:
- Make use of the helper function devm_platform_ioremap_resource{,byname}()
Onenand driver:
* Samsung: Drop Exynos4 and describe driver in KConfig
Raw NAND chip drivers:
* Hynix: Add support for H27UCG8T2ETR-BC MLC NAND
-----BEGIN PGP SIGNATURE-----
iQEzBAABCgAdFiEE9HuaYnbmDhq/XIDIJWrqGEe9VoQFAmGAOJQACgkQJWrqGEe9
VoTDugf/Z3rfv0A5rEZou8qKVzdLeQ2Aun3HgZZZ3NMMLYTigLiR8iQqR4fvN8KR
xivkbowMoDNgPiVABvTMhD73hPjVokGGE9/KnwY0w6/0tw+pNzkuPBSVO6uYK9Pc
d51cc8OY1Gb9DdYyQB+RI/heox1H8OYskla1lTG14hP69UAqPmj8O43o4OnW/l1T
S8NyBgMAm6YaStAKfJoDNiaDYIYNCMxQVmOB643O71xBkG30cXxPefNDe31yUFVj
b2lL1E2zQT6xc44a9tu5T9WbXBz9oj+r9JTmrAXdtqcqWnC903ap6QYttc5jLoqd
IJw25CZqZjy+/3j8smtxxmIu3kptCA==
=zb5E
-----END PGP SIGNATURE-----
Merge tag 'nand/for-5.16' into mtd/next
Core:
* Remove obsolete macros only used by the old nand_ecclayout struct
* MAINTAINERS: Add entry for Qualcomm NAND controller driver
Raw NAND controller drivers:
* Arasan:
- Prevent an unsupported configuration
* Xway, Socrates: plat_nand, Pasemi, Orion, mpc5121, GPIO, Au1550nd, AMS-Delta:
- Keep the driver compatible with on-die ECC engines
* cs553x, lpc32xx_slc, ndfc, sharpsl, tmio, txx9ndfmc:
- Revert the commits: "Fix external use of SW Hamming ECC helper"
- And let callers use the bare Hamming helpers
* Fsmc: Fix use of SM ORDER
* Intel:
- Fix potential buffer overflow in probe
* xway, vf610, txx9ndfm, tegra, stm32, plat_nand, oxnas, omap, mtk, hisi504,
gpmi, gpio, denali, bcm6368, atmel:
- Make use of the helper function devm_platform_ioremap_resource{,byname}()
Onenand driver:
* Samsung: Drop Exynos4 and describe driver in KConfig
Raw NAND chip drivers:
* Hynix: Add support for H27UCG8T2ETR-BC MLC NAND
Merge misc updates from Andrew Morton:
"257 patches.
Subsystems affected by this patch series: scripts, ocfs2, vfs, and
mm (slab-generic, slab, slub, kconfig, dax, kasan, debug, pagecache,
gup, swap, memcg, pagemap, mprotect, mremap, iomap, tracing, vmalloc,
pagealloc, memory-failure, hugetlb, userfaultfd, vmscan, tools,
memblock, oom-kill, hugetlbfs, migration, thp, readahead, nommu, ksm,
vmstat, madvise, memory-hotplug, rmap, zsmalloc, highmem, zram,
cleanups, kfence, and damon)"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (257 commits)
mm/damon: remove return value from before_terminate callback
mm/damon: fix a few spelling mistakes in comments and a pr_debug message
mm/damon: simplify stop mechanism
Docs/admin-guide/mm/pagemap: wordsmith page flags descriptions
Docs/admin-guide/mm/damon/start: simplify the content
Docs/admin-guide/mm/damon/start: fix a wrong link
Docs/admin-guide/mm/damon/start: fix wrong example commands
mm/damon/dbgfs: add adaptive_targets list check before enable monitor_on
mm/damon: remove unnecessary variable initialization
Documentation/admin-guide/mm/damon: add a document for DAMON_RECLAIM
mm/damon: introduce DAMON-based Reclamation (DAMON_RECLAIM)
selftests/damon: support watermarks
mm/damon/dbgfs: support watermarks
mm/damon/schemes: activate schemes based on a watermarks mechanism
tools/selftests/damon: update for regions prioritization of schemes
mm/damon/dbgfs: support prioritization weights
mm/damon/vaddr,paddr: support pageout prioritization
mm/damon/schemes: prioritize regions within the quotas
mm/damon/selftests: support schemes quotas
mm/damon/dbgfs: support quotas of schemes
...
Call bdi_unregister explicitly instead of relying on the automatic
unregistration.
Link: https://lkml.kernel.org/r/20211021124441.668816-3-hch@lst.de
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Miquel Raynal <miquel.raynal@bootlin.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We never checked for errors on add_disk() as this function
returned void. Now that this is fixed, use the shiny new
error handling.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
Link: https://lore.kernel.org/r/20211103230437.1639990-10-mcgrof@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We never checked for errors on add_disk() as this function
returned void. Now that this is fixed, use the shiny new
error handling.
Acked-by: Miquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
Link: https://lore.kernel.org/r/20211015233028.2167651-10-mcgrof@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move inode_to_bdi out of line to avoid having to include blkdev.h.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20210920123328.1399408-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Previously, if del_mtd_device() failed with -EBUSY due to a non-zero
usecount, a subsequent call to attempt the deletion again would try to
remove a debugfs directory that had already been removed and panic.
With this change the second call can instead proceed safely.
Fixes: e8e3edb95c ("mtd: create per-device and module-scope debugfs entries")
Signed-off-by: Zev Weiss <zev@bewilderbeest.net>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20211014203953.5424-1-zev@bewilderbeest.net
This patch adds support for an optional MTD label for mtd2block emulated
MTD devices. Useful when, e.g., testing device images using Qemu.
The following line in /etc/fstab can then be used to mount a file system
regardless if running on an embedded system, or emulated with block2mtd:
mtd:Config /mnt jffs2 noatime,nodiratime 0 0
Kernel command line syntax in the emulated case:
block2mtd.block2mtd=/dev/sda,,Config
Notice the ',,' it is the optional erase_size, which like before this
patch, defaults to PAGE_SIZE when omitted. Hence the strlen() check.
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20211009060955.552636-3-troglobit@gmail.com
drivers/mtd/maps/ixp4xx.c requires MTD_CFI_BE_BYTE_SWAP to be set
in order to compile.
drivers/mtd/maps/ixp4xx.c:57:4: error: #error CONFIG_MTD_CFI_BE_BYTE_SWAP required
This patch avoids the #error output by enforcing the policy in
Kconfig. Not sure if this is the right approach, but it helps doing
randconfig builds.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210927141045.1597593-1-arnd@kernel.org
Under the following conditions:
* after rounding up by 4 the number of bytes to transfer (this is
related to the controller's internal constraints),
* if this (rounded) amount of data is situated beyond the end of the
device,
* and only in NV-DDR mode,
the Arasan NAND controller timeouts.
This currently can happen in a particular helper used when picking
software ECC algorithms. Let's prevent this situation by refusing to use
the NV-DDR interface with software engines.
Fixes: 4edde60314 ("mtd: rawnand: arasan: Support NV-DDR interface")
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20211008163640.1753821-1-miquel.raynal@bootlin.com
Add support for the H27UCG8T2ETR-BC MLC NAND. The NAND is used widely
in the NTC CHIP, is an MLC type NAND, and is 8GB in size. Neither
JEDEC nor ONFI detection identifies it correctly, so the ID is added
to the nand_ids.c file. Additionally, per the datasheet this NAND
appears to use the same paired pages scheme as the Toshiba
TC58TEG5DCLTA00 (dist3), so add support for that to enable use in
SLC emulation mode.
Tested on a NTC CHIP the device is able to write to a ubifs formatted
partition, and then have U-Boot (with proposed patches) boot from a
kernel located on that ubifs formatted partition.
Signed-off-by: Chris Morgan <macromorgan@hotmail.com>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210930162402.344-1-macroalpha82@gmail.com
Following the introduction of the generic ECC engine infrastructure, it
was necessary to reorganize the code and move the ECC configuration in
the ->attach_chip() hook. Failing to do that properly lead to a first
series of fixes supposed to stabilize the situation. Unfortunately, this
only fixed the use of software ECC engines, preventing any other kind of
engine to be used, including on-die ones.
It is now time to (finally) fix the situation by ensuring that we still
provide a default (eg. software ECC) but will still support different
ECC engines such as on-die ECC engines if properly described in the
device tree.
There are no changes needed on the core side in order to do this, but we
just need to leverage the logic there which allows:
1- a subsystem default (set to Host engines in the raw NAND world)
2- a driver specific default (here set to software ECC engines)
3- any type of engine requested by the user (ie. described in the DT)
As the raw NAND subsystem has not yet been fully converted to the ECC
engine infrastructure, in order to provide a default ECC engine for this
driver we need to set chip->ecc.engine_type *before* calling
nand_scan(). During the initialization step, the core will consider this
entry as the default engine for this driver. This value may of course
be overloaded by the user if the usual DT properties are provided.
Fixes: d525914b5b ("mtd: rawnand: xway: Move the ECC initialization to ->attach_chip()")
Cc: stable@vger.kernel.org
Cc: Jan Hoffmann <jan@3e8.eu>
Cc: Kestrel seventyfour <kestrelseventyfour@gmail.com>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Tested-by: Jan Hoffmann <jan@3e8.eu>
Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-10-miquel.raynal@bootlin.com
Following the introduction of the generic ECC engine infrastructure, it
was necessary to reorganize the code and move the ECC configuration in
the ->attach_chip() hook. Failing to do that properly lead to a first
series of fixes supposed to stabilize the situation. Unfortunately, this
only fixed the use of software ECC engines, preventing any other kind of
engine to be used, including on-die ones.
It is now time to (finally) fix the situation by ensuring that we still
provide a default (eg. software ECC) but will still support different
ECC engines such as on-die ECC engines if properly described in the
device tree.
There are no changes needed on the core side in order to do this, but we
just need to leverage the logic there which allows:
1- a subsystem default (set to Host engines in the raw NAND world)
2- a driver specific default (here set to software ECC engines)
3- any type of engine requested by the user (ie. described in the DT)
As the raw NAND subsystem has not yet been fully converted to the ECC
engine infrastructure, in order to provide a default ECC engine for this
driver we need to set chip->ecc.engine_type *before* calling
nand_scan(). During the initialization step, the core will consider this
entry as the default engine for this driver. This value may of course
be overloaded by the user if the usual DT properties are provided.
Fixes: b36bf0a0fe ("mtd: rawnand: socrates: Move the ECC initialization to ->attach_chip()")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-9-miquel.raynal@bootlin.com
Following the introduction of the generic ECC engine infrastructure, it
was necessary to reorganize the code and move the ECC configuration in
the ->attach_chip() hook. Failing to do that properly lead to a first
series of fixes supposed to stabilize the situation. Unfortunately, this
only fixed the use of software ECC engines, preventing any other kind of
engine to be used, including on-die ones.
It is now time to (finally) fix the situation by ensuring that we still
provide a default (eg. software ECC) but will still support different
ECC engines such as on-die ECC engines if properly described in the
device tree.
There are no changes needed on the core side in order to do this, but we
just need to leverage the logic there which allows:
1- a subsystem default (set to Host engines in the raw NAND world)
2- a driver specific default (here set to software ECC engines)
3- any type of engine requested by the user (ie. described in the DT)
As the raw NAND subsystem has not yet been fully converted to the ECC
engine infrastructure, in order to provide a default ECC engine for this
driver we need to set chip->ecc.engine_type *before* calling
nand_scan(). During the initialization step, the core will consider this
entry as the default engine for this driver. This value may of course
be overloaded by the user if the usual DT properties are provided.
Fixes: 612e048e6a ("mtd: rawnand: plat_nand: Move the ECC initialization to ->attach_chip()")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-8-miquel.raynal@bootlin.com
Following the introduction of the generic ECC engine infrastructure, it
was necessary to reorganize the code and move the ECC configuration in
the ->attach_chip() hook. Failing to do that properly lead to a first
series of fixes supposed to stabilize the situation. Unfortunately, this
only fixed the use of software ECC engines, preventing any other kind of
engine to be used, including on-die ones.
It is now time to (finally) fix the situation by ensuring that we still
provide a default (eg. software ECC) but will still support different
ECC engines such as on-die ECC engines if properly described in the
device tree.
There are no changes needed on the core side in order to do this, but we
just need to leverage the logic there which allows:
1- a subsystem default (set to Host engines in the raw NAND world)
2- a driver specific default (here set to software ECC engines)
3- any type of engine requested by the user (ie. described in the DT)
As the raw NAND subsystem has not yet been fully converted to the ECC
engine infrastructure, in order to provide a default ECC engine for this
driver we need to set chip->ecc.engine_type *before* calling
nand_scan(). During the initialization step, the core will consider this
entry as the default engine for this driver. This value may of course
be overloaded by the user if the usual DT properties are provided.
Fixes: 8fc6f1f042 ("mtd: rawnand: pasemi: Move the ECC initialization to ->attach_chip()")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-7-miquel.raynal@bootlin.com
Following the introduction of the generic ECC engine infrastructure, it
was necessary to reorganize the code and move the ECC configuration in
the ->attach_chip() hook. Failing to do that properly lead to a first
series of fixes supposed to stabilize the situation. Unfortunately, this
only fixed the use of software ECC engines, preventing any other kind of
engine to be used, including on-die ones.
It is now time to (finally) fix the situation by ensuring that we still
provide a default (eg. software ECC) but will still support different
ECC engines such as on-die ECC engines if properly described in the
device tree.
There are no changes needed on the core side in order to do this, but we
just need to leverage the logic there which allows:
1- a subsystem default (set to Host engines in the raw NAND world)
2- a driver specific default (here set to software ECC engines)
3- any type of engine requested by the user (ie. described in the DT)
As the raw NAND subsystem has not yet been fully converted to the ECC
engine infrastructure, in order to provide a default ECC engine for this
driver we need to set chip->ecc.engine_type *before* calling
nand_scan(). During the initialization step, the core will consider this
entry as the default engine for this driver. This value may of course
be overloaded by the user if the usual DT properties are provided.
Fixes: 553508cec2 ("mtd: rawnand: orion: Move the ECC initialization to ->attach_chip()")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-6-miquel.raynal@bootlin.com
Following the introduction of the generic ECC engine infrastructure, it
was necessary to reorganize the code and move the ECC configuration in
the ->attach_chip() hook. Failing to do that properly lead to a first
series of fixes supposed to stabilize the situation. Unfortunately, this
only fixed the use of software ECC engines, preventing any other kind of
engine to be used, including on-die ones.
It is now time to (finally) fix the situation by ensuring that we still
provide a default (eg. software ECC) but will still support different
ECC engines such as on-die ECC engines if properly described in the
device tree.
There are no changes needed on the core side in order to do this, but we
just need to leverage the logic there which allows:
1- a subsystem default (set to Host engines in the raw NAND world)
2- a driver specific default (here set to software ECC engines)
3- any type of engine requested by the user (ie. described in the DT)
As the raw NAND subsystem has not yet been fully converted to the ECC
engine infrastructure, in order to provide a default ECC engine for this
driver we need to set chip->ecc.engine_type *before* calling
nand_scan(). During the initialization step, the core will consider this
entry as the default engine for this driver. This value may of course
be overloaded by the user if the usual DT properties are provided.
Fixes: 6dd09f775b ("mtd: rawnand: mpc5121: Move the ECC initialization to ->attach_chip()")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-5-miquel.raynal@bootlin.com
Following the introduction of the generic ECC engine infrastructure, it
was necessary to reorganize the code and move the ECC configuration in
the ->attach_chip() hook. Failing to do that properly lead to a first
series of fixes supposed to stabilize the situation. Unfortunately, this
only fixed the use of software ECC engines, preventing any other kind of
engine to be used, including on-die ones.
It is now time to (finally) fix the situation by ensuring that we still
provide a default (eg. software ECC) but will still support different
ECC engines such as on-die ECC engines if properly described in the
device tree.
There are no changes needed on the core side in order to do this, but we
just need to leverage the logic there which allows:
1- a subsystem default (set to Host engines in the raw NAND world)
2- a driver specific default (here set to software ECC engines)
3- any type of engine requested by the user (ie. described in the DT)
As the raw NAND subsystem has not yet been fully converted to the ECC
engine infrastructure, in order to provide a default ECC engine for this
driver we need to set chip->ecc.engine_type *before* calling
nand_scan(). During the initialization step, the core will consider this
entry as the default engine for this driver. This value may of course
be overloaded by the user if the usual DT properties are provided.
Fixes: f6341f6448 ("mtd: rawnand: gpio: Move the ECC initialization to ->attach_chip()")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-4-miquel.raynal@bootlin.com
Following the introduction of the generic ECC engine infrastructure, it
was necessary to reorganize the code and move the ECC configuration in
the ->attach_chip() hook. Failing to do that properly lead to a first
series of fixes supposed to stabilize the situation. Unfortunately, this
only fixed the use of software ECC engines, preventing any other kind of
engine to be used, including on-die ones.
It is now time to (finally) fix the situation by ensuring that we still
provide a default (eg. software ECC) but will still support different
ECC engines such as on-die ECC engines if properly described in the
device tree.
There are no changes needed on the core side in order to do this, but we
just need to leverage the logic there which allows:
1- a subsystem default (set to Host engines in the raw NAND world)
2- a driver specific default (here set to software ECC engines)
3- any type of engine requested by the user (ie. described in the DT)
As the raw NAND subsystem has not yet been fully converted to the ECC
engine infrastructure, in order to provide a default ECC engine for this
driver we need to set chip->ecc.engine_type *before* calling
nand_scan(). During the initialization step, the core will consider this
entry as the default engine for this driver. This value may of course
be overloaded by the user if the usual DT properties are provided.
Fixes: dbffc8ccdf ("mtd: rawnand: au1550: Move the ECC initialization to ->attach_chip()")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-3-miquel.raynal@bootlin.com
Following the introduction of the generic ECC engine infrastructure, it
was necessary to reorganize the code and move the ECC configuration in
the ->attach_chip() hook. Failing to do that properly lead to a first
series of fixes supposed to stabilize the situation. Unfortunately, this
only fixed the use of software ECC engines, preventing any other kind of
engine to be used, including on-die ones.
It is now time to (finally) fix the situation by ensuring that we still
provide a default (eg. software ECC) but will still support different
ECC engines such as on-die ECC engines if properly described in the
device tree.
There are no changes needed on the core side in order to do this, but we
just need to leverage the logic there which allows:
1- a subsystem default (set to Host engines in the raw NAND world)
2- a driver specific default (here set to software ECC engines)
3- any type of engine requested by the user (ie. described in the DT)
As the raw NAND subsystem has not yet been fully converted to the ECC
engine infrastructure, in order to provide a default ECC engine for this
driver we need to set chip->ecc.engine_type *before* calling
nand_scan(). During the initialization step, the core will consider this
entry as the default engine for this driver. This value may of course
be overloaded by the user if the usual DT properties are provided.
Fixes: 59d9347332 ("mtd: rawnand: ams-delta: Move the ECC initialization to ->attach_chip()")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-2-miquel.raynal@bootlin.com
This reverts commit 56a8d3fd1f.
Before the introduction of the ECC framework infrastructure, many
drivers used the ->calculate/correct() Hamming helpers directly. The
point of this framework was to avoid this kind of hackish calls and use a
proper and generic API but it is true that in certain cases, drivers
still need to use these helpers in order to do ECC computations on
behalf of their limited hardware.
Right after the introduction of the ECC engine core introduction, it was
spotted that it was not possible to use the shiny rawnand software ECC
helpers so easily because an ECC engine object should have been
allocated and initialized first. While this works well in most cases,
for these drivers just leveraging the power of a single helper in
conjunction with some pretty old and limited hardware, it did not fit.
The idea back then was to declare intermediate helpers which would make
use of the exported software ECC engine bare functions while keeping the
rawnand layer compatibility. As there was already functions with the
rawnand_sw_hamming_ prefix it was decided to declare new local helpers
for this purpose in each driver needing one.
Besides being far from optimal, this design choice was blamed by Linus
when he pulled the "fixes" pull request [1] so that is why now it is
time to clean this mess up.
The implementation of the rawnand_ecc_sw_* helpers has now been enhanced
to support both cases, when the ECC object is instantiated and when it is
not. This way, we can still use the existing and exported rawnand
helpers while avoiding the need for each driver to declare its own
helper, thus this fix from [2] can now be safely reverted.
[1] https://lore.kernel.org/lkml/CAHk-=wh_ZHF685Fni8V9is17mj=pFisUaZ_0=gq6nbK+ZcyQmg@mail.gmail.com/
[2] https://lore.kernel.org/linux-mtd/20210413161840.345208-1-miquel.raynal@bootlin.com
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928221507.199198-9-miquel.raynal@bootlin.com
This reverts commit c4b7d7c480.
Before the introduction of the ECC framework infrastructure, many
drivers used the ->calculate/correct() Hamming helpers directly. The
point of this framework was to avoid this kind of hackish calls and use a
proper and generic API but it is true that in certain cases, drivers
still need to use these helpers in order to do ECC computations on
behalf of their limited hardware.
Right after the introduction of the ECC engine core introduction, it was
spotted that it was not possible to use the shiny rawnand software ECC
helpers so easily because an ECC engine object should have been
allocated and initialized first. While this works well in most cases,
for these drivers just leveraging the power of a single helper in
conjunction with some pretty old and limited hardware, it did not fit.
The idea back then was to declare intermediate helpers which would make
use of the exported software ECC engine bare functions while keeping the
rawnand layer compatibility. As there was already functions with the
rawnand_sw_hamming_ prefix it was decided to declare new local helpers
for this purpose in each driver needing one.
Besides being far from optimal, this design choice was blamed by Linus
when he pulled the "fixes" pull request [1] so that is why now it is
time to clean this mess up.
The implementation of the rawnand_ecc_sw_* helpers has now been enhanced
to support both cases, when the ECC object is instantiated and when it is
not. This way, we can still use the existing and exported rawnand
helpers while avoiding the need for each driver to declare its own
helper, thus this fix from [2] can now be safely reverted.
[1] https://lore.kernel.org/lkml/CAHk-=wh_ZHF685Fni8V9is17mj=pFisUaZ_0=gq6nbK+ZcyQmg@mail.gmail.com/
[2] https://lore.kernel.org/linux-mtd/20210413161840.345208-1-miquel.raynal@bootlin.com
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928221507.199198-8-miquel.raynal@bootlin.com
This reverts commit 3e09c02525.
Before the introduction of the ECC framework infrastructure, many
drivers used the ->calculate/correct() Hamming helpers directly. The
point of this framework was to avoid this kind of hackish calls and use a
proper and generic API but it is true that in certain cases, drivers
still need to use these helpers in order to do ECC computations on
behalf of their limited hardware.
Right after the introduction of the ECC engine core introduction, it was
spotted that it was not possible to use the shiny rawnand software ECC
helpers so easily because an ECC engine object should have been
allocated and initialized first. While this works well in most cases,
for these drivers just leveraging the power of a single helper in
conjunction with some pretty old and limited hardware, it did not fit.
The idea back then was to declare intermediate helpers which would make
use of the exported software ECC engine bare functions while keeping the
rawnand layer compatibility. As there was already functions with the
rawnand_sw_hamming_ prefix it was decided to declare new local helpers
for this purpose in each driver needing one.
Besides being far from optimal, this design choice was blamed by Linus
when he pulled the "fixes" pull request [1] so that is why now it is
time to clean this mess up.
The implementation of the rawnand_ecc_sw_* helpers has now been enhanced
to support both cases, when the ECC object is instantiated and when it is
not. This way, we can still use the existing and exported rawnand
helpers while avoiding the need for each driver to declare its own
helper, thus this fix from [2] can now be safely reverted.
[1] https://lore.kernel.org/lkml/CAHk-=wh_ZHF685Fni8V9is17mj=pFisUaZ_0=gq6nbK+ZcyQmg@mail.gmail.com/
[2] https://lore.kernel.org/linux-mtd/20210413161840.345208-1-miquel.raynal@bootlin.com
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928221507.199198-7-miquel.raynal@bootlin.com
This reverts commit 46fcb57e6b.
Before the introduction of the ECC framework infrastructure, many
drivers used the ->calculate/correct() Hamming helpers directly. The
point of this framework was to avoid this kind of hackish calls and use a
proper and generic API but it is true that in certain cases, drivers
still need to use these helpers in order to do ECC computations on
behalf of their limited hardware.
Right after the introduction of the ECC engine core introduction, it was
spotted that it was not possible to use the shiny rawnand software ECC
helpers so easily because an ECC engine object should have been
allocated and initialized first. While this works well in most cases,
for these drivers just leveraging the power of a single helper in
conjunction with some pretty old and limited hardware, it did not fit.
The idea back then was to declare intermediate helpers which would make
use of the exported software ECC engine bare functions while keeping the
rawnand layer compatibility. As there was already functions with the
rawnand_sw_hamming_ prefix it was decided to declare new local helpers
for this purpose in each driver needing one.
Besides being far from optimal, this design choice was blamed by Linus
when he pulled the "fixes" pull request [1] so that is why now it is
time to clean this mess up.
The implementation of the rawnand_ecc_sw_* helpers has now been enhanced
to support both cases, when the ECC object is instantiated and when it is
not. This way, we can still use the existing and exported rawnand
helpers while avoiding the need for each driver to declare its own
helper, thus this fix from [2] can now be safely reverted.
[1] https://lore.kernel.org/lkml/CAHk-=wh_ZHF685Fni8V9is17mj=pFisUaZ_0=gq6nbK+ZcyQmg@mail.gmail.com/
[2] https://lore.kernel.org/linux-mtd/20210413161840.345208-1-miquel.raynal@bootlin.com
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928221507.199198-6-miquel.raynal@bootlin.com
This reverts commit 6a4c5ada57.
Before the introduction of the ECC framework infrastructure, many
drivers used the ->calculate/correct() Hamming helpers directly. The
point of this framework was to avoid this kind of hackish calls and use a
proper and generic API but it is true that in certain cases, drivers
still need to use these helpers in order to do ECC computations on
behalf of their limited hardware.
Right after the introduction of the ECC engine core introduction, it was
spotted that it was not possible to use the shiny rawnand software ECC
helpers so easily because an ECC engine object should have been
allocated and initialized first. While this works well in most cases,
for these drivers just leveraging the power of a single helper in
conjunction with some pretty old and limited hardware, it did not fit.
The idea back then was to declare intermediate helpers which would make
use of the exported software ECC engine bare functions while keeping the
rawnand layer compatibility. As there was already functions with the
rawnand_sw_hamming_ prefix it was decided to declare new local helpers
for this purpose in each driver needing one.
Besides being far from optimal, this design choice was blamed by Linus
when he pulled the "fixes" pull request [1] so that is why now it is
time to clean this mess up.
The implementation of the rawnand_ecc_sw_* helpers has now been enhanced
to support both cases, when the ECC object is instantiated and when it is
not. This way, we can still use the existing and exported rawnand
helpers while avoiding the need for each driver to declare its own
helper, thus this fix from [2] can now be safely reverted.
[1] https://lore.kernel.org/lkml/CAHk-=wh_ZHF685Fni8V9is17mj=pFisUaZ_0=gq6nbK+ZcyQmg@mail.gmail.com/
[2] https://lore.kernel.org/linux-mtd/20210413161840.345208-1-miquel.raynal@bootlin.com
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928221507.199198-5-miquel.raynal@bootlin.com
This reverts commit 3d227a0b0c.
Before the introduction of the ECC framework infrastructure, many
drivers used the ->calculate/correct() Hamming helpers directly. The
point of this framework was to avoid this kind of hackish calls and use a
proper and generic API but it is true that in certain cases, drivers
still need to use these helpers in order to do ECC computations on
behalf of their limited hardware.
Right after the introduction of the ECC engine core introduction, it was
spotted that it was not possible to use the shiny rawnand software ECC
helpers so easily because an ECC engine object should have been
allocated and initialized first. While this works well in most cases,
for these drivers just leveraging the power of a single helper in
conjunction with some pretty old and limited hardware, it did not fit.
The idea back then was to declare intermediate helpers which would make
use of the exported software ECC engine bare functions while keeping the
rawnand layer compatibility. As there was already functions with the
rawnand_sw_hamming_ prefix it was decided to declare new local helpers
for this purpose in each driver needing one.
Besides being far from optimal, this design choice was blamed by Linus
when he pulled the "fixes" pull request [1] so that is why now it is
time to clean this mess up.
The implementation of the rawnand_ecc_sw_* helpers has now been enhanced
to support both cases, when the ECC object is instantiated and when it is
not. This way, we can still use the existing and exported rawnand
helpers while avoiding the need for each driver to declare its own
helper, thus this fix from [2] can now be safely reverted.
[1] https://lore.kernel.org/lkml/CAHk-=wh_ZHF685Fni8V9is17mj=pFisUaZ_0=gq6nbK+ZcyQmg@mail.gmail.com/
[2] https://lore.kernel.org/linux-mtd/20210413161840.345208-1-miquel.raynal@bootlin.com
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928221507.199198-4-miquel.raynal@bootlin.com
Before the introduction of the ECC framework infrastructure, many
drivers used the ->calculate/correct() Hamming helpers directly. The
point of this framework was to avoid this kind of hackish calls and use a
proper and generic API but it is true that in certain cases, drivers
still need to use these helpers in order to do ECC computations on
behalf of their limited hardware.
Right after the introduction of the ECC engine core introduction, it was
spotted that it was not possible to use the shiny rawnand software ECC
helpers so easily because an ECC engine object should have been
allocated and initialized first. While this works well in most cases,
for these drivers just leveraging the power of a single helper in
conjunction with some pretty old and limited hardware, it did not fit.
The idea back then was to declare intermediate helpers which would make
use of the exported software ECC engine bare functions while keeping the
rawnand layer compatibility. As there was already functions with the
rawnand_sw_hamming_ prefix it was decided to declare new local helpers
for this purpose in each driver needing one.
Besides being far from optimal, this design choice was blamed by Linus
when he pulled the "fixes" pull request [1] so that is why now it is
time to clean this mess up.
Enhancing the implementation of the rawnand_ecc_sw_* helpers to support
both cases, when the ECC object is instantiated and when it is not is a
quite elegant way to solve this situation. This way, we can still use
the existing and exported rawnand helpers while avoiding the need for
each driver to declare its own helper.
Following this change, most of the fixes sent in [2] can now be safely
reverted. Only the fsmc fix will need to be kept because there is
actually something specific to the driver to do in its ->correct()
helper.
[1] https://lore.kernel.org/lkml/CAHk-=wh_ZHF685Fni8V9is17mj=pFisUaZ_0=gq6nbK+ZcyQmg@mail.gmail.com/
[2] https://lore.kernel.org/linux-mtd/20210413161840.345208-1-miquel.raynal@bootlin.com/
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928221507.199198-3-miquel.raynal@bootlin.com
The introduction of the generic ECC engine API lead to a number of
changes in various drivers which broke some of them. Here is a typical
example: I expected the SM_ORDER option to be handled by the Hamming ECC
engine internals. Problem: the fsmc driver does not instantiate (yet) a
real ECC engine object so we had to use a 'bare' ECC helper instead of
the shiny rawnand functions. However, when not intializing this engine
properly and using the bare helpers, we do not get the SM ORDER feature
handled automatically. It looks like this was lost in the process so
let's ensure we use the right SM ORDER now.
Fixes: ad9ffdce45 ("mtd: rawnand: fsmc: Fix external use of SW Hamming ECC helper")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/linux-mtd/20210928221507.199198-2-miquel.raynal@bootlin.com
