nvmem: fixes for 6.16

PATCH Superseded [email protected] 3 participants 9 messages Jul 12, 2025 – Jul 21, 2025

Subsystems (4)

Stable branch Arm i.MX and Layerscape NVMEM framework U-boot environment variables

lore

Share & tools

lore

[email protected]

author maintainer

2025-07-12 18:17 UTC #0 lore raw

cover nvmem: fixes for 6.16

From: Srinivas Kandagatla <[email protected]>

Hi Greg,
Here are two nvmem fixes for 6.16, Could you queue these for 6.16

Fixes include:
- crc32 endianness conversion in u-boot-env
- fix the mac address length in imx-ocotp provider

thanks,
Srini
.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Michael C. Pratt (1):
nvmem: layouts: u-boot-env: remove crc32 endianness conversion

Steffen Bätz (1):
nvmem: imx-ocotp: fix MAC address byte length

 drivers/nvmem/imx-ocotp-ele.c      | 5 ++++-
 drivers/nvmem/imx-ocotp.c          | 5 ++++-
 drivers/nvmem/layouts/u-boot-env.c | 6 +++---
 3 files changed, 11 insertions(+), 5 deletions(-)

Signature · 1 line

2.43.0

[email protected]

author maintainer

2025-07-12 18:17 UTC #1 lore raw

PATCH 1/2 nvmem: layouts: u-boot-env: remove crc32 endianness conversion

From: "Michael C. Pratt" <[email protected]>

On 11 Oct 2022, it was reported that the crc32 verification of the u-boot environment failed only on big-endian systems for the u-boot-env nvmem layout driver with the following error.

Invalid calculated CRC32: 0x88cd6f09 (expected: 0x096fcd88)

This problem has been present since the driver was introduced, and before it was made into a layout driver.

The suggested fix at the time was to use further endianness conversion macros in order to have both the stored and calculated crc32 values to compare always represented in the system's endianness. This was not accepted due to sparse warnings
and some disagreement on how to handle the situation.
Later on in a newer revision of the patch, it was proposed to use cpu_to_le32() for both values to compare instead of le32_to_cpu() and store the values as __le32 type to remove compilation errors.

The necessity of this is based on the assumption that the use of crc32() requires endianness conversion because the algorithm uses little-endian, however, this does not prove to be the case and the issue is unrelated.

Upon inspecting the current kernel code,
there already is an existing use of le32_to_cpu() in this driver, which suggests there already is special handling for big-endian systems, however, it is big-endian systems that have the problem.

This, being the only functional difference between architectures in the driver combined with the fact that the suggested fix was to use the exact same endianness conversion for the values brings up the possibility that it was not necessary to begin with, as the same endianness conversion for two values expected to be the same is expected to be equivalent to no conversion at all.

After inspecting the u-boot environment of devices of both endianness and trying to remove the existing endianness conversion,
the problem is resolved in an equivalent way as the other suggested fixes.

Ultimately, it seems that u-boot is agnostic to endianness at least for the purpose of environment variables.
In other words, u-boot reads and writes the stored crc32 value with the same endianness that the crc32 value is calculated with in whichever endianness a certain architecture runs on.

Therefore, the u-boot-env driver does not need to convert endianness. Remove the usage of endianness macros in the u-boot-env driver, and change the type of local variables to maintain the same return type.

If there is a special situation in the case of endianness,
it would be a corner case and should be handled by a unique "compatible".

Even though it is not necessary to use endianness conversion macros here, it may be useful to use them in the future for consistent error printing.

Fixes d5542923f200 ("nvmem: add driver handling U-Boot environment variables")

Reported-by INAGAKI Hiroshi <[email protected]>

Link https://lore.kernel.org/all/[email protected] open here

Cc [email protected] # 6.12.x

Cc [email protected] # 6.6.x: f4cf4e5: Revert "nvmem: add new config option"

Cc [email protected] # 6.6.x: 7f38b70: of: device: Export of_device_make_bus_id()

Cc [email protected] # 6.6.x: 4a1a402: nvmem: Move of_nvmem_layout_get_container() in another header

Cc [email protected] # 6.6.x: fc29fd8: nvmem: core: Rework layouts to become regular devices

Cc [email protected] # 6.6.x: 0331c61: nvmem: core: Expose cells through sysfs

Cc [email protected] # 6.6.x: 401df0d: nvmem: layouts: refactor .add_cells() callback arguments

Cc [email protected] # 6.6.x: 6d0ca4a: nvmem: layouts: store owner from modules with nvmem_layout_driver_register()

Cc [email protected] # 6.6.x: 5f15811: nvmem: layouts: add U-Boot env layout

Cc [email protected] # 6.6.x

Signed-off-by Michael C. Pratt <[email protected]>

Signed-off-by Srinivas Kandagatla <[email protected]>

---
 drivers/nvmem/layouts/u-boot-env.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

drivers/nvmem/layouts/u-boot-env.c +3 -3

index 436426d4e8f9..8571aac56295 100644
@@ -92,7 +92,7 @@ int u_boot_env_parse(struct device *dev, struct nvmem_device *nvmem,
 	size_t crc32_data_offset;
 	size_t crc32_data_len;
 	size_t crc32_offset;
-	__le32 *crc32_addr;
+	uint32_t *crc32_addr;
 	size_t data_offset;
 	size_t data_len;
 	size_t dev_size;

@@ -143,8 +143,8 @@ int u_boot_env_parse(struct device *dev, struct nvmem_device *nvmem,
 		goto err_kfree;
 	}
 
-	crc32_addr = (__le32 *)(buf + crc32_offset);
-	crc32 = le32_to_cpu(*crc32_addr);
+	crc32_addr = (uint32_t *)(buf + crc32_offset);
+	crc32 = *crc32_addr;
 	crc32_data_len = dev_size - crc32_data_offset;
 	data_len = dev_size - data_offset;

Signature · 1 line

2.43.0

Greg KH

maintainer

2025-07-13 15:41 UTC #2 lore raw

Re: [PATCH 1/2] nvmem: layouts: u-boot-env: remove crc32 endianness conversion

Quoted 50 lines · On Sat, Jul 12, 2025 at 07:17:26PM +0100, [email protected] wrote:

From: "Michael C. Pratt" <[email protected]>
On 11 Oct 2022, it was reported that the crc32 verification of the u-boot environment failed only on big-endian systems for the u-boot-env nvmem layout driver with the following error.
Invalid calculated CRC32: 0x88cd6f09 (expected: 0x096fcd88)
This problem has been present since the driver was introduced, and before it was made into a layout driver.
The suggested fix at the time was to use further endianness conversion macros in order to have both the stored and calculated crc32 values to compare always represented in the system's endianness. This was not accepted due to sparse warnings
and some disagreement on how to handle the situation.
Later on in a newer revision of the patch, it was proposed to use cpu_to_le32() for both values to compare instead of le32_to_cpu() and store the values as __le32 type to remove compilation errors.
The necessity of this is based on the assumption that the use of crc32() requires endianness conversion because the algorithm uses little-endian, however, this does not prove to be the case and the issue is unrelated.
Upon inspecting the current kernel code,
there already is an existing use of le32_to_cpu() in this driver, which suggests there already is special handling for big-endian systems, however, it is big-endian systems that have the problem.
This, being the only functional difference between architectures in the driver combined with the fact that the suggested fix was to use the exact same endianness conversion for the values brings up the possibility that it was not necessary to begin with, as the same endianness conversion for two values expected to be the same is expected to be equivalent to no conversion at all.
After inspecting the u-boot environment of devices of both endianness and trying to remove the existing endianness conversion,
the problem is resolved in an equivalent way as the other suggested fixes.
Ultimately, it seems that u-boot is agnostic to endianness at least for the purpose of environment variables.
In other words, u-boot reads and writes the stored crc32 value with the same endianness that the crc32 value is calculated with in whichever endianness a certain architecture runs on.
Therefore, the u-boot-env driver does not need to convert endianness. Remove the usage of endianness macros in the u-boot-env driver, and change the type of local variables to maintain the same return type.

Quoted 7 lines

If there is a special situation in the case of endianness,
it would be a corner case and should be handled by a unique "compatible".
Even though it is not necessary to use endianness conversion macros here, it may be useful to use them in the future for consistent error printing.
Fixes d5542923f200 ("nvmem: add driver handling U-Boot environment variables")

Note, this is a 6.1 commit id, but:

Quoted 4 lines

Reported-by INAGAKI Hiroshi <[email protected]>
Link https://lore.kernel.org/all/[email protected] open here
Cc [email protected] # 6.12.x
Cc [email protected] # 6.6.x: f4cf4e5: Revert "nvmem: add new config option"

Quoted 8 lines

Cc [email protected] # 6.6.x: 7f38b70: of: device: Export of_device_make_bus_id()
Cc [email protected] # 6.6.x: 4a1a402: nvmem: Move of_nvmem_layout_get_container() in another header
Cc [email protected] # 6.6.x: fc29fd8: nvmem: core: Rework layouts to become regular devices
Cc [email protected] # 6.6.x: 0331c61: nvmem: core: Expose cells through sysfs
Cc [email protected] # 6.6.x: 401df0d: nvmem: layouts: refactor .add_cells() callback arguments
Cc [email protected] # 6.6.x: 6d0ca4a: nvmem: layouts: store owner from modules with nvmem_layout_driver_register()
Cc [email protected] # 6.6.x: 5f15811: nvmem: layouts: add U-Boot env layout
Cc [email protected] # 6.6.x

That's a load of (short) git ids for just 6.6.y? What about 6.1.y?

thanks,

greg k-h

Greg KH

maintainer

2025-07-13 15:42 UTC #3 lore raw

Re: [PATCH 1/2] nvmem: layouts: u-boot-env: remove crc32 endianness conversion

Quoted 74 lines · Greg KH

Quoted 50 lines · On Sat, Jul 12, 2025 at 07:17:26PM +0100, [email protected] wrote:
From: "Michael C. Pratt" <[email protected]>
On 11 Oct 2022, it was reported that the crc32 verification of the u-boot environment failed only on big-endian systems for the u-boot-env nvmem layout driver with the following error.
Invalid calculated CRC32: 0x88cd6f09 (expected: 0x096fcd88)
This problem has been present since the driver was introduced, and before it was made into a layout driver.
The suggested fix at the time was to use further endianness conversion macros in order to have both the stored and calculated crc32 values to compare always represented in the system's endianness. This was not accepted due to sparse warnings
and some disagreement on how to handle the situation.
Later on in a newer revision of the patch, it was proposed to use cpu_to_le32() for both values to compare instead of le32_to_cpu() and store the values as __le32 type to remove compilation errors.
The necessity of this is based on the assumption that the use of crc32() requires endianness conversion because the algorithm uses little-endian, however, this does not prove to be the case and the issue is unrelated.
Upon inspecting the current kernel code,
there already is an existing use of le32_to_cpu() in this driver, which suggests there already is special handling for big-endian systems, however, it is big-endian systems that have the problem.
This, being the only functional difference between architectures in the driver combined with the fact that the suggested fix was to use the exact same endianness conversion for the values brings up the possibility that it was not necessary to begin with, as the same endianness conversion for two values expected to be the same is expected to be equivalent to no conversion at all.
After inspecting the u-boot environment of devices of both endianness and trying to remove the existing endianness conversion,
the problem is resolved in an equivalent way as the other suggested fixes.
Ultimately, it seems that u-boot is agnostic to endianness at least for the purpose of environment variables.
In other words, u-boot reads and writes the stored crc32 value with the same endianness that the crc32 value is calculated with in whichever endianness a certain architecture runs on.
Therefore, the u-boot-env driver does not need to convert endianness. Remove the usage of endianness macros in the u-boot-env driver, and change the type of local variables to maintain the same return type.
Quoted 7 lines
If there is a special situation in the case of endianness,
it would be a corner case and should be handled by a unique "compatible".
Even though it is not necessary to use endianness conversion macros here, it may be useful to use them in the future for consistent error printing.
Fixes d5542923f200 ("nvmem: add driver handling U-Boot environment variables")
Note, this is a 6.1 commit id, but:
Quoted 4 lines
Reported-by INAGAKI Hiroshi <[email protected]>
Link https://lore.kernel.org/all/[email protected] open here
Cc [email protected] # 6.12.x
Cc [email protected] # 6.6.x: f4cf4e5: Revert "nvmem: add new config option"
Quoted 8 lines
Cc [email protected] # 6.6.x: 7f38b70: of: device: Export of_device_make_bus_id()
Cc [email protected] # 6.6.x: 4a1a402: nvmem: Move of_nvmem_layout_get_container() in another header
Cc [email protected] # 6.6.x: fc29fd8: nvmem: core: Rework layouts to become regular devices
Cc [email protected] # 6.6.x: 0331c61: nvmem: core: Expose cells through sysfs
Cc [email protected] # 6.6.x: 401df0d: nvmem: layouts: refactor .add_cells() callback arguments
Cc [email protected] # 6.6.x: 6d0ca4a: nvmem: layouts: store owner from modules with nvmem_layout_driver_register()
Cc [email protected] # 6.6.x: 5f15811: nvmem: layouts: add U-Boot env layout
Cc [email protected] # 6.6.x

Quoted 1 line

That's a load of (short) git ids for just 6.6.y? What about 6.1.y?

And really, ALL of those commits are needed for this very tiny patch? Reverting a config option? sysfs apis being added? Huh?

confused,

greg k-h

Srinivas Kandagatla

author maintainer

2025-07-15 15:44 UTC #4 lore raw

Re: [PATCH 1/2] nvmem: layouts: u-boot-env: remove crc32 endianness conversion

Quoted 75 lines · Greg KH

Quoted 74 lines · Greg KH
Quoted 50 lines · On Sat, Jul 12, 2025 at 07:17:26PM +0100, [email protected] wrote:
From: "Michael C. Pratt" <[email protected]>
On 11 Oct 2022, it was reported that the crc32 verification of the u-boot environment failed only on big-endian systems for the u-boot-env nvmem layout driver with the following error.
Invalid calculated CRC32: 0x88cd6f09 (expected: 0x096fcd88)
This problem has been present since the driver was introduced, and before it was made into a layout driver.
The suggested fix at the time was to use further endianness conversion macros in order to have both the stored and calculated crc32 values to compare always represented in the system's endianness. This was not accepted due to sparse warnings
and some disagreement on how to handle the situation.
Later on in a newer revision of the patch, it was proposed to use cpu_to_le32() for both values to compare instead of le32_to_cpu() and store the values as __le32 type to remove compilation errors.
The necessity of this is based on the assumption that the use of crc32() requires endianness conversion because the algorithm uses little-endian, however, this does not prove to be the case and the issue is unrelated.
Upon inspecting the current kernel code,
there already is an existing use of le32_to_cpu() in this driver, which suggests there already is special handling for big-endian systems, however, it is big-endian systems that have the problem.
This, being the only functional difference between architectures in the driver combined with the fact that the suggested fix was to use the exact same endianness conversion for the values brings up the possibility that it was not necessary to begin with, as the same endianness conversion for two values expected to be the same is expected to be equivalent to no conversion at all.
After inspecting the u-boot environment of devices of both endianness and trying to remove the existing endianness conversion,
the problem is resolved in an equivalent way as the other suggested fixes.
Ultimately, it seems that u-boot is agnostic to endianness at least for the purpose of environment variables.
In other words, u-boot reads and writes the stored crc32 value with the same endianness that the crc32 value is calculated with in whichever endianness a certain architecture runs on.
Therefore, the u-boot-env driver does not need to convert endianness. Remove the usage of endianness macros in the u-boot-env driver, and change the type of local variables to maintain the same return type.
Quoted 7 lines
If there is a special situation in the case of endianness,
it would be a corner case and should be handled by a unique "compatible".
Even though it is not necessary to use endianness conversion macros here, it may be useful to use them in the future for consistent error printing.
Fixes d5542923f200 ("nvmem: add driver handling U-Boot environment variables")
Note, this is a 6.1 commit id, but:
Quoted 12 lines
Reported-by INAGAKI Hiroshi <[email protected]>
Link https://lore.kernel.org/all/[email protected] open here
Cc [email protected] # 6.12.x
Cc [email protected] # 6.6.x: f4cf4e5: Revert "nvmem: add new config option"
Cc [email protected] # 6.6.x: 7f38b70: of: device: Export of_device_make_bus_id()
Cc [email protected] # 6.6.x: 4a1a402: nvmem: Move of_nvmem_layout_get_container() in another header
Cc [email protected] # 6.6.x: fc29fd8: nvmem: core: Rework layouts to become regular devices
Cc [email protected] # 6.6.x: 0331c61: nvmem: core: Expose cells through sysfs
Cc [email protected] # 6.6.x: 401df0d: nvmem: layouts: refactor .add_cells() callback arguments
Cc [email protected] # 6.6.x: 6d0ca4a: nvmem: layouts: store owner from modules with nvmem_layout_driver_register()
Cc [email protected] # 6.6.x: 5f15811: nvmem: layouts: add U-Boot env layout
Cc [email protected] # 6.6.x

Quoted 3 lines

Quoted 1 line
That's a load of (short) git ids for just 6.6.y? What about 6.1.y?
And really, ALL of those commits are needed for this very tiny patch?

May be not, AFAIU Fixes: d5542923f200 ("nvmem: add driver handling U-Boot environment variables") should be enough.

--srini

Quoted 5 lines

Reverting a config option? sysfs apis being added? Huh?
confused,
greg k-h

Greg KH

maintainer

2025-07-16 09:42 UTC #5 lore raw

Re: [PATCH 1/2] nvmem: layouts: u-boot-env: remove crc32 endianness conversion

Quoted 78 lines · Srinivas Kandagatla

Quoted 75 lines · Greg KH
Quoted 74 lines · Greg KH
Quoted 50 lines · On Sat, Jul 12, 2025 at 07:17:26PM +0100, [email protected] wrote:
From: "Michael C. Pratt" <[email protected]>
On 11 Oct 2022, it was reported that the crc32 verification of the u-boot environment failed only on big-endian systems for the u-boot-env nvmem layout driver with the following error.
Invalid calculated CRC32: 0x88cd6f09 (expected: 0x096fcd88)
This problem has been present since the driver was introduced, and before it was made into a layout driver.
The suggested fix at the time was to use further endianness conversion macros in order to have both the stored and calculated crc32 values to compare always represented in the system's endianness. This was not accepted due to sparse warnings
and some disagreement on how to handle the situation.
Later on in a newer revision of the patch, it was proposed to use cpu_to_le32() for both values to compare instead of le32_to_cpu() and store the values as __le32 type to remove compilation errors.
The necessity of this is based on the assumption that the use of crc32() requires endianness conversion because the algorithm uses little-endian, however, this does not prove to be the case and the issue is unrelated.
Upon inspecting the current kernel code,
there already is an existing use of le32_to_cpu() in this driver, which suggests there already is special handling for big-endian systems, however, it is big-endian systems that have the problem.
This, being the only functional difference between architectures in the driver combined with the fact that the suggested fix was to use the exact same endianness conversion for the values brings up the possibility that it was not necessary to begin with, as the same endianness conversion for two values expected to be the same is expected to be equivalent to no conversion at all.
After inspecting the u-boot environment of devices of both endianness and trying to remove the existing endianness conversion,
the problem is resolved in an equivalent way as the other suggested fixes.
Ultimately, it seems that u-boot is agnostic to endianness at least for the purpose of environment variables.
In other words, u-boot reads and writes the stored crc32 value with the same endianness that the crc32 value is calculated with in whichever endianness a certain architecture runs on.
Therefore, the u-boot-env driver does not need to convert endianness. Remove the usage of endianness macros in the u-boot-env driver, and change the type of local variables to maintain the same return type.
Quoted 7 lines
If there is a special situation in the case of endianness,
it would be a corner case and should be handled by a unique "compatible".
Even though it is not necessary to use endianness conversion macros here, it may be useful to use them in the future for consistent error printing.
Fixes d5542923f200 ("nvmem: add driver handling U-Boot environment variables")
Note, this is a 6.1 commit id, but:
Quoted 12 lines
Reported-by INAGAKI Hiroshi <[email protected]>
Link https://lore.kernel.org/all/[email protected] open here
Cc [email protected] # 6.12.x
Cc [email protected] # 6.6.x: f4cf4e5: Revert "nvmem: add new config option"
Cc [email protected] # 6.6.x: 7f38b70: of: device: Export of_device_make_bus_id()
Cc [email protected] # 6.6.x: 4a1a402: nvmem: Move of_nvmem_layout_get_container() in another header
Cc [email protected] # 6.6.x: fc29fd8: nvmem: core: Rework layouts to become regular devices
Cc [email protected] # 6.6.x: 0331c61: nvmem: core: Expose cells through sysfs
Cc [email protected] # 6.6.x: 401df0d: nvmem: layouts: refactor .add_cells() callback arguments
Cc [email protected] # 6.6.x: 6d0ca4a: nvmem: layouts: store owner from modules with nvmem_layout_driver_register()
Cc [email protected] # 6.6.x: 5f15811: nvmem: layouts: add U-Boot env layout
Cc [email protected] # 6.6.x

Quoted 5 lines

Quoted 3 lines
Quoted 1 line
That's a load of (short) git ids for just 6.6.y? What about 6.1.y?
And really, ALL of those commits are needed for this very tiny patch?
May be not, AFAIU Fixes: d5542923f200 ("nvmem: add driver handling U-Boot environment variables") should be enough.

Great, can you fix this up and resend?

thanks,

greg k-h

Michael Pratt

2025-07-16 21:24 UTC #6 lore raw

Re: [PATCH 1/2] nvmem: layouts: u-boot-env: remove crc32 endianness conversion

In reply to Greg KH

Hi Greg and Srinivas,

Sorry for the delayed response.

Quoted 79 lines · Greg KH

Quoted 75 lines · Greg KH
Quoted 50 lines · On Sat, Jul 12, 2025 at 07:17:26PM +0100, [email protected] wrote:
From: "Michael C. Pratt" [email protected]
On 11 Oct 2022, it was reported that the crc32 verification of the u-boot environment failed only on big-endian systems for the u-boot-env nvmem layout driver with the following error.
Invalid calculated CRC32: 0x88cd6f09 (expected: 0x096fcd88)
This problem has been present since the driver was introduced, and before it was made into a layout driver.
The suggested fix at the time was to use further endianness conversion macros in order to have both the stored and calculated crc32 values to compare always represented in the system's endianness. This was not accepted due to sparse warnings
and some disagreement on how to handle the situation.
Later on in a newer revision of the patch, it was proposed to use cpu_to_le32() for both values to compare instead of le32_to_cpu() and store the values as __le32 type to remove compilation errors.
The necessity of this is based on the assumption that the use of crc32() requires endianness conversion because the algorithm uses little-endian, however, this does not prove to be the case and the issue is unrelated.
Upon inspecting the current kernel code,
there already is an existing use of le32_to_cpu() in this driver, which suggests there already is special handling for big-endian systems, however, it is big-endian systems that have the problem.
This, being the only functional difference between architectures in the driver combined with the fact that the suggested fix was to use the exact same endianness conversion for the values brings up the possibility that it was not necessary to begin with, as the same endianness conversion for two values expected to be the same is expected to be equivalent to no conversion at all.
After inspecting the u-boot environment of devices of both endianness and trying to remove the existing endianness conversion,
the problem is resolved in an equivalent way as the other suggested fixes.
Ultimately, it seems that u-boot is agnostic to endianness at least for the purpose of environment variables.
In other words, u-boot reads and writes the stored crc32 value with the same endianness that the crc32 value is calculated with in whichever endianness a certain architecture runs on.
Therefore, the u-boot-env driver does not need to convert endianness. Remove the usage of endianness macros in the u-boot-env driver, and change the type of local variables to maintain the same return type.
Quoted 7 lines
If there is a special situation in the case of endianness,
it would be a corner case and should be handled by a unique "compatible".
Even though it is not necessary to use endianness conversion macros here, it may be useful to use them in the future for consistent error printing.
Fixes d5542923f200 ("nvmem: add driver handling U-Boot environment variables")
Note, this is a 6.1 commit id, but:
Quoted 12 lines
Reported-by INAGAKI Hiroshi [email protected]
Link https://lore.kernel.org/all/[email protected] open here
Cc [email protected] # 6.12.x
Cc [email protected] # 6.6.x: f4cf4e5: Revert "nvmem: add new config option"
Cc [email protected] # 6.6.x: 7f38b70: of: device: Export of_device_make_bus_id()
Cc [email protected] # 6.6.x: 4a1a402: nvmem: Move of_nvmem_layout_get_container() in another header
Cc [email protected] # 6.6.x: fc29fd8: nvmem: core: Rework layouts to become regular devices
Cc [email protected] # 6.6.x: 0331c61: nvmem: core: Expose cells through sysfs
Cc [email protected] # 6.6.x: 401df0d: nvmem: layouts: refactor .add_cells() callback arguments
Cc [email protected] # 6.6.x: 6d0ca4a: nvmem: layouts: store owner from modules with nvmem_layout_driver_register()
Cc [email protected] # 6.6.x: 5f15811: nvmem: layouts: add U-Boot env layout
Cc [email protected] # 6.6.x

Quoted 2 lines

Quoted 1 line
That's a load of (short) git ids for just 6.6.y? What about 6.1.y?

Sorry for the short tags, I wrongly assumed that what Github provides would not clobber with other commits.

Quoted 2 lines

And really, ALL of those commits are needed for this very tiny patch?

Yes... if we would like to backport to 6.6, (almost) all of them are necessary. There was a lot of development between 6.6 and 6.12 in this area...

This is a long-standing problem since 6.1, but the code is now completely rewritten into a different file, as a "layout driver" instead of a "cell module" if that makes sense...

In order to backport to 6.6, we would have to backport the rewriting of the code to the new layout driver form, which is commit 5f1581128 ("nvmem: layouts: add U-Boot env layout").

Commit 5f1581128 depends on commit 401df0d4f, which strictly depends on commit fc29fd821, and lightly depends (merge conflict) on commit 0331c6119.

Commit fc29fd821 strictly depends on both commit 4a1a40233 and 7f38b7004 for functionality.

I additionally included commit 6d0ca4a2a as it seems to improve function for all layout drivers, and I additionally included commit f4cf4e5db simply because we also backport it, not thinking that an extra one would be a problem.

In summary, the exact set of commits I presented for backporting is well tested, but one or two are indeed not strictly necessary as you pointed out.

Quoted 1 line

Reverting a config option? sysfs apis being added? Huh?

If you prefer, you can skip backporting commits f4cf4e5db and 0331c6119, although, skipping the latter would make you have to resolve the merge conflict.

If backporting to 6.6 is no longer appropriate in your opinion, please at least backport to 6.12 which is very easy.

Perhaps this would be a better set of tags for example:

# 6.12.x
# 6.6.x: 7f38b7004: of: device: Export of_device_make_bus_id()
# 6.6.x: 4a1a40233: nvmem: Move of_nvmem_layout_get_container() in another header # 6.6.x: fc29fd821: nvmem: core: Rework layouts to become regular devices
# 6.6.x: 401df0d4f: nvmem: layouts: refactor .add_cells() callback arguments
# 6.6.x: 6d0ca4a2a: nvmem: layouts: store owner from modules with nvmem_layout_driver_register() # 6.6.x: 5f1581128: nvmem: layouts: add U-Boot env layout
# 6.6.x

Quoted 2 lines

confused,

Sorry for the confusion, thanks.

--
MCP

Greg KH

maintainer

2025-07-21 13:57 UTC #7 lore raw

Re: [PATCH 1/2] nvmem: layouts: u-boot-env: remove crc32 endianness conversion

Quoted 117 lines · Michael Pratt

Hi Greg and Srinivas,
Sorry for the delayed response.
Quoted 79 lines · Greg KH
Quoted 75 lines · Greg KH
Quoted 50 lines · On Sat, Jul 12, 2025 at 07:17:26PM +0100, [email protected] wrote:
From: "Michael C. Pratt" [email protected]
On 11 Oct 2022, it was reported that the crc32 verification of the u-boot environment failed only on big-endian systems for the u-boot-env nvmem layout driver with the following error.
Invalid calculated CRC32: 0x88cd6f09 (expected: 0x096fcd88)
This problem has been present since the driver was introduced, and before it was made into a layout driver.
The suggested fix at the time was to use further endianness conversion macros in order to have both the stored and calculated crc32 values to compare always represented in the system's endianness. This was not accepted due to sparse warnings
and some disagreement on how to handle the situation.
Later on in a newer revision of the patch, it was proposed to use cpu_to_le32() for both values to compare instead of le32_to_cpu() and store the values as __le32 type to remove compilation errors.
The necessity of this is based on the assumption that the use of crc32() requires endianness conversion because the algorithm uses little-endian, however, this does not prove to be the case and the issue is unrelated.
Upon inspecting the current kernel code,
there already is an existing use of le32_to_cpu() in this driver, which suggests there already is special handling for big-endian systems, however, it is big-endian systems that have the problem.
This, being the only functional difference between architectures in the driver combined with the fact that the suggested fix was to use the exact same endianness conversion for the values brings up the possibility that it was not necessary to begin with, as the same endianness conversion for two values expected to be the same is expected to be equivalent to no conversion at all.
After inspecting the u-boot environment of devices of both endianness and trying to remove the existing endianness conversion,
the problem is resolved in an equivalent way as the other suggested fixes.
Ultimately, it seems that u-boot is agnostic to endianness at least for the purpose of environment variables.
In other words, u-boot reads and writes the stored crc32 value with the same endianness that the crc32 value is calculated with in whichever endianness a certain architecture runs on.
Therefore, the u-boot-env driver does not need to convert endianness. Remove the usage of endianness macros in the u-boot-env driver, and change the type of local variables to maintain the same return type.
Quoted 7 lines
If there is a special situation in the case of endianness,
it would be a corner case and should be handled by a unique "compatible".
Even though it is not necessary to use endianness conversion macros here, it may be useful to use them in the future for consistent error printing.
Fixes d5542923f200 ("nvmem: add driver handling U-Boot environment variables")
Note, this is a 6.1 commit id, but:
Quoted 12 lines
Reported-by INAGAKI Hiroshi [email protected]
Link https://lore.kernel.org/all/[email protected] open here
Cc [email protected] # 6.12.x
Cc [email protected] # 6.6.x: f4cf4e5: Revert "nvmem: add new config option"
Cc [email protected] # 6.6.x: 7f38b70: of: device: Export of_device_make_bus_id()
Cc [email protected] # 6.6.x: 4a1a402: nvmem: Move of_nvmem_layout_get_container() in another header
Cc [email protected] # 6.6.x: fc29fd8: nvmem: core: Rework layouts to become regular devices
Cc [email protected] # 6.6.x: 0331c61: nvmem: core: Expose cells through sysfs
Cc [email protected] # 6.6.x: 401df0d: nvmem: layouts: refactor .add_cells() callback arguments
Cc [email protected] # 6.6.x: 6d0ca4a: nvmem: layouts: store owner from modules with nvmem_layout_driver_register()
Cc [email protected] # 6.6.x: 5f15811: nvmem: layouts: add U-Boot env layout
Cc [email protected] # 6.6.x
Quoted 2 lines
Quoted 1 line
That's a load of (short) git ids for just 6.6.y? What about 6.1.y?
Sorry for the short tags, I wrongly assumed that what Github provides would not clobber with other commits.
Quoted 2 lines
And really, ALL of those commits are needed for this very tiny patch?
Yes... if we would like to backport to 6.6, (almost) all of them are necessary. There was a lot of development between 6.6 and 6.12 in this area...
This is a long-standing problem since 6.1, but the code is now completely rewritten into a different file, as a "layout driver" instead of a "cell module" if that makes sense...
In order to backport to 6.6, we would have to backport the rewriting of the code to the new layout driver form, which is commit 5f1581128 ("nvmem: layouts: add U-Boot env layout").
Commit 5f1581128 depends on commit 401df0d4f, which strictly depends on commit fc29fd821, and lightly depends (merge conflict) on commit 0331c6119.
Commit fc29fd821 strictly depends on both commit 4a1a40233 and 7f38b7004 for functionality.
I additionally included commit 6d0ca4a2a as it seems to improve function for all layout drivers, and I additionally included commit f4cf4e5db simply because we also backport it, not thinking that an extra one would be a problem.
In summary, the exact set of commits I presented for backporting is well tested, but one or two are indeed not strictly necessary as you pointed out.

Quoted 7 lines

Quoted 1 line
Reverting a config option? sysfs apis being added? Huh?
If you prefer, you can skip backporting commits f4cf4e5db and 0331c6119, although, skipping the latter would make you have to resolve the merge conflict.
If backporting to 6.6 is no longer appropriate in your opinion, please at least backport to 6.12 which is very easy.

I think that's already done, right?

If not, I'm still confused, can you send patch series for the different stable branches showing what you want applied where?

thanks,

greg k-h

[email protected]

author maintainer

2025-07-12 18:17 UTC #8 lore raw

PATCH 2/2 nvmem: imx-ocotp: fix MAC address byte length

In reply to [email protected]

From: Steffen Bätz <[email protected]>

The commit "13bcd440f2ff nvmem: core: verify cell's raw_len" caused an extension of the "mac-address" cell from 6 to 8 bytes due to word_size of 4 bytes. This led to a required byte swap of the full buffer length, which caused truncation of the mac-address when read.

Previously, the mac-address was incorrectly truncated from 70:B3:D5:14:E9:0E to 00:00:70:B3:D5:14.

Fix the issue by swapping only the first 6 bytes to correctly pass the mac-address to the upper layers.

Fixes 13bcd440f2ff ("nvmem: core: verify cell's raw_len")

Cc [email protected]

Signed-off-by Steffen Bätz <[email protected]>

Tested-by Alexander Stein <[email protected]>

Signed-off-by Srinivas Kandagatla <[email protected]>

---
 drivers/nvmem/imx-ocotp-ele.c | 5 ++++-
 drivers/nvmem/imx-ocotp.c     | 5 ++++-
 2 files changed, 8 insertions(+), 2 deletions(-)

drivers/nvmem/imx-ocotp-ele.c +4 -1

index ca6dd71d8a2e..7807ec0e2d18 100644
@@ -12,6 +12,7 @@
 #include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
+#include <linux/if_ether.h>	/* ETH_ALEN */
 
 enum fuse_type {
 	FUSE_FSB = BIT(0),

@@ -118,9 +119,11 @@ static int imx_ocotp_cell_pp(void *context, const char *id, int index,
 	int i;
 
 	/* Deal with some post processing of nvmem cell data */
-	if (id && !strcmp(id, "mac-address"))
+	if (id && !strcmp(id, "mac-address")) {
+		bytes = min(bytes, ETH_ALEN);
 		for (i = 0; i < bytes / 2; i++)
 			swap(buf[i], buf[bytes - i - 1]);
+	}
 
 	return 0;
 }

drivers/nvmem/imx-ocotp.c +4 -1

index 79dd4fda0329..7bf7656d4f96 100644
@@ -23,6 +23,7 @@
 #include <linux/platform_device.h>
 #include <linux/slab.h>
 #include <linux/delay.h>
+#include <linux/if_ether.h>	/* ETH_ALEN */
 
 #define IMX_OCOTP_OFFSET_B0W0		0x400 /* Offset from base address of the
 					       * OTP Bank0 Word0

@@ -227,9 +228,11 @@ static int imx_ocotp_cell_pp(void *context, const char *id, int index,
 	int i;
 
 	/* Deal with some post processing of nvmem cell data */
-	if (id && !strcmp(id, "mac-address"))
+	if (id && !strcmp(id, "mac-address")) {
+		bytes = min(bytes, ETH_ALEN);
 		for (i = 0; i < bytes / 2; i++)
 			swap(buf[i], buf[bytes - i - 1]);
+	}
 
 	return 0;
 }

Signature · 1 line

2.43.0