[PATCH v6] remoteproc: uclass: Add methods to load firmware to rproc and boot rproc

Add APIs to set a firmware_name to a rproc and boot the rproc with the same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com --- Changes from v5 to v6: *) Collected Acked-by tag from Ravi Gunasekaran r-gunasekaran@ti.com *) Fixed few typos as pointed out by Roger Quadros rogerq@kernel.org *) Added if condition to check if uc_pdata->fw_name exists and free it before the strndup as suggested by Roger Quadros rogerq@kernel.org
Changes from v4 to v5: *) Added Kconfig option REMOTEPROC_MAX_FW_SIZE to set max firmware size that can be loaded to a rproc. *) Added freeing of address in rproc_boot() as pointed out by Ravi. *) Allocating the address at a later point in rproc_boot() *) Rebased on latest u-boot/master [commit 9e00b6993f724da9699ef12573307afea8c19284]
Changes from v3 to v4: *) No functional change. Splitted the patch out of the series as suggested by Nishant. *) Droppped the RFC tag.
v5: https://lore.kernel.org/all/20240217122602.3402774-1-danishanwar@ti.com/ v4: https://lore.kernel.org/all/20240130063322.2345057-1-danishanwar@ti.com/ v3: https://lore.kernel.org/all/20240124064930.1787929-4-danishanwar@ti.com/
drivers/remoteproc/Kconfig | 8 +++ drivers/remoteproc/rproc-uclass.c | 102 ++++++++++++++++++++++++++++++ include/remoteproc.h | 34 ++++++++++ 3 files changed, 144 insertions(+)
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig index 781de530af..9f9877931c 100644 --- a/drivers/remoteproc/Kconfig +++ b/drivers/remoteproc/Kconfig @@ -10,6 +10,7 @@ menu "Remote Processor drivers" # All users should depend on DM config REMOTEPROC bool + select FS_LOADER depends on DM
# Please keep the configuration alphabetically sorted. @@ -102,4 +103,11 @@ config REMOTEPROC_TI_IPU help Say 'y' here to add support for TI' K3 remoteproc driver.
+config REMOTEPROC_MAX_FW_SIZE + hex "Maximum size of firmware file that needs to be loaded to the remote processor" + default 0x10000 + help + Maximum size of the firmware file (elf, binary) that needs to be + loaded to the remote processor. + endmenu diff --git a/drivers/remoteproc/rproc-uclass.c b/drivers/remoteproc/rproc-uclass.c index 28b362c887..f4f22a3851 100644 --- a/drivers/remoteproc/rproc-uclass.c +++ b/drivers/remoteproc/rproc-uclass.c @@ -13,6 +13,7 @@ #include <log.h> #include <malloc.h> #include <virtio_ring.h> +#include <fs_loader.h> #include <remoteproc.h> #include <asm/io.h> #include <dm/device-internal.h> @@ -961,3 +962,104 @@ unsigned long rproc_parse_resource_table(struct udevice *dev, struct rproc *cfg)
return 1; } + +int rproc_set_firmware(struct udevice *rproc_dev, const char *fw_name) +{ + struct dm_rproc_uclass_pdata *uc_pdata; + int len; + char *p; + + if (!rproc_dev || !fw_name) + return -EINVAL; + + uc_pdata = dev_get_uclass_plat(rproc_dev); + if (!uc_pdata) + return -EINVAL; + + len = strcspn(fw_name, "\n"); + if (!len) { + debug("invalid firmware name\n"); + return -EINVAL; + } + + if (uc_pdata->fw_name) + free(uc_pdata->fw_name); + + p = strndup(fw_name, len); + if (!p) + return -ENOMEM; + + uc_pdata->fw_name = p; + + return 0; +} + +int rproc_boot(struct udevice *rproc_dev) +{ + struct dm_rproc_uclass_pdata *uc_pdata; + struct udevice *fs_loader; + int core_id, ret = 0; + char *firmware; + void *addr; + + if (!rproc_dev) + return -EINVAL; + + uc_pdata = dev_get_uclass_plat(rproc_dev); + if (!uc_pdata) + return -EINVAL; + + core_id = dev_seq(rproc_dev); + firmware = uc_pdata->fw_name; + if (!firmware) { + debug("No firmware name set for rproc core %d\n", core_id); + return -EINVAL; + } + + /* Initialize all rproc cores */ + if (!rproc_is_initialized()) { + ret = rproc_init(); + if (ret) { + debug("rproc_init() failed: %d\n", ret); + return ret; + } + } + + /* Loading firmware to a given address */ + ret = get_fs_loader(&fs_loader); + if (ret) { + debug("could not get fs loader: %d\n", ret); + return ret; + } + + if (CONFIG_REMOTEPROC_MAX_FW_SIZE) { + addr = malloc(CONFIG_REMOTEPROC_MAX_FW_SIZE); + if (!addr) + return -ENOMEM; + } else { + debug("CONFIG_REMOTEPROC_MAX_FW_SIZE not defined\n"); + return -EINVAL; + } + + ret = request_firmware_into_buf(fs_loader, firmware, addr, CONFIG_REMOTEPROC_MAX_FW_SIZE, + 0); + if (ret < 0) { + debug("could not request %s: %d\n", firmware, ret); + goto free_buffer; + } + + ret = rproc_load(core_id, (ulong)addr, ret); + if (ret) { + debug("failed to load %s to rproc core %d from addr 0x%08lX err %d\n", + uc_pdata->fw_name, core_id, (ulong)addr, ret); + goto free_buffer; + } + + ret = rproc_start(core_id); + if (ret) + debug("failed to start rproc core %d\n", core_id); + +free_buffer: + free(addr); + return ret; +} diff --git a/include/remoteproc.h b/include/remoteproc.h index 91a88791a4..6f8068e149 100644 --- a/include/remoteproc.h +++ b/include/remoteproc.h @@ -403,6 +403,7 @@ enum rproc_mem_type { * @name: Platform-specific way of naming the Remote proc * @mem_type: one of 'enum rproc_mem_type' * @driver_plat_data: driver specific platform data that may be needed. + * @fw_name: firmware name * * This can be accessed with dev_get_uclass_plat() for any UCLASS_REMOTEPROC * device. @@ -412,6 +413,7 @@ struct dm_rproc_uclass_pdata { const char *name; enum rproc_mem_type mem_type; void *driver_plat_data; + char *fw_name; };
/** @@ -705,6 +707,34 @@ unsigned long rproc_parse_resource_table(struct udevice *dev, struct resource_table *rproc_find_resource_table(struct udevice *dev, unsigned int addr, int *tablesz); +/** + * rproc_set_firmware() - assign a new firmware name + * @rproc_dev: device for which new firmware name is being assigned + * @fw_name: new firmware name to be assigned + * + * This function allows remoteproc drivers or clients to configure a custom + * firmware name. The function does not trigger a remote processor boot, + * only sets the firmware name used for a subsequent boot. + * + * This function sets the fw_name field in uclass pdata of the Remote proc + * + * Return: 0 on success or a negative value upon failure + */ +int rproc_set_firmware(struct udevice *rproc_dev, const char *fw_name); + +/** + * rproc_boot() - boot a remote processor + * @rproc_dev: rproc device to boot + * + * Boot a remote processor (i.e. load its firmware, power it on, ...). + * + * This function first loads the firmware set in the uclass pdata of Remote + * processor to a buffer and then loads firmware to the remote processor + * using rproc_load(). + * + * Return: 0 on success, and an appropriate error value otherwise + */ +int rproc_boot(struct udevice *rproc_dev); #else static inline int rproc_init(void) { return -ENOSYS; } static inline int rproc_dev_init(int id) { return -ENOSYS; } @@ -744,6 +774,10 @@ static inline int rproc_elf_load_rsc_table(struct udevice *dev, ulong fw_addr, ulong fw_size, ulong *rsc_addr, ulong *rsc_size) { return -ENOSYS; } +static inline int rproc_set_firmware(struct udevice *rproc_dev, const char *fw_name) +{ return -ENOSYS; } +static inline int rproc_boot(struct udevice *rproc_dev) +{ return -ENOSYS; } #endif
#endif /* _RPROC_H_ */

On 28/02/2024 14:06, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com
Reviewed-by: Roger Quadros rogerq@kernel.org

On 29/02/24 3:43 pm, Roger Quadros wrote:
On 28/02/2024 14:06, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com
Reviewed-by: Roger Quadros rogerq@kernel.org
Hi Tom,
Can you please pick this patch. There are no active comments on this patch and Roger and Ravi has reviewed / acked the patch.

On Tue, Mar 05, 2024 at 02:39:14PM +0530, MD Danish Anwar wrote:
On 29/02/24 3:43 pm, Roger Quadros wrote:
On 28/02/2024 14:06, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com
Reviewed-by: Roger Quadros rogerq@kernel.org
Hi Tom,
Can you please pick this patch. There are no active comments on this patch and Roger and Ravi has reviewed / acked the patch.
Yes, I'm going to give people a little longer to comment / review and will pull this to -next soon.

On Wed, Feb 28, 2024 at 05:36:45PM +0530, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the
same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org
This breaks building on am64x_evm_r5 am65x_evm_r5_usbdfu am65x_evm_r5_usbmsc in next currently, thanks.

On 3/7/2024 6:16 PM, Tom Rini wrote:
On Wed, Feb 28, 2024 at 05:36:45PM +0530, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the
same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org
This breaks building on am64x_evm_r5 am65x_evm_r5_usbdfu am65x_evm_r5_usbmsc in next currently, thanks.
I will work on fixing this build error and re-spin the patch.

On 11/03/24 10:34 am, Anwar, Md Danish wrote:
On 3/7/2024 6:16 PM, Tom Rini wrote:
On Wed, Feb 28, 2024 at 05:36:45PM +0530, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the
same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org
This breaks building on am64x_evm_r5 am65x_evm_r5_usbdfu am65x_evm_r5_usbmsc in next currently, thanks.
I will work on fixing this build error and re-spin the patch.
Hi Tom, Roger,
This patch adds "request_firmware_into_buf()" in the rproc driver. To use this API, FS_LOADER is needed. So I am adding "select FS_LOADER" in REMOTEPROC Kconfig option. As a result whenever REMOTEPROC is enabled, FS_LOADER also gets enabled.
Now arch/arm/mach-k3/common.c [1] and arch/arm/mach-omap2/boot-common.c [2] has a "load_firmware()" API which calls fs-loader APIs and they have below if condition before calling fs-loader APIs.
if (!IS_ENABLED(CONFIG_FS_LOADER)) return 0;
Till now, CONFIG_FS_LOADER was not set as a result the load_firmware() API in above mentioned files, was returning 0.
Now as this patch enables CONFIG_FS_LOADER, as a result the code after the if check starts getting executed and it tries to look for get_fs_loader() and other fs-loader APIs but this is done at SPL and at this time FS_LOADER is not built yet as a result we see below error. The if checks only checks for CONFIG_FS_LOADER but not for CONFIG_SPL_FS_LOADER.
AR spl/boot/built-in.o LD spl/u-boot-spl arm-none-linux-gnueabihf-ld.bfd: arch/arm/mach-k3/common.o: in function `load_firmware': /home/danish/workspace/u-boot/arch/arm/mach-k3/common.c:184: undefined reference to `get_fs_loader' arm-none-linux-gnueabihf-ld.bfd: /home/danish/workspace/u-boot/arch/arm/mach-k3/common.c:185: undefined reference to `request_firmware_into_buf' make[2]: *** [/home/danish/workspace/u-boot/scripts/Makefile.spl:527: spl/u-boot-spl] Error 1 make[1]: *** [/home/danish/workspace/u-boot/Makefile:2055: spl/u-boot-spl] Error 2 make[1]: Leaving directory '/home/danish/uboot_images/am64x/r5' make: *** [Makefile:177: sub-make] Error 2
This bug has always been there but as CONFIG_FS_LOADER was never enabled, this build error was never seen as the load_firmware() API will return 0 without calling fs-loader APIs.
Now that this patch enables CONFIG_FS_LOADER, the bug gets exposed and build error is seen.
My opinion here would be, to check for CONFIG_IS_ENABLED(FS_LOADER) instead of IS_ENABLED(CONFIG_FS_LOADER) as the former will check for the appropriate config option (CONFIG_SPL_FS_LOADER / CONFIG_FS_LOADER) based on the build stage.
I tested with the below diff and I don't see build errors with am64x_evm_r5, am65x_evm_r5_usbdfu, am65x_evm_r5_usbmsc configs.
diff --git a/arch/arm/mach-k3/common.c b/arch/arm/mach-k3/common.c index f411366778..6792ff7467 100644 --- a/arch/arm/mach-k3/common.c +++ b/arch/arm/mach-k3/common.c @@ -162,7 +162,7 @@ int load_firmware(char *name_fw, char *name_loadaddr, u32 *loadaddr) char *name = NULL; int size = 0;
- if (!IS_ENABLED(CONFIG_FS_LOADER)) + if (!CONFIG_IS_ENABLED(FS_LOADER)) return 0;
*loadaddr = 0; diff --git a/arch/arm/mach-omap2/boot-common.c b/arch/arm/mach-omap2/boot-common.c index 57917da25c..aa0ab13d5f 100644 --- a/arch/arm/mach-omap2/boot-common.c +++ b/arch/arm/mach-omap2/boot-common.c @@ -190,7 +190,7 @@ int load_firmware(char *name_fw, u32 *loadaddr) struct udevice *fsdev; int size = 0;
- if (!IS_ENABLED(CONFIG_FS_LOADER)) + if (!CONFIG_IS_ENABLED(FS_LOADER)) return 0;
if (!*loadaddr)
Tom, Roger, Please let me know if this looks ok. If it's ok, I will post this diff as a separate patch and once that is merged Tom can merge this patch or I can send a v7 if needed.
[1] https://elixir.bootlin.com/u-boot/latest/source/arch/arm/mach-k3/common.c#L1... [2] https://elixir.bootlin.com/u-boot/latest/source/arch/arm/mach-omap2/boot-com...

On Tue, Mar 12, 2024 at 02:02:08PM +0530, MD Danish Anwar wrote:
On 11/03/24 10:34 am, Anwar, Md Danish wrote:
On 3/7/2024 6:16 PM, Tom Rini wrote:
On Wed, Feb 28, 2024 at 05:36:45PM +0530, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the
same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org
This breaks building on am64x_evm_r5 am65x_evm_r5_usbdfu am65x_evm_r5_usbmsc in next currently, thanks.
I will work on fixing this build error and re-spin the patch.
Hi Tom, Roger,
This patch adds "request_firmware_into_buf()" in the rproc driver. To use this API, FS_LOADER is needed. So I am adding "select FS_LOADER" in REMOTEPROC Kconfig option. As a result whenever REMOTEPROC is enabled, FS_LOADER also gets enabled.
Now arch/arm/mach-k3/common.c [1] and arch/arm/mach-omap2/boot-common.c [2] has a "load_firmware()" API which calls fs-loader APIs and they have below if condition before calling fs-loader APIs.
if (!IS_ENABLED(CONFIG_FS_LOADER)) return 0;
Till now, CONFIG_FS_LOADER was not set as a result the load_firmware() API in above mentioned files, was returning 0.
Now as this patch enables CONFIG_FS_LOADER, as a result the code after the if check starts getting executed and it tries to look for get_fs_loader() and other fs-loader APIs but this is done at SPL and at this time FS_LOADER is not built yet as a result we see below error. The if checks only checks for CONFIG_FS_LOADER but not for CONFIG_SPL_FS_LOADER.
AR spl/boot/built-in.o LD spl/u-boot-spl arm-none-linux-gnueabihf-ld.bfd: arch/arm/mach-k3/common.o: in function `load_firmware': /home/danish/workspace/u-boot/arch/arm/mach-k3/common.c:184: undefined reference to `get_fs_loader' arm-none-linux-gnueabihf-ld.bfd: /home/danish/workspace/u-boot/arch/arm/mach-k3/common.c:185: undefined reference to `request_firmware_into_buf' make[2]: *** [/home/danish/workspace/u-boot/scripts/Makefile.spl:527: spl/u-boot-spl] Error 1 make[1]: *** [/home/danish/workspace/u-boot/Makefile:2055: spl/u-boot-spl] Error 2 make[1]: Leaving directory '/home/danish/uboot_images/am64x/r5' make: *** [Makefile:177: sub-make] Error 2
This bug has always been there but as CONFIG_FS_LOADER was never enabled, this build error was never seen as the load_firmware() API will return 0 without calling fs-loader APIs.
Now that this patch enables CONFIG_FS_LOADER, the bug gets exposed and build error is seen.
My opinion here would be, to check for CONFIG_IS_ENABLED(FS_LOADER) instead of IS_ENABLED(CONFIG_FS_LOADER) as the former will check for the appropriate config option (CONFIG_SPL_FS_LOADER / CONFIG_FS_LOADER) based on the build stage.
I tested with the below diff and I don't see build errors with am64x_evm_r5, am65x_evm_r5_usbdfu, am65x_evm_r5_usbmsc configs.
diff --git a/arch/arm/mach-k3/common.c b/arch/arm/mach-k3/common.c index f411366778..6792ff7467 100644 --- a/arch/arm/mach-k3/common.c +++ b/arch/arm/mach-k3/common.c @@ -162,7 +162,7 @@ int load_firmware(char *name_fw, char *name_loadaddr, u32 *loadaddr) char *name = NULL; int size = 0;
if (!IS_ENABLED(CONFIG_FS_LOADER))
if (!CONFIG_IS_ENABLED(FS_LOADER)) return 0; *loadaddr = 0;
diff --git a/arch/arm/mach-omap2/boot-common.c b/arch/arm/mach-omap2/boot-common.c index 57917da25c..aa0ab13d5f 100644 --- a/arch/arm/mach-omap2/boot-common.c +++ b/arch/arm/mach-omap2/boot-common.c @@ -190,7 +190,7 @@ int load_firmware(char *name_fw, u32 *loadaddr) struct udevice *fsdev; int size = 0;
if (!IS_ENABLED(CONFIG_FS_LOADER))
if (!CONFIG_IS_ENABLED(FS_LOADER)) return 0; if (!*loadaddr)
Tom, Roger, Please let me know if this looks ok. If it's ok, I will post this diff as a separate patch and once that is merged Tom can merge this patch or I can send a v7 if needed.
Yes, this seems like the right path, thanks.

On 3/14/2024 6:16 PM, Tom Rini wrote:
On Tue, Mar 12, 2024 at 02:02:08PM +0530, MD Danish Anwar wrote:
On 11/03/24 10:34 am, Anwar, Md Danish wrote:
On 3/7/2024 6:16 PM, Tom Rini wrote:
On Wed, Feb 28, 2024 at 05:36:45PM +0530, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the
same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org
This breaks building on am64x_evm_r5 am65x_evm_r5_usbdfu am65x_evm_r5_usbmsc in next currently, thanks.
I will work on fixing this build error and re-spin the patch.
Hi Tom, Roger,
This patch adds "request_firmware_into_buf()" in the rproc driver. To use this API, FS_LOADER is needed. So I am adding "select FS_LOADER" in REMOTEPROC Kconfig option. As a result whenever REMOTEPROC is enabled, FS_LOADER also gets enabled.
Now arch/arm/mach-k3/common.c [1] and arch/arm/mach-omap2/boot-common.c [2] has a "load_firmware()" API which calls fs-loader APIs and they have below if condition before calling fs-loader APIs.
if (!IS_ENABLED(CONFIG_FS_LOADER)) return 0;
Till now, CONFIG_FS_LOADER was not set as a result the load_firmware() API in above mentioned files, was returning 0.
Now as this patch enables CONFIG_FS_LOADER, as a result the code after the if check starts getting executed and it tries to look for get_fs_loader() and other fs-loader APIs but this is done at SPL and at this time FS_LOADER is not built yet as a result we see below error. The if checks only checks for CONFIG_FS_LOADER but not for CONFIG_SPL_FS_LOADER.
AR spl/boot/built-in.o LD spl/u-boot-spl arm-none-linux-gnueabihf-ld.bfd: arch/arm/mach-k3/common.o: in function `load_firmware': /home/danish/workspace/u-boot/arch/arm/mach-k3/common.c:184: undefined reference to `get_fs_loader' arm-none-linux-gnueabihf-ld.bfd: /home/danish/workspace/u-boot/arch/arm/mach-k3/common.c:185: undefined reference to `request_firmware_into_buf' make[2]: *** [/home/danish/workspace/u-boot/scripts/Makefile.spl:527: spl/u-boot-spl] Error 1 make[1]: *** [/home/danish/workspace/u-boot/Makefile:2055: spl/u-boot-spl] Error 2 make[1]: Leaving directory '/home/danish/uboot_images/am64x/r5' make: *** [Makefile:177: sub-make] Error 2
This bug has always been there but as CONFIG_FS_LOADER was never enabled, this build error was never seen as the load_firmware() API will return 0 without calling fs-loader APIs.
Now that this patch enables CONFIG_FS_LOADER, the bug gets exposed and build error is seen.
My opinion here would be, to check for CONFIG_IS_ENABLED(FS_LOADER) instead of IS_ENABLED(CONFIG_FS_LOADER) as the former will check for the appropriate config option (CONFIG_SPL_FS_LOADER / CONFIG_FS_LOADER) based on the build stage.
I tested with the below diff and I don't see build errors with am64x_evm_r5, am65x_evm_r5_usbdfu, am65x_evm_r5_usbmsc configs.
diff --git a/arch/arm/mach-k3/common.c b/arch/arm/mach-k3/common.c index f411366778..6792ff7467 100644 --- a/arch/arm/mach-k3/common.c +++ b/arch/arm/mach-k3/common.c @@ -162,7 +162,7 @@ int load_firmware(char *name_fw, char *name_loadaddr, u32 *loadaddr) char *name = NULL; int size = 0;
if (!IS_ENABLED(CONFIG_FS_LOADER))
if (!CONFIG_IS_ENABLED(FS_LOADER)) return 0; *loadaddr = 0;
diff --git a/arch/arm/mach-omap2/boot-common.c b/arch/arm/mach-omap2/boot-common.c index 57917da25c..aa0ab13d5f 100644 --- a/arch/arm/mach-omap2/boot-common.c +++ b/arch/arm/mach-omap2/boot-common.c @@ -190,7 +190,7 @@ int load_firmware(char *name_fw, u32 *loadaddr) struct udevice *fsdev; int size = 0;
if (!IS_ENABLED(CONFIG_FS_LOADER))
if (!CONFIG_IS_ENABLED(FS_LOADER)) return 0; if (!*loadaddr)
Tom, Roger, Please let me know if this looks ok. If it's ok, I will post this diff as a separate patch and once that is merged Tom can merge this patch or I can send a v7 if needed.
Yes, this seems like the right path, thanks.
Thanks Tom. Posted this diff as patch https://lore.kernel.org/all/20240314143311.259568-1-danishanwar@ti.com/

On Wed, Feb 28, 2024 at 05:36:45PM +0530, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org
Changes from v5 to v6: *) Collected Acked-by tag from Ravi Gunasekaran r-gunasekaran@ti.com *) Fixed few typos as pointed out by Roger Quadros rogerq@kernel.org *) Added if condition to check if uc_pdata->fw_name exists and free it before the strndup as suggested by Roger Quadros rogerq@kernel.org
Changes from v4 to v5: *) Added Kconfig option REMOTEPROC_MAX_FW_SIZE to set max firmware size that can be loaded to a rproc. *) Added freeing of address in rproc_boot() as pointed out by Ravi. *) Allocating the address at a later point in rproc_boot() *) Rebased on latest u-boot/master [commit 9e00b6993f724da9699ef12573307afea8c19284]
Changes from v3 to v4: *) No functional change. Splitted the patch out of the series as suggested by Nishant. *) Droppped the RFC tag.
v5: https://lore.kernel.org/all/20240217122602.3402774-1-danishanwar@ti.com/ v4: https://lore.kernel.org/all/20240130063322.2345057-1-danishanwar@ti.com/ v3: https://lore.kernel.org/all/20240124064930.1787929-4-danishanwar@ti.com/
drivers/remoteproc/Kconfig | 8 +++ drivers/remoteproc/rproc-uclass.c | 102 ++++++++++++++++++++++++++++++ include/remoteproc.h | 34 ++++++++++ 3 files changed, 144 insertions(+)
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig index 781de530af..9f9877931c 100644 --- a/drivers/remoteproc/Kconfig +++ b/drivers/remoteproc/Kconfig @@ -10,6 +10,7 @@ menu "Remote Processor drivers" # All users should depend on DM config REMOTEPROC bool
- select FS_LOADER depends on DM
# Please keep the configuration alphabetically sorted.
Can we not make the FS_LOADER portion optional? I didn't realize how many non-TI platforms this impacted. And even then it's possible I assume that custom designs will load the firmwares in other manners.

Hi Tom,
On 20/03/24 4:10 am, Tom Rini wrote:
On Wed, Feb 28, 2024 at 05:36:45PM +0530, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org
Changes from v5 to v6: *) Collected Acked-by tag from Ravi Gunasekaran r-gunasekaran@ti.com *) Fixed few typos as pointed out by Roger Quadros rogerq@kernel.org *) Added if condition to check if uc_pdata->fw_name exists and free it before the strndup as suggested by Roger Quadros rogerq@kernel.org
Changes from v4 to v5: *) Added Kconfig option REMOTEPROC_MAX_FW_SIZE to set max firmware size that can be loaded to a rproc. *) Added freeing of address in rproc_boot() as pointed out by Ravi. *) Allocating the address at a later point in rproc_boot() *) Rebased on latest u-boot/master [commit 9e00b6993f724da9699ef12573307afea8c19284]
Changes from v3 to v4: *) No functional change. Splitted the patch out of the series as suggested by Nishant. *) Droppped the RFC tag.
v5: https://lore.kernel.org/all/20240217122602.3402774-1-danishanwar@ti.com/ v4: https://lore.kernel.org/all/20240130063322.2345057-1-danishanwar@ti.com/ v3: https://lore.kernel.org/all/20240124064930.1787929-4-danishanwar@ti.com/
drivers/remoteproc/Kconfig | 8 +++ drivers/remoteproc/rproc-uclass.c | 102 ++++++++++++++++++++++++++++++ include/remoteproc.h | 34 ++++++++++ 3 files changed, 144 insertions(+)
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig index 781de530af..9f9877931c 100644 --- a/drivers/remoteproc/Kconfig +++ b/drivers/remoteproc/Kconfig @@ -10,6 +10,7 @@ menu "Remote Processor drivers" # All users should depend on DM config REMOTEPROC bool
- select FS_LOADER depends on DM
# Please keep the configuration alphabetically sorted.
Can we not make the FS_LOADER portion optional? I didn't realize how many non-TI platforms this impacted. And even then it's possible I assume that custom designs will load the firmwares in other manners.
Yes we can. We can wrap the remoteproc APIs using FS_LOADER in #ifdef CONFIG_FS_LOADER. And instead of REMOTEPROC driver selecting FS_LOADER, the clinet driver (ICSSG in this case) who is calling those remoteproc APIs will select FS_LOADER and enable it.
This will make sure that other platforms (ti or non-ti) that doesn't support ICSSG but enables Remoteproc, will not enable FS_LOADER. This way we are not forcing other platforms using remoteproc to enable FS_LOADER. In this case the APIs will not get built.
Now FS_LOADER will only be enabled when there is a client driver that uses rproc_boot() APIs. It's upto the client driver to enable FS_LOADER
below is the diff,
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig index 9f9877931c..a49802c132 100644 --- a/drivers/remoteproc/Kconfig +++ b/drivers/remoteproc/Kconfig @@ -10,7 +10,6 @@ menu "Remote Processor drivers" # All users should depend on DM config REMOTEPROC bool - select FS_LOADER depends on DM
# Please keep the configuration alphabetically sorted. diff --git a/drivers/remoteproc/rproc-uclass.c b/drivers/remoteproc/rproc-uclass.c index f4f22a3851..a6a8be5009 100644 --- a/drivers/remoteproc/rproc-uclass.c +++ b/drivers/remoteproc/rproc-uclass.c @@ -994,6 +994,7 @@ int rproc_set_firmware(struct udevice *rproc_dev, const char *fw_name) return 0; }
+#ifdef CONFIG_FS_LOADER int rproc_boot(struct udevice *rproc_dev) { struct dm_rproc_uclass_pdata *uc_pdata; @@ -1063,3 +1064,4 @@ free_buffer: free(addr); return ret; } +#endif
Let me know if this looks ok. If it's ok I will post v7 with this change.

On Wed, Mar 20, 2024 at 11:19:01AM +0530, MD Danish Anwar wrote:
Hi Tom,
On 20/03/24 4:10 am, Tom Rini wrote:
On Wed, Feb 28, 2024 at 05:36:45PM +0530, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org
Changes from v5 to v6: *) Collected Acked-by tag from Ravi Gunasekaran r-gunasekaran@ti.com *) Fixed few typos as pointed out by Roger Quadros rogerq@kernel.org *) Added if condition to check if uc_pdata->fw_name exists and free it before the strndup as suggested by Roger Quadros rogerq@kernel.org
Changes from v4 to v5: *) Added Kconfig option REMOTEPROC_MAX_FW_SIZE to set max firmware size that can be loaded to a rproc. *) Added freeing of address in rproc_boot() as pointed out by Ravi. *) Allocating the address at a later point in rproc_boot() *) Rebased on latest u-boot/master [commit 9e00b6993f724da9699ef12573307afea8c19284]
Changes from v3 to v4: *) No functional change. Splitted the patch out of the series as suggested by Nishant. *) Droppped the RFC tag.
v5: https://lore.kernel.org/all/20240217122602.3402774-1-danishanwar@ti.com/ v4: https://lore.kernel.org/all/20240130063322.2345057-1-danishanwar@ti.com/ v3: https://lore.kernel.org/all/20240124064930.1787929-4-danishanwar@ti.com/
drivers/remoteproc/Kconfig | 8 +++ drivers/remoteproc/rproc-uclass.c | 102 ++++++++++++++++++++++++++++++ include/remoteproc.h | 34 ++++++++++ 3 files changed, 144 insertions(+)
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig index 781de530af..9f9877931c 100644 --- a/drivers/remoteproc/Kconfig +++ b/drivers/remoteproc/Kconfig @@ -10,6 +10,7 @@ menu "Remote Processor drivers" # All users should depend on DM config REMOTEPROC bool
- select FS_LOADER depends on DM
# Please keep the configuration alphabetically sorted.
Can we not make the FS_LOADER portion optional? I didn't realize how many non-TI platforms this impacted. And even then it's possible I assume that custom designs will load the firmwares in other manners.
Yes we can. We can wrap the remoteproc APIs using FS_LOADER in #ifdef CONFIG_FS_LOADER. And instead of REMOTEPROC driver selecting FS_LOADER, the clinet driver (ICSSG in this case) who is calling those remoteproc APIs will select FS_LOADER and enable it.
This will make sure that other platforms (ti or non-ti) that doesn't support ICSSG but enables Remoteproc, will not enable FS_LOADER. This way we are not forcing other platforms using remoteproc to enable FS_LOADER. In this case the APIs will not get built.
Now FS_LOADER will only be enabled when there is a client driver that uses rproc_boot() APIs. It's upto the client driver to enable FS_LOADER
below is the diff,
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig index 9f9877931c..a49802c132 100644 --- a/drivers/remoteproc/Kconfig +++ b/drivers/remoteproc/Kconfig @@ -10,7 +10,6 @@ menu "Remote Processor drivers" # All users should depend on DM config REMOTEPROC bool
select FS_LOADER depends on DM
# Please keep the configuration alphabetically sorted. diff --git a/drivers/remoteproc/rproc-uclass.c b/drivers/remoteproc/rproc-uclass.c index f4f22a3851..a6a8be5009 100644 --- a/drivers/remoteproc/rproc-uclass.c +++ b/drivers/remoteproc/rproc-uclass.c @@ -994,6 +994,7 @@ int rproc_set_firmware(struct udevice *rproc_dev, const char *fw_name) return 0; }
+#ifdef CONFIG_FS_LOADER int rproc_boot(struct udevice *rproc_dev) { struct dm_rproc_uclass_pdata *uc_pdata; @@ -1063,3 +1064,4 @@ free_buffer: free(addr); return ret; } +#endif
Let me know if this looks ok. If it's ok I will post v7 with this change.
Yes please, thanks.

On 20/03/24 6:08 pm, Tom Rini wrote:
On Wed, Mar 20, 2024 at 11:19:01AM +0530, MD Danish Anwar wrote:
Hi Tom,
On 20/03/24 4:10 am, Tom Rini wrote:
On Wed, Feb 28, 2024 at 05:36:45PM +0530, MD Danish Anwar wrote:
Add APIs to set a firmware_name to a rproc and boot the rproc with the same firmware.
Clients can call rproc_set_firmware() API to set firmware_name for a rproc whereas rproc_boot() will load the firmware set by rproc_set_firmware() to a buffer by calling request_firmware_into_buf(). rproc_boot() will then load the firmware file to the remote processor and start the remote processor.
Also include "fs-loader.h" and make remoteproc driver select FS_LOADER in Kconfig so that we can call request_firmware_into_buf() from remoteproc driver.
Signed-off-by: MD Danish Anwar danishanwar@ti.com Acked-by: Ravi Gunasekaran r-gunasekaran@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org
Changes from v5 to v6: *) Collected Acked-by tag from Ravi Gunasekaran r-gunasekaran@ti.com *) Fixed few typos as pointed out by Roger Quadros rogerq@kernel.org *) Added if condition to check if uc_pdata->fw_name exists and free it before the strndup as suggested by Roger Quadros rogerq@kernel.org
Changes from v4 to v5: *) Added Kconfig option REMOTEPROC_MAX_FW_SIZE to set max firmware size that can be loaded to a rproc. *) Added freeing of address in rproc_boot() as pointed out by Ravi. *) Allocating the address at a later point in rproc_boot() *) Rebased on latest u-boot/master [commit 9e00b6993f724da9699ef12573307afea8c19284]
Changes from v3 to v4: *) No functional change. Splitted the patch out of the series as suggested by Nishant. *) Droppped the RFC tag.
v5: https://lore.kernel.org/all/20240217122602.3402774-1-danishanwar@ti.com/ v4: https://lore.kernel.org/all/20240130063322.2345057-1-danishanwar@ti.com/ v3: https://lore.kernel.org/all/20240124064930.1787929-4-danishanwar@ti.com/
drivers/remoteproc/Kconfig | 8 +++ drivers/remoteproc/rproc-uclass.c | 102 ++++++++++++++++++++++++++++++ include/remoteproc.h | 34 ++++++++++ 3 files changed, 144 insertions(+)
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig index 781de530af..9f9877931c 100644 --- a/drivers/remoteproc/Kconfig +++ b/drivers/remoteproc/Kconfig @@ -10,6 +10,7 @@ menu "Remote Processor drivers" # All users should depend on DM config REMOTEPROC bool
- select FS_LOADER depends on DM
# Please keep the configuration alphabetically sorted.
Can we not make the FS_LOADER portion optional? I didn't realize how many non-TI platforms this impacted. And even then it's possible I assume that custom designs will load the firmwares in other manners.
Yes we can. We can wrap the remoteproc APIs using FS_LOADER in #ifdef CONFIG_FS_LOADER. And instead of REMOTEPROC driver selecting FS_LOADER, the clinet driver (ICSSG in this case) who is calling those remoteproc APIs will select FS_LOADER and enable it.
This will make sure that other platforms (ti or non-ti) that doesn't support ICSSG but enables Remoteproc, will not enable FS_LOADER. This way we are not forcing other platforms using remoteproc to enable FS_LOADER. In this case the APIs will not get built.
Now FS_LOADER will only be enabled when there is a client driver that uses rproc_boot() APIs. It's upto the client driver to enable FS_LOADER
below is the diff,
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig index 9f9877931c..a49802c132 100644 --- a/drivers/remoteproc/Kconfig +++ b/drivers/remoteproc/Kconfig @@ -10,7 +10,6 @@ menu "Remote Processor drivers" # All users should depend on DM config REMOTEPROC bool
select FS_LOADER depends on DM
# Please keep the configuration alphabetically sorted. diff --git a/drivers/remoteproc/rproc-uclass.c b/drivers/remoteproc/rproc-uclass.c index f4f22a3851..a6a8be5009 100644 --- a/drivers/remoteproc/rproc-uclass.c +++ b/drivers/remoteproc/rproc-uclass.c @@ -994,6 +994,7 @@ int rproc_set_firmware(struct udevice *rproc_dev, const char *fw_name) return 0; }
+#ifdef CONFIG_FS_LOADER int rproc_boot(struct udevice *rproc_dev) { struct dm_rproc_uclass_pdata *uc_pdata; @@ -1063,3 +1064,4 @@ free_buffer: free(addr); return ret; } +#endif
Let me know if this looks ok. If it's ok I will post v7 with this change.
Yes please, thanks.
Posted v7 with the above changes https://lore.kernel.org/all/20240321102819.1011011-1-danishanwar@ti.com/
Please check.
participants (4)
-
Anwar, Md Danish
-
MD Danish Anwar
-
Roger Quadros
-
Tom Rini