[RFC PATCH v3 5/9] EFI: FMP: Add provision to update image's ImageTypeId in image descriptor
Sughosh Ganu
sughosh.ganu at linaro.org
Mon Jan 24 07:27:44 CET 2022
On Mon, 24 Jan 2022 at 08:03, AKASHI Takahiro
<takahiro.akashi at linaro.org> wrote:
>
> On Fri, Jan 21, 2022 at 12:32:04PM +0530, Sughosh Ganu wrote:
> > hi Takahiro,
> >
> > On Thu, 20 Jan 2022 at 10:54, AKASHI Takahiro
> > <takahiro.akashi at linaro.org> wrote:
> > >
> > > On Thu, Jan 20, 2022 at 12:25:44AM +0530, Sughosh Ganu wrote:
> > > > The FWU Multi Banks Update feature allows updating different types of
> > > > updatable firmware images on the platform. These image types are
> > > > identified using the ImageTypeId GUID value. Add support in the
> > > > GetImageInfo function of the FMP protocol to get the GUID values for
> > > > the individual images and populate these in the image descriptor for
> > > > the corresponding images.
> > > >
> > > > Signed-off-by: Sughosh Ganu <sughosh.ganu at linaro.org>
> > > > ---
> > > >
> > > > Changes since V2: None
> > > >
> > > > lib/efi_loader/efi_firmware.c | 90 ++++++++++++++++++++++++++++++++---
> > > > 1 file changed, 83 insertions(+), 7 deletions(-)
> > > >
> > > > diff --git a/lib/efi_loader/efi_firmware.c b/lib/efi_loader/efi_firmware.c
> > > > index a1b88dbfc2..648342ae72 100644
> > > > --- a/lib/efi_loader/efi_firmware.c
> > > > +++ b/lib/efi_loader/efi_firmware.c
> > > > @@ -10,6 +10,7 @@
> > > > #include <charset.h>
> > > > #include <dfu.h>
> > > > #include <efi_loader.h>
> > > > +#include <fwu.h>
> > > > #include <image.h>
> > > > #include <signatures.h>
> > > >
> > > > @@ -96,6 +97,46 @@ efi_status_t EFIAPI efi_firmware_set_package_info_unsupported(
> > > > return EFI_EXIT(EFI_UNSUPPORTED);
> > > > }
> > > >
> > > > +static efi_status_t fill_part_guid_array(const efi_guid_t *guid,
> > > > + efi_guid_t **part_guid_arr)
> > > > +{
> > > > + int i;
> > > > + int dfu_num = 0;
> > > > + efi_guid_t *guid_arr;
> > > > + struct dfu_entity *dfu;
> > > > + efi_status_t ret = EFI_SUCCESS;
> > > > +
> > > > + dfu_init_env_entities(NULL, NULL);
> > > > +
> > > > + dfu_num = 0;
> > > > + list_for_each_entry(dfu, &dfu_list, list) {
> > > > + dfu_num++;
> > > > + }
> > > > +
> > > > + if (!dfu_num) {
> > > > + log_warning("Probably dfu_alt_info not defined\n");
> > > > + ret = EFI_NOT_READY;
> > > > + goto out;
> > > > + }
> > > > +
> > > > + *part_guid_arr = malloc(sizeof(efi_guid_t) * dfu_num);
> > > > + if (!*part_guid_arr) {
> > > > + ret = EFI_OUT_OF_RESOURCES;
> > > > + goto out;
> > > > + }
> > > > +
> > > > + guid_arr = *part_guid_arr;
> > > > + for (i = 0; i < dfu_num; i++) {
> > > > + guidcpy(guid_arr, guid);
> > > > + ++guid_arr;
> > > > + }
> > > > +
> > > > +out:
> > > > + dfu_free_entities();
> > > > +
> > > > + return ret;
> > > > +}
> > > > +
> > > > /**
> > > > * efi_get_dfu_info - return information about the current firmware image
> > > > * @this: Protocol instance
> > > > @@ -104,9 +145,9 @@ efi_status_t EFIAPI efi_firmware_set_package_info_unsupported(
> > > > * @descriptor_version: Pointer to version number
> > > > * @descriptor_count: Pointer to number of descriptors
> > > > * @descriptor_size: Pointer to descriptor size
> > > > - * package_version: Package version
> > > > - * package_version_name: Package version's name
> > > > - * image_type: Image type GUID
> > > > + * @package_version: Package version
> > > > + * @package_version_name: Package version's name
> > > > + * @guid_array: Image type GUID array
> > > > *
> > > > * Return information bout the current firmware image in @image_info.
> > > > * @image_info will consist of a number of descriptors.
> > > > @@ -122,7 +163,7 @@ static efi_status_t efi_get_dfu_info(
> > > > efi_uintn_t *descriptor_size,
> > > > u32 *package_version,
> > > > u16 **package_version_name,
> > > > - const efi_guid_t *image_type)
> > > > + const efi_guid_t *guid_array)
> > > > {
> > > > struct dfu_entity *dfu;
> > > > size_t names_len, total_size;
> > > > @@ -172,7 +213,7 @@ static efi_status_t efi_get_dfu_info(
> > > > next = name;
> > > > list_for_each_entry(dfu, &dfu_list, list) {
> > > > image_info[i].image_index = dfu->alt + 1;
> > > > - image_info[i].image_type_id = *image_type;
> > > > + image_info[i].image_type_id = guid_array[i];
> > > > image_info[i].image_id = dfu->alt;
> > > >
> > > > /* copy the DFU entity name */
> > > > @@ -250,6 +291,7 @@ efi_status_t EFIAPI efi_firmware_fit_get_image_info(
> > > > u16 **package_version_name)
> > > > {
> > > > efi_status_t ret;
> > > > + efi_guid_t *part_guid_arr = NULL;
> > > >
> > > > EFI_ENTRY("%p %p %p %p %p %p %p %p\n", this,
> > > > image_info_size, image_info,
> > > > @@ -264,12 +306,19 @@ efi_status_t EFIAPI efi_firmware_fit_get_image_info(
> > > > !descriptor_size || !package_version || !package_version_name))
> > > > return EFI_EXIT(EFI_INVALID_PARAMETER);
> > > >
> > > > + ret = fill_part_guid_array(&efi_firmware_image_type_uboot_fit,
> > > > + &part_guid_arr);
> > > > + if (ret != EFI_SUCCESS)
> > > > + goto out;
> > >
> > > Why do you not call fwu_plat_fill_partition_guids() for FIT FMP driver?
> > > If you have a specific reason, please describe it.
> >
> > The idea here was to retain the current behaviour for the non FWU use
> > case, where for both raw and FIT images, the same ImageTypeId is being
> > used.
>
> What I meant to say was fwu_plat_fill_partition_guids() is not called
> even in case of CONFIG_FWU_MULTI_BANK_UPDATE.
Ah, that is because I am testing the FWU code with the raw capsules
and not FIT. I will add a weak function. If there is FWU support added
for FIT based capsules in the future, the relevant plat function will
have to be added.
-sughosh
>
> > Do you want to have a weak function which fills the array with a
> > specific ImageTypeId value. Then, if any platform wants to have a
> > different implementation, they can define a function which fills the
> > array with relevant GUID values.
> >
> > >
> > > > +
> > > > ret = efi_get_dfu_info(image_info_size, image_info,
> > > > descriptor_version, descriptor_count,
> > > > descriptor_size,
> > > > package_version, package_version_name,
> > > > - &efi_firmware_image_type_uboot_fit);
> > > > + part_guid_arr);
> > > >
> > > > +out:
> > > > + free(part_guid_arr);
> > > > return EFI_EXIT(ret);
> > > > }
> > > >
> > > > @@ -358,7 +407,10 @@ efi_status_t EFIAPI efi_firmware_raw_get_image_info(
> > > > u32 *package_version,
> > > > u16 **package_version_name)
> > > > {
> > > > + int status;
> > > > efi_status_t ret = EFI_SUCCESS;
> > > > + const efi_guid_t null_guid = NULL_GUID;
> > > > + efi_guid_t *part_guid_arr = NULL;
> > > >
> > > > EFI_ENTRY("%p %p %p %p %p %p %p %p\n", this,
> > > > image_info_size, image_info,
> > > > @@ -373,12 +425,36 @@ efi_status_t EFIAPI efi_firmware_raw_get_image_info(
> > > > !descriptor_size || !package_version || !package_version_name))
> > > > return EFI_EXIT(EFI_INVALID_PARAMETER);
> > > >
> > > > + if (IS_ENABLED(CONFIG_FWU_MULTI_BANK_UPDATE)) {
> > > > + ret = fill_part_guid_array(&null_guid, &part_guid_arr);
> > > > + if (ret != EFI_SUCCESS)
> > > > + goto out;
> > > > +
> > > > + /*
> > > > + * Call the platform function to fill the GUID array
> > > > + * with the corresponding partition GUID values
> > > > + */
> > > > + status = fwu_plat_fill_partition_guids(&part_guid_arr);
> > > > + if (status < 0) {
> > > > + log_err("Unable to get partiion guid's\n");
> > > > + ret = EFI_DEVICE_ERROR;
> > > > + goto out;
> > > > + }
> > > > + } else {
> > > > + ret = fill_part_guid_array(&efi_firmware_image_type_uboot_raw,
> > > > + &part_guid_arr);
> > > > + if (ret != EFI_SUCCESS)
> > > > + goto out;
> > > > + }
> > >
> > > The code:
> > > ret = fill_part_guid_array(&efi_firmware_image_type_uboot_raw,
> > > &part_guid_arr);
> > > if (IS_ENABLED(CONFIG_FWU_MULTI_BANK_UPDATE))
> > > status = fwu_plat_fill_partition_guids(&part_guid_arr);
> > >
> > > would be much simpler here.
> >
> > Yep, will change.
> >
> > >
> > > But I don't know why you want to call fwu_plat_fill_partition_guids()
> > > only in case of CONFIG_FWU_MULTI_BANK_UPDATE. The functionality should
> > > be the same whether A/B update or not.
> >
> > Let me know if you are okay with what I have proposed above.
>
> Yes. The weak "plat" function will be expected to be optimized away.
>
> -Takahiro Akashi
>
>
> > -sughosh
> >
> > >
> > > -Takahiro Akashi
> > >
> > > > +
> > > > ret = efi_get_dfu_info(image_info_size, image_info,
> > > > descriptor_version, descriptor_count,
> > > > descriptor_size,
> > > > package_version, package_version_name,
> > > > - &efi_firmware_image_type_uboot_raw);
> > > > + part_guid_arr);
> > > >
> > > > +out:
> > > > + free(part_guid_arr);
> > > > return EFI_EXIT(ret);
> > > > }
> > > >
> > > > --
> > > > 2.17.1
> > > >
More information about the U-Boot
mailing list