[U-Boot] [RFC PATCH 04/11] SPL: FIT: allow loading multiple images

[Michal Simek]

On 23.1.2017 03:49, Kever Yang wrote:
> Hi Andre,
> 
> On 01/22/2017 06:58 PM, André Przywara wrote:
>> On 22/01/17 07:08, Kever Yang wrote:
>>> Hi Andre,
>>>
>>>      Thanks for your patches, this is great help for enable ATF on
>>> U-Boot
>>> SPL.
>>> For ATF use case, we would like to identify which one is bl31 for we
>>> need to
>>> get entry point for it while we only need load address for other image.
>>> Any idea on get this information from different "loadables"?
>> So I thought this use case is covered by the current scheme?
>>
>> See below ...
>>
>>> On 01/20/2017 09:53 AM, Andre Przywara wrote:
>>>> So far we were not using the FIT image format to its full potential:
>>>> The SPL FIT loader was just loading the first image from the /images
>>>> node plus one of the listed DTBs.
>>>> Now with the refactored loader code it's easy to load an arbitrary
>>>> number of images in addition to the two mentioned above.
>>>> As described in the FIT image source file format description, iterate
>>>> over all images listed at the "loadables" property in the configuration
>>>> node and load every image at its desired location.
>>>> This allows to load any kind of images:
>>>> - firmware images to execute before U-Boot proper (for instance
>>>>     ARM Trusted Firmware (ATF))
>>>> - firmware images for management processors (SCP, arisc, ...)
>>>> - firmware images for devices like WiFi controllers
>>>> - bit files for FPGAs
>>>> - additional configuration data
>>>> - kernels and/or ramdisks
>>>> The actual usage of this feature would be platform and/or board
>>>> specific.
>>>>
>>>> Signed-off-by: Andre Przywara <andre.przywara at arm.com>
>>>> ---
>>>>    common/spl/spl_fit.c | 27 +++++++++++++++++++++++++++
>>>>    1 file changed, 27 insertions(+)
>>>>
>>>> diff --git a/common/spl/spl_fit.c b/common/spl/spl_fit.c
>>>> index d4149c5..18269f7 100644
>>>> --- a/common/spl/spl_fit.c
>>>> +++ b/common/spl/spl_fit.c
>>>> @@ -190,6 +190,7 @@ int spl_load_simple_fit(struct spl_image_info
>>>> *spl_image,
>>>>        struct spl_image_info image_info;
>>>>        int node, images;
>>>>        int base_offset, align_len = ARCH_DMA_MINALIGN - 1;
>>>> +    int index = 0;
>>>>          /*
>>>>         * Figure out where the external images start. This is the base
>>>> for the
>>>> @@ -234,6 +235,11 @@ int spl_load_simple_fit(struct spl_image_info
>>>> *spl_image,
>>>>        if (node < 0) {
>>>>            debug("could not find firmware image, trying
>>>> loadables...\n");
>>>>            node = spl_fit_get_image_node(fit, images, "loadables", 0);
>>>> +        /*
>>>> +         * If we pick the U-Boot image from "loadables", start at
>>>> +         * the second image when later loading additional images.
>>>> +         */
>>>> +        index = 1;
>>>>        }
>>>>        if (node < 0) {
>>>>            debug("%s: Cannot find u-boot image node: %d\n",
>>>> @@ -259,5 +265,26 @@ int spl_load_simple_fit(struct spl_image_info
>>>> *spl_image,
>>>>        image_info.load_addr = spl_image->load_addr + spl_image->size;
>>>>        spl_load_fit_image(info, sector, fit, base_offset, node,
>>>> &image_info);
>>>>    +    /* Now check if there are more images for us to load */
>>>> +    for (; ; index++) {
>>>> +        node = spl_fit_get_image_node(fit, images, "loadables",
>>>> index);
>>>> +        if (node < 0)
>>>> +            break;
>>>> +
>>>> +        spl_load_fit_image(info, sector, fit, base_offset, node,
>>>> +                   &image_info);
>>>> +
>>>> +        /*
>>>> +         * If the "firmware" image did not provide an entry point,
>>>> +         * use the first valid entry point from the loadables.
>>>> +         */
>>>> +        if (spl_image->entry_point == -1U &&
>>>> +            image_info.entry_point != -1U)
>>>> +            spl_image->entry_point = image_info.entry_point;
>> So this part here saves the entry point from the first image which
>> provides an "entry" property in the loadables section, given that
>> "firmware" didn't provide (a legal) one.
>>
>> So depending on what you want, these are the options:
>> a) You want to branch to U-Boot (default case if mkimage -f auto is
>> used): You either don't specify an explicit entry point at all or
>> provide an "entry" property for the U-Boot "firmware" image.
>> The code below at the end of this function takes care of this.
>> b) You want to branch to something else (bl31.bin): You make sure there
>> is no explicit entry point in "firmware", instead put one "entry"
>> property in the bl31 image description you want to boot. This will then
>> branch to that instead of U-Boot.
>> Check the example .its in patch 10/11 for an example of this.
>>
>> So does this fit for you? Or is there a problem with this?
> 
> So that means only one 'entry' can be present even if there are more
> than one image, right?
> I think every "firmware" image is possible to have both "load" and
> "entry" node,
> for example if we have bl31, bl32 and bl33(U-Boot) in one FIT image,
> then we need to fill the bl31_params_t structure with bl32 entry point
> and U-Boot entry point,
> so that bl31 able to call bl32 and bl33.
> That's why I think we need to identify each of the "firmware", but not
> only know they are
> "loadables".

You need put this information somewhere. It means node itself should
contain that information. And there is no field for it. It means if you
want support it you need to extend it. Definitely for "full" boot you
would need it.

Thanks,
Michal