Re: [PATCH v5 4/5] mmc: sdhci-tegra: Implement alternative_gpt_sector()

[Dmitry Osipenko <digetx@gmail.com>]

18.08.2021 16:35, Thierry Reding пишет:
>>> +static int sdhci_tegra_alternative_gpt_sector(struct mmc_card *card,
>>> +                                             sector_t *gpt_sector)
>>> +{
>>> +       unsigned int boot_sectors_num;
>>> +
>>> +       /* filter out unrelated cards */
>>> +       if (card->ext_csd.rev < 3 ||
>>> +           !mmc_card_mmc(card) ||
>>> +           !mmc_card_is_blockaddr(card) ||
>>> +            mmc_card_is_removable(card->host))
>>> +               return -ENOENT;
>>> +
>>> +       /*
>>> +        * eMMC storage has two special boot partitions in addition to the
>>> +        * main one.  NVIDIA's bootloader linearizes eMMC boot0->boot1->main
>>> +        * accesses, this means that the partition table addresses are shifted
>>> +        * by the size of boot partitions.  In accordance with the eMMC
>>> +        * specification, the boot partition size is calculated as follows:
>>> +        *
>>> +        *      boot partition size = 128K byte x BOOT_SIZE_MULT
>>> +        *
>>> +        * Calculate number of sectors occupied by the both boot partitions.
>>> +        */
>>> +       boot_sectors_num = card->ext_csd.raw_boot_mult * SZ_128K /
>>> +                          SZ_512 * MMC_NUM_BOOT_PARTITION;
>>> +
>>> +       /* Defined by NVIDIA and used by Android devices. */
>>> +       *gpt_sector = card->ext_csd.sectors - boot_sectors_num - 1;
>>> +
>>> +       return 0;
>>> +}
>> I suggest you move this code into the mmc core/block layer instead (it
>> better belongs there).
>>
>> Additionally, let's add a new host cap, MMC_CAP_ALTERNATIVE_GPT, to
>> let the core know when it should use the code above.
> Couldn't a generic "alternative GPT" mean pretty much anything? As far
> as I know this is very specific to a series of Tegra chips and firmware
> running on them. On some of these devices you can even replace the OEM
> firmware by something custom that's less quirky.
> 
> I'm not aware of anyone else employing this kind of quirk, so I don't
> want anyone to get any ideas that this is a good thing. Putting it into
> the core runs the risk of legitimizing this.

I also think it's better to keep it internal to Tegra. Ulf, could you
please clarify why do you want to have it moved into the core? Are you
aware of any other platforms that want exactly the same quirk? Thierry
should be correct that it's relevant only to Tegra SoCs.

Regarding the 'legitimizing', it's not a bad thing to me at all. If we
want to run more devices with a mainline kernel, then such quirks are
inevitable.