On Fri, Apr 11, 2014 at 1:15 AM, Ulf Hansson <ulf.hansson@xxxxxxxxxx> wrote:
> On 11 April 2014 01:31, Tim Kryger <tim.kryger@xxxxxxxxxx> wrote:
>> +static unsigned int sdhci_get_ocr_avail_from_vmmc(struct sdhci_host *host)
>> +{
>> + unsigned int ocr_avail = 0;
>> + struct regulator *vmmc = host->vmmc;
>> +
>> + if (regulator_is_supported_voltage(vmmc, 1650000, 1950000) > 0)
>> + ocr_avail |= MMC_VDD_165_195;
>> +
>> + if (regulator_is_supported_voltage(vmmc, 2900000, 3000000) > 0)
>> + ocr_avail |= MMC_VDD_29_30;
>> +
>> + if (regulator_is_supported_voltage(vmmc, 3000000, 3100000) > 0)
>> + ocr_avail |= MMC_VDD_30_31;
>> +
>> + if (regulator_is_supported_voltage(vmmc, 3200000, 3300000) > 0)
>> + ocr_avail |= MMC_VDD_32_33;
>> +
>> + if (regulator_is_supported_voltage(vmmc, 3300000, 3400000) > 0)
>> + ocr_avail |= MMC_VDD_33_34;
>> +
>> + return ocr_avail;
>> +}
>> +
>
> There is an API called mmc_regulator_get_ocrmask() for this.
Great. Thanks for pointing this out.
>> + ocr_avail = sdhci_get_ocr_avail_from_vmmc(host);
>> + } else {
>> + if (caps[0] & SDHCI_CAN_VDD_330)
>> + ocr_avail |= MMC_VDD_32_33 | MMC_VDD_33_34;
>> + if (caps[0] & SDHCI_CAN_VDD_300)
>> + ocr_avail |= MMC_VDD_29_30 | MMC_VDD_30_31;
>> + if (caps[0] & SDHCI_CAN_VDD_180)
>> + ocr_avail |= MMC_VDD_165_195;
>> + }
>> +
>> + if (host->ocr_mask)
>> + ocr_avail = host->ocr_mask;
>> +
>> + mmc->ocr_avail = ocr_avail;
>> + mmc->ocr_avail_sdio = ocr_avail;
>> + if (host->ocr_avail_sdio)
>> + mmc->ocr_avail_sdio &= host->ocr_avail_sdio;
>> + mmc->ocr_avail_sd = ocr_avail;
>> + if (host->ocr_avail_sd)
>> + mmc->ocr_avail_sd &= host->ocr_avail_sd;
>> + else /* normal SD controllers don't support 1.8V */
>> + mmc->ocr_avail_sd &= ~MMC_VDD_165_195;
>> + mmc->ocr_avail_mmc = ocr_avail;
>> + if (host->ocr_avail_mmc)
>> + mmc->ocr_avail_mmc &= host->ocr_avail_mmc;
>> +
>> + if (mmc->ocr_avail == 0) {
>> + pr_err("%s: Hardware doesn't report any support voltages.\n",
>> + mmc_hostname(mmc));
>> + return -ENODEV;
>> + }
>
> I have not fully understand why you have different ocr masks depending
> on what card you initialize. Is that really supported by the
> controller?
Would you mind clarifying your question? I'm not sure what you are
after. Also this logic is mostly unchanged, I have simply moved it up
earlier in the function to keep all the ocr parts together.
> I have seen some patches around lately touching the code for handling
> the regulators (vcc and vccq) in sdhci.
>
> A few times I have suggested to switch to use the
> mmc_regulator_get_supply() API to simplify and consolidate code. Could
> you please have a look at that?
Sure, I'll take a look. Thanks for the helpful comments.
-Tim
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
