Hi Shimoda-san, thank you for this bugfix! Looks like a nice catch. > + > + /* > + * According to eMMC specification, we should issue CMD1 Maybe it is good to specify where? I found it in the standard v5.1, section 6.4.3. > + * repeatidly in the idle state until the eMMC is ready even if "repeatedly" > + * the mmc_attach_mmc() calls this function with ocr = 0. I would suggest to remove the "if ..." part from the comment. Because you remove the 'if (ocr == 0)' code, so it becomes less obvious what that part of the comment relates to. > + * Otherwise some eMMC devices seem to enter the inactive mode > + * after mmc_init_card() issued CMD0 when the eMMC device is > + * busy. > + */ Other than that, good documentation! > + if (!ocr && !mmc_host_is_spi(host)) > + cmd.arg = cmd.resp[0] | BIT(30); I think BIT(30) is okay for this bugfix. But as a follow up, we should turn this into a define. It could be used in mmc_init_card() as well and will make the code way more readable. All the best, Wolfram
Attachment:
signature.asc
Description: PGP signature
