On 5 March 2018 at 15:47, Shawn Lin <shawn.lin@xxxxxxxxxxxxxx> wrote:
> On 2018/3/5 22:27, Ulf Hansson wrote:
>>
>> On 5 March 2018 at 14:41, Harish Jenny K N <harish_kandiga@xxxxxxxxxx>
>> wrote:
>>>
>>> This patch parses mmc host caps and caps2 exported via debugfs.
>>>
>>> Signed-off-by: Harish Jenny K N <harish_kandiga@xxxxxxxxxx>
>>> ---
>>>
>>> Changes in v2:
>>> - Added support for parsing caps2
>>>
>>> mmc.c | 10 +++
>>> mmc_cmds.c | 222
>>> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>> mmc_cmds.h | 2 +
>>> 3 files changed, 234 insertions(+)
>>>
>>> diff --git a/mmc.c b/mmc.c
>>> index 50c9c9e..dab81bd 100644
>>> --- a/mmc.c
>>> +++ b/mmc.c
>>> @@ -53,6 +53,16 @@ static struct Command commands[] = {
>>> /*
>>> * avoid short commands different for the case only
>>> */
>>> + { do_read_hostcaps, -1,
>>> + "hostcaps read", "<path to debugfs host caps file>\n"
>>> + "Prints parsed data from host caps file>.",
>>> + NULL
>>> + },
>>> + { do_read_hostcaps2, -1,
>>> + "hostcaps2 read", "<path to debugfs host caps2 file>\n"
>>> + "Prints parsed data from host caps2 file>.",
>>> + NULL
>>> + },
>>> { do_read_extcsd, -1,
>>> "extcsd read", "<device>\n"
>>> "Print extcsd data from <device>.",
>>> diff --git a/mmc_cmds.c b/mmc_cmds.c
>>> index 038dbd4..55a27f1 100644
>>> --- a/mmc_cmds.c
>>> +++ b/mmc_cmds.c
>>> @@ -54,6 +54,61 @@
>>> #define WPTYPE_PWRON 2
>>> #define WPTYPE_PERM 3
>>>
>>> +#define BIT(x) (1ul << x)
>>> +
>>> +#define MMCUTIL_4_BIT_DATA BIT(0) /*Can the host do 4 bit transfers*/
>>> +#define MMCUTIL_MMC_HIGHSPEED BIT(1) /* Can do MMC high-speed timing */
>>> +#define MMCUTIL_SD_HIGHSPEED BIT(2) /* Can do SD high-speed timing */
>>> +#define MMCUTIL_SDIO_IRQ BIT(3) /* Can signal pending SDIO IRQs */
>>> +#define MMCUTIL_SPI BIT(4) /* Talks only SPI protocols */
>>> +#define MMCUTIL_NEEDS_POLL BIT(5) /* Needs polling for card-detection */
>>> +#define MMCUTIL_8_BIT_DATA BIT(6) /* Can the host do 8 bit transfers */
>>> +#define MMCUTIL_AGGRESSIVE_PM BIT(7) /* Suspend (e)MMC/SD at idle */
>>> +#define MMCUTIL_NONREMOVABLE BIT(8) /* Nonremovable e.g. eMMC */
>>> +#define MMCUTIL_WAIT_WHILE_BUSY BIT(9) /* Waits while card is busy */
>>> +#define MMCUTIL_ERASE BIT(10) /* Allow erase/trim commands */
>>> +#define MMCUTIL_3_3V_DDR BIT(11) /* Host supports eMMC DDR 3.3V */
>>> +#define MMCUTIL_1_8V_DDR BIT(12) /* Host supports eMMC DDR 1.8V */
>>> +#define MMCUTIL_1_2V_DDR BIT(13) /* Host supports eMMC DDR 1.2V */
>>> +#define MMCUTIL_POWER_OFF_CARD BIT(14) /* Can power off after boot */
>>> +#define MMCUTIL_BUS_WIDTH_TEST BIT(15) /* CMD14/CMD19 bus width ok */
>>> +#define MMCUTIL_UHS_SDR12 BIT(16) /* Host supports UHS SDR12 mode */
>>> +#define MMCUTIL_UHS_SDR25 BIT(17) /* Host supports UHS SDR25 mode */
>>> +#define MMCUTIL_UHS_SDR50 BIT(18) /* Host supports UHS SDR50 mode */
>>> +#define MMCUTIL_UHS_SDR104 BIT(19) /* Host supports UHS SDR104 mode */
>>> +#define MMCUTIL_UHS_DDR50 BIT(20) /* Host supports UHS DDR50 mode */
>>> +/* BIT(21) is free for reuse */
>>> +#define MMCUTIL_DRIVER_TYPE_A BIT(23) /* Host supports Driver Type A */
>>> +#define MMCUTIL_DRIVER_TYPE_C BIT(24) /* Host supports Driver Type C */
>>> +#define MMCUTIL_DRIVER_TYPE_D BIT(25) /* Host supports Driver Type D */
>>> +#define MMCUTIL_DONE_COMPLETE BIT(27) /* RW reqs can be completed */
>>> +/* within mmc_request_done() */
>>> +#define MMCUTIL_CD_WAKE BIT(28) /* Enable card detect wake */
>>> +#define MMCUTIL_CMD_DURING_TFR BIT(29) /* Commands during data
>>> transfer*/
>>> +#define MMCUTIL_CMD23 BIT(30) /* CMD23 supported. */
>>> +#define MMCUTIL_HW_RESET BIT(31) /* Hardware reset */
>>> +
>>> +#define MMCUTIL2_BOOTPART_NOACC BIT(0) /* Boot partition no access */
>>> +#define MMCUTIL2_FULL_PWR_CYCLE BIT(2) /* Can do full power cycle */
>>> +#define MMCUTIL2_HS200_1_8V_SDR BIT(5) /* can support */
>>> +#define MMCUTIL2_HS200_1_2V_SDR BIT(6) /* can support */
>>> +#define MMCUTIL2_CD_ACTIVE_HIGH BIT(10) /* Card-detect signal active
>>> high */
>>> +#define MMCUTIL2_RO_ACTIVE_HIGH BIT(11) /* Write-protect signal active
>>> high */
>>> +#define MMCUTIL2_NO_PRESCAN_POWERUP BIT(14) /* Don't power up before
>>> scan */
>>> +#define MMCUTIL2_HS400_1_8V BIT(15) /* Can support HS400 1.8V */
>>> +#define MMCUTIL2_HS400_1_2V BIT(16) /* Can support HS400 1.2V */
>>> +#define MMCUTIL2_SDIO_IRQ_NOTHREAD BIT(17)
>>> +#define MMCUTIL2_NO_WRITE_PROTECT BIT(18) /* No physical write protect
>>> pin,*/
>>> +/* assume that card is always read-write */
>>> +#define MMCUTIL2_NO_SDIO BIT(19) /* Don't send SDIO commands */
>>> +/* during initialization*/
>>> +#define MMCUTIL2_HS400_ES BIT(20) /* Host supports enhanced strobe */
>>> +#define MMCUTIL2_NO_SD BIT(21) /* Do not send SD commands */
>>> +/* during initialization*/
>>> +#define MMCUTIL2_NO_MMC BIT(22) /* Don't send (e)MMC commands*/
>>> +/* during initialization */
>>> +#define MMCUTIL2_CQE BIT(23) /* Has eMMC command queue engine */
>>> +#define MMCUTIL2_CQE_DCMD BIT(24) /* CQE can issue a direct command */
>>
>>
>>
>> Sorry, but this doesn't work. We may very well be changing the bit
>> offsets for the caps and caps2 in the mmc kernel header, keeping a
>> copy of them is not a good idea. It's just a matter of *when* it will
>> break.
>
>
> I guess we could create a header file under include/uapi? This's how
> other drivers, for instance, tty, PCI or graphic subsystem to keep user
> space consistent with kernel side.
Well, yes, for cases where this about providing a userspace ABI.
This is about debugfs, which don't have this constraint. We may decide
to break/drop/change any debugfs support, whenever we want.
So, overall, no I don't like it. If this were about the mmc ioctl or
the sysfs nodes, then yes.
[...]
Kind regards
Uffe
--
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
