On Wed, 5 Apr 2023 at 13:57, Christian Löhle <CLoehle@xxxxxxxxxxxxxx> wrote:
>
> SPI doesn't have the usual PROG path we can check for error bits
> after moving back to TRAN. Instead it holds the line LOW until
> completion. We can then check if the card shows any errors or
> is in IDLE state, indicating the line is no longer LOW because
> the card was reset.
>
> Signed-off-by: Christian Loehle <cloehle@xxxxxxxxxxxxxx>
> ---
> drivers/mmc/core/block.c | 9 +++++++--
> 1 file changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
> index 16e262ddc954..35ff7101cbb1 100644
> --- a/drivers/mmc/core/block.c
> +++ b/drivers/mmc/core/block.c
> @@ -182,6 +182,7 @@ static void mmc_blk_rw_rq_prep(struct mmc_queue_req *mqrq,
> int recovery_mode,
> struct mmc_queue *mq);
> static void mmc_blk_hsq_req_done(struct mmc_request *mrq);
> +static int mmc_spi_err_check(struct mmc_card *card);
>
> static struct mmc_blk_data *mmc_blk_get(struct gendisk *disk)
> {
> @@ -553,7 +554,7 @@ static int __mmc_blk_ioctl_cmd(struct mmc_card *card, struct mmc_blk_data *md,
> /* If it's an R1B response we need some more preparations. */
> busy_timeout_ms = idata->ic.cmd_timeout_ms ? : MMC_BLK_TIMEOUT_MS;
> r1b_resp = (cmd.flags & MMC_RSP_R1B) == MMC_RSP_R1B;
> - if (r1b_resp)
> + if (r1b_resp && !mmc_host_is_spi(card->host))
> use_r1b_resp = mmc_prepare_busy_cmd(card->host, &cmd,
> busy_timeout_ms);
>
> @@ -612,8 +613,12 @@ static int __mmc_blk_ioctl_cmd(struct mmc_card *card, struct mmc_blk_data *md,
> if ((card->host->caps & MMC_CAP_WAIT_WHILE_BUSY) && use_r1b_resp)
> return 0;
>
> + if (mmc_host_is_spi(card->host)) {
> + if (idata->ic.write_flag)
> + err = mmc_spi_err_check(card);
> + }
Nitpick: I would prefer to use an "early return" rather than an
else-if here. Would you mind changing that?
> /* Ensure RPMB/R1B command has completed by polling with CMD13. */
> - if (idata->rpmb || r1b_resp)
> + else if (idata->rpmb || r1b_resp)
> err = mmc_poll_for_busy(card, busy_timeout_ms, false,
> MMC_BUSY_IO);
>
Other than the nitpick, this looks good to me!
Kind regards
Uffe
