On 11/14/2016 03:50 AM, Hans de Goede wrote:
Hi,
On 04-11-16 16:18, Maxime Ripard wrote:
Some eMMC will fail at the first switch, but would succeed in a subsequent one.
Make sure we try several times to cover those cases. The number of retries (and the behaviour) is currently what is being used in Linux.
Signed-off-by: Maxime Ripard maxime.ripard@free-electrons.com
LGTM:
Reviewed-by: Hans de Goede hdegoede@redhat.com
Pantelis or Tom, can you pick this up please ?
Applied on u-boot-mmc. Thanks!
Best Regards, Jaehoon Chung
Regards,
Hans
drivers/mmc/mmc.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index 4380c7c195a6..d6b7e4f510c9 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -494,6 +494,7 @@ int mmc_switch(struct mmc *mmc, u8 set, u8 index, u8 value) { struct mmc_cmd cmd; int timeout = 1000;
int retries = 3; int ret;
cmd.cmdidx = MMC_CMD_SWITCH;
@@ -502,11 +503,17 @@ int mmc_switch(struct mmc *mmc, u8 set, u8 index, u8 value) (index << 16) | (value << 8);
- ret = mmc_send_cmd(mmc, &cmd, NULL);
- while (retries > 0) {
ret = mmc_send_cmd(mmc, &cmd, NULL);
- /* Waiting for the ready status */
- if (!ret)
ret = mmc_send_status(mmc, timeout);
/* Waiting for the ready status */if (!ret) {ret = mmc_send_status(mmc, timeout);return ret;}retries--;}
return ret;