On 01/14/14 01:49, Minkyu Kang wrote: > On 09/01/14 23:31, Mateusz Zalega wrote: >> In some cases MMC was still uninitialized while media capacity check, >> leading to broken ums command. >> >> Change-Id: I4b86c2c59e430fb8b55272ea14f00316d8cb3dca >> Signed-off-by: Mateusz Zalega <m.zal...@samsung.com> >> Cc: Lukasz Majewski <l.majew...@samsung.com> >> Cc: Minkyu Kang <mk7.k...@samsung.com> >> Cc: Kyungmin Park <kyungmin.p...@samsung.com> >> --- >> board/samsung/common/ums.c | 3 +++ >> 1 file changed, 3 insertions(+) >> >> diff --git a/board/samsung/common/ums.c b/board/samsung/common/ums.c >> index dc155ad..0d8f30d 100644 >> --- a/board/samsung/common/ums.c >> +++ b/board/samsung/common/ums.c >> @@ -37,6 +37,9 @@ static struct ums ums_dev = { >> >> static struct ums *ums_disk_init(struct mmc *mmc) >> { >> + if (mmc_init(mmc)) >> + return NULL; >> + >> uint64_t mmc_end_sector = mmc->capacity / SECTOR_SIZE; >> uint64_t ums_end_sector = UMS_NUM_SECTORS + UMS_START_SECTOR; >> >> > > Why don't you init the mmc before call this function? > > diff --git a/board/samsung/common/ums.c b/board/samsung/common/ums.c > index dc155ad..7017c2a 100644 > --- a/board/samsung/common/ums.c > +++ b/board/samsung/common/ums.c > @@ -72,5 +72,8 @@ struct ums *ums_init(unsigned int dev_num) > if (!mmc) > return NULL; > > + if (mmc_init(mmc)) > + return NULL; > + > return ums_disk_init(mmc); > }
I agree, ACK. -- Mateusz Zalega Samsung R&D Institute Poland _______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot