mmc: Do not send status of send_status is false
PATCH 1/1 mmc: Do not send status of send_status is false
Commit 44645f87de5 ("mmc: Fix mmc_switch excessive timeout") introduced a side effect where CMD13 SEND_STATUS is issued in case mmc_wait_dat0() does not return -ENOSYS and $send_status is not set. This happens on all hardware which does implement .mmc_wait_dat0 callback, e.g. i.MX8M .
This leads to lengthy timeout before booting OS in case of eMMC in one of the HS200/HS400 modes, since the card cannot respond to CMD13 while downgrading from HS200/HS400 to regular HS mode.
Fix this by adding the missing conditional.
--- drivers/mmc/mmc.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/mmc/mmc.c +3 -0
@@ -831,6 +831,9 @@ static int __mmc_switch(struct mmc *mmc, u8 set, u8 index, u8 value,return 0 ; } +if (!send_status) +return 0 ; +/* Finally wait until the card is ready or indicates a failure * to switch. It doesn't hurt to use CMD13 here even if send_status * is false, because by now (after 'timeout_ms' ms) the bus should be
Signature · 1 line
2.35.1
Quoted 11 lines · On Fri, 15 Jul 2022 01:58:24 +0200 Marek Vasut <[email protected]> wrote:
Commit
44645f87de5("mmc: Fixmmc_switchexcessive timeout") introduced a side effect where CMD13SEND_STATUSis issued in casemmc_wait_dat0()does not return-ENOSYSand $send_status is not set. This happens on all hardware which does implement .mmc_wait_dat0 callback, e.g. i.MX8M .This leads to lengthy timeout before booting OS in case of eMMC in one of the HS200/HS400 modes, since the card cannot respond to CMD13 while downgrading from HS200/HS400 to regular HS mode.
Fix this by adding the missing conditional.
Quoted 7 lines
Fixes 44645f87de5 ("mmc: Fix mmc_switch excessive timeout")Signed-off-by Marek Vasut <[email protected]>Cc Jaehoon Chung <[email protected]>Cc Kirill Kapranov <[email protected]>Cc Marek Behún <[email protected]>Cc Pantelis Antoniou <[email protected]>Cc Ye Li <[email protected]>
Does the lenghty timeout occur even on xenon with the patch
https://source.denx.de/u-boot/u-boot/-/commit/0f3466f52fbacce67e147b9234e6323edff26a6d
?
Marek
Quoted 22 lines · Marek Behún
Quoted 11 lines · On Fri, 15 Jul 2022 01:58:24 +0200 Marek Vasut <[email protected]> wrote:
Commit
44645f87de5("mmc: Fixmmc_switchexcessive timeout") introduced a side effect where CMD13SEND_STATUSis issued in casemmc_wait_dat0()does not return-ENOSYSand $send_status is not set. This happens on all hardware which does implement .mmc_wait_dat0 callback, e.g. i.MX8M .This leads to lengthy timeout before booting OS in case of eMMC in one of the HS200/HS400 modes, since the card cannot respond to CMD13 while downgrading from HS200/HS400 to regular HS mode.
Fix this by adding the missing conditional.
Quoted 7 lines
Fixes 44645f87de5 ("mmc: Fix mmc_switch excessive timeout")Signed-off-by Marek Vasut <[email protected]>Cc Jaehoon Chung <[email protected]>Cc Kirill Kapranov <[email protected]>Cc Marek Behún <[email protected]>Cc Pantelis Antoniou <[email protected]>Cc Ye Li <[email protected]>
Quoted 3 lines
Does the lenghty timeout occur even on xenon with the patch
https://source.denx.de/u-boot/u-boot/-/commit/0f3466f52fbacce67e147b9234e6323edff26a6d
?
Unlikely, you will get -ENOSYS from wait on dat0 and bail out, i.e. you'd hit the condition before this one because wait on dat0 is not implemented.
Quoted 11 lines · Marek Vasut
Commit
44645f87de5("mmc: Fixmmc_switchexcessive timeout") introduced a side effect where CMD13SEND_STATUSis issued in casemmc_wait_dat0()does not return-ENOSYSand $send_status is not set. This happens on all hardware which does implement .mmc_wait_dat0 callback, e.g. i.MX8M .This leads to lengthy timeout before booting OS in case of eMMC in one of the HS200/HS400 modes, since the card cannot respond to CMD13 while downgrading from HS200/HS400 to regular HS mode.
Fix this by adding the missing conditional.
Quoted 7 lines
Fixes 44645f87de5 ("mmc: Fix mmc_switch excessive timeout")Signed-off-by Marek Vasut <[email protected]>Cc Jaehoon Chung <[email protected]>Cc Kirill Kapranov <[email protected]>Cc Marek Behún <[email protected]>Cc Pantelis Antoniou <[email protected]>Cc Ye Li <[email protected]>
Best Regards,
Jaehoon Chung
Quoted 18 lines
--- drivers/mmc/mmc.c | 3 +++ 1 file changed, 3 insertions(+)drivers/mmc/mmc.c +3
-0@@ -831,6 +831,9 @@ static int __mmc_switch(struct mmc *mmc, u8 set, u8 index, u8 value,return 0 ; } +if (!send_status) +return 0 ; +/* Finally wait until the card is ready or indicates a failure * to switch. It doesn't hurt to use CMD13 here even if send_status * is false, because by now (after 'timeout_ms' ms) the bus should be
Quoted 11 lines · Marek Vasut
Commit
44645f87de5("mmc: Fixmmc_switchexcessive timeout") introduced a side effect where CMD13SEND_STATUSis issued in casemmc_wait_dat0()does not return-ENOSYSand $send_status is not set. This happens on all hardware which does implement .mmc_wait_dat0 callback, e.g. i.MX8M .This leads to lengthy timeout before booting OS in case of eMMC in one of the HS200/HS400 modes, since the card cannot respond to CMD13 while downgrading from HS200/HS400 to regular HS mode.
Fix this by adding the missing conditional.
Quoted 8 lines
Fixes 44645f87de5 ("mmc: Fix mmc_switch excessive timeout")Signed-off-by Marek Vasut <[email protected]>Cc Jaehoon Chung <[email protected]>Cc Kirill Kapranov <[email protected]>Cc Marek Behún <[email protected]>Cc Pantelis Antoniou <[email protected]>Cc Ye Li <[email protected]>Reviewed-by Jaehoon Chung <[email protected]>
Applied to u-boot/master, thanks!
Signature · 1 line
Tom