On 2/22/20 2:05 AM, Tom Rini wrote: > On Fri, Feb 21, 2020 at 04:21:52PM +0900, Jaehoon Chung wrote: >> On 2/21/20 10:07 AM, Jaehoon Chung wrote: >>> Hi Tom, >>> >>> On 2/21/20 4:57 AM, Tom Rini wrote: >>>> On Thu, Feb 20, 2020 at 01:45:31PM +0900, Jaehoon Chung wrote: >>>> >>>>> CONFIG_MMC_BROKEN_CD needs not to define to Kconfig. >>>>> broken-cd is already provide to dt-property. >>>>> If want to poll card-detect, set to broken-cd instead of enabling >>>>> CONFIG_MMC_BROKEN_CD. >>>>> >>>>> When checked the boards that is eabled CONFIG_MMC_BROKEN_CD, >>>>> it also used the value of dt as broken-cd. >>>>> >>>>> Jaehoon Chung (3): >>>>> mmc: jz_mmc; add MMC_CAP_NEEDS_POLL by default >>>>> mmc: check the flags of host_caps about broken-cd >>>>> mmc: Kconfig: remove MMC_BROKEN_CD configuration >>>>> >>>>> configs/brppt2_defconfig | 1 - >>>>> configs/ci20_mmc_defconfig | 1 - >>>>> configs/meerkat96_defconfig | 1 - >>>>> drivers/mmc/Kconfig | 5 ----- >>>>> drivers/mmc/jz_mmc.c | 6 ++++-- >>>>> drivers/mmc/mmc.c | 10 +++++----- >>>>> 6 files changed, 9 insertions(+), 15 deletions(-) >>>> >>>> Did you size-test this change? ci20 is extremely tight on space. >>> >>> I didn't check size-test about this. Is there any check-tool? >>> If there is check-tool, let me know, plz. Then i will check it. (In future, >>> I will check before sending patch.) >>> I just checked this patch with CI. >> >> u-boot.img : 327573 -> 327625 >> u-boot-spl.bin : 10336 -> 10464 >> >> What is ci20's limitation size? > > > The ci20 limit is ((14 * 1024) - 0xa00) for SPL. So we grow by 100 > bytes. And with that small of an SPL we're not using DT, so, is the > driver still acting correctly? Can we restructure the change such that > growth is absolutely minimal? Thanks!
I will re-check this side. I didn't know that ci20 limit is too smaller. As you mentioned, ci20 SPL doesn't use dt. So it needs to set to NEEDS_POLL capability. After checking more, i will resend patches. Thanks for reviewing. Best Regards, Jaehoon Chung >
