On 11.04.22 18:03, Tom Rini wrote:
On Mon, Apr 11, 2022 at 05:56:52PM +0200, Stefano Babic wrote:
Hi Ariel,
On 11.04.22 13:56, Ariel D'Alessandro wrote:
Hi,
Gentle re-ping. What's blocking this? This has been reviewed and queued for some time already. Please, be clear on the status or what's the plan for this patchset.
After first run, build is blocked by CI due to CONFIG_ that should be se in configs/imx8mn_bsh_smm_s2_defconfig instead of include/configs/imx8mn_bsh_smm_s2_common.h. I can recognize at least CONFIG_SYS_MMCSD_RAW_MODE_U_BOOT_SECTOR (and family), but it looks like there are some more.
See error here:
https://source.denx.de/u-boot/custodians/u-boot-imx/-/jobs/420878
Since this is going to pop up a bit I suspect, https://source.denx.de/u-boot/custodians/u-boot-imx/-/jobs/420878#L24 shows the shell command to run that gives "Unmigrated symbols found in include/configs/imx8mn_bsh_smm_s2_common.h". And if instead of: NUM=`comm -123 --total --output-delimiter=, ${KSYMLST} ${KUSEDLST} | cut -d , -f 3`; if [[ $NUM -ne 0 ]]; then echo "Unmigrated symbols found in $CFG"; exit 1; fi
You do: comm -12 ${KSYMLST} ${KUSEDLST}
It is much better because CI does not report which CONFIG_ were found.
You'll see the whole list of unmigrated symbols. Many of which are easy to migrate and it's just a matter of ./tools/moveconfig.py -y FOO and then sanity checking the output.
And a currently failing patchset is also maybe motivation to expand the CI check to note them too? As I can certainly see how it would be good to have the what failed more visible.
Sure, at least to get the whole list of CONFIG_ that clash.
Regards, Stefano