On Thu, Jul 09, 2026 at 09:27:07AM +0200, Jonas Karlman wrote: > Hi Tom, > > On 7/9/2026 3:53 AM, Tom Rini wrote: > > On Wed, Jul 08, 2026 at 10:05:35PM +0000, Jonas Karlman wrote: > > > >> Change xPL_SERIAL_PRESENT symbols to depend on the xPL_DM_SERIAL symbol > >> intended for the related xPL phase instead of the DM_SERIAL symbol > >> used for U-Boot proper to make CONFIG_IS_ENABLED(SERIAL_PRESENT) guards > >> work as intended. > >> > >> Signed-off-by: Jonas Karlman <[email protected]> > >> --- > >> drivers/serial/Kconfig | 6 +++--- > >> 1 file changed, 3 insertions(+), 3 deletions(-) > >> > >> diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig > >> index e221800d5d06..f125061fe184 100644 > >> --- a/drivers/serial/Kconfig > >> +++ b/drivers/serial/Kconfig > >> @@ -71,7 +71,7 @@ config SERIAL_PRESENT > >> > >> config SPL_SERIAL_PRESENT > >> bool "Provide a serial driver in SPL" > >> - depends on DM_SERIAL && SPL > >> + depends on SPL_DM_SERIAL > >> default y > >> help > >> In very space-constrained devices even the full UART driver is too > > > > Using the database option to qconfig.py, this is not equivalent. There's > > 901 platforms today that enable SPL_SERIAL_PRESENT, and 344 that enable > > SPL_SERIAL_PRESENT but not SPL_DM_SERIAL. Now it's possible that there's > > no functional change here. Something like "pinetab" would be a single > > platform to check the size changes on as it's SPL_SERIAL_PRESENT=y and > > SPL_DM_SERIAL=n. > > > > This discrepancy is likely due to the 'default y' and the strange > 'depend on', based on my observation xPL_SERIAL_PRESENT is mostly used > to guard U_BOOT_DRIVER() and other DM related functions, so I do not > expect any functional change after these changes. > > Also did quick build tests of pinetab_defconfig and there was no binary > change to spl/ before or after this series.
OK, good to know, thanks. I'll grab the series and do the world build before/after test. -- Tom
signature.asc
Description: PGP signature
