On Wed, 29 May 2024 10:09:28 +0200 Nuno Sá <noname.nuno@xxxxxxxxx> wrote: > On Wed, 2024-05-29 at 09:23 +0200, Arnd Bergmann wrote: > > On Wed, May 29, 2024, at 09:20, Nuno Sá wrote: > > > On Tue, 2024-05-28 at 14:12 +0200, Arnd Bergmann wrote: > > > > From: Arnd Bergmann <arnd@xxxxxxxx> > > > > > > > > Build testing without SPI results in a warning: > > > > > > > > WARNING: unmet direct dependencies detected for REGMAP_SPI > > > > Depends on [n]: SPI [=n] > > > > Selected by [m]: > > > > - AD9739A [=m] && IIO [=m] && (SPI [=n] || COMPILE_TEST [=y]) > > > > > > > > There is no need for this particular COMPILE_TEST option, as allmodconfig > > > > and randconfig testing can just assume that SPI is enabled separately. > > > > > > > > Drop it to avoid the warning. > > > > > > > > Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx> > > > > --- > > > > > > Reviewed-by: Nuno Sa <nuno.sa@xxxxxxxxxx> > > > > > > Only for my understanding. When does COMPILE_TEST makes sense to add? > > > > The most common use is to enable building a driver > > that is platform specific on x86 allmodconfig > > or on randconfig, e.g. > > > > config LPC18XX_DAC > > tristate "NXP LPC18xx DAC driver" > > depends on ARCH_LPC18XX || COMPILE_TEST > > depends on HAS_IOMEM > > > > Since ARCH_LPC18XX is only visible on arch/arm, the > > driver would never be build tested on anything else > > without the ||COMPILE_TEST. > > Makes sense... Thanks! > > - Nuno Sá > Applied. Thanks