mmc: sdhci-of-aspeed: test: Fix dependencies when KUNIT=m
authorDavid Gow <davidgow@google.com>
Fri, 15 Jul 2022 04:03:54 +0000 (12:03 +0800)
committerShuah Khan <skhan@linuxfoundation.org>
Mon, 15 Aug 2022 19:50:56 +0000 (13:50 -0600)
commitd52788b3da2750ebc617891cf260b8f8912f522b
treee56d9c8ff732048e7f05fdc3930a627f47718952
parent568035b01cfb107af8d2e4bd2fb9aea22cf5b868
mmc: sdhci-of-aspeed: test: Fix dependencies when KUNIT=m

While the sdhci-of-aspeed KUnit tests do work when builtin, and do work
when KUnit itself is being built as a module, the two together break.

This is because the KUnit tests (understandably) depend on KUnit, so a
built-in test cannot build if KUnit is a module.

Fix this by adding a dependency on (MMC_SDHCI_OF_ASPEED=m || KUNIT=y),
which only excludes this one problematic configuration.

This was reported on a nasty openrisc-randconfig run by the kernel test
robot, though for some reason (compiler optimisations removing the test
code?) I wasn't able to reproduce it locally on x86:
https://lore.kernel.org/linux-mm/202207140122.fzhlf60k-lkp@intel.com/T/

Fixes: 291cd54e5b05 ("mmc: sdhci-of-aspeed: test: Use kunit_test_suite() macro")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: David Gow <davidgow@google.com>
Acked-by: Andrew Jeffery <andrew@aj.id.au>
Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
Acked-by: Brendan Higgins <brendanhiggins@google.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
drivers/mmc/host/Kconfig