[PATCH 16/26] spl: Don't cache devices when UNIT_TEST is enabled

[Sean Anderson]

On 10/12/23 03:23, Heinrich Schuchardt wrote:
> On 10/12/23 03:56, Sean Anderson wrote:
>> Several SPL functions try to avoid performing initialization twice by
>> caching devices. This is fine for regular boot, but does not work with
>> UNIT_TEST, since all devices are torn down after each test. Disable caching
>> so we don't use stale devices.
>>
>> Signed-off-by: Sean Anderson <seanga2 at gmail.com>
>> ---
>>
>>   common/spl/spl_fat.c | 2 +-
>>   common/spl/spl_mmc.c | 3 ++-
>>   2 files changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/common/spl/spl_fat.c b/common/spl/spl_fat.c
>> index c6e2526ade1..8bec9cce5ca 100644
>> --- a/common/spl/spl_fat.c
>> +++ b/common/spl/spl_fat.c
>> @@ -24,7 +24,7 @@ static int spl_register_fat_device(struct blk_desc *block_dev, int partition)
>>   {
>>       int err = 0;
>>
>> -    if (fat_registered)
>> +    if (!CONFIG_IS_ENABLED(UNIT_TEST) && fat_registered)
> 
> Please, avoid separate code paths depending on CONFIG_UNIT_TESTS.
> 
> We shouldn't change normal behavior if unit tests are not running. It is
> preferable to let the unit test framework call a new function resetting
> fat_registered when needed. This will allow future tests to test all
> code paths separately.

OK, that sounds reasonable to me.

--Sean