On 4/16/24 23:06, Daniel Wagner wrote: > On Wed, Apr 17, 2024 at 12:58:53AM +0000, Shinichiro Kawasaki wrote: >> Yes, "mark them they are injected from the environment" was the one reason to >> have the parameters in upper cases. The other reason was the consistency across >> the all parameters described in Documentation/running-tests.md. >> >>> can we please keep the small letter similar to nvme_trtype ? >> I'm fine to have small letter, lower cases for the new parameter, but I would >> like to clarify the reason to have lower cases. Do you mean to indicate that >> "the parameters are test group local" using the lower cases? what I meant that for nvme we have lowecase global variables, in case we add new upper case variables they will create confusion ... > Lower cased environment variables are not very common, in fact > POSIX.1-2017 mandates upper cased environment variables [1]. Also only > the nvme part of the framework is using the lower case ones, thus I > agree with Shinichiro to streamline these nvme variables to upper cased > versions. > > [1] https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.html agree they should be uppercase not denying that, the only reason I asked since existing variables are lowercase, it'd be very confusing to have some variables in nvme category with lower case and some upper case ... -ck
