On Sat, Jul 27, 2019 at 1:46 PM Paul Walmsley <paul.walms...@sifive.com> wrote: > > On Sat, 27 Jul 2019, Anup Patel wrote: > > > If your only objection is uppercase letter not agreeing with YMAL schema > > then why not fix the YMAL schema to have regex for RISC-V ISA string? > > I don't agree with you that the specification compels software to accept > arbitrary case combinations in the riscv,isa DT string.
DT describes HW and HW follows RISC-V spec. Enforcing software choices in DT YMAL schema is not correct approach. Some other OS (such as FreeBSD, NetBSD, etc) might choose to go with upper-case characters only in their DTS files. > > > The YMAL schema should not enforce any artificial restriction which is > > theoretically allowed in the RISC-V spec. > > Unless someone can come up with a compelling reason for why restricting > the DT ISA strings to all lowercase letters and numbers is insufficient to > express the full range of options in the spec, the additional complexity > to add mixed-case parsing, both in this patch and in the other patches in > this series, seems pointless. So, using strncasecmp() in-place of strncmp() and using tolower() for each character comparison is complex for you ? Why do we need a pointless restriction in YAML schema ? Regards, Anup
