On Tue, 2019-10-01 at 00:02 -0700, Christoph Hellwig wrote: > On Mon, Sep 30, 2019 at 05:23:18PM -0700, Atish Patra wrote: > > /proc/cpuinfo should just print all the isa string as an > > information > > instead of determining what is supported or not. ELF hwcap can be > > used by the userspace to figure out that. > > > > Simplify the isa string printing by removing the unsupported isa > > string > > print and all related code. > > > > The relevant discussion can be found at > > http://lists.infradead.org/pipermail/linux-riscv/2019-September/006702.html > > This looks good, but can you also rename the orig_isa argument to isa > now that we never modify it? > Sure. I will do that.
> > /* > > * Linux doesn't support rv32e or rv128i, and we only support > > booting > > * kernels on harts with the same ISA that the kernel is > > compiled for. > > */ > > #if defined(CONFIG_32BIT) > > - if (strncmp(isa, "rv32i", 5) != 0) > > + if (strncmp(orig_isa, "rv32i", 5) != 0) > > return; > > #elif defined(CONFIG_64BIT) > > - if (strncmp(isa, "rv64i", 5) != 0) > > + if (strncmp(orig_isa, "rv64i", 5) != 0) > > return; > > #endif > > And I don't think having these checks here makes much sense. Correct. As we are dumping the isa information as it is, we should get rid of these checks as well. > If we want > to check this at all we should do it somewhere in the boot process. riscv_of_processor_hartid() or seems to be a better candidate. We already check if "rv" is present in isa string or not. I will extend that to check for rv64i or rv32i. Is that okay ? -- Regards, Atish
