Neel Gala
Neel Gala
Hi.. sorry for the delay in getting to this. @jmerdich thanks for those inputs - I am in agreement with all your suggestions and if you could submit a PR...
> Hello, > > For the standard/regular instructions, I think checking for encoding conflict is necessary. Note that only checking encoding conflict for instructions that belong to the same base...
> And I also think it should be disabled that intrusions with the same name and same base extension to define multiple times. Ahh.. yes.. this is probably something I...
yes.. that's because the encoding.out.h generated (and used by spike) needs to have all the instructions decoded (even if the encodings overlap) because spike has another layer of "decode"/"asserts" to...
> It probably would’ve been less confusing if the static file in this repo were named encoding.in.h, to make it clear it isn’t the complete file, but rather just one...
so what should be the legal ISA string for that device ? RV32I or RV32E ?
I ask because.. in the arch-test framework we choose the I or E tests based on which character is present in the ISA string (from riscv-config yaml). For devices that...
@pawks is refering to the [supported_xlen](https://riscv-config.readthedocs.io/en/stable/yaml-specs.html#supported-xlen) field in the isa yaml which can be used to select rv32/64 tests for a device supporting both.
could you please update pypi? Thanks
@tovine This definitely looks cleaner and better. I will need sometime to verify this at my end. However, does this still produce the same tables as currently seen in the...