Johan Euphrosine
Johan Euphrosine
I looks like https://github.com/google/xls/blob/main/xls/modules/zstd/ is the only place where they are currently used: https://github.com/google/xls/blob/00eeaf43b42c80f8533b2ff02d068508ef6afa83/xls/modules/zstd/block_dec.x#L56-L61
Does this replace #1057?
> We can discard the previous PR, if you think that it is easier to merge the cocotb support with the AXI tests I think it'll be easier to maintain...
> 2) compose the actual value with multiple inputs to satisfy the constraint. This may affect the statistical distribution compared to a properly constrained input. can you give an example?...
> One of the weirdnesses here is you'll end up with DSLX errors reported on your "constraint string". +1, it might be a bit annoying to have to define intermediate...
Sharing a small helper that we've been using to simplify the expression of preconditions for checking quick check properties. So that instead of writing long chain of `||` negated preconditions...
> However, there is no way to leverage the problem with "sub-optimal" delay model for the scheduler @QuantamHD @richmckeever couldn't this also be enabled by fdo? https://github.com/google/xls/blob/f02c9999c3a739cbcbd6f414ea19fb10439f3cf1/xls/tools/scheduling_options_flags.proto#L47-L58
@tmichalak https://github.com/google/xls/commit/bf28838aa0fc0e00ac6165511ffbc5e5f4430195 fyi (for some reason it seems that I can't mention `@antmicro/xls`)
Note this fails with the following error: ``` 0002: 0003: fn f(x: u32) -> u32 { N + x } 0004: fn main() { map(u32[4]:[0, 1, 2, 3], f) }...