Closed WonHoYoo closed 3 years ago
Thanks @WonHoYoo for reporting this! Do you by any chance have the output of Moore handy and can post it here for quick reference / sake of completeness?
Hello, @fabianschuiki. Sorry for the late reply because of my heavy work schedule in my workplace. Below is the output from the moore command.
entity @FU (i1$ %clk, i1$ %rst_n, [1 x i32]$ %A_ADD, [1 x i32]$ %B_ADD, [1 x i32]$ %A_MUL, [1 x i32]$ %B_MUL) -> ([1 x i32]$ %C_ADD, [1 x i32]$ %C_MUL) {
%0 = const i32 0
%1 = [i32 %0]
%2 = const time 0s
drv [1 x i32]$ %C_ADD, %1, %2
drv [1 x i32]$ %C_MUL, %1, %2
}
It seems the generate statement is ignored in the moore toolchain. Would there be anything I am missing on the moore options to apply for getting the generate statement work right? Thanks
Hey @WonHoYoo, thanks for getting back to this with the output! This should definitely contain the vAdd
and vMul
instances as you expect, without the need to pass any special flags to Moore. Let me look into this bug.
This is fixed in version 0.12.1 :+1:
Hi, I just tried moore today, and got the error on the below systemverilog module
with below command
When printing the ast with the option, --dump-ast, it seems the moore compiler doesn't generate the module instantiation as much as the number of the for loop. Could you check on this? Thank you