[libre-riscv-dev] [Bug 325] create POWER9 TRAP pipeline

[bugzilla-daemon at libre-soc.org]

https://bugs.libre-soc.org/show_bug.cgi?id=325

--- Comment #22 from Cole Poirier <colepoirier at gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #21)
>  comb += self.o.msr.data.eq(Cat(b[63:31], b[26:22], b[15:0]))
> 
> this says "concatenate bits and put thrm into the LSBs of MSR".
> 
> look again at the VHDL.  it says, "transfer 3 sets of bits into their exact
> same corresponding bit places.
> 
> also, python numbering is start:end+1
> 
> you have specified end:start
> 
> so the ranges are 0:16
> 22:27
> 31:64
> 
> make 3 separate assignments ok?

Thanks, will do. 

> 
>  comb += self.o.msr.ok.eq(a)
> 
> 
> the ok field is 1 bit, saying "this reg is to be written".  by placing the
> LSB of a into it (because msr.ok is only 1 wide), you end up with random
> situations where the MSB will randomly get written.
> 
> msr.ok should be set to 1, to gyarantee it will be written.

Ah thank you that was a typo, had meant to put 1.

> also remember to translate this bit:
> 
>           if c_in(MSR_PR) = '1' then
>  131                         ctrl_tmp.msr(MSR_EE) <= '1';
>  132                         ctrl_tmp.msr(MSR_IR) <= '1';
>  133                         ctrl_tmp.msr(MSR_DR) <= '1';
> 
> 
> 
> OP_SC looks great.

Thanks. I'm not sure how to tranlate c_in and in OP_MTMSR this seems to be done
by just:

```
comb += self.o.msr.data.eq(a)
comb += self.o.msr.ok.eq(1)
```

Is this correct?

-- 
You are receiving this mail because:
You are on the CC list for the bug.