[Libre-soc-dev] loadstore.py and dcbz modifications

[Tobias Platen]

On Sun, 2021-10-03 at 00:45 +0100, Luke Kenneth Casson Leighton wrote:
>  
> https://git.libre-soc.org/?p=soc.git;a=commitdiff;h=8f78d84e722d30efd553a5caa43ea2f96037ada6
> 
> tobias: i didn't see any messages on either the bugtracker or the
> mailing
> list or the #libre-soc IRC channel about what exactly you planned to
> do
> here.
I'm also joining the #libre-soc IRC channel now.
> 
> if consulted, i would have said that it is better to add an extra
> dcbz parameter
> to the set_wr_addr API rather than duplicate exact functionality of
> set_wr_addr
> unnecessarily, across what is a key and fundamental part of the
> memory
> infrastructure.
I agree, when I first started planning on Friday evening I was unaware
that this would be redundant. I now clean up my code, run my unit tests
again. 
> 
> i have therefore disabled the set_dcbz_addr function and indicated
> in comments what action needs to be taken in its place.
When writing my unit tests, I first wanted to verify that dcbz is
correct. I did not think about redundancies when writing set_dcbz_addr.
> 
> can you please remove it and adapt set_wr_addr by adding (at the end)
> an extra dcbz parameter.
I did, but not commited yet.
> 
> this will need to be done in *all six* places where set_wr_addr is
> used.
> please do all of them *in one single commit*, with an appropriate and
> descriptive commit message.
Just doing a grep shoud be fine, to find *all six* places.
> 
> please also *document* that in an appropriate bugreport, and use the
> bugtracker
> (as well as the mailing list) to communicate what you plan to do,
> before you do it.
Bugreport will be created after doing the commit.
> 
> tia,
> 
> l.
> 
> 
> ---
> crowd-funded eco-conscious hardware:     
> https://www.crowdsupply.com/eoma68
> 
> _______________________________________________
> Libre-soc-dev mailing list
> Libre-soc-dev at lists.libre-soc.org
> http://lists.libre-soc.org/mailman/listinfo/libre-soc-dev
Tobias