Re: [PATCHv3 0/9] Add new gdbarch::displaced_step_buffer_length field

[Simon Marchi <simark@simark.ca>]

On 3/10/23 13:43, Andrew Burgess via Gdb-patches wrote:
> Changes in V3:
> 
>   - Have now run 'black' on each patch in this series, so everything
>     should be formatted correctly,
> 
>   - Have fixed the out of date names in the last patch of this series.
>     Everything buildds and runs fine for me now,
> 
>   - Have removed the out of date text from the commit message in patch
>     #3,
> 
>   - Have fixed the missing comma Tom spotted in patch #4,
> 
>   - I have NOT removed the comments Simon pointed out in patch #4.
>     Removing these would require changing the generated code for more
>     Components than I already change in this commit.  And so, I think,
>     if those comments are to go that would require a separate patch.
> 
>     That said, we do generate validation code within the getters for
>     many components, so I think having a comment in the components
>     where we don't generate validation makes sense.  So for me, I
>     think the comments do add value, I'd suggest we should keep them.
> 
>   - And the big one.  I've changed the 'invalid' default from False to
>     True.  I know Simon suggested that False was the correct choice,
>     but I actually think that True makes more sense.  I'd rather we
>     assume we should generate the validity checks and require than new
>     components explicitly choose to not have that check, rather than
>     just assuming we shouldn't check.
> 
>     However, in order to get to the default True state I ended up
>     having to fix some other issues.  And, so, incase people really
>     would rather see False as the default, I've left the patch which
>     changes to default True as the penultimate patch in the series.
>     If you feel really strongly that False is best, I can just drop
>     the patch that switches over to use True.  Just let me know.
> 
> Let me know what you think,
> 
> Thanks,
> Andrew

Thanks Andrew, it all LGTM:

Approved-By: Simon Marchi <simon.marchi@efficios.com>

Simon