Re: using scratchpads to enhance RTL-level if-conversion: the new patch now passes bootstrap with the default BUILD_CONFIG [i.e. no stage2-to-stage3 comparison errors even with debugging info off in stage2 and on in stage3] AND passes "make check" testing with no new regressions

[Abe Skolnik <abe_skolnik at yahoo dot com>]

[Bernd Schmidt wrote:]


> This is currently not really reviewable due to
> broken indentation, [...] coding guidelines.

Thanks for all your feedback, including that to which I have not explicitly replied below.  I will work on this aspect of the code very soon, probably today.


> Please ensure your code is formatted
> the same way as all other code in gcc.

I understand that this is important, and already made some effort to comply, but clearly not enough yet.


> https://www.gnu.org/prep/standards/html_node/Writing-C.html

Thanks for the link!



[Abe wrote:]


>> +        /* Abe`s note: do we need to do the following after getting
>> +           a new pseudo-reg., as shown elsewhere in this file?
>> +          if (max_regno < max_reg_num ())  max_regno = max_reg_num ();
>> +        */


[Bernd Schmidt wrote:]

> Avoid referencing yourself in the comments with things like "Abe's note".

I understand now that I should have put this in the body of the email rather than in the patch.  What I meant by this comment was to solicit feedback on whether or not that commented-out statement is needed in that position in the code; perhaps since the patch now passes bootstrap and regression testing, I can safely assume -- since the line is still commented out -- that it is not needed.


> Spurious whitespace change.

Sorry; I thought I had fixed that one.



I will reply separately to the algorithmic comment so that the resulting email and mailing-list archive threads are not mixed in with the above.


Thanks again for your helpful advice.

Regards,

Abe