Re: [Qemu-devel] [RFC PATCH v3 1/5] coccinelle: add a script to optimize tcg op using tcg_gen_extract()

[SF Markus Elfring]

>  create mode 100644 scripts/coccinelle/tcg_gen_extract.cocci

Will an other subdirectory be more appropriate for this SmPL script?


> +// Coccinelle helpful issue:
> +// https://github.com/coccinelle/coccinelle/issues/86

I am curious if such an information source will trigger further
software evolution.
How do you think about to mention also the corresponding topic
“Propagating values back from Python script to SmPL rule with other metavariable
type than “identifier”” just for the case that the issue number can be fragile?


> address@hidden@ // match shri*+andi* pattern, calls script verify_len
> +identifier ret, arg;
> +constant ofs, len;
> +identifier shr_fn =~ "^tcg_gen_shri_";
> +identifier and_fn =~ "^tcg_gen_andi_";
> +position shr_p;
> +position and_p;
> +@@
> +(
> address@hidden(ret, arg, ofs);
> address@hidden(ret, ret, len);
> +)

My software development attention was caught also a bit by this specification.
How much do you care for coding style there?

* Two repeated SmPL key words while using the variable list functionality 
before.

* I wonder about the relevance for the parentheses.
  Did you try to express a disjunction for the semantic patch language
  besides the usage of two function (or macro) calls?


> +        print "  candidate", "IS" if is_optimizable else "is NOT", 
> "optimizable"

Would you like to move this information display into a separate function?

Do you care if the “print” is the usage of a function call or a statement?
https://docs.python.org/3.0/whatsnew/3.0.html#print-is-a-function


> address@hidden(ret, arg, ofs);
> address@hidden(ret, ret, len);
> ++extract_fn(ret, arg, ofs, len);

Are there any more cases to consider for the sown function call replacement?

Regards,
Markus