Re: [Cocci] Determination of an usage statistic for macro calls “for_each…node…”

From: Markus Elfring <Markus.Elfring@web.de>
To: Sumera Priyadarsini <sylphrenadin@gmail.com>,
	Coccinelle <cocci@systeme.lip6.fr>
Cc: Michal Marek <michal.lkml@markovi.net>,
	Gilles Muller <Gilles.Muller@lip6.fr>,
	kernel-janitors@vger.kernel.org,
	Nicolas Palix <nicolas.palix@imag.fr>,
	linux-kernel@vger.kernel.org, Julia Lawall <Julia.Lawall@lip6.fr>
Subject: Re: [Cocci] Determination of an usage statistic for macro calls “for_each…node…”
Date: Sun, 18 Oct 2020 13:16:34 +0200	[thread overview]
Message-ID: <1727796.Akdtb4oiQ5@sonne> (raw)
In-Reply-To: <1980390.yx7ZeylQeb@sonne>

> Would you like to look for software configuration alternatives for better parallel data processing?

I would like to share another source code analysis approach.
I hope that this contribution can trigger further helpful software development ideas.

@initialize:python@
@@
import sys

def write_identifier(source, loop):
    names = []
    for x in source:
       names.append(loop)

    sys.stdout.write("\n".join(names) + "\n")

@find@
identifier fe, x;
iterator name for_each_node_by_name,
              for_each_node_by_type,
              for_each_node_with_property,
              for_each_matching_node,
              for_each_matching_node_and_match,
              for_each_compatible_node,
              for_each_child_of_node,
              for_each_available_child_of_node;
position pos;
statement s;
type t;
@@
 t x(...)
 {
 <+...
(for_each_child_of_node@fe@pos(...) s
|for_each_available_child_of_node@fe@pos(...) s
|for_each_compatible_node@fe@pos(...) s
|for_each_node_by_name@fe@pos(...) s
|for_each_node_by_type@fe@pos(...) s
|for_each_matching_node@fe@pos(...) s
|for_each_matching_node_and_match@fe@pos(...) s
|for_each_node_with_property@fe@pos(...) s
)
 ...+>
 }

@script:python collection@
fe << find.fe;
place << find.pos;
@@
write_identifier(place, fe)

Test result:
elfring@Sonne:~/Projekte/Linux/next-patched> git checkout next-20201016 && XX=$(date) && time spatch --python python3 --jobs 4 --include-headers --no-includes --dir . ~/Projekte/Coccinelle/janitor/report_for_each_node_macro_calls5.cocci | echo "$(echo 'call' && cat)" | csvsql --query 'select call, count(*) from stdin group by call'; YY=$(date) && echo "$XX | $YY"
…
523 files match
…
call,count(*)
for_each_available_child_of_node,158
for_each_child_of_node,359
for_each_compatible_node,80
for_each_matching_node,22
for_each_matching_node_and_match,16
for_each_node_by_name,59
for_each_node_by_type,53
for_each_node_with_property,6

real	0m47,779s
user	2m19,285s
sys	0m1,541s
So 18. Okt 13:13:02 CEST 2020 | So 18. Okt 13:13:50 CEST 2020

Can such facts influence the specification of efficient SmPL disjunctions any more?

Regards,
Markus

_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci