Checking incidence for SmPL initialisation rules after parallel script execution
Hello,
I noticed some software development progress for an issue like ?SmPL constraints are broken for type metavariables.?.
https://github.com/coccinelle/coccinelle/issues/32
Now I am checking further test results from a command like the following for the software combination ?Coccinelle 1.0.6-00033-g23cca0a2 (OCaml 4.03.0)?.
spatch.opt -timeout 23 -j 4 --chunksize 1 -sp-file ~/Projekte/Coccinelle/janitor/list_non-void_and_non-static_functions2.cocci -dir ~/geladen/Freeware/Quellen/NetBSD/7.0.2 > ~/Projekte/Bau/NetBSD/Coccinelle/functions1/API2.txt 2> ~/Projekte/Bau/NetBSD/Coccinelle/functions1/API2-errors.txt
This script contains also a SmPL rule ?initialize:python? which should output a header line (usually once?).
elfring at Sonne:~/Projekte/Coccinelle/janitor> ack 'function|"data type"|"source file"|line|column' ~/Projekte/Bau/NetBSD/Coccinelle/functions1/API2.txt | wc -l 2022
Is it really appropriate that this header appears so often in the shown output file?
Regards,
Markus
Quoted 13 lines · SF Markus Elfring
Hello,
I noticed some software development progress for an issue like ?SmPL constraints are broken for type metavariables.?.
https://github.com/coccinelle/coccinelle/issues/32Now I am checking further test results from a command like the following for the software combination ?Coccinelle 1.0.6-00033-g23cca0a2 (OCaml 4.03.0)?.
spatch.opt -timeout 23 -j 4 --chunksize 1 -sp-file
~/Projekte/Coccinelle/janitor/list_non-void_and_non-static_functions2.cocci-dir ~/geladen/Freeware/Quellen/NetBSD/7.0.2 >~/Projekte/Bau/NetBSD/Coccinelle/functions1/API2.txt2>~/Projekte/Bau/NetBSD/Coccinelle/functions1/API2-errors.txt
Quoted 8 lines
This script contains also a SmPL rule ?initialize:python? which should output a header line (usually once?).
elfring at Sonne:~/Projekte/Coccinelle/janitor> ack 'function|"data type"|"source file"|line|column'
~/Projekte/Bau/NetBSD/Coccinelle/functions1/API2.txt| wc -l 2022Is it really appropriate that this header appears so often in the shown output file?
Without having the semantic patch and the .c file, ideally with both cut down to a minimal version that shows the problem, I can't answer this question.
julia
Quoted 5 lines
Quoted 1 line
Is it really appropriate that this header appears so often in the shown output file?
Without having the semantic patch and the .c file, ideally with both cut down to a minimal version that shows the problem,
I can't answer this question.
Does the follow display look a bit better?
elfring at Sonne:~/Projekte/Coccinelle/janitor> grep 'function|"data type"|"source file"|line|column' ~/Projekte/Bau/NetBSD/Coccinelle/functions1/API2.txt | wc -l 5
(Four jobs plus one)
I am looking for another update on the consequences from the parallel execution of scripts for the semantic patch language.
How often should the specified initialisation be performed for the embedded programming language scripts?
Regards,
Markus
Quoted 11 lines · SF Markus Elfring
Quoted 5 lines
Quoted 1 line
Is it really appropriate that this header appears so often in the shown output file?
Without having the semantic patch and the .c file, ideally with both cut down to a minimal version that shows the problem,
I can't answer this question.Does the follow display look a bit better?
elfring at Sonne:~/Projekte/Coccinelle/janitor> grep 'function|"data type"|"source file"|line|column'
~/Projekte/Bau/NetBSD/Coccinelle/functions1/API2.txt| wc -l 5
Quoted 7 lines
(Four jobs plus one)
I am looking for another update on the consequences from the parallel execution of scripts for the semantic patch language.
How often should the specified initialisation be performed for the embedded programming language scripts?
Without having the semantic patch and the .c file, ideally with both cut down to a minimal version that shows the problem,
I can't answer this question.
julia
Quoted 3 lines
Without having the semantic patch and the .c file, ideally with both cut down to a minimal version that shows the problem,
I can't answer this question.
Can you generally clarify current consequences from usage of a SmPL rule ?initialize:?? together with the parameter ?--jobs? for the program ?spatch??
Regards,
Markus
Quoted 6 lines · SF Markus Elfring
Quoted 3 lines
Without having the semantic patch and the .c file, ideally with both cut down to a minimal version that shows the problem,
I can't answer this question.Can you generally clarify current consequences from usage of a SmPL rule ?initialize:?? together with the parameter ?--jobs? for the program ?spatch??
It should happen only once. Thus without your semantic patch and your .c file, I can't tell why it doesn't behave as expected.
julia
Quoted 4 lines
Quoted 2 lines
Can you generally clarify current consequences from usage of a SmPL rule ?initialize:?? together with the parameter ?--jobs? for the program ?spatch??
It should happen only once.
Thanks for this information.
I would like to know if the script initialisation should happen for each background process and/or after the start of the program ?spatch?.
Quoted 2 lines
Thus without your semantic patch and your .c file, I can't tell why it doesn't behave as expected.
You got the corresponding data from my source code analysis try for NetBSD. How often do you observe a simple text line output from initialisation code after parallel execution of one from your test scripts?
Regards,
Markus
Quoted 11 lines · SF Markus Elfring
Quoted 4 lines
Quoted 2 lines
Can you generally clarify current consequences from usage of a SmPL rule ?initialize:?? together with the parameter ?--jobs? for the program ?spatch??
It should happen only once.
Thanks for this information.
I would like to know if the script initialisation should happen for each background process and/or after the start of the program ?spatch?.
Quoted 6 lines
Quoted 2 lines
Thus without your semantic patch and your .c file, I can't tell why it doesn't behave as expected.
You got the corresponding data from my source code analysis try for NetBSD. How often do you observe a simple text line output from initialisation code after parallel execution of one from your test scripts?
It seems to be a problem with python initialization only. Here is my example script:
@initialize:python@
@@
print "hello"
@@
@@
- of_node_put(...);
Note how few lines it contains, how it is designed to illustrate the specific problem, how easy it is to adapt to any software project one has handy, how it is included in the email for easy reference etc. Please do the same.
julia
Quoted 1 line
It seems to be a problem with python initialization only.
Thanks for such a feedback.
Quoted 1 line
Please do the same.
Your SmPL script example is smaller (than mine) and is sufficient to some degree for the clarification of desired software initialisation behaviour.
Regards,
Markus
Julia:
Quoted 9 lines
@initialize:python@
@@print "hello"
@@
@@-
of_node_put(...);
This is now fixed: the initialization code was executed once, but the output was not flushed before the fork, leading the buffered output to be duplicated among all the child processes. Thanks for reporting this.
--
Thierry.
Julia Lawall (2016/12/19, 06:54 UTC)?:
Quoted 42 lines · SF Markus Elfring
Quoted 12 lines
Quoted 5 lines
Quoted 3 lines
Can you generally clarify current consequences from usage of a SmPL rule ?initialize:?? together with the parameter ?--jobs? for the program ?spatch??
It should happen only once.
Thanks for this information.
I would like to know if the script initialisation should happen for each background process and/or after the start of the program ?spatch?.
Quoted 6 lines
Quoted 2 lines
Thus without your semantic patch and your .c file, I can't tell why it doesn't behave as expected.
You got the corresponding data from my source code analysis try for NetBSD. How often do you observe a simple text line output from initialisation code after parallel execution of one from your test scripts?
It seems to be a problem with python initialization only. Here is my example script:
@initialize:python@
@@print "hello"
@@
@@-
of_node_put(...);Note how few lines it contains, how it is designed to illustrate the specific problem, how easy it is to adapt to any software project one has handy, how it is included in the email for easy reference etc. Please do the same.
julia_______________________________________________Cocci mailing list
Cocci at systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci