logo separator

[mkgmap-dev] Commit r4136: fix missing img file in gmapsupp when --max-jobs option is not given

From Gerd Petermann gpetermann_muenchen at hotmail.com on Thu Mar 15 08:17:09 GMT 2018

Hi Mike,

I've just noticed that both old and new versions don't work well when the first job compiles a typ file
with a command like this
java -jar mkgmap.jar -c options typfile.txt -c split/template.args
It probably requires much less memory to compile the typ file compared to an OSM file.
I guess polish (*.mp) files are somewhere in the middle.
Also my quick fix to handle the case that memoryUsage.getUsed() gives 0 when compiling *.img is probably not working well,
it doesn't compute reasonable values when someone combines preprocessed *.img (maybe with contour lines) with other osm files.
with a command like
java -jar mkgmap.jar contours/*.img -c split/template.args

Maybe you have better a solution for these cases?

Gerd




________________________________________
Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von Mike Baggaley <mike at tvage.co.uk>
Gesendet: Mittwoch, 14. März 2018 17:30:30
An: 'Development list for mkgmap'
Betreff: Re: [mkgmap-dev] Commit r4136: fix missing img file in gmapsupp        when --max-jobs option is not given

Hi Gerd, it was intentional that the first task was removed from the list of
tasks to be performed, so that it didn't run twice. However, I hadn't
realised that there was some mop up code that copied the filename at the end
of the task, so that bit didn't get called for the first task. The attached
patch fixes the problem. I received notification of your update that fixes
the problem by removing the line futures.remove(0); whilst looking into it.
This patch moves the task result handling to a new function, that is called
both by the first task and by the subsequent multithreaded tasks, and does
not require the first task to be run twice. I've also added a test of the
number of processors before attempting to determine how many processors to
use, so that the max-jobs code won't get executed if the system only has one
processor.

Regards,
Mike

-----Original Message-----
From: Gerd Petermann [mailto:gpetermann_muenchen at hotmail.com]
Sent: 13 March 2018 13:53
To: Development list for mkgmap <mkgmap-dev at lists.mkgmap.org.uk>
Subject: Re: [mkgmap-dev] Commit r4136: fix missing img file in gmapsupp
when --max-jobs option is not given

Mike,

please review, I have no idea why the original patch contained the line
futures.remove(0);
I should have noticed this error earlier :-(

The problem always happened when creating a gmapsupp without specifying
max-jobs option.
I assume you just wanted to make sure that the job is not executed twice?

Gerd

________________________________________
Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von svn
commit <svn at mkgmap.org.uk>
Gesendet: Dienstag, 13. März 2018 14:44:03
An: mkgmap-svn at lists.mkgmap.org.uk; mkgmap-dev at lists.mkgmap.org.uk
Betreff: [mkgmap-dev] Commit r4136: fix missing img file in gmapsupp when
--max-jobs option is not given

Version mkgmap-r4136 was committed by gerd on Tue, 13 Mar 2018

fix missing img file in gmapsupp when --max-jobs option is not given

http://www.mkgmap.org.uk/websvn/revision.php?repname=mkgmap&rev=4136
_______________________________________________
mkgmap-dev mailing list
mkgmap-dev at lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev



More information about the mkgmap-dev mailing list