logo separator

[mkgmap-dev] Error in memory calculation

From Gerd Petermann gpetermann_muenchen at hotmail.com on Fri Feb 7 15:40:11 GMT 2020

Hi Ticker,

oh, sorry. The corresponding patch was from Mike Baggaley, not from you.

@Mike: What do you think?

Gerd

________________________________________
Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von Ticker Berkin <rwb-mkgmap at jagit.co.uk>
Gesendet: Freitag, 7. Februar 2020 16:36
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] Error in memory calculation

Hi Gerd

I really don't know anything about Java memory management and how it
interacts multi-tasking/max-jobs, so I'm not a good person to review
this in any meaningful way.

Sorry
Ticker

On Fri, 2020-02-07 at 02:55 -0700, Gerd Petermann wrote:
> Hi Ticker,
>
> I think the problem occurs when the G1 garbage collector is active
> and no
> garbage collection was performed. In this case memoryUsage.getMax()
> may
> return -1 while memoryUsage.getUsed() returns a positive value.
> In Java 1.8 you can force the usage of G1 with -XX:+UseG1GC .
> So, this is really a very special case. It happened for John Thorne
> who is
> using several rather small *.mp files as input, but I think it can
> also
> occur for others when the first input file is a typ file in *.txt
> format.
>
> Please review my patch:
> mem.patch <http://gis.19327.n8.nabble.com/file/t318326/mem.patch>  ;
>
> Gerd
>
>
> Ticker Berkin wrote
> > Hi Gerd
> >
> > I don't have a 64bit environment or that much memory. In my normal
> > environment, I get:
> >
> > $ java -Xms1500M -Xmx1540M -jar ../mkgmap.trunk/mkgmap.jar \
> >   dummy.osm dummy.osm dummy.osm
> > Time started: Wed Feb 05 12:56:28 GMT 2020
> > Setting max-jobs to 4
> > Number of MapFailedExceptions: 0
> > Number of ExitExceptions: 0
> > Time finished: Wed Feb 05 12:56:29 GMT 2020
> > Total time taken: 1 second
> >
> > -Xmx1540M is the maximum value that allows Java to start up on my
> > machine.
> >
> > With a newer version I loaded in October, but it didn't have any
> > obvious advantages:
> >
> > $ /opt/jre1.8.0_231/bin/java -Xms1500M -Xmx1540M \
> >   -jar ../mkgmap.trunk/mkgmap.jar \
> >   dummy.osm dummy.osm dummy.osm
> > Time started: Wed Feb 05 12:58:29 GMT 2020
> > Setting max-jobs to 4
> > Number of MapFailedExceptions: 0
> > Number of ExitExceptions: 0
> > Time finished: Wed Feb 05 12:58:30 GMT 2020
> > Total time taken: 1 second
> >
> > Ticker
> >
> > On Fri, 2020-01-31 at 16:15 +0000, Gerd Petermann wrote:
> > > please try this command with openjdk 11 or 13 64Bit:
> > > java -Xms4G -Xmx10G -jar mkgmap.jar f:\osm\dummy.osm
> > > f:\osm\dummy.osm
> > > f:\osm\dummy.osm
> > > This is what I get:
> > > Time started: Fri Jan 31 17:12:53 CET 2020
> > > Setting max-jobs to 1
> > > Number of MapFailedExceptions: 0
> > > To reduce the run time, consider increasing the amnount of memory
> > > available for use by mkgmap by using the Java -Xmx flag to set
> > > the
> > > memory to more than 41000 MB, providing this is less than the
> > > amount
> > > of physical memory installed.
> > > Number of ExitExceptions: 0
> > > Time finished: Fri Jan 31 17:12:54 CET 2020
> > > Total time taken: 1 second
> > _______________________________________________
> > mkgmap-dev mailing list
>
> > mkgmap-dev at .org
>
> > http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
>
>
>
>
>
> --
> Sent from:
> http://gis.19327.n8.nabble.com/Mkgmap-Development-f5324443.html
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
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