[mkgmap-dev] [PATCH V3]mkgmap performance

Wed Dec 28 12:25:53 GMT 2011

Hello Steve,

Steve Ratcliffe wrote
> 
> On 12/27/2011 01:49 PM, GerdP wrote:
>>> I don't know why this happens now, and it really shouldn't (apart from
>>> >  the timestamps which are easy to ignore), I will try to find out why.
>>> >
>> thanks, tt would be great if this could be changed.
> 
> I discovered that it is only boundary lines that are affected.
> 
> I didn't quite track down the exact cause, but I guess that the
> relation members are held in a hash map somewhere and are read out in
> an indeterminate order.
> 
> Whatever the exact reason, adding a hashCode() method to the Way class
> prevents the random ordering, so I shall commit that change.
> 

your fix helps when I use max-jobs=1, only a few bytes are different then, 
but I still get random output with max-jobs > 1. Is this expected?

Regarding those different bytes for max-jobs=1: You said they are
time-stamps. Would it be possible to write a constant value (from a
parameter) to these bytes for debugging purposes? 

Gerd

--
View this message in context: http://gis.638310.n2.nabble.com/PATCH-mkgmap-performance-part-2-tp7123938p7132545.html
Sent from the Mkgmap Development mailing list archive at Nabble.com.