[postgis-devel] Winnie is whining
Nicklas Avén
nicklas.aven at jordogskog.no
Sun May 31 11:55:54 PDT 2015
On Sun, 2015-05-31 at 13:42 -0400, Paragon Corporation wrote:
> Brian,
>
>
>
> Very cool. Fails in the same exact spot as where Winnie is failing
> sometimes and yours has more information. So I guess not just a
> windows 32-bit issue whatever it is.
>
> Now can you try running again and see if it fails. Make sure to do a
> make clean and all that stuff.
>
>
>
> Nicklas,
>
>
>
> Does that help any.
Hmm, I don't know.
That "free(): invalid next size (fast):"
seems to be a quite generic memory error:
http://stackoverflow.com/questions/4729395/error-free-invalid-next-size-fast
As the answer says in the link it doesn't even have to be happening just
where it crashes. You said something about that the error moved when you
removed that test, right?
That could point in the direction that there already is some dirty
memory when getting that far.
I cannot replicate this so I have trouble testing and searching.
But what I would try is removing the last part of "test_lwgeom_tcpa"
row 1093 to 1102 in cu_measures.c
It is just for testing. But the M-value in that run is quite big. I
haven't read all parts of the code. But if that value is squared
somewhere it doesn't fit in a double any more. And I guess if it is
stored in a double from the heap strange things might happen.
But I don't know, just a thought. I looked into some of the code, but
couldn't see anything obvious.
If I recall right someone (Paul or Sandro) removed the sqrt from some
comparing distance-calculations not so long time ago. That could maybe
also cause some very big numbers to store somewhere.
I remember that when I was working with the distance function rewriting
in 2009 I tested something like that. I tried to avoid sqrt internally
before the smallest distance was found. And then just do that
calculation on the result. But I ran into strange phenomenas which I
suspected had to do with storing too big numbers. But I was never sure
about what the problem was. Just realized it wasn't stable code :-)
I think Paul or Sandro might have more ideas about what it can be.
Thanks
Nicklas
>
>
>
> Thanks,
>
> Regina
>
>
>
> From: Brian M Hamlin [mailto:maplabs at light42.com]
> Sent: Sunday, May 31, 2015 2:49 AM
> To: PostGIS Development Discussion; Paragon Corporation
> Subject: RE: [postgis-devel] Winnie is whining
>
>
>
>
> ah right - this is better
>
>
>
> http://paste.debian.net/190185/
>
>
>
> On Sun, 31 May 2015 00:42:27 -0400, Paragon Corporation <lr at pcorp.us>
> wrote:
>
> Brian,
>
>
>
>
>
> You seem to have a bunch of issues here not related
>
>
>
>
>
> 1) Looks like you don't have Cunit installed so tests
> don't even try to run.
>
>
>
> 2) Something wrong with your postgres setup. So
> postgresql tests are failing – better check
>
>
>
> /tmp/pgis_reg/regress_log
>
>
>
>
>
> To see what is up there.
>
>
>
>
>
>
>
> Anyway seems Winnie is still intermittently failing despite my
> commit that fixed my issue. I just purged her local bin to
> rule out stale liblwgeom getting in the way and after that it
> passed, but running the whole thing again to see if it fails
> again.
>
>
>
>
>
> Only difference between my setup and winnies aside from hers
> runs under Jenkins (and doing multiple sequential runs) is
> I'm on windows 7 64-bit and she's windows 2012 64-bit. Can't
> see how those would make a difference and why it's just the
> 32-bit and intermittent.
>
>
>
>
>
> The 32bit is at gcc 4.8.1 and the 64-bit is at gcc 4.8.3 so I
> suppose it could be a gcc difference. I think I'll just purge
> the whole postgis folder to see if something sticking there.
>
>
>
>
>
>
>
> Thanks,
>
>
>
> Regina
>
>
>
>
>
>
>
> From: Brian M Hamlin [mailto:maplabs at light42.com]
> Sent: Sunday, May 31, 2015 12:28 AM
> To: PostGIS Development Discussion; Paragon Corporation
> Subject: Re: [postgis-devel] Winnie is whining
>
>
>
>
>
>
> this is a 32bit build just now ..
>
>
>
>
>
>
>
> http://paste.debian.net/190006/
>
>
>
>
>
> On Sat, 30 May 2015 23:18:27 -0400, Paragon Corporation
> <lr at pcorp.us> wrote:
>
>
>
> I was thinking it might be liblwgeom got stuck
> somewhere. I think that has
> happened to me before and it causes random problems
> cause it installs itself
> in usr/local/bin or some such annoying place and an
> old one gets used with
> new code.
> Seems isolated to the 32-bit version. I think 64-bit
> is fine.
>
> Anyway I tried to replicate on my dev box running
> windows 7 64-bit (using
> the same mingw64 32-bit gcc 4.8.1) and sadly it
> crashes on my desktop.
> Though location is slightly different. I get this
>
> liblwgeom code interrupted
> liblwgeom code interrupted
> liblwgeom code interrupted
> passed
> Test: test_lwgeom_locate_along ...passed
> Test: test_lw_dist2d_pt_arc ..passed
> Test: test_lw_dist2d_seg_arc ...passed
> Test: test_lw_dist2d_arc_arc ..passed
> Test: test_lw_arc_length ...passed
> Test: test_lw_dist2d_pt_ptarrayarc ...passed
> Test: test_lw_dist2d_ptarray_ptarrayarc ...passed
> Test: test_lwgeom_tcpa .passed
> Test: test_lwgeom_is_trajectory ...Geometry is not a
> LINESTRING
> Line does not have M dimension
> Measure of vertex 1 (1) not bigger than measure of
> vertex 0 (1)
> Measure of vertex 1 (0) not bigger than measure of
> vertex 0 (1)
> Measure of vertex 2 (2) not bigger than measure of
> vertex 1 (3)
> passed
> Suite: effectivearea
> Test: do_test_lwgeom_effectivearea_lines ...passed
> Test: do_test_lwgeom_effectivearea_polys ...passed
> Suite: miscellaneous
> Test: test_misc_force_2d ...Makefile:80: recipe for
> target `check' failed
> make[2]: *** [check] Error 5
> make[2]: Leaving directory
> `/projects/postgis/branches/2.2/liblwgeom/cunit'
> Makefile:168: recipe for target `check' failed
> make[1]: *** [check] Error 2
> make[1]: Leaving directory
> `/projects/postgis/branches/2.2/liblwgeom'
> GNUmakefile:14: recipe for target `check' failed
>
>
>
>
> -----Original Message-----
> From: postgis-devel-bounces at lists.osgeo.org
> [mailto:postgis-devel-bounces at lists.osgeo.org] On
> Behalf Of Nicklas Avén
> Sent: Saturday, May 30, 2015 6:27 PM
> To: PostGIS Development Discussion
> Subject: [postgis-devel] Winnie is whining
>
> Hallo
>
> Regina, I see winnie is unhappy. But I cannot
> understand how my last commit
> could cause that.
>
> Test: test_lw_dist2d_seg_arc ...passed
> Test: test_lw_dist2d_arc_arc ..passed
> Test: test_lw_arc_length ...passed
> Test: test_lw_dist2d_pt_ptarrayarc ...passed
> Test: test_lw_dist2d_ptarray_ptarrayarc ...passed
> Test: test_lwgeom_tcpa .Makefile:80: recipe for target
> `check'
> failed
> make[2]: *** [check] Error 5
> make[2]: Leaving directory
> `/projects/postgis/branches/2.2/liblwgeom/cunit'
> Makefile:168: recipe for target `check' failed
> make[1]: *** [check] Error 2
> make[1]: Leaving directory
> `/projects/postgis/branches/2.2/liblwgeom'
> GNUmakefile:14: recipe for target `check' failed
> make: *** [check] Error 1
>
> Can it be something else causing it?
>
>
> Thanks
>
> Nicklas
>
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at lists.osgeo.org
> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-devel
>
>
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at lists.osgeo.org
> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-devel
>
>
>
>
>
>
>
> --
> Brian M Hamlin
> OSGeo California Chapter
> blog.light42.com
>
>
>
>
>
>
>
>
>
>
>
>
> --
> Brian M Hamlin
> OSGeo California Chapter
> blog.light42.com
>
>
>
>
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at lists.osgeo.org
> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-devel
More information about the postgis-devel
mailing list