[postgis-tickets] [PostGIS] #4990: getfacecontainingpoint fails on i386

PostGIS trac at osgeo.org
Wed Sep 22 21:58:30 PDT 2021


#4990: getfacecontainingpoint fails on i386
-----------------------------+---------------------------
  Reporter:  Bas Couwenberg  |      Owner:  strk
      Type:  defect          |     Status:  new
  Priority:  medium          |  Milestone:  PostGIS 3.2.0
 Component:  topology        |    Version:  master
Resolution:                  |   Keywords:
-----------------------------+---------------------------

Comment (by Bas Couwenberg):

 > Bas I don't think I can run a i386 Docker image on an x86_64 host
 system, right ?

 I don't use Docker, but their docs suggests you can:

  https://docs.docker.com/desktop/multi-arch/

 > Do you have access on a 32bit host to run that image on and let me in ?

 No, I only have amd64 systems with i386 chroots.

 You can create those chroots on your own system:
 {{{
 # On a Debian system
 sudo cowbuilder --create \
                 --distribution=sid \
                 --architecture=i386 \
                 --basepath=/var/cache/pbuilder/base-sid-i386.cow

 # On an Ubuntu system
 sudo apt-get install debian-archive-keyring
 sudo cowbuilder --create \
                 --distribution=sid \
                 --architecture=i386 \
                 --basepath=/var/cache/pbuilder/base-sid-i386.cow \
                 --mirror http://deb.debian.org/debian/ \
                 --debootstrapopts "--keyring=/usr/share/keyrings/debian-
 archive-keyring.gpg"

 # After creating the chroot
 sudo cowbuilder --login \
                 --basepath=/var/cache/pbuilder/base-sid-i386.cow

 # Within the chroot session after login
 sed -i 's/^#deb-src/deb-src/g' /etc/apt/sources.list
 apt-get build-dep postgis
 apt-get source postgis
 cd postgis-*
 <install vim/emacs/nano/etc>
 <hack/test/repeat>
 }}}

  https://debian-gis-team.pages.debian.net/policy/packaging.html#git-
 pbuilder

 > Merge request ​https://gitlab.com/postgis/postgis/-/merge_requests/58
 contains a partial change to avoid unaligned reads to at least the
 computed shortestLine. I think there are more possible unaligned reads BUT
 they could all be made against unserialized geoms which should be safe.

 Still fails with those changes applied to 3.2.0-alpha1:
 {{{
  ./topology/test/regress/getfacecontainingpoint .. failed (diff expected
 obtained: /tmp/pgis_reg/test_167_diff)
 -----------------------------------------------------------------------------
 --- ./topology/test/regress/getfacecontainingpoint_expected     2021-09-23
 03:55:19.000000000 +0000
 +++ /tmp/pgis_reg/test_167_out  2021-09-23 04:27:47.149083888 +0000
 @@ -21,21 +21,4 @@
  ERROR:  Two or more faces found
  ERROR:  Two or more faces found
  ERROR:  Two or more faces found
 -t1|1
 -t2|1
 -t3|9
 -t4|2
 -t5|2
 -t6|0
 -t7|11
 -t8|4
 -t9|1
 -t10|10
 -t11|10
 -t12|10
 -t13|10
 -t14|1
 -t15|12
 -t16|15
 -t17|16
 -t18|19
 +ERROR:  Closest point is first point of closest segment, unexpectedly
 -----------------------------------------------------------------------------
 }}}
  https://salsa.debian.org/debian-gis-team/postgis/-/jobs/1978768/raw

  https://salsa.debian.org/debian-gis-team/postgis/-/pipelines/293026

 The package build on the buildd likewise failed:

 https://buildd.debian.org/status/fetch.php?pkg=postgis&arch=i386&ver=3.2.0%7Ealpha1%2Bdfsg-1%7Eexp2&stamp=1632372717&raw=0

  https://buildd.debian.org/status/package.php?p=postgis&suite=experimental

-- 
Ticket URL: <https://trac.osgeo.org/postgis/ticket/4990#comment:10>
PostGIS <http://trac.osgeo.org/postgis/>
The PostGIS Trac is used for bug, enhancement & task tracking, a user and developer wiki, and a view into the subversion code repository of PostGIS project.


More information about the postgis-tickets mailing list