FTBFS on armel: error: unable to find a register to spill in class 'VFP_LO_REGS'

Bug #803798 reported by Stefano Rivera
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Linaro GCC
Won't Fix
Undecided
Michael Collison
gcc-4.6 (Ubuntu)
Fix Released
Undecided
Unassigned
libkdcraw (Ubuntu)
Fix Released
Undecided
Unassigned
sauerbraten (Ubuntu)
Fix Released
Undecided
Unassigned
vtk (Ubuntu)
Fix Released
Medium
Unassigned

Bug Description

vtk 5.6.1-6ubuntu2 failed to build on armel with GCC 4.6.0-14ubuntu1. It's a no-change rebuild of 5.6.1-6ubuntu1 which build successfully with GCC
4.6.0-10ubuntu2.

Full build log: https://launchpadlibrarian.net/74326600/buildlog_ubuntu-oneiric-armel.vtk_5.6.1-6ubuntu2_FAILEDTOBUILD.txt.gz

[ 27%] Building CXX object Imaging/CMakeFiles/vtkImaging.dir/vtkImageQuantizeRGBToIndex.cxx.o
/build/buildd/vtk-5.6.1/Imaging/vtkImageQuantizeRGBToIndex.cxx: In function 'void vtkImageQuantizeRGBToIndexHistogram(T*, int*, vtkIdType*, int, int*, int**) [with T = double, vtkIdType = int]':
/build/buildd/vtk-5.6.1/Imaging/vtkImageQuantizeRGBToIndex.cxx:220:1: error: unable to find a register to spill in class 'VFP_LO_REGS'
/build/buildd/vtk-5.6.1/Imaging/vtkImageQuantizeRGBToIndex.cxx:220:1: error: this is the insn:
(insn 189 201 181 17 (set (subreg:SI (reg:QI 409) 0)
        (unsigned_fix:SI (fix:DF (reg:DF 73 s10 [orig:206 D.41537 ] [206])))) /build/buildd/vtk-5.6.1/Imaging/vtkImageQuantizeRGBToIndex.cxx:182 671 {fixuns_truncdfsi2}
     (expr_list:REG_DEAD (reg:DF 73 s10 [orig:206 D.41537 ] [206])
        (nil)))
/build/buildd/vtk-5.6.1/Imaging/vtkImageQuantizeRGBToIndex.cxx:220: confused by earlier errors, bailing out
make[3]: *** [Imaging/CMakeFiles/vtkImaging.dir/vtkImageQuantizeRGBToIndex.cxx.o] Error 1

Changed in vtk (Ubuntu):
importance: Undecided → Medium
description: updated
Revision history for this message
Jani Monoses (jani) wrote :
Revision history for this message
Michael Hope (michaelh1) wrote :

Thank you for the bug report. Could you provide the following information so that we can reproduce the fault?
 * The command line that shows the fault
 * Preprocessed source of the failing file

To generate the preprocessed source, add '-save-temps' to the GCC command line. This will produce a .i file that you can compress and attach to this ticket.

I suspect this is the same as LP: #807573.

Changed in gcc-linaro:
status: New → Incomplete
Revision history for this message
Stefano Rivera (stefanor) wrote :

>Could you provide the following information so that we can reproduce the fault?

 Sorry, I'm afraid I can't do that. I don't have a physical ARM box, and this package won't play ball under qemu.

Revision history for this message
Michael Hope (michaelh1) wrote :

OK. Let's see if the fix to LP: #807573 also fixes this problem and go from there.

Revision history for this message
Ricardo Salveti (rsalveti) wrote :

Fixed with with the workaround: adding -mfloat-abi=soft on armel

Changed in libkdcraw (Ubuntu):
status: New → Fix Released
Changed in vtk (Ubuntu):
status: New → Confirmed
Revision history for this message
Jani Monoses (jani) wrote :
Revision history for this message
Matthias Klose (doko) wrote :

sauerbraten built

Changed in sauerbraten (Ubuntu):
status: New → Fix Released
Revision history for this message
Matthias Klose (doko) wrote :

gcc-4.6 fixed

Changed in gcc-4.6 (Ubuntu):
status: New → Fix Released
Revision history for this message
Matthias Klose (doko) wrote :

vtk built

Changed in vtk (Ubuntu):
status: Confirmed → Fix Released
Revision history for this message
Michael Collison (michael-collison) wrote :

Test case not provided to reproduce.

Changed in gcc-linaro:
assignee: nobody → Michael Collison (michael-collison)
status: Incomplete → Won't Fix
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.