Bug#805323: graphite2: FTBFS on hppa: Segmentation faults in awamicmp1 and awamicmp2
Source: graphite2
Version: 1.3.4-1
Severity: normal
Dear Maintainer,
Here is buildd log:
https://buildd.debian.org/status/fetch.php?pkg=graphite2&arch=hppa&ver=1.3.4-1&stamp=1447508131
There is a segmentation fault in python:
do_page_fault() command='python' type=15 address=0x005cf000 in libgraphite2.so.3.0.1[faafc000+22000]
vm_start = 0x0032a000, vm_end = 0x005cf000
Gdb gives following info:
Core was generated by `/usr/bin/python /home/dave/debian/graphite2/graphite2-1.3.4/tests/fnttxtrender'.
Program terminated with signal SIGUSR1, User defined signal 1.
#0 0xfab08a38 in graphite2::Vector<float>::insert (x=<optimized out>,
n=2147116631, p=0x5cf000, this=0xfd705ef8)
at /home/dave/debian/graphite2/graphite2-1.3.4/src/inc/List.h:142
142 for (; n; --n, ++p) { new (p) T(x); }
(gdb) disass $pc-16,$pc+16
Dump of assembler code from 0xfab08a28 to 0xfab08a48:
0xfab08a28 <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Position const&, int, float, float, graphite2::json*)+1876>: add,l ret0,r9,r9
0xfab08a2c <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Position const&, int, float, float, graphite2::json*)+1880>: cmpib,= 0,r7,0xfab08a48 <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Position const&, int, float, float, graphite2::json*)+1908>
0xfab08a30 <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Position const&, int, float, float, graphite2::json*)+1884>: stw r9,4(r16)
0xfab08a34 <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Position const&, int, float, float, graphite2::json*)+1888>: cmpiclr,= 0,r3,r0
=> 0xfab08a38 <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Position const&, int, float, float, graphite2::json*)+1892>: fstw fr12,0(r3)
0xfab08a3c <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Po---Type <return> to continue, or q <return> to quit---
sition const&, int, float, float, graphite2::json*)+1896>: ldo -1(r7),r7
0xfab08a40 <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Position const&, int, float, float, graphite2::json*)+1900>: cmpib,<> 0,r7,0xfab08a34 <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Position const&, int, float, float, graphite2::json*)+1888>
0xfab08a44 <graphite2::KernCollider::initSlot(graphite2::Segment*, graphite2::Slot*, graphite2::Rect const&, float, graphite2::Position const&, graphite2::Position const&, int, float, float, graphite2::json*)+1904>: ldo 4(r3),r3
End of assembler dump.
(gdb) p/x $r3
$1 = 0x5cf000
Probably, this is some kind of big-endian issue. Store is to data location
just outside of vm region for libgraphite2.so.3.0.1.
Regards,
Dave Anglin
-- System Information:
Debian Release: stretch/sid
APT prefers unreleased
APT policy: (500, 'unreleased'), (500, 'unstable')
Architecture: hppa (parisc64)
Kernel: Linux 3.18.24+ (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) (ignored: LC_ALL set to en_CA.utf8)
Shell: /bin/sh linked to /bin/dash
Init: sysvinit (via /sbin/init)
Reply to: