GitLab

Created by: bchretien

I noticed that after copying a gdb backtrace inside a code block to a comment as follows:

Program received signal SIGFPE, Arithmetic exception.
0x00007ffff6109895 in ?? () from /usr/lib/libcuda.so
(cuda-gdb) bt
#0 0x00007ffff6109895 in ?? () from /usr/lib/libcuda.so
#1 0x00007ffff610b309 in ?? () from /usr/lib/libcuda.so
#2 0x00007ffff610c832 in ?? () from /usr/lib/libcuda.so
#3 0x00007ffff60a958c in ?? () from /usr/lib/libcuda.so
#4 0x00007ffff609fb8e in ?? () from /usr/lib/libcuda.so
#5 0x00007ffff5f4364f in ?? () from /usr/lib/libcuda.so
#6 0x00007ffff5ff8829 in ?? () from /usr/lib/libcuda.so
#7 0x00007ffff7b97976 in ?? () from /opt/cuda/lib64/libcudart.so.5.0
#8 0x00007ffff7b9cbe0 in ?? () from /opt/cuda/lib64/libcudart.so.5.0
#9 0x00007ffff7b9545b in ?? () from /opt/cuda/lib64/libcudart.so.5.0
#10 0x00007ffff7b9e4ac in ?? () from /opt/cuda/lib64/libcudart.so.5.0
#11 0x00007ffff7ba1ffa in ?? () from /opt/cuda/lib64/libcudart.so.5.0
#12 0x00007ffff7b9b323 in ?? () from /opt/cuda/lib64/libcudart.so.5.0
#13 0x00007ffff7b8b759 in ?? () from /opt/cuda/lib64/libcudart.so.5.0
#14 0x00007ffff7bb0f2a in cudaMalloc () from /opt/cuda/lib64/libcudart.so.5.0
#15 0x000000000040093f in main () at texture.cu:14

The issue where this comment was made is now referenced in issues 1 to 15 (mentioned in issue #...). I guess that code blocks should be ignored when checking for such references.

This was observed with GitLab 7.4.2 (477743a), so it may have been fixed since then, but I haven't been able to find any mentions on this bug in the issue tracker (looking up "issue", "mention", "reference" and so on makes it hard to find relevant issues).