Closed GoogleCodeExporter closed 9 years ago
Hi,
I can't really tell what the problem is from the information provided. Please
report this to the Arch Linux package maintainer.
Thanks
Original comment by fatbob.s...@gmail.com
on 30 May 2012 at 7:38
Sure, probably should have done that to start, thanks for the recommendation.
Original comment by stelleg@gmail.com
on 30 May 2012 at 7:42
So after communicating for a while with the package manager, I'm still not sure
what the issue is. The backtrace is the following, with the latest openssl:
uMurmur version 0.2.10 ('Bats in the Belfry')
#0 0x400d0e00 in _armv7_neon_probe () from /usr/lib/libcrypto.so.1.0.0
#1 0x400cd3c4 in OPENSSL_cpuid_setup () from /usr/lib/libcrypto.so.1.0.0
#2 0x4000f6dc in call_init () from /lib/ld-linux.so.3
#3 0x4000f7a0 in _dl_init_internal () from /lib/ld-linux.so.3
#4 0x40000d94 in _dl_start_user () from /lib/ld-linux.so.3
#5 0x40000d94 in _dl_start_user () from /lib/ld-linux.so.3
In the end I gave up and compiled it against polarssl, which is now working
fine.
Original comment by stelleg@gmail.com
on 30 May 2012 at 10:09
Any updates on the matter?
I have a similar issue with a framework:
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/libthread_db.so.1".
Program received signal SIGILL, Illegal instruction.
0x406a9e00 in _armv7_neon_probe () from /usr/lib/libcrypto.so.1.0.0
Original comment by roman.fl...@gmail.com
on 2 Jul 2012 at 9:23
From the trace above, this is most likely an issue with OpenSSL.
A quick google on '_armv7_neon_probe' reveals this:
http://archlinuxarm.org/forum/viewtopic.php?f=31&t=3004&start=50#p18353
I haven't delved into the details of the thread, but there seem to be some kind
of misdetection of CPU (ARM7 vs ARM6) that leads to an illegal instruction for
the CPU you're running it on. Or something to that extent...
Original comment by fatbob.s...@gmail.com
on 3 Jul 2012 at 6:23
Original issue reported on code.google.com by
stelleg@gmail.com
on 26 May 2012 at 12:27