Arch Linux

Description:

A trivial C program fails instrumentation due to
superflous addr16 prefix in first call to a C
library function.

This bug is known and has been fixed in:

https://bugs.kde.org/show_bug.cgi?id=384230

In Archlinux32 there is a patch, see also:

https://bbs.archlinux32.org/viewtopic.php?pid=501

The question would be: does it make sense to take this bug into
a fix in the i686 valgrind package upstreams?

Additional info:
* package version(s): 3.13.0-3
* config and/or log files etc.


Steps to reproduce:

Simple program fails with:

cat >test.c <<EOF
#include <stdio.h>

int main( void )
{
puts( "hello" );
return 0;
}
EOF

gcc -g -o test test.c

valgrind ./test
vex x86->IR: unhandled instruction bytes: 0x67 0xE8 0xD3 0x8B