I followed your advice and looked at the registers. Here is what I found:
The function that behaves inconsistently is _dl_start in ld-2.8.so. Most
of the time the first variation in the flow occurs as early as 296
instructions down the road, namely at the jump
30b3b0:0f 86 d7 fd ff
More than one user has hit the -EEXIST problem when using
utrace_attach_task and UTRACE_ATTACH_EXCLUSIVE without
UTRACE_ATTACH_MATCH_DATA|_OPS. Document that a bit more.
Signed-off-by: Ananth N Mavinakayanahalli ana...@in.ibm.com
---
kernel/utrace.c |4
1 file changed, 4 insertions(+)