This might not seem like much, but it's already better than Mac OS'
backtrace(3).
(cherry picked from commit 2a46f6293f19b9d6100187fbd53088088e671b88)
Change-Id: I197b06c4c19f728ad6f194b2f473bd9dcd99b8a6
Backtracing through eh_frame section is more effective allowing to reuse
ebp register for other purposes within routine. GCC with turned on
optimizations (-O1 and above) implicitly defines -fomit-frame-pointer
anyway. eh_frame sections are generated by default with GCC on any
optimization level.
This change implements remote unwinding (separate process unwinding).
Local unwinding is already implemented through _Unwind_Backtrace call
which is implemented in libgcc.
Change-Id: I1aea1ecd19c21710f9cf5f05dc272fc51b67b7aa
Signed-off-by: Pavel Chupin <pavel.v.chupin@intel.com>
-Werror is used for this project so these warnings were causing the
clang build to fail.
Change-Id: I18c447ce239645e05f59c3cf0e2b8bb17d9d3030
Author: Edwin Vane <edwin.vane@intel.com>
Reviewed-by: Tarea A Siraj <tareq.a.siraj@intel.com>
This is a forward-compatibility patch used to avoid two problems:
1/ The C library <signal.h> is going to be updated to define 'struct sigcontext'
properly soon. Avoid redefining this structure here when it's not really needed
to prevent a type conflict.
2/ Similarly, proper ucontext_t support is going to be added, prevent conflict by
using a macro renaming trick. Mainly because there is a slight chance that the
C library definition will follow a slightly different layout/naming.
For context, see details at:
http://code.google.com/p/android/issues/detail?id=34784https://android-review.googlesource.com/#/c/38875/1
Change-Id: Ie94eb5d77297f8e0bafd146b65a648bd8805a551
Longer term, we might want to offer <ucontext.h> and have that be like glibc's,
but not today. (Note that POSIX says nothing about mcontext_t and glibc,
Mac OS, and the Linux kernel headers are all mutually incompatible.)
Change-Id: Ia074b51f4567cd54e06bbe29721389c8d1278614
Keep track of whether memory maps are readable. Use the information
in try_get_word to try to avoid accidentally dereferencing an invalid
pointer within the current process. (Note that I haven't ever
seen that happen during normal unwinding, but it pays to be
a little more careful.)
Refactored try_get_word a little to make it easier to pass it the
needed state for validation checks by way of a little memory_t struct.
Improved how the memory map for the current process is cached. This is
important because we need up to date information about readable maps.
Use a 5 second cache expiration.
Improved the PC -> LR fallback logic in the unwinder so we can
eke out an extra frame sometimes.
Fixed a bug reading ELF program headers. The phnum & phentsize
fields are half-words. We were incorrectly interpreting
phnum as a whole word.
Used android_atomic_* operations carefully in the unwinder
to prevent possible memory races between the dumper and the dumpee.
This was highly unlikely (or even impossible due to the presence
of other barriers along the way) but the code is clearer now about
its invariants.
Fixed a bug in debuggerd where the pid was being passed to have
its stack dump taken instead of the tid, resulting in short
stacks because ptrace couldn't read the data if pid != tid.
Did a full sweep to ensure that we use pid / tid correctly everywhere.
Ported old code from debuggerd to rewind the program counter back
one instruction so that it points to the branch instruction itself
instead of the return address.
Change-Id: Icc4eb08320052975a4ae7f0f5f0ac9308a2d33d7
Elliott Hughes
Pavel Chupin
Jean-Baptiste Queru
Edwin Vane
David 'Digit' Turner
Jeff Brown