[PATCH] SafeStack documentation improvements

[Volodymyr Kuznetsov]

Thanks for the comments! I've addressed them in the updated patch.


================
Comment at: docs/SafeStack.rst:107
@@ +106,3 @@
+breaking the probabilistic security guarantees through information leaks, may
+be not complete yet. System library functions such as ``setjmp``, exception
+handling mechanisms, intrinsics such as ``__buildin_frame_address``, or
----------------
jfb wrote:
> "may be not complete yet" is pretty awkward phrasing.
I've rephrased this and the previous paragraphs to make it sound less awkward.

================
Comment at: docs/SafeStack.rst:111
@@ +110,3 @@
+such safe stack pointer leaks could be detected by a static binary analysis or
+a dynamic binary instrumentation based tools.
+
----------------
jfb wrote:
> I'd strengthen this last statement: at the moment safe stack assumes that the compiler's implementation is correct. This has not been verified except through code inspection, and could always regress in the future. It's therefore desirable to have a separate static or dynamic binary analysis / checker.
Great point! I think it applies to the entire SafeStack instrumentation, not just the hiding aspect, hence I added this note at the end of the Security Limitations section.

http://reviews.llvm.org/D10598

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/