<div dir="ltr">I've addressed the comments in r202849, thanks!</div><div class="gmail_extra"><br><br><div class="gmail_quote">On Tue, Mar 4, 2014 at 5:32 PM, Yury Gribov <span dir="ltr"><<a href="mailto:tetra2005@gmail.com" target="_blank">tetra2005@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
  More or less what I was thinking about. Minor comments inlined.<br>
<br>
<br>
================<br>
Comment at: lib/asan/asan_stack.h:63<br>
@@ +62,3 @@<br>
<div class="">+    }                                                                          \<br>
+    if (max_size > 1)                                                          \<br>
+      stack.trace[1] = GET_CALLER_PC();                                        \<br>
</div>----------------<br>
Perhaps move this into previous if?<br>
<br>
================<br>
Comment at: test/asan/TestCases/malloc_context_size.cc:6<br>
@@ -5,2 +5,3 @@<br>
<div class=""> // RUN: ASAN_OPTIONS=malloc_context_size=1:fast_unwind_on_malloc=1 not %t 2>&1 | FileCheck %s --check-prefix=CHECK-%os<br>
+// RUN: ASAN_OPTIONS=malloc_context_size=2 not %t 2>&1 | FileCheck %s --check-prefix=TWO<br>
<br>
</div>----------------<br>
Note that all tests with malloc_context_size <= 2 will be using fast path now i.e. normal unwinder may be lacking tests.<br>
<br>
<br>
<a href="http://llvm-reviews.chandlerc.com/D2936" target="_blank">http://llvm-reviews.chandlerc.com/D2936</a><br>
<br>
COMMIT<br>
  <a href="http://llvm-reviews.chandlerc.com/rL202845" target="_blank">http://llvm-reviews.chandlerc.com/rL202845</a><br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div>Alexey Samsonov, MSK</div>
</div>