[PATCH] [asan] Update the llvm webpage describing ASan

Alexey Samsonov vonosmas at gmail.com
Fri Jun 19 16:29:26 PDT 2015

Thanks for working on this!

Comment at: docs/AddressSanitizer.rst:64
@@ +63,3 @@
+exit with a non-zero exit code. AddressSanitizer exits on the first detected
+error. This is by design. One reason: it makes the generated code smaller and
+faster (both by ~5%). Another reason: this makes fixing bugs unavoidable. With
I would prefer bullet points for reasons.

Comment at: docs/AddressSanitizer.rst:66
@@ +65,3 @@
+faster (both by ~5%). Another reason: this makes fixing bugs unavoidable. With
+Valgrind, it is often the case that users treat Valgrind warnings as false
+positives (which they are not) and don't fix them.
I think we don't have to mention Valgrind, and just claim that all ASan errors are true positives, and
the program can be in inconsistent state, causing confusing results, and (potentially) misleading
subsequent reports.

Comment at: docs/AddressSanitizer.rst:228
@@ -186,3 +227,3 @@
 * Linux i386/x86\_64 (tested on Ubuntu 12.04);
-* MacOS 10.6 - 10.9 (i386/x86\_64).
+* OS X 10.6 - 10.11 (i386/x86\_64).
 * Android ARM
I wonder if we can bump it to 10.7...

Comment at: docs/AddressSanitizer.rst:249
@@ +248,3 @@
+when initialization of globals defined in one translation unit uses
+globals defined in another translation unit. To enable this check at runtime,
+you should set environment variable
That's weird to have this at the very bottom. Maybe, put between limitations and platforms? 

Comment at: docs/AddressSanitizer.rst:256
@@ +255,3 @@
+For the experimental memory leak detector in AddressSanitizer, see
While you're here.... it's not experimental anymore - it's on by default for Linux. (not implemented on Mac)



More information about the llvm-commits mailing list