<div dir="ltr">I've recently played with the GCC implementation of pointer checker on a real hardware, <div>my recent impressions are here: <a href="https://github.com/google/sanitizers/wiki/AddressSanitizerIntelMemoryProtectionExtensions">https://github.com/google/sanitizers/wiki/AddressSanitizerIntelMemoryProtectionExtensions</a> </div><div>(there is also some old pre-hardware content). </div><div><br></div><div>In short, I totally agree with what David says above: MPX is a disaster. </div><div>(Usual disclaimer: my opinion here is too biased)<br></div><div><br></div><div>I am glad that LLVM already has the support for MPX instructions, but I see no good reason to add the MPX-based checker to LLVM. </div><div>Yes, it will allow us to detect intra-object overflows, something that asan <a href="https://github.com/google/sanitizers/wiki/AddressSanitizerIntraObjectOverflow">can not do by default</a>, but it's not worth the extreme complexity of the MPX-based checker. </div><div><br></div><div>--kcc </div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Jan 28, 2016 at 9:14 AM, David Chisnall via llvm-dev <span dir="ltr"><<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On 28 Jan 2016, at 17:03, Dmitrii Kuvaiskii via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.llvm.org</a>> wrote:<br>
><br>
> Interestingly, my<br>
> preliminary experiments indicate that adding MPX bounds checking via<br>
> Pointer Checker in gcc is usually slower than using software-only<br>
> AddressSanitizer.<br>
<br>
</span>This corresponds with other results that I have seen. The last time I looked at the output from gcc, it also did not generate pointer updates that were safe in the presence of concurrency (they must be bracketed in transactions if you want the MPX metadata and the pointer updates to be atomic) and the overhead of this is likely to be even more.<br>
<br>
I am particularly impressed with Intel for creating a hardware implementation that is both slower than a software-only version and can not (due to its fail-open policy being embedded in the hardware) be used for security.<br>
<br>
David<br>
<div class="HOEnZb"><div class="h5"><br>
_______________________________________________<br>
LLVM Developers mailing list<br>
<a href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.llvm.org</a><br>
<a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
</div></div></blockquote></div><br></div>