[compiler-rt] r244101 - [UBSan] Fix UBSan-vptr false positive.

[Hans Wennborg via llvm-commits]

On Fri, Aug 14, 2015 at 3:23 PM, Richard Smith <richard at metafoo.co.uk> wrote:
> On Wed, Aug 5, 2015 at 1:07 PM, Hans Wennborg <hans at chromium.org> wrote:
>>
>> I'm fine with merging this if Richard agrees.
>>
>> Richard, what say you?
>
>
> Sorry I missed this. Yes, this looks fine to me.

Thanks! r245114.


>> On Wed, Aug 5, 2015 at 12:52 PM, Alexey Samsonov <vonosmas at gmail.com>
>> wrote:
>> > +Hans, Richard
>> >
>> > Is it too late to merge this into 3.7rc2? This patch should be
>> > relatively
>> > safe, and fixes a bug that was reported looong time ago.
>> >
>> > On Wed, Aug 5, 2015 at 12:35 PM, Alexey Samsonov <vonosmas at gmail.com>
>> > wrote:
>> >>
>> >> Author: samsonov
>> >> Date: Wed Aug  5 14:35:46 2015
>> >> New Revision: 244101
>> >>
>> >> URL: http://llvm.org/viewvc/llvm-project?rev=244101&view=rev
>> >> Log:
>> >> [UBSan] Fix UBSan-vptr false positive.
>> >>
>> >> Offset from vptr to the start of most-derived object can actually
>> >> be positive in some virtual base class vtables.
>> >>
>> >> Patch by Stephan Bergmann!