[llvm-dev] ilist/iplist are broken (maybe I'll fix them?)

Reid Kleckner via llvm-dev llvm-dev at lists.llvm.org
Tue Oct 20 11:23:37 PDT 2015

I think the implicit iterator conversions are much less important now that
we have range based for loops, but I still like having them.

On Tue, Oct 20, 2015 at 11:13 AM, Duncan P. N. Exon Smith via llvm-dev <
llvm-dev at lists.llvm.org> wrote:

> > On 2015-Oct-07, at 17:57, Duncan P. N. Exon Smith <dexonsmith at apple.com>
> wrote:
> >
> > I've been digging into some undefined behaviour stemming from how ilist
> > is typically configured.  r247937, r247944, and r247978 caused a UBSan
> > failure to start firing on our Green Dragon bots, and after an IRC
> > conversation between David and Nick and Mehdi, we added a blacklist:
> > --
> > $echo "src:$WORKSPACE/llvm/include/llvm/CodeGen/MachineFunction.h" >>
> sanitize.blacklist
> > --
> >
> > ilist/iplist is a pretty strange list, and the more I dig into it (to
> > fix the UB) the more broken I think it is.
> >
> > I want to change a few things about it, but it'll be somewhat
> > intrusive (pun not really intended), so I want to get some buy-in before
> > really diving in.  I've CC'ed the people in the IRC conversation and a
> > couple of others that seem to care about ADT and/or UB.
> A quick update on this.
> The first problem I hit was that there are callers that *rely* on
> `getNextNode()` returning the sentinel instead of `nullptr`.  If you
> look at the implementation of `getNextNode()`, that's kind of insane.
> The only way I could think to root out all the similar issues was to
> look at every call to the implicit conversions and confirm they aren't
> doing anything strange.  Most easily done by applying the attached
> patch, and getting this compiling again.  I have some more commentary
> in, e.g., r249767 and r249782.
> Some of the problems I've uncovered include r249758, r249763, r249764,
> and more scary cases like r249925 and r250211.
> I've almost finished LLVM proper, but I haven't touched clang yet, or
> other LLVM projects.
> Is there any contention about this?  Do we eventually want to commit
> this patch, or should we go back to our old implicit ways once I've
> cleaned up ilist and iplist?  (Basically, should we make clang clean
> too and commit this patch, or should I just fix the bugs?)
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20151020/f9a42202/attachment.html>

More information about the llvm-dev mailing list