[llvm-dev] RFC: LTO should use -disable-llvm-verifier

Tue Sep 8 13:16:05 PDT 2015

On Tue, Sep 08, 2015 at 10:52:29AM -0700, Duncan P. N. Exon Smith wrote:
> 
> > On 2015-Sep-04, at 15:53, Peter Collingbourne via llvm-dev <llvm-dev at lists.llvm.org> wrote:
> > 
> > On Fri, Sep 04, 2015 at 03:11:39PM -0700, Mehdi Amini wrote:
> >> 
> >>> On Sep 4, 2015, at 11:38 AM, Peter Collingbourne <peter at pcc.me.uk> wrote:
> >>> 
> >>> On Fri, Sep 04, 2015 at 11:13:43AM -0700, Mehdi Amini wrote:
> >>>> 
> >>>>> On Sep 4, 2015, at 11:03 AM, Eric Christopher <echristo at gmail.com> wrote:
> >>>>> 
> >>>>> 
> >>>>> 
> >>>>> On Fri, Sep 4, 2015 at 12:48 AM Mehdi Amini <mehdi.amini at apple.com <mailto:mehdi.amini at apple.com>> wrote:
> >>>>>> On Sep 4, 2015, at 12:22 AM, Eric Christopher <echristo at gmail.com <mailto:echristo at gmail.com>> wrote:
> >>>>>> 
> >>>>>> 
> >>>>>> 
> >>>>>> On Thu, Sep 3, 2015 at 11:45 PM Mehdi Amini <mehdi.amini at apple.com <mailto:mehdi.amini at apple.com>> wrote:
> >>>>>> Hi,
> >>>>>> 
> >>>>>>> On Sep 2, 2015, at 7:31 PM, Peter Collingbourne via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
> >>>>>>> 
> >>>>>>> On Thu, Sep 03, 2015 at 01:10:42AM +0000, Eric Christopher wrote:
> >>>>>>>> On Tue, Sep 1, 2015 at 10:43 AM Duncan P. N. Exon Smith <
> >>>>>>>> dexonsmith at apple.com <mailto:dexonsmith at apple.com>> wrote:
> >>>>>>>> 
> >>>>>>>>> 
> >>>>>>>>>> On 2015-Aug-31, at 18:09, Eric Christopher <echristo at gmail.com <mailto:echristo at gmail.com>> wrote:
> >>>>>>>>>> 
> >>>>>>>>>> 
> >>>>>>>>>> 
> >>>>>>>>>> On Mon, Aug 31, 2015 at 5:50 PM Duncan P. N. Exon Smith <
> >>>>>>>>> dexonsmith at apple.com <mailto:dexonsmith at apple.com>> wrote:
> >>>>>>>>>> 
> >>>>>>>>>>> On 2015-Aug-31, at 12:21, Eric Christopher <echristo at gmail.com <mailto:echristo at gmail.com>> wrote:
> >>>>>>>>>>> Yep. This is where I was going :)
> >>>>>>>>>> 
> >>>>>>>>>> Glad I found consensus, but I want to double-check that this makes
> >>>>>>>>>> sense to add to the driver.  I didn't quite think through the
> >>>>>>>>>> implications myself.
> >>>>>>>>>> 
> >>>>>>>>>> Since the driver doesn't know if there's any bitcode, or if LTO is
> >>>>>>>>>> going to be invoked, it seems like I'll have to change the noasserts
> >>>>>>>>>> driver to *always* pass the option to the linker just in case we are
> >>>>>>>>>> doing LTO.  Is this reasonable?
> >>>>>>>>>> 
> >>>>>>>>>> Also, I realized that passing `-mllvm -disable-llvm-verifier` to ld64
> >>>>>>>>>> is redundant... so I'm thinking `-mllvm -disable-verify`.  Make
> >>>>>>>>>> sense?
> >>>>>>>>>> 
> >>>>>>>>>> *sigh* Reasons to hate the driver interface again...
> >>>>>>>>>> 
> >>>>>>>>>> I guess this is ok. Could possibly add it to the existing terrible
> >>>>>>>>> "enable this pass" interface on liblto as well.
> >>>>>>>>> 
> >>>>>>>>> The linker doesn't know whether clang was built with asserts, though.
> >>>>>>>>> 
> >>>>>>>>> We could just make it implicit: move the decision to libLTO itself.  Given
> >>>>>>>>> that clang and libLTO.dylib are different executables anyway -- and you
> >>>>>>>>> might be interposing an asserts libLTO.dylib to use with an installed clang
> >>>>>>>>> -- maybe it's even better?
> >>>>>>>>> 
> >>>>>>>>> 
> >>>>>>>> *nod* We could do that. Seems better than the alternative.
> >>>>>>> 
> >>>>>>> +1
> >>>>>>> 
> >>>>>>>>>> I don't suppose ld64 could move to a model like we're talking about with
> >>>>>>>>> lld that pcc is working on?
> >>>>>>>>> 
> >>>>>>>>> What specifically?
> >>>>>>>> 
> >>>>>>>> 
> >>>>>>>> Ah, using the C++ interface to handle everything and not using libLTO at
> >>>>>>>> all.
> >>>>>>>> 
> >>>>>>>> He can speak more to this though.
> >>>>>>> 
> >>>>>>> The C++ interface is much more convenient for a C++ program to use, but
> >>>>>>> clients need to revlock themselves to LLVM in order to use it.
> >>>>>>> 
> >>>>>>> In theory the same does not apply to libLTO, but we do end up changing it
> >>>>>>> occasionally to add new APIs which ld64 promptly starts using, so it isn't
> >>>>>>> clear how much ld64 gains by relying on libLTO,
> >>>>>> 
> >>>>>> Backward/Forward compatibility?
> >>>>>> Drop any version of clang/libLTO and still be able to use the system provided linker on any version of OS X?
> >>>>>> Sounds like a valuable feature to me, which is what I believe was (is?) sought by the C API in general (but that’s another story).
> >>>>>> 
> >>>>>> 
> >>>>>> Static linking against the parts of llvm you care about? There's nothing in the ld64 build system that means it can't do this :)
> >>>>> 
> >>>>> I’m not sure how is that supposed to work?
> >>>>> I drop a new clang/libLTO on the system, clang generates bitcode, and libLTO handles it.
> >>>>> 
> >>>>> If ld64 was statically linked against LLVM, it couldn’t read the new bitcode right?
> >>>>> 
> >>>>> 
> >>>>> Well sure, that's what Peter meant when he said revlocked. :)
> >>>> 
> >>>> I’m not sure we’re talking about the same thing, I may have misunderstood his point (“In theory the same does not apply to libLTO, but we do end up changing it
> >>>> occasionally to add new APIs which ld64 promptly starts using, so it isn’t clear how much ld64 gains by relying on libLTO”) ; my take from that was that since the new ld64 starts using the new libLTO interface, there is a revlock anyway and then there is no advantage in keeping the old libLTO entry points.
> >>>> I just described a use case that shows how useful it can be.
> >>>> 
> >>>> Apologize if I misunderstood the point.
> >>> 
> >>> That seems like a potentially valid use case, but if it doesn't actually
> >>> work right now it doesn't really buy you much.
> >> 
> >> It is working right now AFAIK. At least my understand is that the current situation is that you can drop a clang/libLTO built today on a machine that has ld64 from a few years ago and it will work.
> 
> BTW, a brand new ld64 will work with an old libLTO.dylib, too.  I'm
> not aware of any revlock at all.

lto_codegen_set_module? I imagine that a new ld64 wouldn't work with an old
libLTO that doesn't define that symbol.

> While it's fairly easy for external people (like you) to compile a
> new version of clang/libLTO.dylib, without the ld64 sources it would
> be hard to compile a new version of the linker.  At the least, ld64's
> use of the stable LTO C API enables compiler hackers to hack away.

I've been able to successfully build the ld64 sources from opensource.apple.com
against an open source LLVM build. While it could definitely be made easier,
it is possible.

> We take advantage of this internally, since it gives us freedom to
> stage ld64 and clang releases independently.
>
> (Once we bring up lld, I think the story can be quite different.)

If you may be willing to give up this ability if you adopt lld:
1) I think that says something about how important it is
2) Is there any harm in tying the linker to clang now as opposed to when you
   adopt lld? Because ld64 is admittedly a legacy thing we should avoid
   jumping through hoops for it if possible. Surely you have some way of
   staging internal projects that use unstable LLVM APIs (swift?)

Thanks,
-- 
Peter