[llvm-dev] Shift-by-signext - sext is bad for analysis - ignore it's use count?

Roman Lebedev via llvm-dev llvm-dev at lists.llvm.org
Mon Oct 7 10:45:54 PDT 2019 

On Mon, Oct 7, 2019 at 11:32 AM Roman Lebedev <lebedev.ri at gmail.com> wrote:
>
> Bump. Any further thoughts here?
>
> To recap - i don't really see how this can be a demandedbits problem - we do
> demand all those bits, we just know they must be zero.
> (i would love to be proven wrong though!)
>
> Roman.
>
> On Tue, Oct 1, 2019 at 11:17 PM Roman Lebedev <lebedev.ri at gmail.com> wrote:
> >
> > The thing is, we *don't* "not demand" those high bits.
> > We *don't* not care what's in those bits - IR shifts don't mask their
> > shift amounts.
> > I.e we can't replace `x >> (32-y)` with `x >> (-y)`,
> > which would be legal transform should we not demand those bits.
> > We very much demand them. We just know those bits to be zero.
> >
> > And i'm not sure how to convey that to SimplifyDemandedBits().
> > I can't pass the Known down the stack, the function resets it first thing,
> > so Known can only be passed from callee to the caller.
> >
> > This is why i'm asking whether anyone is concerned if we proceed with
> > https://reviews.llvm.org/D68150
> >
> > On Tue, Oct 1, 2019 at 10:44 PM Nikita Popov <nikita.ppv at gmail.com> wrote:
> > >
> > > On Tue, Oct 1, 2019 at 8:57 PM Roman Lebedev via llvm-dev <llvm-dev at lists.llvm.org> wrote:
> > >>
> > >> Thanks for taking a look!
> > >>
> > >> On Tue, Oct 1, 2019 at 9:09 PM Philip Reames <listmail at philipreames.com> wrote:
> > >> > On 9/27/19 1:40 PM, Roman Lebedev via llvm-dev wrote:
> > >> > > In https://reviews.llvm.org/D68103 the InstCombine learned that shift-by-sext
> > >> > > is simply a shift-by-zext.
> > >> >
> > >> > Just to make sure I'm following, the reasoning here is that the shift
> > >> > amount must be positive or the shift would produce poison? And thus,
> > >> > it's safe to assume that the sext == zext because we've (at worst)
> > >> > removed UB in the original program?
> > >> Yes.
> > >> zext and sext are equivalent for non-negative inputs.
> > >> For negative inputs, sext will produce negative output.
> > >> And interpreted as unsigned number, such negative shift amount
> > >> is *always* bigger than largest legal shift amount:
> > >> * i1 can only be shifted by 0
> > >> * i2 can only be shifted by 1 (positive); 2,3 (negative) are not legal
> > >> shift amounts
> > >> * i3 can only be shifted by 1,2 (positive); 3 (positive) and 4-7
> > >> (negative) are not legal shift amounts
> > >> * i4 can only be shifted by 1,2,3 (positive); 4-7 (positive) and 8-15
> > >> (negative) are not legal shift amounts
> > >> * and so on.
> > >> Therefore we must not have negative shift amount (or we have UB), so
> > >> we can just use zext.
> > >>
> > >> > If so, two slightly off topic ideas.
> > >> >
> > >> > 1) This feels like a demanded bits problem.  We know that any shift
> > >> > value outside of a given range is UB, and thus only need to demand the
> > >> > bits necessary to represent the defined range.  Might be an interesting
> > >> > extension.
> > >> Most specifically, we demand only low
> > >> bitwidth(shift)-ctlz(bitwidth(shift)-1) bits
> > >> of shift amount, yes. I did think of it as a demanded bits problem,
> > >> but didn't really succeed, let me see again..
> > >
> > >
> > > While the demanded bits analysis in InstCombine doesn't handle multi-use scenarios (apart from cases where a specific use may be reduced to an instruction operand), it should be possible to handle this as part of BDCE. This would require a) computing demanded bits for the shift amount in the DemandedBits analysis, which currently doesn't happen, and b) replacing a sext with a zext in BDCE if the top bit of the source operand is not demanded.
> > >
> > > Nikita
> > >

> > >> > 2) Are we possibly missing opportunities by not exploiting knowledge of
> > >> > the a known negative shift amount?
> > >> Anything specific you are thinking of?
Okay, i can now tell that the issue is *somewhat* larger, and we do
certainly miss some opportunities:
https://godbolt.org/z/6ucL9u

int shl0(int x, signed char y) {
  return x << (32 - y);
}
Produces
  %o0 = sext i8 %6 to i32
  %o1 = sub i32 32, %o0
  %r = shl i32 %5, %o1

While
int shl1(int x, unsigned char y) {
  return x << (32 - y);
}
produces
  %n0 = zext i8 %6 to i32
  %n1 = sub i32 32, %n0
  %r = shl i32 %5, %n1
(as one would expect)

And here we again don't need sign-extension:
https://rise4fun.com/Alive/ymk

In my eyes, that doesn't completely invalidate
https://reviews.llvm.org/D68150 approach,
i still think it is a good idea, but clearly we *also* need
something else, more generic.

My first thought would be CVP and LVI, but i did not look yet..

Roman

> > >> > > But the transform is limited to single-use sext.
> > >> > > We can quite trivially get a case where there are two shifts by the same sext:
> > >> > > https://godbolt.org/z/j6mO3t  <- We should handle those cases.
> > >> > >
> > >> > > In https://reviews.llvm.org/D68103#1686130 Sanjay Patel notes that this
> > >> > > sext is intrusive for analysis, that we will gain far better analysis with zext,
> > >> > > so we should just ignore forego of the one-use check,
> > >> > > and simply replace all shift-by-sext with shift-by-zext.
> > >> >
> > >> > Doing the multi-use case is unfortunately complicated.  Your limited use
> > >> > scan might be a reasonable option in practice, but the need for cutoffs
> > >> > creates undesirable dynamics.
> > >> To reiterate, the proposal is to just always transform that sext into zext,
> > >> with no care to the use count. So i'm not sure what cutoffs you mean.
> > >>
> > >> > A couple ideas on how to possibly approach the problem:
> > >> >
> > >> > 1) If we can prove that one shift dominates the other uses, then if we
> > >> > can find UB which triggers based on overflow, we can do the replacement.
> > >> I'm not sure what all this means.
> > >> Here i only want to get rid of *all* of those sext's for good, so everything
> > >> else that may have to deal with looking past extension of shift amount
> > >> doesn't need to be modified.
> > >>
> > >> > 2) Having a general multiple use demanded use routine would be very
> > >> > powerful.  Is it worth exploring the harder topic for generality?
> > >> There is some support for multi-use demandedbits in backend.
> > >> As for middle-end i'm not sure.
> > >> For sure, having that powerful mechanism may be useful, in general.
> > >>
> > >> > 3) If we had an anyextend IR node, it might be reasonable to eagerly
> > >> > produce the duplicate nodes, and rely on later CSE.  I keep running
> > >> > across cases where we have an extend where we know the high bits don't
> > >> > matter, maybe it's time to represent that?
> > >> >
> > >> > > I implemented this proposed suggestion here:
> > >> > > https://reviews.llvm.org/D68150
> > >> > >
> > >> > > Does anyone see any problems with that trade-off?
> > >> > >
> > >> Roman.
> > >>
> > >> > > _______________________________________________
> > >> > > LLVM Developers mailing list
> > >> > > llvm-dev at lists.llvm.org
> > >> > > https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
> > >> _______________________________________________
> > >> LLVM Developers mailing list
> > >> llvm-dev at lists.llvm.org
> > >> https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev

More information about the llvm-dev mailing list