similar to: Why does FPBinOp(X, undef) -> NaN?

On Fri, Feb 7, 2020 at 12:29 PM Nuno Lopes <nunoplopes at sapo.pt> wrote: > > It's not correct (output of Alive2): > > define half @fn(half %a) { > %b = fadd half %a, undef > ret half %b > } > => > define half @fn(half %a) { > ret half undef > } > Transformation doesn't verify! > ERROR: Value mismatch > > Example: > half %a

%y = fadd float %x, undef Can we simplify this? Currently in IR, we do nothing for fadd/fsub/fmul. For fdiv/frem, we propagate undef. The code comment for fdiv/frem says: "the undef could be a snan" If that's correct, then shouldn't it be the same for fadd/fsub/fmul? But this can't be correct because we support targets that don't raise exceptions...and even targets

I’m not sure the transformation happening with fdiv is correct. If we have “%y = fdiv float %x, undef” and %x is a NaN then the result will be NaN for any value of the undef, right? So if I understand the undef rules correctly (never a certainty) then we can’t safely replace the expression with undef. We could, I think, replace it with “%y = %x” though. I think the same is true for fadd, fsub,

Hello All, the patch below adds a target lowering hook to state that the target supports (or not) floating point exception behavior. The patch is small and contains one possible use for the hook (folding potentially exception raising fp operations). Any comments? Thanks Pedro -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: opensource1.txt

Yes, if %x is a NaN, we should expect that NaN is propagated. I'm still not sure what to do here. We can take comfort in knowing that whatever we do is likely an improvement over the current situation though. :) That's because the code in InstSimplify is inconsistent with the LangRef: http://llvm.org/docs/LangRef.html#undefined-values (UB for fdiv by 0?) ...and both of those are

Why is NaN “just ‘undef’ in IR”? NaN is a specific value with well-defined behavior. I would think that unless the no-NaNs flag is used we need to preserve the behavior of NaNs. From: Sanjay Patel [mailto:spatel at rotateright.com] Sent: Wednesday, February 28, 2018 12:08 PM To: Kaylor, Andrew <andrew.kaylor at intel.com> Cc: llvm-dev <llvm-dev at lists.llvm.org>; Nuno Lopes

I assume you meant to ask for ports that *don’t* support floating point exceptions. To my knowledge, neither R600 nor NVPTX support floating point exceptions. —Owen > On Aug 8, 2014, at 2:41 PM, Eric Christopher <echristo at gmail.com> wrote: > > There's a lot of cut and paste in those routines. Can you do something > to unify it a bit? Also, do we have any ports that

Correct - NaN is not undef in IR. But we don't have a NaN in this example. We have its moral equivalent in LLVM - an uninitialized value, undef. So we're not introducing any extra uncertainty by propagating the undef. The backend can choose whatever encoding of undef makes sense when lowering? And yes, I don't know why FP-div-by-zero would ever be UB. I think that text in the LangRef

What I’m saying is that if we have one operand that is not an undef value then that operand might be NaN and if it is then the result must be NaN. So while it may be true that we don’t have a NaN, it is not true that we definitely do not have a NaN in the example. This is analogous to the example in the language reference where it says “%A = or %X, undef” -> “%A = undef” is unsafe because any

Ah, thanks for explaining. So given that any of these ops will return NaN with a NaN operand, let's choose the undef operand value to be NaN. That means we can fold all of these to a NaN constant in the general case. But if we have 'nnan' FMF, then we can fold harder to undef? nnan - Allow optimizations to assume the arguments and result are not NaN. Such optimizations are required to

I'm pretty sure that isn't what nnan is supposed to mean. If the result of nnan math were undefined in the sense of "undef", programs using nnan could have undefined behavior if the result is used in certain ways which would not be undefined for any actual float value (e.g. converting the result to a string), which seems like a surprising result.  And I don't think we

On Mar 1, 2018, at 10:07 AM, Kaylor, Andrew <andrew.kaylor at intel.com> wrote: > So you don’t think sNaNs can just be treated as if they were qNaNs? I understand why we would want to ignore the signaling part of things, but the rules for operating on NaNs are pretty clear and reasonable to implement. The signaling aspect can, I think, be safely ignored when we are in the mode of assuming

Per subject, this patch adding an additional pass to handle vector operations; the idea is that this allows removing the code from LegalizeDAG that handles illegal types, which should be a significant simplification. There are still some issues with this patch, but does the approach look sane? -Eli -------------- next part -------------- Index: lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp

We can do "add %x, undef" => "undef" because for any value of %x, we can always find a value that when added to %x produces any value in the domain of integers. This is not the case with floats since with some inputs, e.g., NaNs, we are not able to produce some values in the domain (e.g., there's no value of %x that makes "fadd NaN, %x" return 42.0). In

So you don’t think sNaNs can just be treated as if they were qNaNs? I understand why we would want to ignore the signaling part of things, but the rules for operating on NaNs are pretty clear and reasonable to implement. The signaling aspect can, I think, be safely ignored when we are in the mode of assuming the default FP environment. As for the distinction between IEEE and LLVM IR, I would

For the first part of Sanjay’s question, I think the answer is, “Yes, we can fold all of these to NaN in the general case.” For the second part, which the nnan FMF is present, I’m not sure. The particulars of the semantics of nnan are unclear to me. But let me explore what Eli is saying. It sounds reasonable, but I have a question about it. Suppose we have the nnan FMF set, and we encounter

On Thu, Mar 1, 2018 at 2:08 AM, Nuno Lopes <nunoplopes at sapo.pt> wrote: > We can do "add %x, undef" => "undef" because for any value of %x, we can > always find a value that when added to %x produces any value in the domain > of integers. > > This is not the case with floats since with some inputs, e.g., NaNs, we > are not able to produce some

There's a question about the behavior of funnel shift [1] + select and poison here that reminds me of previous discussions about select and poison [2]: https://github.com/AliveToolkit/alive2/pull/32#discussion_r257528880 Example: define i8 @fshl_zero_shift_guard(i8 %x, i8 %y, i8 %sh) { %c = icmp eq i8 %sh, 0 %f = fshl i8 %x, i8 %y, i8 %sh %s = select i1 %c, i8 %x, i8 %f ; shift amount is 0

Hi Sanjay, Thanks, I saw this flag and it's definitely should be considered, but it appeared to me to be static characteristic of target platform. I'm not sure how appropriate it would be to change its value from a front-end. It says "Has", while optional flag would rather say "Uses" meaning that implementation cares about floating point exceptions. Regards, Sergey

On Wed, May 20, 2009 at 4:55 PM, Dan Gohman <gohman at apple.com> wrote: > Can you explain why you chose the approach of using a new pass? > I pictured removing LegalizeDAG's type legalization code would > mostly consist of finding all the places that use TLI.getTypeAction > and just deleting code for handling its Expand and Promote. Are you > anticipating something more