similar to: Issues with using scalar evolution with newer versions of LLVM IR

Thank You.. I used following command to generate .bc or .ll /Documents/clang+llvm-4.0.0-x86_64-linux-gnu-ubuntu-16.04/bin/clang -O0 -emit-llvm -S -o vec4.ll vecsum.c /Documents/clang+llvm-7.0.0-x86_64-linux-gnu-ubuntu-16.04/bin/clang -O0 -emit-llvm -S -o vec7.ll vecsum.c On Wed, Jan 16, 2019 at 6:49 AM Sanjoy Das <sanjoy at playingwithpointers.com> wrote: > It is hard to tell

Hello. I tried to get the trip count of a loop with Scalar evolution. I got inspired from http://stackoverflow.com/questions/13834364/how-to-get-loop-bounds-in-llvm . However the analysis described there doesn't work well for the second inner loop of thes function below (although if we declare Bcols a short it works well): void MatMul(int Arows, int Acols, int Brows, int

The Problem? Nowaday, SCEV called "Scalar Evolution" does only evolate instructions that has predictable operand, Constant-Based operand. such as that can evolute as a constant. otherwise we couldn't evolate it as SCEV node, evolated as SCEVUnknown. important thing that we remember is, we do not use SCEV only for Loop Deletion, which that doesn't really needed on nature loops

Hi, Scalar evolution seems to be wrapping around the trip count in the following loop. void add (int *restrict a, int *restrict b, int *restrict c) { char i; for (i = 0; i < 255; i++) a[i] = b[i] + c[i]; } When I run scalar evolution on the bit code, I get a backedge-taken count which is obviously wrong. $> cat loop.ll ; Function Attrs: nounwind define void @add(i32* noalias

I've been doing some digging in this area (scev, wrapping arithmetic), learning as much as I can, and have reached a point where I'm fairly confused about the semantics of nuw in scalar evolution expressions. Consider the following program: define void @foo(i32 %begin) { entry: br label %loop loop: %idx = phi i32 [ %begin, %entry ], [ %idx.dec, %loop ] %idx.dec = sub nuw i32

Thank You. i used -polly-ast-detect-parallel but there is no coincident info generated; my c code is simple vec-sum as follows; #include <stdio.h> int a[2048], b[2048], c[2048]; foo () { int i; for (i=0; i<2048; i++) { a[i]=b[5] + c[i]; } } i executed following commands; $clang -S -emit-llvm vec-sum.cpp -march=native -O3 -mllvm -disable-llvm-optzns -o vec-sum.s $opt -S

Hi, I'm trying to get the following loop to vectorize (simple reduction): unsigned int sum2(unsigned int *a, int len){ unsigned int s = 0; for (int i = 0; i < len; i += 4) s += *a++; return s; } The loop fails to vectorize because SCEV could not compute the loop exit count. It appears SCEV cannot handle the non-unit increment of the loop counter. Is this a known limitation of

Hi All, In the attached test case there, is an unnested loop with 2 iterations. The loop latch block is terminated by an unconditional branch, so simplifycfg folds the almost empty latch block into its predecessor which is the loop header. This results in an additional backedge in the CFG, so when LoopRotate pass is called it canonicalizes the loop into a nested loop. However, now the loop

On 6/30/2017 7:48 AM, Balaram Makam via llvm-dev wrote: > > Edit. Predecessor -> successor. > > *From:* llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org] *On Behalf > Of *Balaram Makam via llvm-dev > *Sent:* Friday, June 30, 2017 10:47 AM > *To:* llvm-dev at lists.llvm.org > *Subject:* [llvm-dev] LoopSimplify pass prevents loop unrolling > > Hi All, > >

Hello, If I run clang on the following code: void func(unsigned n) { > for (unsigned long x = 1; x < n; ++x) > dummy(x); > } I get the following llvm ir: define void @func(i32 %n) { > entry: > %conv = zext i32 %n to i64 > %cmp5 = icmp ugt i32 %n, 1 > br i1 %cmp5, label %for.body, label %for.cond.cleanup > for.cond.cleanup:

Is that why we do not deduce +<nsw> from "add nsw" either? Is that an intrinsic limitation of creating a context-invariant expressions from a Value* or is that a limitation of our implementation (our unification not considering the nsw flags)? On Wed, Aug 15, 2018 at 12:39 PM Friedman, Eli <efriedma at codeaurora.org> wrote: > On 8/15/2018 12:21 PM, Alexandre Isoard via

I'm not sure I understand the poison/undef/UB distinctions. But on this example: define i32 @func(i1 zeroext %b, i32 %x, i32 %y) { > entry: > %adds = add nsw i32 %x, %y > %addu = add nuw i32 %x, %y > %cond = select i1 %b, i32 %adds, i32 %addu > ret i32 %cond > } It is important to not propagate the nsw/nuw between the two SCEV expressions (which unification would

Ok. To go back to the original issue, would it be meaningful to add a SCEVUMax(0, BTC) on the final BTC computed by SCEV? So that it does not use "negative values"? On Wed, Aug 15, 2018 at 2:40 PM Friedman, Eli <efriedma at codeaurora.org> wrote: > On 8/15/2018 2:27 PM, Alexandre Isoard wrote: > > I'm not sure I understand the poison/undef/UB distinctions. >

Hi Ayal, Let me start with commenting on this: > A dedicated intrinsic that freezes the compare instruction, for no apparent reason, may potentially cripple subsequent passes from further optimizing the vectorized loop. The point is we have a very good reason, which is that it passes on the right information on the backend, enabling opimisations as opposed to crippling them. The compare

> The compare of interest is clear, I think. It compares a Vector Induction Variable with a broadcasted loop invariant value, aka the BTC. Obtaining the latter operand is the goal, clearly, but to do so, the former operand needs to be recognized as a VIV. Yep, exactly that. > What if this compare is not generated by LV’s fold-tail-by-masking transformation? Not sure I completely follow

Hi all, It looks like the induction variable simplification pass prefers doing a zero-extension to compute the wider trip count of loops when extending the IV. This can sometimes result in loss of information making ScalarEvolution's analysis conservative which can lead to missed performance opportunities. For example, consider this loopnest- int i, j; for(i=0; i< 40; i++) for(j=0;

Hi Gerolf, I think we have several (perhaps separable) issues here: 1. Do we have a canonical form for loops, preserved through the optimizer, that allows naturally-constructed loop nests to remain separable? 2. Do we forbid non-lowering transformations that turn vectorizable loops into non-vectorizable loops? 3. How do we detect cases where transformations cause a negative answer to either

Hi, I am looking at two very simple kernels. They implement the following loops: constant_bound(): for (int i = 0; i < 100; i+=4); parameteric_bound(): for (int i = 0; i < n; i+=4); For the first loop SCEV is able to derive the number of loop iterations, for the second loop it returns 'Unpredictable backedge-taken count'. Is this expected because it is a difficult problem or

Hello, I want to study the individual O3 optimizations. For this I am using following commands, but unable to replicate O3 behavior. 1. Documents/clang+llvm-9.0.0-x86_64-linux-gnu-ubuntu-18.04/bin/clang -O1 -Xclang -disable-llvm-passes -emit-llvm -S vecsum.c -o vecsum-noopt.ll 2. Documents/clang+llvm-9.0.0-x86_64-linux-gnu-ubuntu-18.04/bin/clang -O3 -mllvm -debug-pass=Arguments -emit-llvm -S

Hello, What is the fastest way to do this? I has to be done quite a few times. Basically I have sets of 3 numbers (as characters) and sets of 3 dashes and I have to store them in named columns. The order of the sets and the column name they fall under is important. The actual numbers and the pattern/order of the sets should be considered random/unpredictable. Sample data: vec =