search for: movddup

...me C++ code that looks like this: // MatchVectorShuffle - Matches a shuffle node against the available instructions, // returning the lowest cost one as well as the actual cost of it. unsigned MatchVectorShuffle(VectorShuffleSDNode *N) { unsigned LowestCost = ~0; if (N can be matched by movddup) { unsigned movddupcost = ... // can be either constant, or callback into subtarget info if (LowestCost > movddupcost) LowestCost = movddupcost; operands = [whatever] opcode = X86::MOVDDUP; } } if (N can be matched by movhlps) { unsigned movhlps...

On Friday 01 May 2009 13:46, Chris Lattner wrote: > Right, a lot of these problems can be solved by some nice refactoring > stuff. I'm also hoping that some of the complexity in defining > shuffle matching code can be helped by making the definition of the > shuffle patterns more declarative within the td file. It would be > really nice to say that "this shuffle does a

On May 1, 2009, at 3:50 PM, Chris Lattner wrote: > > The goal is to replace the pattern fragment and the C++ code for > X86::isMOVDDUPMask with something like: > > def movddup : PatFrag<(ops node:$lhs, node:$rhs), > (vector_shuffle node:$lhs, node:$rhs, > 0, 1, 0, 1, Cost<42>) > > Alternatively, the cost could be put on the instructions etc, wha...

...ign 8 _DA: .quad 4599075939470750515 # double 3.000000e-01 .comm _Y,800,3 # @Y .comm _X,800,3 # @X gcc -S -O3 -o test2.s test.c -march=native result: .file "test.c" .text .p2align 4,,15 .globl _f .def _f; .scl 2; .type 32; .endef _f: pushl %ebp movddup _DA, %xmm2 movl %esp, %ebp xorl %eax, %eax .p2align 4,,10 L2: movapd _Y(%eax), %xmm0 movapd _X(%eax), %xmm1 mulpd %xmm2, %xmm1 subpd %xmm1, %xmm0 movapd %xmm0, _Y(%eax) addl $16, %eax cmpl $800, %eax jne L2 xorw %ax, %ax leave ret .globl _DA .data .align 16 _DA: .long 858993459 .l...

On Tue, Sep 9, 2014 at 11:39 PM, Chandler Carruth <chandlerc at google.com> wrote: > Awesome, thanks for all the information! > > See below: > > On Tue, Sep 9, 2014 at 6:13 AM, Andrea Di Biagio <andrea.dibiagio at gmail.com> > wrote: >> >> You have already mentioned how the new shuffle lowering is missing >> some features; for example, you explicitly

...------ next part -------------- 002E00D0 push ebp 002E00D1 mov ebp,esp 002E00D3 push ebx 002E00D4 push edi 002E00D5 push esi 002E00D6 and esp,0FFFFFFF0h 002E00DC sub esp,110h 002E00E2 mov eax,dword ptr [ebp+8] 002E00E5 movddup xmm0,mmword ptr [eax+10h] 002E00EA movapd xmmword ptr [esp+80h],xmm0 002E00F3 movddup xmm0,mmword ptr [eax+8] 002E00F8 movapd xmmword ptr [esp+70h],xmm0 002E00FE movddup xmm0,mmword ptr [eax] 002E0102 movapd xmmword ptr [esp+60h],xmm0 002E0108 xorpd xmm0...

Hmm, I'm not able to get those .ll files to compile if I disable SSE and I end up with SSE instructions(including sqrtpd) if I don't disable it. On Thu, Jul 18, 2013 at 10:53 PM, Peter Newman <peter at uformia.com> wrote: > Is there something specifically required to enable SSE? If it's not > detected as available (based from the target triple?) then I don't think

...uild/clang/debug/bin/./FileCheck /usr/local/google/home/blaikie/dev/llvm/src/test/CodeGen/X86/vec_splat.ll -check-prefix=SSE3 -- Exit Code: 1 Command Output (stderr): -- /usr/local/google/home/blaikie/dev/llvm/src/test/CodeGen/X86/vec_splat.ll:33:9: error: expected string not found in input ; SSE3: movddup ^ <stdin>:21:12: note: scanning from here test_v2sd: # @test_v2sd ^ <stdin>:23:2: note: possible intended match here movsd 12(%esp), %xmm0 ^ -- ******************** FAIL: LLVM :: CodeGen/X86/sibcall.ll (48 of 51) ******************** TEST 'LLVM :: CodeGen/X86/s...