similar to: VSHUFFLE Splitting

Hello, i am experimenting with the increase in register/ vector width to 64 elements of 32 bits instead of 16 in x86 backend. for eg. i have a loop with 65 iterations; if my IR generates v64i32 and 1 scalar, still the backend breaks the v64i32 into 4 v16i32. i want it to retain v64i32. like if there are 128 elements in loop then it should break it into 2 v64i32 instructions. in order to do this i

Thank You. On Fri, Jul 7, 2017 at 10:03 AM, Craig Topper <craig.topper at gmail.com> wrote: > Yes, that error is from instruction selection. I think your legalization > changes worked fine. > > ~Craig > > On Thu, Jul 6, 2017 at 8:21 PM, hameeza ahmed via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > >> also i further run the following command;

also i further run the following command; llc -debug filer-knl_o3.ll and its output is attached here. by looking at the output can we say that legalization runs fine and the error is due to instruction selection/ pattern matching which is not yet implemented? so do i need to worry and try to correct it at this stage or should i move forward to implement instruction selection/ pattern matching?

Thank you; i have changed as follows.is it fine now? def VADD_256B : I<0xFE, MRMDestReg, (outs VR2048:$dst), (ins VR2048:$src1, VR2048:$src2), "VADD_256B\t{$src, $dst|$dst, $src}", [(set VR2048:$dst, (add VR2048:$src1, VR2048:$src2))]]>; Also here i have changed class RI to I. Does it make any difference? On Sat, Jul 8, 2017 at 9:38 AM, Craig Topper

Thank you. i understood how avx512 vector instructions are written in x86instravx512. i need to define my vector instructions so i wrote; def VMOV_256B_RM : I<0x6F, MRMSrcMem, (outs VR2048:$dst), (ins i32mem:$src), "vmov_256B_rm\t{$src, $dst|$dst, $src}", [(set VR2048:$dst, (v64i32 (scalar_to_vector (loadi32 addr:$src))))],

Have you read http://llvm.org/docs/WritingAnLLVMBackend.html and http://llvm.org/docs/CodeGenerator.html ? http://llvm.org/docs/WritingAnLLVMBackend.html#instruction-selector describes how to define a store instruction. -Eli On 7/6/2017 6:51 PM, hameeza ahmed via llvm-dev wrote: > Please correct me i m stuck at this point. > > On Jul 6, 2017 5:18 PM, "hameeza ahmed"

Thank You. I have seen the opcode is 8 bits and all the combinations are already used in llvm x86. Now what to do? On Sat, Jul 8, 2017 at 10:57 AM, Craig Topper <craig.topper at gmail.com> wrote: > Yes its an opcode conflict. You'll have to look through Intel documents > and find an unused opcode. I've only added instructions based on a real > spec so I don't know

Hello, I did as you said, Please tell me whether the following correct now?? def GATHER_256B : I<0x68, MRMSrcMem, (outs VR_2048:$dst, _.KRCWM:$mask_wb), (VR_2048:$src1, _.KRCWM:$mask, ins i2048mem:$src2), "GATHER_256B\t{$src2, {$dst}{${mask}}|${dst} {${mask}}, $src2}"), [(set VR_2048:$dst, _.KRCWM:$mask_wb, (v64i32 (GatherNode

i want to implement gather for v64i32. i wrote following code. def GATHER_256B : I<0x68, MRMSrcMem, (outs VR_2048:$dst), (ins i2048mem:$src), "GATHER_256B\t{$src, $dst|$dst, $src}", [(set VR_2048:$dst, (v64i32 (masked_gather addr:$src)))], IIC_MOV_MEM>, TA; def: Pat<(v64f32 (masked_gather addr:$src)), (GATHER_256B

Thank You. Still getting errors.I have modified my instructions as you said as follows: def GATHER_256B : I<0x68, MRMSrcMem, (outs VR_2048:$dst, VK64WM:$mask_wb), (ins VR_2048:$src1, VK64WM:$mask, i2048mem:$src2), "GATHER_256B\t{$src2, {$dst} {${mask}}|${dst} {${mask}}, $src2}", [(set VR_2048:$dst, VK64WM:$mask_wb, (v64i32 (masked_gather

I would be very grateful if you specify whether there is some way to allocate registers (different order) / from different register sets to the same instruction based on the vector width/ no of iterations. I have tried several alternatives but could not succeed. Also I have asked this question many times but no one responds. Is there something wrong with this?? Kindly guide me. Thank You On

Please tell me whether the following implementation is correct..... My target supports 64 bit mask means immediate(0-2^63) I have implemented it but i dont know whether its correct or not. Please see the changes below that i have made in x86isellowering.cpp static SDValue lower2048BitVectorShuffle(const SDLoc &DL, ArrayRef<int> Mask, MVT VT,

Thank you so much. it run fine. Can you please resolve following issue; I now have support for v2048i32 but my backend supports v64i32 so ultimately v2048i32 needs to be split into 32 v64i32 instructions. the only difference between 2 is if its orginally v2048i32 i want my registers assignment from REG_A set. if its v64i32 originally, then i want registers from set REG_B. How to accomplish

Sorry to ask but what does it mean to put both? On Tue, Sep 5, 2017 at 4:01 AM, Craig Topper <craig.topper at gmail.com> wrote: > Leave TA. Put both. > > ~Craig > > On Mon, Sep 4, 2017 at 4:00 PM, hameeza ahmed <hahmed2305 at gmail.com> > wrote: > >> You are right. But when i defined my instruction as follows: >> def P_256B_VADD : I<0xE1,

You are right. But when i defined my instruction as follows: def P_256B_VADD : I<0xE1, MRMDestReg, (outs VRP_2048:$dst), (ins VRP_2048:$src1, VRPIM_2048:$src2),"P_256B_VADD\t{$src1, $src2, $dst|$dst, $src1, $src2}", [(set VRP_2048:$dst, (add (v64i32 VRP_2048:$src1), (v64i32 VRP_2048:$src2)))]>, VEX_4V; I get opcode conflicts? Then what to do? On Tue, Sep 5, 2017 at 3:51 AM,

Thank You. I used EVEX_4V with all the instructions. I replaced TA and EVEX both with EVEX_4V. Now, I am getting following error: llvm-tblgen: /utils/TableGen/X86RecognizableInstr.cpp:687: void llvm::X86Disassembler::RecognizableInstr::emitInstructionSpecifier(): Assertion `numPhysicalOperands >= 2 + additionalOperands && numPhysicalOperands <= 4 + additionalOperands &&

Thank You, I changed TA to EVEX or EVEX_4V. But now i am getting following error: Invalid prefix! UNREACHABLE executed at /lib/Target/X86/MCTargetDesc/X86MCCodeEmitter.cpp:647! On Tue, Sep 5, 2017 at 4:36 AM, Craig Topper <craig.topper at gmail.com> wrote: > Not all instructions can use EVEX_4V. Move instructions in particular > cannot because they don't have 2 sources. >

I was getting same error when i keep both EVEX/EVEX_4V and TA. So, i restored my original instructions and for that i have to include bool HasTA = TSFlags & X86II::TA; in x86MCCodeEmitter.cpp then used this condition; if(HasTA) ++SrcRegNum; in order to emit binary correctly. Is it right? On Tue, Sep 5, 2017 at 5:45 AM, Craig Topper <craig.topper at gmail.com> wrote: >

Hello, I m working towards backend. Here i need to define vector load and stores for 64 i32 elements. so in x86instrinfo.td i wrote; def VMOV_256B_RM : I<0x6F, MRMSrcMem, (outs VR2048:$dst), (ins i32mem:$src), "vmov_256B_rm\t{$src, $dst|$dst, $src}", [(set VR2048:$dst, (v64i32 (scalar_to_vector (loadi32 addr:$src))))],

Thank You. My add instruction has TA as follows: def P_256B_VADD : I<0xE1, MRMDestReg, (outs VRP_2048:$dst), (ins VRP_2048:$src1, VRPIM_2048:$src2),"P_256B_VADD\t{$src1, $src2, $dst|$dst, $src1, $src2}", [(set VRP_2048:$dst, (add (v64i32 VRP_2048:$src1), (v64i32 VRP_2048:$src2)))]>, TA; so i defined; bool HasTA = TSFlags & X86II::TA; in x86MCCodeEmitter.cpp then used