I don't see an ideal solution to these 2 related, potentially large, perf regressions:
https://bugs.llvm.org/show_bug.cgi?id=42708
https://bugs.llvm.org/show_bug.cgi?id=43146
We decided that load combining was unsuitable for IR because it could obscure other optimizations in IR. So we removed the LoadCombiner pass and deferred to the backend. Therefore, preventing SLP from destroying load combine opportunities requires that it recognizes patterns that could be combined later, but not do the optimization itself (it's not a vector combine anyway, so it's probably out-of-scope for SLP).
In the x86 tests shown (and discussed in more detail in the bug reports), SDAG combining will produce a single instruction on these tests like:
movbe rax, qword ptr [rdi]
or:
mov rax, qword ptr [rdi]
Not some (half) vector monstrosity as we currently do using SLP:
vpmovzxbq ymm0, dword ptr [rdi + 1] # ymm0 = mem[0],zero,zero,zero,zero,zero,zero,zero,mem[1],zero,zero,zero,zero,zero,zero,zero,mem[2],zero,zero,zero,zero,zero,zero,zero,mem[3],zero,zero,zero,zero,zero,zero,zero vpsllvq ymm0, ymm0, ymmword ptr [rip + .LCPI0_0] movzx eax, byte ptr [rdi] movzx ecx, byte ptr [rdi + 5] shl rcx, 40 movzx edx, byte ptr [rdi + 6] shl rdx, 48 or rdx, rcx movzx ecx, byte ptr [rdi + 7] shl rcx, 56 or rcx, rdx or rcx, rax vextracti128 xmm1, ymm0, 1 vpor xmm0, xmm0, xmm1 vpshufd xmm1, xmm0, 78 # xmm1 = xmm0[2,3,0,1] vpor xmm0, xmm0, xmm1 vmovq rax, xmm0 or rax, rcx vzeroupper ret
Ooops, sorry, my fault, pointed to the wrong function. Better to modify a function in include/llvm/CodeGen/BasicTTIImpl.h, BasicTTIImplBase::getArithmeticInstrCost, if this is common limitation.