[X86] X86FixupVectorConstants generates SSE2 instructions on SSE1 targets

[RKSimon]

As detailed here: #134547

define void @store_v2f32_constant(ptr %v) {
  store <2 x float> <float 2.560000e+02, float 5.120000e+02>, ptr %v, align 4
  ret void
}

X86FixupVectorConstants will attempt to load the v2f32 using a MOVSD load, which is only available on SSE2+ targets

[llvmbot]

@llvm/issue-subscribers-backend-x86

Author: Simon Pilgrim (RKSimon)

As detailed here: #134547

define void @<!-- -->store_v2f32_constant(ptr %v) {
  store &lt;2 x float&gt; &lt;float 2.560000e+02, float 5.120000e+02&gt;, ptr %v, align 4
  ret void
}

X86FixupVectorConstants will attempt to load the v2f32 using a MOVSD load, which is only available on SSE2+ targets

[RKSimon]

/cherry-pick 08e080e

[llvmbot]

/pull-request #135191