[llvm] [AMDGPU][GISel] Add RegBankLegalize support for G_STRICT_{FADD|FSUB|FMUL} (PR #169406)

[Petar Avramovic via llvm-commits]

================
@@ -953,7 +962,9 @@ RegBankLegalizeRules::RegBankLegalizeRules(const GCNSubtarget &_ST,
            hasSALUFloat)
       .Div(V2S16, {{VgprV2S16}, {VgprV2S16, VgprV2S16}})
       .Any({{UniV2S32}, {{UniInVgprV2S32}, {VgprV2S32, VgprV2S32}}})
-      .Any({{DivV2S32}, {{VgprV2S32}, {VgprV2S32, VgprV2S32}}});
+      .Any({{DivV2S32}, {{VgprV2S32}, {VgprV2S32, VgprV2S32}}})
+      .Any({{UniV2S64}, {{UniInVgprV2S64}, {VgprV2S64, VgprV2S64}}})
+      .Any({{DivV2S64}, {{VgprV2S64}, {VgprV2S64, VgprV2S64}}});
----------------
petar-avramovic wrote:

Missing test for v2s64, afaik there is no pattern for them.
Missing test for v2s32 think this is for GFX942 and similar targets
missing test for hasSALUFloat for example gfx12

https://github.com/llvm/llvm-project/pull/169406