[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

Essentially the same as the GEP change in r230786.

A similar migration script can be used to update test cases, though a few more
test case improvements/changes were required this time around: (r229269-r229278)

import fileinput
import sys
import re

pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)")

for line in sys.stdin:
  sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line))

Reviewers: rafael, dexonsmith, grosser

Differential Revision: http://reviews.llvm.org/D7649

llvm-svn: 230794
diff --git a/llvm/test/CodeGen/R600/fma.ll b/llvm/test/CodeGen/R600/fma.ll
index 3c874b2..d6024aa 100644
--- a/llvm/test/CodeGen/R600/fma.ll
+++ b/llvm/test/CodeGen/R600/fma.ll
@@ -14,9 +14,9 @@
 ; EG: FMA {{\*? *}}[[RES]]
 define void @fma_f32(float addrspace(1)* %out, float addrspace(1)* %in1,
                      float addrspace(1)* %in2, float addrspace(1)* %in3) {
-  %r0 = load float addrspace(1)* %in1
-  %r1 = load float addrspace(1)* %in2
-  %r2 = load float addrspace(1)* %in3
+  %r0 = load float, float addrspace(1)* %in1
+  %r1 = load float, float addrspace(1)* %in2
+  %r2 = load float, float addrspace(1)* %in3
   %r3 = tail call float @llvm.fma.f32(float %r0, float %r1, float %r2)
   store float %r3, float addrspace(1)* %out
   ret void
@@ -31,9 +31,9 @@
 ; EG-DAG: FMA {{\*? *}}[[RES]].[[CHHI]]
 define void @fma_v2f32(<2 x float> addrspace(1)* %out, <2 x float> addrspace(1)* %in1,
                        <2 x float> addrspace(1)* %in2, <2 x float> addrspace(1)* %in3) {
-  %r0 = load <2 x float> addrspace(1)* %in1
-  %r1 = load <2 x float> addrspace(1)* %in2
-  %r2 = load <2 x float> addrspace(1)* %in3
+  %r0 = load <2 x float>, <2 x float> addrspace(1)* %in1
+  %r1 = load <2 x float>, <2 x float> addrspace(1)* %in2
+  %r2 = load <2 x float>, <2 x float> addrspace(1)* %in3
   %r3 = tail call <2 x float> @llvm.fma.v2f32(<2 x float> %r0, <2 x float> %r1, <2 x float> %r2)
   store <2 x float> %r3, <2 x float> addrspace(1)* %out
   ret void
@@ -52,9 +52,9 @@
 ; EG-DAG: FMA {{\*? *}}[[RES]].W
 define void @fma_v4f32(<4 x float> addrspace(1)* %out, <4 x float> addrspace(1)* %in1,
                        <4 x float> addrspace(1)* %in2, <4 x float> addrspace(1)* %in3) {
-  %r0 = load <4 x float> addrspace(1)* %in1
-  %r1 = load <4 x float> addrspace(1)* %in2
-  %r2 = load <4 x float> addrspace(1)* %in3
+  %r0 = load <4 x float>, <4 x float> addrspace(1)* %in1
+  %r1 = load <4 x float>, <4 x float> addrspace(1)* %in2
+  %r2 = load <4 x float>, <4 x float> addrspace(1)* %in3
   %r3 = tail call <4 x float> @llvm.fma.v4f32(<4 x float> %r0, <4 x float> %r1, <4 x float> %r2)
   store <4 x float> %r3, <4 x float> addrspace(1)* %out
   ret void
@@ -68,8 +68,8 @@
   %in.b.gep = getelementptr float, float addrspace(1)* %in.b, i32 %tid
   %out.gep = getelementptr float, float addrspace(1)* %out, i32 %tid
 
-  %a = load float addrspace(1)* %in.a.gep, align 4
-  %b = load float addrspace(1)* %in.b.gep, align 4
+  %a = load float, float addrspace(1)* %in.a.gep, align 4
+  %b = load float, float addrspace(1)* %in.b.gep, align 4
 
   %fma = call float @llvm.fma.f32(float %a, float 2.0, float %b)
   store float %fma, float addrspace(1)* %out.gep, align 4
@@ -83,8 +83,8 @@
   %in.b.gep = getelementptr float, float addrspace(1)* %in.b, i32 %tid
   %out.gep = getelementptr float, float addrspace(1)* %out, i32 %tid
 
-  %a = load float addrspace(1)* %in.a.gep, align 4
-  %c = load float addrspace(1)* %in.b.gep, align 4
+  %a = load float, float addrspace(1)* %in.a.gep, align 4
+  %c = load float, float addrspace(1)* %in.b.gep, align 4
 
   %fma = call float @llvm.fma.f32(float %a, float %b, float %c)
   store float %fma, float addrspace(1)* %out.gep, align 4