[AArch64][SVE] Test more types in sve-fixed-length-subvector.ll

[joechrisellis]

Summary

Previously only the i32 type was tested. Now, the {i,f}{16,32,64} types
are tested.

The v8{i,f}16 cases lower differently to the other cases, which is worth
defending. The lowering for the other cases is currently identical, but
probably worth having for the better coverage.

[peterwaller-arm]

I think this needs a closer look at the CHECK labels - I see 'CHECK_GE' used where I think 'VBITS_GE' is intended. Also, 'CHECKT', which I take it is not intended.

I'm also not sure about introducing a test beyond the fundamental limitation of 2048 bits. I think it should be dropped for now, at a minimum it would be inconsistent with other files.

[joechrisellis]

Remove the -aarch64-sve-vector-bits-min=4096 case; the architectural maximum is 2048 bits, so this case is erroneous.

[joechrisellis]

Amend labels.

[david-arm]

llvm/test/CodeGen/AArch64/sve-fixed-length-subvector.ll

50

Should this just be a normal CHECK line consistent with the v8i32 case (same size as v16i16)?

64

VBITS_GE_512 consistent with the v16i32 case?

78

VBITS_GE_1024?

202

Same comments as the v16i16 and other cases.

[joechrisellis]

Address review comments.

• @david-arm: thanks! I had forgotten to update the check line to reflect the different vector size. Should be correct now.

[david-arm]

LGTM!

llvm/test/CodeGen/AArch64/sve-fixed-length-subvector.ll

 ; how fixed length operation are lowered to scalable ones, with multiple blocks
 ; ensuring insert/extract sequences are not folded away.
 
 target triple = "aarch64-unknown-linux-gnu"
 
 ; Don't use SVE when its registers are no bigger than NEON.
 ; NO_SVE-NOT: ptrue
 
+define void @subvector_v8i16(<8 x i16> *%in, <8 x i16>* %out) #0 {
+; CHECK-LABEL: subvector_v8i16:
+; CHECK: ldr [[DATA:q[0-9]+]], [x0]
+; CHECK: str [[DATA]], [x1]
+; CHECK: ret
+  %a = load <8 x i16>, <8 x i16>* %in
+  br label %bb1
+
+bb1:
+  store <8 x i16> %a, <8 x i16>* %out
+  ret void
+}
+
+define void @subvector_v16i16(<16 x i16> *%in, <16 x i16>* %out) #0 {
+; CHECK-LABEL: subvector_v16i16:
+; VBITS_GE_512: ptrue [[PG:p[0-9]+]].h, vl16
+; VBITS_GE_512: ld1h { [[DATA:z[0-9]+.h]] }, [[PG]]/z, [x0]
+; VBITS_GE_512: st1h { [[DATA]] }, [[PG]], [x1]
+; VBITS_GE_512: ret
+  %a = load <16 x i16>, <16 x i16>* %in
+  br label %bb1
+
+bb1:
+  store <16 x i16> %a, <16 x i16>* %out
+  ret void
+}
+
+define void @subvector_v32i16(<32 x i16> *%in, <32 x i16>* %out) #0 {
+; CHECK-LABEL: subvector_v32i16:
+; VBITS_GE_1024: ptrue [[PG:p[0-9]+]].h, vl32
+; VBITS_GE_1024: ld1h { [[DATA:z[0-9]+.h]] }, [[PG]]/z, [x0]
+; VBITS_GE_1024: st1h { [[DATA]] }, [[PG]], [x1]
+; CHECK: ret
+  %a = load <32 x i16>, <32 x i16>* %in
+  br label %bb1
+
+bb1:
+  store <32 x i16> %a, <32 x i16>* %out
+  ret void
+}
+
+define void @subvector_v64i16(<64 x i16> *%in, <64 x i16>* %out) #0 {
+; CHECK-LABEL: subvector_v64i16:
+; VBITS_GE_2048: ptrue [[PG:p[0-9]+]].h, vl64
+; VBITS_GE_2048: ld1h { [[DATA:z[0-9]+.h]] }, [[PG]]/z, [x0]
+; VBITS_GE_2048: st1h { [[DATA]] }, [[PG]], [x1]
+; CHECK: ret
+  %a = load <64 x i16>, <64 x i16>* %in
+  br label %bb1
+
+bb1:
+  store <64 x i16> %a, <64 x i16>* %out
+  ret void
+}
+
 define void @subvector_v8i32(<8 x i32> *%in, <8 x i32>* %out) #0 {
 ; CHECK-LABEL: subvector_v8i32:
 ; CHECK: ptrue [[PG:p[0-9]+]].s, vl8
 ; CHECK: ld1w { [[DATA:z[0-9]+.s]] }, [[PG]]/z, [x0]
 ; CHECK: st1w { [[DATA]] }, [[PG]], [x1]
 ; CHECK: ret
   %a = load <8 x i32>, <8 x i32>* %in
   br label %bb1
 
 bb1:
   store <8 x i32> %a, <8 x i32>* %out
   ret void
 }
 
 define void @subvector_v16i32(<16 x i32> *%in, <16 x i32>* %out) #0 {
 ; CHECK-LABEL: subvector_v16i32:
 ; VBITS_GE_512: ptrue [[PG:p[0-9]+]].s, vl16
 ; VBITS_GE_512: ld1w { [[DATA:z[0-9]+.s]] }, [[PG]]/z, [x0]
 ; VBITS_GE_512: st1w { [[DATA]] }, [[PG]], [x1]
-; CHECKT: ret
+; VBITS_GE_512: ret
   %a = load <16 x i32>, <16 x i32>* %in
   br label %bb1
 
 bb1:
   store <16 x i32> %a, <16 x i32>* %out
   ret void
 }
 
; CHECK: ret
   %a = load <64 x i32>, <64 x i32>* %in
   br label %bb1
 
 bb1:
   store <64 x i32> %a, <64 x i32>* %out
   ret void
 }
 
+
+define void @subvector_v8i64(<8 x i64> *%in, <8 x i64>* %out) #0 {
+; CHECK-LABEL: subvector_v8i64:
+; VBITS_GE_512: ptrue [[PG:p[0-9]+]].d, vl8
+; VBITS_GE_512: ld1d { [[DATA:z[0-9]+.d]] }, [[PG]]/z, [x0]
+; VBITS_GE_512: st1d { [[DATA]] }, [[PG]], [x1]
+; VBITS_GE_512: ret
+  %a = load <8 x i64>, <8 x i64>* %in
+  br label %bb1
+
+bb1:
+  store <8 x i64> %a, <8 x i64>* %out
+  ret void
+}
+
+define void @subvector_v16i64(<16 x i64> *%in, <16 x i64>* %out) #0 {
+; CHECK-LABEL: subvector_v16i64:
+; VBITS_GE_1024: ptrue [[PG:p[0-9]+]].d, vl16
+; VBITS_GE_1024: ld1d { [[DATA:z[0-9]+.d]] }, [[PG]]/z, [x0]
+; VBITS_GE_1024: st1d { [[DATA]] }, [[PG]], [x1]
+; VBITS_GE_1024: ret
+  %a = load <16 x i64>, <16 x i64>* %in
+  br label %bb1
+
+bb1:
+  store <16 x i64> %a, <16 x i64>* %out
+  ret void
+}
+
+define void @subvector_v32i64(<32 x i64> *%in, <32 x i64>* %out) #0 {
+; CHECK-LABEL: subvector_v32i64:
+; VBITS_GE_2048: ptrue [[PG:p[0-9]+]].d, vl32
+; VBITS_GE_2048: ld1d { [[DATA:z[0-9]+.d]] }, [[PG]]/z, [x0]
+; VBITS_GE_2048: st1d { [[DATA]] }, [[PG]], [x1]
+; CHECK: ret
+  %a = load <32 x i64>, <32 x i64>* %in
+  br label %bb1
+
+bb1:
+  store <32 x i64> %a, <32 x i64>* %out
+  ret void
+}
+
+define void @subvector_v8f16(<8 x half> *%in, <8 x half>* %out) #0 {
+; CHECK-LABEL: subvector_v8f16:
+; CHECK: ldr [[DATA:q[0-9]+]], [x0]
+; CHECK: str [[DATA]], [x1]
+; CHECK: ret
+  %a = load <8 x half>, <8 x half>* %in
+  br label %bb1
+
+bb1:
+  store <8 x half> %a, <8 x half>* %out
+  ret void
+}
+
+define void @subvector_v16f16(<16 x half> *%in, <16 x half>* %out) #0 {
+; CHECK-LABEL: subvector_v16f16:
+; VBITS_GE_512: ptrue [[PG:p[0-9]+]].h, vl16
+; VBITS_GE_512: ld1h { [[DATA:z[0-9]+.h]] }, [[PG]]/z, [x0]
+; VBITS_GE_512: st1h { [[DATA]] }, [[PG]], [x1]
+; VBITS_GE_512: ret
+  %a = load <16 x half>, <16 x half>* %in
+  br label %bb1
+
+bb1:
+  store <16 x half> %a, <16 x half>* %out
+  ret void
+}
+
+define void @subvector_v32f16(<32 x half> *%in, <32 x half>* %out) #0 {
+; CHECK-LABEL: subvector_v32f16:
+; VBITS_GE_1024: ptrue [[PG:p[0-9]+]].h, vl32
+; VBITS_GE_1024: ld1h { [[DATA:z[0-9]+.h]] }, [[PG]]/z, [x0]
+; VBITS_GE_1024: st1h { [[DATA]] }, [[PG]], [x1]
+; CHECK: ret
+  %a = load <32 x half>, <32 x half>* %in
+  br label %bb1
+
+bb1:
+  store <32 x half> %a, <32 x half>* %out
+  ret void
+}
+
+define void @subvector_v64f16(<64 x half> *%in, <64 x half>* %out) #0 {
+; CHECK-LABEL: subvector_v64f16:
+; VBITS_GE_2048: ptrue [[PG:p[0-9]+]].h, vl64
+; VBITS_GE_2048: ld1h { [[DATA:z[0-9]+.h]] }, [[PG]]/z, [x0]
+; VBITS_GE_2048: st1h { [[DATA]] }, [[PG]], [x1]
+; CHECK: ret
+  %a = load <64 x half>, <64 x half>* %in
+  br label %bb1
+
+bb1:
+  store <64 x half> %a, <64 x half>* %out
+  ret void
+}
+
+define void @subvector_v8f32(<8 x float> *%in, <8 x float>* %out) #0 {
+; CHECK-LABEL: subvector_v8f32:
+; CHECK: ptrue [[PG:p[0-9]+]].s, vl8
+; CHECK: ld1w { [[DATA:z[0-9]+.s]] }, [[PG]]/z, [x0]
+; CHECK: st1w { [[DATA]] }, [[PG]], [x1]
+; CHECK: ret
+  %a = load <8 x float>, <8 x float>* %in
+  br label %bb1
+
+bb1:
+  store <8 x float> %a, <8 x float>* %out
+  ret void
+}
+
+define void @subvector_v16f32(<16 x float> *%in, <16 x float>* %out) #0 {
+; CHECK-LABEL: subvector_v16f32:
+; VBITS_GE_512: ptrue [[PG:p[0-9]+]].s, vl16
+; VBITS_GE_512: ld1w { [[DATA:z[0-9]+.s]] }, [[PG]]/z, [x0]
+; VBITS_GE_512: st1w { [[DATA]] }, [[PG]], [x1]
+; VBITS_GE_512: ret
+  %a = load <16 x float>, <16 x float>* %in
+  br label %bb1
+
+bb1:
+  store <16 x float> %a, <16 x float>* %out
+  ret void
+}
+
+define void @subvector_v32f32(<32 x float> *%in, <32 x float>* %out) #0 {
+; CHECK-LABEL: subvector_v32f32:
+; VBITS_GE_1024: ptrue [[PG:p[0-9]+]].s, vl32
+; VBITS_GE_1024: ld1w { [[DATA:z[0-9]+.s]] }, [[PG]]/z, [x0]
+; VBITS_GE_1024: st1w { [[DATA]] }, [[PG]], [x1]
+; CHECK: ret
+  %a = load <32 x float>, <32 x float>* %in
+  br label %bb1
+
+bb1:
+  store <32 x float> %a, <32 x float>* %out
+  ret void
+}
+
+define void @subvector_v64f32(<64 x float> *%in, <64 x float>* %out) #0 {
+; CHECK-LABEL: subvector_v64f32:
+; VBITS_GE_2048: ptrue [[PG:p[0-9]+]].s, vl64
+; VBITS_GE_2048: ld1w { [[DATA:z[0-9]+.s]] }, [[PG]]/z, [x0]
+; VBITS_GE_2048: st1w { [[DATA]] }, [[PG]], [x1]
+; CHECK: ret
+  %a = load <64 x float>, <64 x float>* %in
+  br label %bb1
+
+bb1:
+  store <64 x float> %a, <64 x float>* %out
+  ret void
+}
+define void @subvector_v8f64(<8 x double> *%in, <8 x double>* %out) #0 {
+; CHECK-LABEL: subvector_v8f64:
+; VBITS_GE_512: ptrue [[PG:p[0-9]+]].d, vl8
+; VBITS_GE_512: ld1d { [[DATA:z[0-9]+.d]] }, [[PG]]/z, [x0]
+; VBITS_GE_512: st1d { [[DATA]] }, [[PG]], [x1]
+; VBITS_GE_512: ret
+  %a = load <8 x double>, <8 x double>* %in
+  br label %bb1
+
+bb1:
+  store <8 x double> %a, <8 x double>* %out
+  ret void
+}
+
+define void @subvector_v16f64(<16 x double> *%in, <16 x double>* %out) #0 {
+; CHECK-LABEL: subvector_v16f64:
+; VBITS_GE_1024: ptrue [[PG:p[0-9]+]].d, vl16
+; VBITS_GE_1024: ld1d { [[DATA:z[0-9]+.d]] }, [[PG]]/z, [x0]
+; VBITS_GE_1024: st1d { [[DATA]] }, [[PG]], [x1]
+; VBITS_GE_1024: ret
+  %a = load <16 x double>, <16 x double>* %in
+  br label %bb1
+
+bb1:
+  store <16 x double> %a, <16 x double>* %out
+  ret void
+}
+
+define void @subvector_v32f64(<32 x double> *%in, <32 x double>* %out) #0 {
+; CHECK-LABEL: subvector_v32f64:
+; VBITS_GE_2048: ptrue [[PG:p[0-9]+]].d, vl32
+; VBITS_GE_2048: ld1d { [[DATA:z[0-9]+.d]] }, [[PG]]/z, [x0]
+; VBITS_GE_2048: st1d { [[DATA]] }, [[PG]], [x1]
+; CHECK: ret
+  %a = load <32 x double>, <32 x double>* %in
+  br label %bb1
+
+bb1:
+  store <32 x double> %a, <32 x double>* %out
+  ret void
+}
+
 define <8 x i1> @no_warn_dropped_scalable(<8 x i32>* %in) #0 {
 ; CHECK-LABEL: no_warn_dropped_scalable:
 ; CHECK: ptrue [[PG:p[0-9]+]].s, vl8
 ; CHECK: ld1w { [[A:z[0-9]+]].s }, [[PG]]/z, [x0]
 ; CHECK: cmpgt p{{[0-9]}}.s, [[PG]]/z, [[A]].s, #0
 ; CHECK: ret
   %a = load <8 x i32>, <8 x i32>* %in
   br label %bb1
 
 bb1:
   %cond = icmp sgt <8 x i32> %a, zeroinitializer
   ret <8 x i1> %cond
 }
 
 ; binop(insert_subvec(a), insert_subvec(b)) -> insert_subvec(binop(a,b)) like
 ; combines remove redundant subvector operations. This test ensures it's not
 ; performed when the input idiom is the result of operation legalisation. When
 ; not prevented the test triggers infinite combine->legalise->combine->...
 define void @no_subvector_binop_hang(<8 x i32>* %in, <8 x i32>* %out, i1 %cond) #0 {
 ; CHECK-LABEL: no_subvector_binop_hang:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ptrue [[PG:p[0-9]+]].s, vl8
 ; CHECK-NEXT:    ld1w { [[A:z[0-9]+]].s }, [[PG]]/z, [x0]
 ; CHECK-NEXT:    ld1w { [[B:z[0-9]+]].s }, [[PG]]/z, [x1]
-; CHECK-NEXT:    tbz w2, #0, .LBB5_2
+; CHECK-NEXT:    tbz w2, #0, [[LABEL:\.[A-z0-9_]+]]
 ; CHECK-NEXT:  // %bb.1: // %bb.1
 ; CHECK-NEXT:    orr [[OR:z[0-9]+]].d, [[A]].d, [[B]].d
 ; CHECK-NEXT:    st1w { [[OR]].s }, [[PG]], [x1]
-; CHECK-NEXT:  .LBB5_2: // %bb.2
+; CHECK-NEXT:  [[LABEL]]: // %bb.2
 ; CHECK-NEXT:    ret
   %a = load <8 x i32>, <8 x i32>* %in
   %b = load <8 x i32>, <8 x i32>* %out
   br i1 %cond, label %bb.1, label %bb.2
 
 bb.1:
   %or = or <8 x i32> %a, %b
   store <8 x i32> %or, <8 x i32>* %out
   br label %bb.2
 
 bb.2:
   ret void
 }
 
 attributes #0 = { "target-features"="+sve" }