[RISCV][Clang] Add some RVV Floating-Point intrinsic functions. (vfclass, vfmerge, vfrec7, vfrsqrt7, vfsqrt)

[khchen]

Summary

Support vfclass, vfmerge, vfrec7, vfrsqrt7, vfsqrt instructions.

[craig.topper]

LGTM

[khchen]

rebase

[thakis]

One of the ten commits

[thakis]

One of the ten commits that landed here (https://github.com/llvm/llvm-project/compare/a3bfddbb6a27...59d5b8c27b43), this being one of them, slowed down check-clang by over 20%: https://bugs.llvm.org/show_bug.cgi?id=49962

Since recovering test runtime regressions is very difficult, it's probably a good idea to revert this first and then reland in smaller chunks to see which of the commits causes the problem (?)

[craig.topper]

In D99741#2690124, @thakis wrote:

One of the ten commits that landed here (https://github.com/llvm/llvm-project/compare/a3bfddbb6a27...59d5b8c27b43), this being one of them, slowed down check-clang by over 20%: https://bugs.llvm.org/show_bug.cgi?id=49962

Since recovering test runtime regressions is very difficult, it's probably a good idea to revert this first and then reland in smaller chunks to see which of the commits causes the problem (?)

The fact that we're running the optimization pipeline in these tests might be to blame. Might also be that riscv_vector.h is currently about 71000 lines which probably isn't quick to parse.

[khchen]

The fact that we're running the optimization pipeline in these tests might be to blame. Might also be that riscv_vector.h is currently about 71000 lines which probably isn't quick to parse.

I found removing all overloaded api definition functions in riscv_vector.h could save the time more than 100% (run llvm-lit with one file vloxei.c, cost 3.57s-> 1.56s)
I'm not sure how to improve the riscv_vector.h, is it possible to generate a PCH for riscv_vector.h?

[rogfer01]

In D99741#2690156, @craig.topper wrote:

In D99741#2690124, @thakis wrote:

One of the ten commits that landed here (https://github.com/llvm/llvm-project/compare/a3bfddbb6a27...59d5b8c27b43), this being one of them, slowed down check-clang by over 20%: https://bugs.llvm.org/show_bug.cgi?id=49962

Since recovering test runtime regressions is very difficult, it's probably a good idea to revert this first and then reland in smaller chunks to see which of the commits causes the problem (?)

The fact that we're running the optimization pipeline in these tests might be to blame. Might also be that riscv_vector.h is currently about 71000 lines which probably isn't quick to parse.

I hacked the RISCVVEmitter.cpp file to emit a riscv_vector_testing.h that wraps all the definitions with macros like RVV_TEST_vadd and RVV_TEST_vadd_mask (I used the IRName as it was easier for the experiment).

The testing header looks like this

#if defined(RVV_TEST_vadd)                                                       
__rvv_overloaded vint8m1_t vadd(vint8m1_t op0, vint8m1_t op1, size_t op2){       
  return vadd_vv_i8m1(op0, op1, op2);                                            
}                                                                                
#endif // RVV_TEST_vadd                                                          
                                                                                 
#if defined(RVV_TEST_vadd_mask)                                                  
__rvv_overloaded vint8m1_t vadd(vbool8_t op0, vint8m1_t op1, vint8m1_t op2, vint8m1_t op3, size_t op4){
  return vadd_vv_i8m1_m(op0, op1, op2, op3, op4);                                
}                                                                                
#endif // RVV_TEST_vadd_mask                                                     

In a release build (in a machine that is not very fast) looks like this:

$ ./bin/llvm-lit --time-test -sv ../llvm-src/clang/test/CodeGen/RISCV/rvv-intrinsics/vadd.c ../llvm-src/clang/test/CodeGen/RISCV/rvv-intrinsics/vadd_fast.c 
llvm-lit: /home/rferrer/fio/upstream/llvm-src/llvm/utils/lit/lit/llvm/config.py:428: note: using clang: /home/rferrer/fio/upstream/llvm-build-release/bin/clang
Slowest Tests:
--------------------------------------------------------------------------
8.91s: Clang :: CodeGen/RISCV/rvv-intrinsics/vadd.c
2.11s: Clang :: CodeGen/RISCV/rvv-intrinsics/vadd_fast.c

Where vadd.c is the current file and vadd_fast.c is the same as vadd.c with this change

--- vadd.c	2021-04-15 06:51:30.554996630 +0000
+++ vadd_fast.c	2021-04-15 07:00:38.684074973 +0000
@@ -8,7 +8,9 @@
 // RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
 
 // ASM-NOT: warning
-#include <riscv_vector.h>
+#define RVV_TEST_vadd
+#define RVV_TEST_vadd_mask
+#include <riscv_vector_testing.h>
 
 // CHECK-RV32-LABEL: @test_vadd_vv_i8mf8(
 // CHECK-RV32-NEXT:  entry:

This seems to suggest that making sure the tested header contains only the necessary will significantly speed up the testing.

We would want to make sure riscv_vector.h and riscv_vector_testing.h are equivalent when all the macros are enabled. I guess such a test is feasible if we preprocess the testing header with all the RVV_TEST_nnn macros enabled and then we do some form of textual comparison (which will likely have to ignore whitespace). This way we can test riscv_vector.h using its proxy riscv_vector_testing.h.

[rogfer01]

I've left a PoC of my approach here https://reviews.llvm.org/D100529 in case we can use that

[thakis]

Was there any more progress on this? If not, let's revert for now to not permanently slow down tests by over 20%.

[khchen]

In D99741#2694344, @thakis wrote:

Was there any more progress on this? If not, let's revert for now to not permanently slow down tests by over 20%.

https://reviews.llvm.org/D100611
https://reviews.llvm.org/D100617

Do you mean revert https://github.com/llvm/llvm-project/compare/a3bfddbb6a27...59d5b8c27b43?

[thakis]

In D99741#2694369, @khchen wrote:

In D99741#2694344, @thakis wrote:

Was there any more progress on this? If not, let's revert for now to not permanently slow down tests by over 20%.

https://reviews.llvm.org/D100611
https://reviews.llvm.org/D100617

Great, let's wait and see for a few days to see if it lands, and if so how much that recovers then.

Do you mean revert https://github.com/llvm/llvm-project/compare/a3bfddbb6a27...59d5b8c27b43?

Yes.

[khchen]

In D99741#2694435, @thakis wrote:

In D99741#2694369, @khchen wrote:

In D99741#2694344, @thakis wrote:

Was there any more progress on this? If not, let's revert for now to not permanently slow down tests by over 20%.

https://reviews.llvm.org/D100611
https://reviews.llvm.org/D100617

Great, let's wait and see for a few days to see if it lands, and if so how much that recovers then.

Do you mean revert https://github.com/llvm/llvm-project/compare/a3bfddbb6a27...59d5b8c27b43?

Yes.

Please wait a few days.
Those commits add almost total 300k line from all tests.
I think maybe reverting is not a good idea if we have a good progress on this slow down issue.

[thakis]

Please wait a few days.
Those commits add almost total 300k line from all tests.
I think maybe reverting is not a good idea if we have a good progress on this slow down issue.

Yes, if we can fix the regression in a few days, then all is good.

[thakis]

I think maybe reverting is not a good idea if we have a good progress on this slow down issue.

Yes, if we can fix the regression in a few days, then all is good.

Another ping here. check-clang is down to 5 min now, but it's still a 11% regression (used to be 4.5 min, went to 5.5 min, now 5 min).

[craig.topper]

In D99741#2701398, @thakis wrote:

I think maybe reverting is not a good idea if we have a good progress on this slow down issue.

Yes, if we can fix the regression in a few days, then all is good.

Another ping here. check-clang is down to 5 min now, but it's still a 11% regression (used to be 4.5 min, went to 5.5 min, now 5 min).

https://reviews.llvm.org/D100611 has not landed yet. Is the 5 min number with or without that patch?

[thakis]

In D99741#2701918, @craig.topper wrote:

In D99741#2701398, @thakis wrote:

I think maybe reverting is not a good idea if we have a good progress on this slow down issue.

Yes, if we can fix the regression in a few days, then all is good.

Another ping here. check-clang is down to 5 min now, but it's still a 11% regression (used to be 4.5 min, went to 5.5 min, now 5 min).

https://reviews.llvm.org/D100611 has not landed yet. Is the 5 min number with or without that patch?

Without. I'm just looking at my bot cycle times. Ok, let's wait for that to play out :)

[khchen]

@thakis
https://reviews.llvm.org/D100611 had landed, could you check your bot to see the cycle time?

[thakis]

No effect, or even worse. We're back at 5:30 cycle time and we'v been in that state for over 2 weeks now. I think it's time to revert back to green, so this slowdown doesn't become permanent.

[thakis]

(here's a build form a few minutes ago: http://45.33.8.238/macm1/8325/summary.html)

[craig.topper]

I believe we had reductions at builds 7601 and 8175.

It looks like there may have been an increase between builds 7979 and 7991 the build failed for a while there. I suspect "e951b04 [AArch64][SVE] Regression test all ACLE tests with C++" as it added a quite a few RUN lines.

I think we also caused an increase at build 8202 when these 4 commits went in
bd32c2d [RISCV] Implement the vwcvt{u}.x.x.v/vncvt.x.x.w builtin.
645c5f2 [RISCV] Implement the pseudo compare builtin.
bfb3fca [RISCV] Implement the vfabs.v/vfneg.v builtin.
4b24341 [RISCV] Implement the vmmv.m/vmnot.m builtin.

[thakis]

In D99741#2721235, @craig.topper wrote:

I believe we had reductions at builds 7601 and 8175.

It looks like there may have been an increase between builds 7979 and 7991 the build failed for a while there. I suspect "e951b04 [AArch64][SVE] Regression test all ACLE tests with C++" as it added a quite a few RUN lines.

I think we also caused an increase at build 8202 when these 4 commits went in
bd32c2d [RISCV] Implement the vwcvt{u}.x.x.v/vncvt.x.x.w builtin.
645c5f2 [RISCV] Implement the pseudo compare builtin.
bfb3fca [RISCV] Implement the vfabs.v/vfneg.v builtin.
4b24341 [RISCV] Implement the vmmv.m/vmnot.m builtin.

Any disagreement about reverting riscv builtin changes until there's some way to test them without increasing check-clang time by 20%?

[craig.topper]

In D99741#2721596, @thakis wrote:

In D99741#2721235, @craig.topper wrote:

I believe we had reductions at builds 7601 and 8175.

It looks like there may have been an increase between builds 7979 and 7991 the build failed for a while there. I suspect "e951b04 [AArch64][SVE] Regression test all ACLE tests with C++" as it added a quite a few RUN lines.

I think we also caused an increase at build 8202 when these 4 commits went in
bd32c2d [RISCV] Implement the vwcvt{u}.x.x.v/vncvt.x.x.w builtin.
645c5f2 [RISCV] Implement the pseudo compare builtin.
bfb3fca [RISCV] Implement the vfabs.v/vfneg.v builtin.
4b24341 [RISCV] Implement the vmmv.m/vmnot.m builtin.

Any disagreement about reverting riscv builtin changes until there's some way to test them without increasing check-clang time by 20%?

We're preparing a patch to remove to stop testing both rv32 and rv64 on every test. That should reduce the time by half. What is an acceptable number?

[HsiangKai]

In D99741#2721669, @craig.topper wrote:

In D99741#2721596, @thakis wrote:

In D99741#2721235, @craig.topper wrote:

I believe we had reductions at builds 7601 and 8175.

It looks like there may have been an increase between builds 7979 and 7991 the build failed for a while there. I suspect "e951b04 [AArch64][SVE] Regression test all ACLE tests with C++" as it added a quite a few RUN lines.

I think we also caused an increase at build 8202 when these 4 commits went in
bd32c2d [RISCV] Implement the vwcvt{u}.x.x.v/vncvt.x.x.w builtin.
645c5f2 [RISCV] Implement the pseudo compare builtin.
bfb3fca [RISCV] Implement the vfabs.v/vfneg.v builtin.
4b24341 [RISCV] Implement the vmmv.m/vmnot.m builtin.

Any disagreement about reverting riscv builtin changes until there's some way to test them without increasing check-clang time by 20%?

We're preparing a patch to remove to stop testing both rv32 and rv64 on every test. That should reduce the time by half. What is an acceptable number?

I have removed rv32 test cases for vector intrinsics in https://github.com/llvm/llvm-project/commit/b358a2be52480c008a0789c78f5df1bb2053bfd8.

[thakis]

! In D99741#2721669, @craig.topper wrote:
We're preparing a patch to remove to stop testing both rv32 and rv64 on every test. That should reduce the time by half. What is an acceptable number?

You tell me. I'd say well below 1%?

I wanted to prepare a graph that shows check-clang time over the last year – it stayed basically constant and the jumped by 20% (now 10% now that half the tests are disabled). It'd look dramatic, but I haven't had time to make a graph. I'm sure you can imagine it though :) (Also, I feel the burden of proof shouldn't be on people pointing out regressions.)

@rnk suggested putting these tests behind some kind of REQUIRES: expensive-checks thing that's off by default, instead of reverting everything, until perf is sorted out. That sounds like a great idea to me – wdyt?

[craig.topper]

In D99741#2730989, @thakis wrote:

! In D99741#2721669, @craig.topper wrote:
We're preparing a patch to remove to stop testing both rv32 and rv64 on every test. That should reduce the time by half. What is an acceptable number?

You tell me. I'd say well below 1%?

I wanted to prepare a graph that shows check-clang time over the last year – it stayed basically constant and the jumped by 20% (now 10% now that half the tests are disabled). It'd look dramatic, but I haven't had time to make a graph. I'm sure you can imagine it though :) (Also, I feel the burden of proof shouldn't be on people pointing out regressions.)

@rnk suggested putting these tests behind some kind of REQUIRES: expensive-checks thing that's off by default, instead of reverting everything, until perf is sorted out. That sounds like a great idea to me – wdyt?

I just uploaded D101700 which takes lit on clang/test/CodeGen/RISCV/rvv* from ~8 seconds to ~5.5 seconds as measured on a 48 core server I use at work. Not sure how that translates to a less capable machine with less cores, slower disk, etc.

For comparison, here are some times for running lit on subsets of tests on this machine before the patch. Mostly looked at clang/test/CodeGen as I was curious to see how close we are to other targets.

clang/test/                 - 68.75 seconds
clang/test/CodeGen          - 24.59 seconds
clang/test/CodeGen/X86      - 2.68 seconds
clang/test/CodeGen/aarch64* - 12.33 seconds
clang/test/CodeGen/arm*     - 2.04 seconds

We're also considering D100529 to further reduce the RISCV test time. Another option we've talked about is concatenating some of the tests into a single file to cut down the number of times we parse riscv_vector.h.

At the time the check-clang time increase was first raised, lit on clang/test/CodeGen/RISCV/rvv* was 22.57 seconds on my machine. clang/test/CodeGen/ was 34.15 seconds.

clang/include/clang/Basic/riscv_vector.td

class RVVBuiltin<string suffix, string prototype, string type_range,
   string HeaderCode = "";
 
 }
 
 //===----------------------------------------------------------------------===//
 // Basic classes with automatic codegen.
 //===----------------------------------------------------------------------===//
 
+class RVVOutBuiltin<string suffix, string prototype, string type_range>
+    : RVVBuiltin<suffix, prototype, type_range> {
+  let IntrinsicTypes = [-1];
+}
+
+class RVVOp0Builtin<string suffix, string prototype, string type_range>
+    : RVVBuiltin<suffix, prototype, type_range> {
+  let IntrinsicTypes = [0];
+}
+
 class RVVOutOp1Builtin<string suffix, string prototype, string type_range>
     : RVVBuiltin<suffix, prototype, type_range> {
   let IntrinsicTypes = [-1, 1];
 }
 
 class RVVOutOp0Op1Builtin<string suffix, string prototype, string type_range>
     : RVVBuiltin<suffix, prototype, type_range> {
   let IntrinsicTypes = [-1, 0, 1];
multiclass RVVFloatingMaskOutBuiltinSet
     : RVVOp0Op1BuiltinSet<NAME, "fd",
                           [["vv", "vm", "mvv"],
                            ["vf", "vm", "mve"]]>;
 
 multiclass RVVFloatingMaskOutVFBuiltinSet
     : RVVOp0Op1BuiltinSet<NAME, "fd",
                           [["vf", "vm", "mve"]]>;
 
+class RVVFloatingUnaryBuiltin<string builtin_suffix, string ir_suffix,
+                              string prototype>
+    : RVVOutBuiltin<ir_suffix, prototype, "fd"> {
+  let Name = NAME # "_" # builtin_suffix;
+}
+
+class RVVFloatingUnaryVVBuiltin : RVVFloatingUnaryBuiltin<"v", "v", "vv">;
+
 // For widen operation which has different mangling name.
 multiclass RVVWidenBuiltinSet<string intrinsic_name, string type_range,
                               list<list<string>> suffixes_prototypes> {
   let Log2LMUL = [-3, -2, -1, 0, 1, 2],
       IRName = intrinsic_name, IRNameMask = intrinsic_name # "_mask" in {
     foreach s_p = suffixes_prototypes in {
       let Name = NAME # "_" # s_p[0],
           MangledName = NAME # "_" # s_p[0] in {

 
 // 14.7. Vector Widening Floating-Point Fused Multiply-Add Instructions
 defm vfwmacc  : RVVFloatingWidenTerBuiltinSet;
 defm vfwnmacc : RVVFloatingWidenTerBuiltinSet;
 defm vfwmsac  : RVVFloatingWidenTerBuiltinSet;
 defm vfwnmsac : RVVFloatingWidenTerBuiltinSet;
 
 // 14.8. Vector Floating-Point Square-Root Instruction
-// TODO
+def vfsqrt : RVVFloatingUnaryVVBuiltin;
 
 // 14.9. Vector Floating-Point Reciprocal Square-Root Estimate Instruction
-// TODO
+def vfrsqrt7 : RVVFloatingUnaryVVBuiltin;
 
 // 14.10. Vector Floating-Point Reciprocal Estimate Instruction
-// TODO
+def vfrec7 : RVVFloatingUnaryVVBuiltin;
 
 // 14.11. Vector Floating-Point MIN/MAX Instructions
 defm vfmin : RVVFloatingBinBuiltinSet;
 defm vfmax : RVVFloatingBinBuiltinSet;
 
 // 14.12. Vector Floating-Point Sign-Injection Instructions
 defm vfsgnj  : RVVFloatingBinBuiltinSet;
 defm vfsgnjn : RVVFloatingBinBuiltinSet;
 defm vfsgnjx : RVVFloatingBinBuiltinSet;
 
 // 14.13. Vector Floating-Point Compare Instructions
 defm vmfeq : RVVFloatingMaskOutBuiltinSet;
 defm vmfne : RVVFloatingMaskOutBuiltinSet;
 defm vmflt : RVVFloatingMaskOutBuiltinSet;
 defm vmfle : RVVFloatingMaskOutBuiltinSet;
 defm vmfgt : RVVFloatingMaskOutVFBuiltinSet;
 defm vmfge : RVVFloatingMaskOutVFBuiltinSet;
 
 // 14.14. Vector Floating-Point Classify Instruction
-// TODO
+let Name = "vfclass_v" in
+  def vfclass : RVVOp0Builtin<"Uv", "Uvv", "fd">;
 
 // 14.15. Vector Floating-Point Merge Instructio
-// TODO
+let Name = "vfmerge_vfm", HasMask = false, PermuteOperands = [2, 0, 1] in
+  def vfmerge : RVVOutOp1Builtin<"v", "vvem", "fd">;
 
 // 14.16. Vector Floating-Point Move Instruction
 // TODO
 
 // 14.17. Single-Width Floating-Point/Integer Type-Convert Instructions
 // TODO
 
 // 14.18. Widening Floating-Point/Integer Type-Convert Instructions
 // TODO
 
 // 14.19. Narrowing Floating-Point/Integer Type-Convert Instructions
 // TODO

clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfclass.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vfclass.nxv1f32.i32(<vscale x 1 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vfclass.nxv1f32.i64(<vscale x 1 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vfclass_v_u32mf2(vfloat32mf2_t op1, size_t vl) {
+  return vfclass(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vfclass.nxv2f32.i32(<vscale x 2 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vfclass.nxv2f32.i64(<vscale x 2 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vfclass_v_u32m1(vfloat32m1_t op1, size_t vl) {
+  return vfclass(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vfclass.nxv4f32.i32(<vscale x 4 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vfclass.nxv4f32.i64(<vscale x 4 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vfclass_v_u32m2(vfloat32m2_t op1, size_t vl) {
+  return vfclass(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vfclass.nxv8f32.i32(<vscale x 8 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vfclass.nxv8f32.i64(<vscale x 8 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vfclass_v_u32m4(vfloat32m4_t op1, size_t vl) {
+  return vfclass(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vfclass.nxv16f32.i32(<vscale x 16 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vfclass.nxv16f32.i64(<vscale x 16 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vfclass_v_u32m8(vfloat32m8_t op1, size_t vl) {
+  return vfclass(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vfclass.nxv1f64.i32(<vscale x 1 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vfclass.nxv1f64.i64(<vscale x 1 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vfclass_v_u64m1(vfloat64m1_t op1, size_t vl) {
+  return vfclass(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vfclass.nxv2f64.i32(<vscale x 2 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vfclass.nxv2f64.i64(<vscale x 2 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vfclass_v_u64m2(vfloat64m2_t op1, size_t vl) {
+  return vfclass(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vfclass.nxv4f64.i32(<vscale x 4 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vfclass.nxv4f64.i64(<vscale x 4 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vfclass_v_u64m4(vfloat64m4_t op1, size_t vl) {
+  return vfclass(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vfclass.nxv8f64.i32(<vscale x 8 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vfclass.nxv8f64.i64(<vscale x 8 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vfclass_v_u64m8(vfloat64m8_t op1, size_t vl) {
+  return vfclass(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vfclass.mask.nxv1f32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vfclass.mask.nxv1f32.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vfclass_v_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff,
+                                     vfloat32mf2_t op1, size_t vl) {
+  return vfclass(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vfclass.mask.nxv2f32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vfclass.mask.nxv2f32.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vfclass_v_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff,
+                                   vfloat32m1_t op1, size_t vl) {
+  return vfclass(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vfclass.mask.nxv4f32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vfclass.mask.nxv4f32.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vfclass_v_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff,
+                                   vfloat32m2_t op1, size_t vl) {
+  return vfclass(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vfclass.mask.nxv8f32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vfclass.mask.nxv8f32.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vfclass_v_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff,
+                                   vfloat32m4_t op1, size_t vl) {
+  return vfclass(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vfclass.mask.nxv16f32.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vfclass.mask.nxv16f32.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vfclass_v_u32m8_m(vbool4_t mask, vuint32m8_t maskedoff,
+                                   vfloat32m8_t op1, size_t vl) {
+  return vfclass(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vfclass.mask.nxv1f64.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vfclass.mask.nxv1f64.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vfclass_v_u64m1_m(vbool64_t mask, vuint64m1_t maskedoff,
+                                   vfloat64m1_t op1, size_t vl) {
+  return vfclass(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vfclass.mask.nxv2f64.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vfclass.mask.nxv2f64.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vfclass_v_u64m2_m(vbool32_t mask, vuint64m2_t maskedoff,
+                                   vfloat64m2_t op1, size_t vl) {
+  return vfclass(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vfclass.mask.nxv4f64.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vfclass.mask.nxv4f64.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vfclass_v_u64m4_m(vbool16_t mask, vuint64m4_t maskedoff,
+                                   vfloat64m4_t op1, size_t vl) {
+  return vfclass(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vfclass.mask.nxv8f64.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vfclass.mask.nxv8f64.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vfclass_v_u64m8_m(vbool8_t mask, vuint64m8_t maskedoff,
+                                   vfloat64m8_t op1, size_t vl) {
+  return vfclass(mask, maskedoff, op1, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmerge.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmerge.nxv1f32.f32.i32(<vscale x 1 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmerge.nxv1f32.f32.i64(<vscale x 1 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfmerge_vfm_f32mf2(vbool64_t mask, vfloat32mf2_t op1,
+                                      float op2, size_t vl) {
+  return vfmerge(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmerge.nxv2f32.f32.i32(<vscale x 2 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmerge.nxv2f32.f32.i64(<vscale x 2 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfmerge_vfm_f32m1(vbool32_t mask, vfloat32m1_t op1, float op2,
+                                    size_t vl) {
+  return vfmerge(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmerge.nxv4f32.f32.i32(<vscale x 4 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmerge.nxv4f32.f32.i64(<vscale x 4 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfmerge_vfm_f32m2(vbool16_t mask, vfloat32m2_t op1, float op2,
+                                    size_t vl) {
+  return vfmerge(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmerge.nxv8f32.f32.i32(<vscale x 8 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmerge.nxv8f32.f32.i64(<vscale x 8 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfmerge_vfm_f32m4(vbool8_t mask, vfloat32m4_t op1, float op2,
+                                    size_t vl) {
+  return vfmerge(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmerge.nxv16f32.f32.i32(<vscale x 16 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmerge.nxv16f32.f32.i64(<vscale x 16 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfmerge_vfm_f32m8(vbool4_t mask, vfloat32m8_t op1, float op2,
+                                    size_t vl) {
+  return vfmerge(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmerge.nxv1f64.f64.i32(<vscale x 1 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmerge.nxv1f64.f64.i64(<vscale x 1 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfmerge_vfm_f64m1(vbool64_t mask, vfloat64m1_t op1,
+                                    double op2, size_t vl) {
+  return vfmerge(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmerge.nxv2f64.f64.i32(<vscale x 2 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmerge.nxv2f64.f64.i64(<vscale x 2 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfmerge_vfm_f64m2(vbool32_t mask, vfloat64m2_t op1,
+                                    double op2, size_t vl) {
+  return vfmerge(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmerge.nxv4f64.f64.i32(<vscale x 4 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmerge.nxv4f64.f64.i64(<vscale x 4 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfmerge_vfm_f64m4(vbool16_t mask, vfloat64m4_t op1,
+                                    double op2, size_t vl) {
+  return vfmerge(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmerge.nxv8f64.f64.i32(<vscale x 8 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmerge.nxv8f64.f64.i64(<vscale x 8 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfmerge_vfm_f64m8(vbool8_t mask, vfloat64m8_t op1, double op2,
+                                    size_t vl) {
+  return vfmerge(mask, op1, op2, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfrec7.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrec7.nxv1f32.i32(<vscale x 1 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrec7.nxv1f32.i64(<vscale x 1 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfrec7_v_f32mf2(vfloat32mf2_t op1, size_t vl) {
+  return vfrec7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrec7.nxv2f32.i32(<vscale x 2 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrec7.nxv2f32.i64(<vscale x 2 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfrec7_v_f32m1(vfloat32m1_t op1, size_t vl) {
+  return vfrec7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrec7.nxv4f32.i32(<vscale x 4 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrec7.nxv4f32.i64(<vscale x 4 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfrec7_v_f32m2(vfloat32m2_t op1, size_t vl) {
+  return vfrec7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrec7.nxv8f32.i32(<vscale x 8 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrec7.nxv8f32.i64(<vscale x 8 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfrec7_v_f32m4(vfloat32m4_t op1, size_t vl) {
+  return vfrec7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrec7.nxv16f32.i32(<vscale x 16 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrec7.nxv16f32.i64(<vscale x 16 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfrec7_v_f32m8(vfloat32m8_t op1, size_t vl) {
+  return vfrec7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrec7.nxv1f64.i32(<vscale x 1 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrec7.nxv1f64.i64(<vscale x 1 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfrec7_v_f64m1(vfloat64m1_t op1, size_t vl) {
+  return vfrec7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrec7.nxv2f64.i32(<vscale x 2 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrec7.nxv2f64.i64(<vscale x 2 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfrec7_v_f64m2(vfloat64m2_t op1, size_t vl) {
+  return vfrec7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrec7.nxv4f64.i32(<vscale x 4 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrec7.nxv4f64.i64(<vscale x 4 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfrec7_v_f64m4(vfloat64m4_t op1, size_t vl) {
+  return vfrec7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrec7.nxv8f64.i32(<vscale x 8 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrec7.nxv8f64.i64(<vscale x 8 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfrec7_v_f64m8(vfloat64m8_t op1, size_t vl) {
+  return vfrec7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrec7.mask.nxv1f32.i32(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrec7.mask.nxv1f32.i64(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfrec7_v_f32mf2_m(vbool64_t mask, vfloat32mf2_t maskedoff,
+                                     vfloat32mf2_t op1, size_t vl) {
+  return vfrec7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrec7.mask.nxv2f32.i32(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrec7.mask.nxv2f32.i64(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfrec7_v_f32m1_m(vbool32_t mask, vfloat32m1_t maskedoff,
+                                   vfloat32m1_t op1, size_t vl) {
+  return vfrec7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrec7.mask.nxv4f32.i32(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrec7.mask.nxv4f32.i64(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfrec7_v_f32m2_m(vbool16_t mask, vfloat32m2_t maskedoff,
+                                   vfloat32m2_t op1, size_t vl) {
+  return vfrec7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrec7.mask.nxv8f32.i32(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrec7.mask.nxv8f32.i64(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfrec7_v_f32m4_m(vbool8_t mask, vfloat32m4_t maskedoff,
+                                   vfloat32m4_t op1, size_t vl) {
+  return vfrec7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrec7.mask.nxv16f32.i32(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrec7.mask.nxv16f32.i64(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfrec7_v_f32m8_m(vbool4_t mask, vfloat32m8_t maskedoff,
+                                   vfloat32m8_t op1, size_t vl) {
+  return vfrec7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrec7.mask.nxv1f64.i32(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrec7.mask.nxv1f64.i64(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfrec7_v_f64m1_m(vbool64_t mask, vfloat64m1_t maskedoff,
+                                   vfloat64m1_t op1, size_t vl) {
+  return vfrec7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrec7.mask.nxv2f64.i32(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrec7.mask.nxv2f64.i64(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfrec7_v_f64m2_m(vbool32_t mask, vfloat64m2_t maskedoff,
+                                   vfloat64m2_t op1, size_t vl) {
+  return vfrec7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrec7.mask.nxv4f64.i32(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrec7.mask.nxv4f64.i64(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfrec7_v_f64m4_m(vbool16_t mask, vfloat64m4_t maskedoff,
+                                   vfloat64m4_t op1, size_t vl) {
+  return vfrec7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrec7.mask.nxv8f64.i32(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrec7.mask.nxv8f64.i64(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfrec7_v_f64m8_m(vbool8_t mask, vfloat64m8_t maskedoff,
+                                   vfloat64m8_t op1, size_t vl) {
+  return vfrec7(mask, maskedoff, op1, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfrsqrt7.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrsqrt7.nxv1f32.i32(<vscale x 1 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrsqrt7.nxv1f32.i64(<vscale x 1 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfrsqrt7_v_f32mf2(vfloat32mf2_t op1, size_t vl) {
+  return vfrsqrt7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrsqrt7.nxv2f32.i32(<vscale x 2 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrsqrt7.nxv2f32.i64(<vscale x 2 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfrsqrt7_v_f32m1(vfloat32m1_t op1, size_t vl) {
+  return vfrsqrt7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrsqrt7.nxv4f32.i32(<vscale x 4 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrsqrt7.nxv4f32.i64(<vscale x 4 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfrsqrt7_v_f32m2(vfloat32m2_t op1, size_t vl) {
+  return vfrsqrt7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrsqrt7.nxv8f32.i32(<vscale x 8 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrsqrt7.nxv8f32.i64(<vscale x 8 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfrsqrt7_v_f32m4(vfloat32m4_t op1, size_t vl) {
+  return vfrsqrt7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrsqrt7.nxv16f32.i32(<vscale x 16 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrsqrt7.nxv16f32.i64(<vscale x 16 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfrsqrt7_v_f32m8(vfloat32m8_t op1, size_t vl) {
+  return vfrsqrt7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrsqrt7.nxv1f64.i32(<vscale x 1 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrsqrt7.nxv1f64.i64(<vscale x 1 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfrsqrt7_v_f64m1(vfloat64m1_t op1, size_t vl) {
+  return vfrsqrt7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrsqrt7.nxv2f64.i32(<vscale x 2 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrsqrt7.nxv2f64.i64(<vscale x 2 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfrsqrt7_v_f64m2(vfloat64m2_t op1, size_t vl) {
+  return vfrsqrt7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrsqrt7.nxv4f64.i32(<vscale x 4 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrsqrt7.nxv4f64.i64(<vscale x 4 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfrsqrt7_v_f64m4(vfloat64m4_t op1, size_t vl) {
+  return vfrsqrt7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrsqrt7.nxv8f64.i32(<vscale x 8 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrsqrt7.nxv8f64.i64(<vscale x 8 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfrsqrt7_v_f64m8(vfloat64m8_t op1, size_t vl) {
+  return vfrsqrt7(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrsqrt7.mask.nxv1f32.i32(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrsqrt7.mask.nxv1f32.i64(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfrsqrt7_v_f32mf2_m(vbool64_t mask, vfloat32mf2_t maskedoff,
+                                       vfloat32mf2_t op1, size_t vl) {
+  return vfrsqrt7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrsqrt7.mask.nxv2f32.i32(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrsqrt7.mask.nxv2f32.i64(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfrsqrt7_v_f32m1_m(vbool32_t mask, vfloat32m1_t maskedoff,
+                                     vfloat32m1_t op1, size_t vl) {
+  return vfrsqrt7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrsqrt7.mask.nxv4f32.i32(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrsqrt7.mask.nxv4f32.i64(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfrsqrt7_v_f32m2_m(vbool16_t mask, vfloat32m2_t maskedoff,
+                                     vfloat32m2_t op1, size_t vl) {
+  return vfrsqrt7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrsqrt7.mask.nxv8f32.i32(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrsqrt7.mask.nxv8f32.i64(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfrsqrt7_v_f32m4_m(vbool8_t mask, vfloat32m4_t maskedoff,
+                                     vfloat32m4_t op1, size_t vl) {
+  return vfrsqrt7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrsqrt7.mask.nxv16f32.i32(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrsqrt7.mask.nxv16f32.i64(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfrsqrt7_v_f32m8_m(vbool4_t mask, vfloat32m8_t maskedoff,
+                                     vfloat32m8_t op1, size_t vl) {
+  return vfrsqrt7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrsqrt7.mask.nxv1f64.i32(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrsqrt7.mask.nxv1f64.i64(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfrsqrt7_v_f64m1_m(vbool64_t mask, vfloat64m1_t maskedoff,
+                                     vfloat64m1_t op1, size_t vl) {
+  return vfrsqrt7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrsqrt7.mask.nxv2f64.i32(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrsqrt7.mask.nxv2f64.i64(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfrsqrt7_v_f64m2_m(vbool32_t mask, vfloat64m2_t maskedoff,
+                                     vfloat64m2_t op1, size_t vl) {
+  return vfrsqrt7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrsqrt7.mask.nxv4f64.i32(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrsqrt7.mask.nxv4f64.i64(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfrsqrt7_v_f64m4_m(vbool16_t mask, vfloat64m4_t maskedoff,
+                                     vfloat64m4_t op1, size_t vl) {
+  return vfrsqrt7(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrsqrt7.mask.nxv8f64.i32(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrsqrt7.mask.nxv8f64.i64(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfrsqrt7_v_f64m8_m(vbool8_t mask, vfloat64m8_t maskedoff,
+                                     vfloat64m8_t op1, size_t vl) {
+  return vfrsqrt7(mask, maskedoff, op1, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfsqrt.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfsqrt.nxv1f32.i32(<vscale x 1 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfsqrt.nxv1f32.i64(<vscale x 1 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6:#.*]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfsqrt_v_f32mf2(vfloat32mf2_t op1, size_t vl) {
+  return vfsqrt(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfsqrt.nxv2f32.i32(<vscale x 2 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfsqrt.nxv2f32.i64(<vscale x 2 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfsqrt_v_f32m1(vfloat32m1_t op1, size_t vl) {
+  return vfsqrt(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfsqrt.nxv4f32.i32(<vscale x 4 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfsqrt.nxv4f32.i64(<vscale x 4 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfsqrt_v_f32m2(vfloat32m2_t op1, size_t vl) {
+  return vfsqrt(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfsqrt.nxv8f32.i32(<vscale x 8 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfsqrt.nxv8f32.i64(<vscale x 8 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfsqrt_v_f32m4(vfloat32m4_t op1, size_t vl) {
+  return vfsqrt(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfsqrt.nxv16f32.i32(<vscale x 16 x float> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfsqrt.nxv16f32.i64(<vscale x 16 x float> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfsqrt_v_f32m8(vfloat32m8_t op1, size_t vl) {
+  return vfsqrt(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfsqrt.nxv1f64.i32(<vscale x 1 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfsqrt.nxv1f64.i64(<vscale x 1 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfsqrt_v_f64m1(vfloat64m1_t op1, size_t vl) {
+  return vfsqrt(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfsqrt.nxv2f64.i32(<vscale x 2 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfsqrt.nxv2f64.i64(<vscale x 2 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfsqrt_v_f64m2(vfloat64m2_t op1, size_t vl) {
+  return vfsqrt(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfsqrt.nxv4f64.i32(<vscale x 4 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfsqrt.nxv4f64.i64(<vscale x 4 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfsqrt_v_f64m4(vfloat64m4_t op1, size_t vl) {
+  return vfsqrt(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfsqrt.nxv8f64.i32(<vscale x 8 x double> [[OP1:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfsqrt.nxv8f64.i64(<vscale x 8 x double> [[OP1:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfsqrt_v_f64m8(vfloat64m8_t op1, size_t vl) {
+  return vfsqrt(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfsqrt.mask.nxv1f32.i32(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfsqrt.mask.nxv1f32.i64(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfsqrt_v_f32mf2_m(vbool64_t mask, vfloat32mf2_t maskedoff,
+                                     vfloat32mf2_t op1, size_t vl) {
+  return vfsqrt(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfsqrt.mask.nxv2f32.i32(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfsqrt.mask.nxv2f32.i64(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfsqrt_v_f32m1_m(vbool32_t mask, vfloat32m1_t maskedoff,
+                                   vfloat32m1_t op1, size_t vl) {
+  return vfsqrt(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfsqrt.mask.nxv4f32.i32(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfsqrt.mask.nxv4f32.i64(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfsqrt_v_f32m2_m(vbool16_t mask, vfloat32m2_t maskedoff,
+                                   vfloat32m2_t op1, size_t vl) {
+  return vfsqrt(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfsqrt.mask.nxv8f32.i32(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfsqrt.mask.nxv8f32.i64(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfsqrt_v_f32m4_m(vbool8_t mask, vfloat32m4_t maskedoff,
+                                   vfloat32m4_t op1, size_t vl) {
+  return vfsqrt(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfsqrt.mask.nxv16f32.i32(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfsqrt.mask.nxv16f32.i64(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfsqrt_v_f32m8_m(vbool4_t mask, vfloat32m8_t maskedoff,
+                                   vfloat32m8_t op1, size_t vl) {
+  return vfsqrt(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfsqrt.mask.nxv1f64.i32(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfsqrt.mask.nxv1f64.i64(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfsqrt_v_f64m1_m(vbool64_t mask, vfloat64m1_t maskedoff,
+                                   vfloat64m1_t op1, size_t vl) {
+  return vfsqrt(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfsqrt.mask.nxv2f64.i32(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfsqrt.mask.nxv2f64.i64(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfsqrt_v_f64m2_m(vbool32_t mask, vfloat64m2_t maskedoff,
+                                   vfloat64m2_t op1, size_t vl) {
+  return vfsqrt(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfsqrt.mask.nxv4f64.i32(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfsqrt.mask.nxv4f64.i64(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfsqrt_v_f64m4_m(vbool16_t mask, vfloat64m4_t maskedoff,
+                                   vfloat64m4_t op1, size_t vl) {
+  return vfsqrt(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfsqrt.mask.nxv8f64.i32(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfsqrt.mask.nxv8f64.i64(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR6]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfsqrt_v_f64m8_m(vbool8_t mask, vfloat64m8_t maskedoff,
+                                   vfloat64m8_t op1, size_t vl) {
+  return vfsqrt(mask, maskedoff, op1, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics/vfclass.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vfclass.nxv1f32.i32(<vscale x 1 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vfclass.nxv1f32.i64(<vscale x 1 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vfclass_v_u32mf2(vfloat32mf2_t op1, size_t vl) {
+  return vfclass_v_u32mf2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vfclass.nxv2f32.i32(<vscale x 2 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vfclass.nxv2f32.i64(<vscale x 2 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vfclass_v_u32m1(vfloat32m1_t op1, size_t vl) {
+  return vfclass_v_u32m1(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vfclass.nxv4f32.i32(<vscale x 4 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vfclass.nxv4f32.i64(<vscale x 4 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vfclass_v_u32m2(vfloat32m2_t op1, size_t vl) {
+  return vfclass_v_u32m2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vfclass.nxv8f32.i32(<vscale x 8 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vfclass.nxv8f32.i64(<vscale x 8 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vfclass_v_u32m4(vfloat32m4_t op1, size_t vl) {
+  return vfclass_v_u32m4(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vfclass.nxv16f32.i32(<vscale x 16 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vfclass.nxv16f32.i64(<vscale x 16 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vfclass_v_u32m8(vfloat32m8_t op1, size_t vl) {
+  return vfclass_v_u32m8(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vfclass.nxv1f64.i32(<vscale x 1 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vfclass.nxv1f64.i64(<vscale x 1 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vfclass_v_u64m1(vfloat64m1_t op1, size_t vl) {
+  return vfclass_v_u64m1(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vfclass.nxv2f64.i32(<vscale x 2 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vfclass.nxv2f64.i64(<vscale x 2 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vfclass_v_u64m2(vfloat64m2_t op1, size_t vl) {
+  return vfclass_v_u64m2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vfclass.nxv4f64.i32(<vscale x 4 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vfclass.nxv4f64.i64(<vscale x 4 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vfclass_v_u64m4(vfloat64m4_t op1, size_t vl) {
+  return vfclass_v_u64m4(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vfclass.nxv8f64.i32(<vscale x 8 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vfclass.nxv8f64.i64(<vscale x 8 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vfclass_v_u64m8(vfloat64m8_t op1, size_t vl) {
+  return vfclass_v_u64m8(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vfclass.mask.nxv1f32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vfclass.mask.nxv1f32.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vfclass_v_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff,
+                                     vfloat32mf2_t op1, size_t vl) {
+  return vfclass_v_u32mf2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vfclass.mask.nxv2f32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vfclass.mask.nxv2f32.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vfclass_v_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff,
+                                   vfloat32m1_t op1, size_t vl) {
+  return vfclass_v_u32m1_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vfclass.mask.nxv4f32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vfclass.mask.nxv4f32.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vfclass_v_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff,
+                                   vfloat32m2_t op1, size_t vl) {
+  return vfclass_v_u32m2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vfclass.mask.nxv8f32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vfclass.mask.nxv8f32.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vfclass_v_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff,
+                                   vfloat32m4_t op1, size_t vl) {
+  return vfclass_v_u32m4_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vfclass.mask.nxv16f32.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vfclass.mask.nxv16f32.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vfclass_v_u32m8_m(vbool4_t mask, vuint32m8_t maskedoff,
+                                   vfloat32m8_t op1, size_t vl) {
+  return vfclass_v_u32m8_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vfclass.mask.nxv1f64.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vfclass.mask.nxv1f64.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vfclass_v_u64m1_m(vbool64_t mask, vuint64m1_t maskedoff,
+                                   vfloat64m1_t op1, size_t vl) {
+  return vfclass_v_u64m1_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vfclass.mask.nxv2f64.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vfclass.mask.nxv2f64.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vfclass_v_u64m2_m(vbool32_t mask, vuint64m2_t maskedoff,
+                                   vfloat64m2_t op1, size_t vl) {
+  return vfclass_v_u64m2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vfclass.mask.nxv4f64.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vfclass.mask.nxv4f64.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vfclass_v_u64m4_m(vbool16_t mask, vuint64m4_t maskedoff,
+                                   vfloat64m4_t op1, size_t vl) {
+  return vfclass_v_u64m4_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfclass_v_u64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vfclass.mask.nxv8f64.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfclass_v_u64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vfclass.mask.nxv8f64.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vfclass_v_u64m8_m(vbool8_t mask, vuint64m8_t maskedoff,
+                                   vfloat64m8_t op1, size_t vl) {
+  return vfclass_v_u64m8_m(mask, maskedoff, op1, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics/vfmerge.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmerge.nxv1f32.f32.i32(<vscale x 1 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmerge.nxv1f32.f32.i64(<vscale x 1 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfmerge_vfm_f32mf2(vbool64_t mask, vfloat32mf2_t op1,
+                                      float op2, size_t vl) {
+  return vfmerge_vfm_f32mf2(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmerge.nxv2f32.f32.i32(<vscale x 2 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmerge.nxv2f32.f32.i64(<vscale x 2 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfmerge_vfm_f32m1(vbool32_t mask, vfloat32m1_t op1, float op2,
+                                    size_t vl) {
+  return vfmerge_vfm_f32m1(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmerge.nxv4f32.f32.i32(<vscale x 4 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmerge.nxv4f32.f32.i64(<vscale x 4 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfmerge_vfm_f32m2(vbool16_t mask, vfloat32m2_t op1, float op2,
+                                    size_t vl) {
+  return vfmerge_vfm_f32m2(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmerge.nxv8f32.f32.i32(<vscale x 8 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmerge.nxv8f32.f32.i64(<vscale x 8 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfmerge_vfm_f32m4(vbool8_t mask, vfloat32m4_t op1, float op2,
+                                    size_t vl) {
+  return vfmerge_vfm_f32m4(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmerge.nxv16f32.f32.i32(<vscale x 16 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmerge.nxv16f32.f32.i64(<vscale x 16 x float> [[OP1:%.*]], float [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfmerge_vfm_f32m8(vbool4_t mask, vfloat32m8_t op1, float op2,
+                                    size_t vl) {
+  return vfmerge_vfm_f32m8(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmerge.nxv1f64.f64.i32(<vscale x 1 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmerge.nxv1f64.f64.i64(<vscale x 1 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfmerge_vfm_f64m1(vbool64_t mask, vfloat64m1_t op1,
+                                    double op2, size_t vl) {
+  return vfmerge_vfm_f64m1(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmerge.nxv2f64.f64.i32(<vscale x 2 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmerge.nxv2f64.f64.i64(<vscale x 2 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfmerge_vfm_f64m2(vbool32_t mask, vfloat64m2_t op1,
+                                    double op2, size_t vl) {
+  return vfmerge_vfm_f64m2(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmerge.nxv4f64.f64.i32(<vscale x 4 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmerge.nxv4f64.f64.i64(<vscale x 4 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfmerge_vfm_f64m4(vbool16_t mask, vfloat64m4_t op1,
+                                    double op2, size_t vl) {
+  return vfmerge_vfm_f64m4(mask, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfmerge_vfm_f64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmerge.nxv8f64.f64.i32(<vscale x 8 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfmerge_vfm_f64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmerge.nxv8f64.f64.i64(<vscale x 8 x double> [[OP1:%.*]], double [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfmerge_vfm_f64m8(vbool8_t mask, vfloat64m8_t op1, double op2,
+                                    size_t vl) {
+  return vfmerge_vfm_f64m8(mask, op1, op2, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics/vfrec7.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrec7.nxv1f32.i32(<vscale x 1 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrec7.nxv1f32.i64(<vscale x 1 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfrec7_v_f32mf2(vfloat32mf2_t op1, size_t vl) {
+  return vfrec7_v_f32mf2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrec7.nxv2f32.i32(<vscale x 2 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrec7.nxv2f32.i64(<vscale x 2 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfrec7_v_f32m1(vfloat32m1_t op1, size_t vl) {
+  return vfrec7_v_f32m1(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrec7.nxv4f32.i32(<vscale x 4 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrec7.nxv4f32.i64(<vscale x 4 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfrec7_v_f32m2(vfloat32m2_t op1, size_t vl) {
+  return vfrec7_v_f32m2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrec7.nxv8f32.i32(<vscale x 8 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrec7.nxv8f32.i64(<vscale x 8 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfrec7_v_f32m4(vfloat32m4_t op1, size_t vl) {
+  return vfrec7_v_f32m4(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrec7.nxv16f32.i32(<vscale x 16 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrec7.nxv16f32.i64(<vscale x 16 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfrec7_v_f32m8(vfloat32m8_t op1, size_t vl) {
+  return vfrec7_v_f32m8(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrec7.nxv1f64.i32(<vscale x 1 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrec7.nxv1f64.i64(<vscale x 1 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfrec7_v_f64m1(vfloat64m1_t op1, size_t vl) {
+  return vfrec7_v_f64m1(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrec7.nxv2f64.i32(<vscale x 2 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrec7.nxv2f64.i64(<vscale x 2 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfrec7_v_f64m2(vfloat64m2_t op1, size_t vl) {
+  return vfrec7_v_f64m2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrec7.nxv4f64.i32(<vscale x 4 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrec7.nxv4f64.i64(<vscale x 4 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfrec7_v_f64m4(vfloat64m4_t op1, size_t vl) {
+  return vfrec7_v_f64m4(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrec7.nxv8f64.i32(<vscale x 8 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrec7.nxv8f64.i64(<vscale x 8 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfrec7_v_f64m8(vfloat64m8_t op1, size_t vl) {
+  return vfrec7_v_f64m8(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrec7.mask.nxv1f32.i32(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrec7.mask.nxv1f32.i64(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfrec7_v_f32mf2_m(vbool64_t mask, vfloat32mf2_t maskedoff,
+                                     vfloat32mf2_t op1, size_t vl) {
+  return vfrec7_v_f32mf2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrec7.mask.nxv2f32.i32(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrec7.mask.nxv2f32.i64(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfrec7_v_f32m1_m(vbool32_t mask, vfloat32m1_t maskedoff,
+                                   vfloat32m1_t op1, size_t vl) {
+  return vfrec7_v_f32m1_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrec7.mask.nxv4f32.i32(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrec7.mask.nxv4f32.i64(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfrec7_v_f32m2_m(vbool16_t mask, vfloat32m2_t maskedoff,
+                                   vfloat32m2_t op1, size_t vl) {
+  return vfrec7_v_f32m2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrec7.mask.nxv8f32.i32(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrec7.mask.nxv8f32.i64(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfrec7_v_f32m4_m(vbool8_t mask, vfloat32m4_t maskedoff,
+                                   vfloat32m4_t op1, size_t vl) {
+  return vfrec7_v_f32m4_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrec7.mask.nxv16f32.i32(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrec7.mask.nxv16f32.i64(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfrec7_v_f32m8_m(vbool4_t mask, vfloat32m8_t maskedoff,
+                                   vfloat32m8_t op1, size_t vl) {
+  return vfrec7_v_f32m8_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrec7.mask.nxv1f64.i32(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrec7.mask.nxv1f64.i64(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfrec7_v_f64m1_m(vbool64_t mask, vfloat64m1_t maskedoff,
+                                   vfloat64m1_t op1, size_t vl) {
+  return vfrec7_v_f64m1_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrec7.mask.nxv2f64.i32(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrec7.mask.nxv2f64.i64(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfrec7_v_f64m2_m(vbool32_t mask, vfloat64m2_t maskedoff,
+                                   vfloat64m2_t op1, size_t vl) {
+  return vfrec7_v_f64m2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrec7.mask.nxv4f64.i32(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrec7.mask.nxv4f64.i64(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfrec7_v_f64m4_m(vbool16_t mask, vfloat64m4_t maskedoff,
+                                   vfloat64m4_t op1, size_t vl) {
+  return vfrec7_v_f64m4_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrec7_v_f64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrec7.mask.nxv8f64.i32(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrec7_v_f64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrec7.mask.nxv8f64.i64(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfrec7_v_f64m8_m(vbool8_t mask, vfloat64m8_t maskedoff,
+                                   vfloat64m8_t op1, size_t vl) {
+  return vfrec7_v_f64m8_m(mask, maskedoff, op1, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics/vfrsqrt7.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrsqrt7.nxv1f32.i32(<vscale x 1 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrsqrt7.nxv1f32.i64(<vscale x 1 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfrsqrt7_v_f32mf2(vfloat32mf2_t op1, size_t vl) {
+  return vfrsqrt7_v_f32mf2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrsqrt7.nxv2f32.i32(<vscale x 2 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrsqrt7.nxv2f32.i64(<vscale x 2 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfrsqrt7_v_f32m1(vfloat32m1_t op1, size_t vl) {
+  return vfrsqrt7_v_f32m1(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrsqrt7.nxv4f32.i32(<vscale x 4 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrsqrt7.nxv4f32.i64(<vscale x 4 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfrsqrt7_v_f32m2(vfloat32m2_t op1, size_t vl) {
+  return vfrsqrt7_v_f32m2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrsqrt7.nxv8f32.i32(<vscale x 8 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrsqrt7.nxv8f32.i64(<vscale x 8 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfrsqrt7_v_f32m4(vfloat32m4_t op1, size_t vl) {
+  return vfrsqrt7_v_f32m4(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrsqrt7.nxv16f32.i32(<vscale x 16 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrsqrt7.nxv16f32.i64(<vscale x 16 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfrsqrt7_v_f32m8(vfloat32m8_t op1, size_t vl) {
+  return vfrsqrt7_v_f32m8(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrsqrt7.nxv1f64.i32(<vscale x 1 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrsqrt7.nxv1f64.i64(<vscale x 1 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfrsqrt7_v_f64m1(vfloat64m1_t op1, size_t vl) {
+  return vfrsqrt7_v_f64m1(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrsqrt7.nxv2f64.i32(<vscale x 2 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrsqrt7.nxv2f64.i64(<vscale x 2 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfrsqrt7_v_f64m2(vfloat64m2_t op1, size_t vl) {
+  return vfrsqrt7_v_f64m2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrsqrt7.nxv4f64.i32(<vscale x 4 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrsqrt7.nxv4f64.i64(<vscale x 4 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfrsqrt7_v_f64m4(vfloat64m4_t op1, size_t vl) {
+  return vfrsqrt7_v_f64m4(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrsqrt7.nxv8f64.i32(<vscale x 8 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrsqrt7.nxv8f64.i64(<vscale x 8 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfrsqrt7_v_f64m8(vfloat64m8_t op1, size_t vl) {
+  return vfrsqrt7_v_f64m8(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrsqrt7.mask.nxv1f32.i32(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfrsqrt7.mask.nxv1f32.i64(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfrsqrt7_v_f32mf2_m(vbool64_t mask, vfloat32mf2_t maskedoff,
+                                       vfloat32mf2_t op1, size_t vl) {
+  return vfrsqrt7_v_f32mf2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrsqrt7.mask.nxv2f32.i32(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfrsqrt7.mask.nxv2f32.i64(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfrsqrt7_v_f32m1_m(vbool32_t mask, vfloat32m1_t maskedoff,
+                                     vfloat32m1_t op1, size_t vl) {
+  return vfrsqrt7_v_f32m1_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrsqrt7.mask.nxv4f32.i32(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfrsqrt7.mask.nxv4f32.i64(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfrsqrt7_v_f32m2_m(vbool16_t mask, vfloat32m2_t maskedoff,
+                                     vfloat32m2_t op1, size_t vl) {
+  return vfrsqrt7_v_f32m2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrsqrt7.mask.nxv8f32.i32(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfrsqrt7.mask.nxv8f32.i64(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfrsqrt7_v_f32m4_m(vbool8_t mask, vfloat32m4_t maskedoff,
+                                     vfloat32m4_t op1, size_t vl) {
+  return vfrsqrt7_v_f32m4_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrsqrt7.mask.nxv16f32.i32(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfrsqrt7.mask.nxv16f32.i64(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfrsqrt7_v_f32m8_m(vbool4_t mask, vfloat32m8_t maskedoff,
+                                     vfloat32m8_t op1, size_t vl) {
+  return vfrsqrt7_v_f32m8_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrsqrt7.mask.nxv1f64.i32(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfrsqrt7.mask.nxv1f64.i64(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfrsqrt7_v_f64m1_m(vbool64_t mask, vfloat64m1_t maskedoff,
+                                     vfloat64m1_t op1, size_t vl) {
+  return vfrsqrt7_v_f64m1_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrsqrt7.mask.nxv2f64.i32(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfrsqrt7.mask.nxv2f64.i64(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfrsqrt7_v_f64m2_m(vbool32_t mask, vfloat64m2_t maskedoff,
+                                     vfloat64m2_t op1, size_t vl) {
+  return vfrsqrt7_v_f64m2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrsqrt7.mask.nxv4f64.i32(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfrsqrt7.mask.nxv4f64.i64(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfrsqrt7_v_f64m4_m(vbool16_t mask, vfloat64m4_t maskedoff,
+                                     vfloat64m4_t op1, size_t vl) {
+  return vfrsqrt7_v_f64m4_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfrsqrt7_v_f64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrsqrt7.mask.nxv8f64.i32(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfrsqrt7_v_f64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfrsqrt7.mask.nxv8f64.i64(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfrsqrt7_v_f64m8_m(vbool8_t mask, vfloat64m8_t maskedoff,
+                                     vfloat64m8_t op1, size_t vl) {
+  return vfrsqrt7_v_f64m8_m(mask, maskedoff, op1, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics/vfsqrt.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +f -target-feature +d -target-feature +experimental-v \
+// RUN:   -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfsqrt.nxv1f32.i32(<vscale x 1 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfsqrt.nxv1f32.i64(<vscale x 1 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfsqrt_v_f32mf2(vfloat32mf2_t op1, size_t vl) {
+  return vfsqrt_v_f32mf2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfsqrt.nxv2f32.i32(<vscale x 2 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfsqrt.nxv2f32.i64(<vscale x 2 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfsqrt_v_f32m1(vfloat32m1_t op1, size_t vl) {
+  return vfsqrt_v_f32m1(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfsqrt.nxv4f32.i32(<vscale x 4 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfsqrt.nxv4f32.i64(<vscale x 4 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfsqrt_v_f32m2(vfloat32m2_t op1, size_t vl) {
+  return vfsqrt_v_f32m2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfsqrt.nxv8f32.i32(<vscale x 8 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfsqrt.nxv8f32.i64(<vscale x 8 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfsqrt_v_f32m4(vfloat32m4_t op1, size_t vl) {
+  return vfsqrt_v_f32m4(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfsqrt.nxv16f32.i32(<vscale x 16 x float> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfsqrt.nxv16f32.i64(<vscale x 16 x float> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfsqrt_v_f32m8(vfloat32m8_t op1, size_t vl) {
+  return vfsqrt_v_f32m8(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfsqrt.nxv1f64.i32(<vscale x 1 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfsqrt.nxv1f64.i64(<vscale x 1 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfsqrt_v_f64m1(vfloat64m1_t op1, size_t vl) {
+  return vfsqrt_v_f64m1(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfsqrt.nxv2f64.i32(<vscale x 2 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfsqrt.nxv2f64.i64(<vscale x 2 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfsqrt_v_f64m2(vfloat64m2_t op1, size_t vl) {
+  return vfsqrt_v_f64m2(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfsqrt.nxv4f64.i32(<vscale x 4 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfsqrt.nxv4f64.i64(<vscale x 4 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfsqrt_v_f64m4(vfloat64m4_t op1, size_t vl) {
+  return vfsqrt_v_f64m4(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfsqrt.nxv8f64.i32(<vscale x 8 x double> [[OP1:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfsqrt.nxv8f64.i64(<vscale x 8 x double> [[OP1:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfsqrt_v_f64m8(vfloat64m8_t op1, size_t vl) {
+  return vfsqrt_v_f64m8(op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfsqrt.mask.nxv1f32.i32(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfsqrt.mask.nxv1f32.i64(<vscale x 1 x float> [[MASKEDOFF:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
+//
+vfloat32mf2_t test_vfsqrt_v_f32mf2_m(vbool64_t mask, vfloat32mf2_t maskedoff,
+                                     vfloat32mf2_t op1, size_t vl) {
+  return vfsqrt_v_f32mf2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfsqrt.mask.nxv2f32.i32(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfsqrt.mask.nxv2f32.i64(<vscale x 2 x float> [[MASKEDOFF:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
+//
+vfloat32m1_t test_vfsqrt_v_f32m1_m(vbool32_t mask, vfloat32m1_t maskedoff,
+                                   vfloat32m1_t op1, size_t vl) {
+  return vfsqrt_v_f32m1_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfsqrt.mask.nxv4f32.i32(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfsqrt.mask.nxv4f32.i64(<vscale x 4 x float> [[MASKEDOFF:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
+//
+vfloat32m2_t test_vfsqrt_v_f32m2_m(vbool16_t mask, vfloat32m2_t maskedoff,
+                                   vfloat32m2_t op1, size_t vl) {
+  return vfsqrt_v_f32m2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfsqrt.mask.nxv8f32.i32(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfsqrt.mask.nxv8f32.i64(<vscale x 8 x float> [[MASKEDOFF:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
+//
+vfloat32m4_t test_vfsqrt_v_f32m4_m(vbool8_t mask, vfloat32m4_t maskedoff,
+                                   vfloat32m4_t op1, size_t vl) {
+  return vfsqrt_v_f32m4_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfsqrt.mask.nxv16f32.i32(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfsqrt.mask.nxv16f32.i64(<vscale x 16 x float> [[MASKEDOFF:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
+//
+vfloat32m8_t test_vfsqrt_v_f32m8_m(vbool4_t mask, vfloat32m8_t maskedoff,
+                                   vfloat32m8_t op1, size_t vl) {
+  return vfsqrt_v_f32m8_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfsqrt.mask.nxv1f64.i32(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfsqrt.mask.nxv1f64.i64(<vscale x 1 x double> [[MASKEDOFF:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
+//
+vfloat64m1_t test_vfsqrt_v_f64m1_m(vbool64_t mask, vfloat64m1_t maskedoff,
+                                   vfloat64m1_t op1, size_t vl) {
+  return vfsqrt_v_f64m1_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfsqrt.mask.nxv2f64.i32(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfsqrt.mask.nxv2f64.i64(<vscale x 2 x double> [[MASKEDOFF:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
+//
+vfloat64m2_t test_vfsqrt_v_f64m2_m(vbool32_t mask, vfloat64m2_t maskedoff,
+                                   vfloat64m2_t op1, size_t vl) {
+  return vfsqrt_v_f64m2_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfsqrt.mask.nxv4f64.i32(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfsqrt.mask.nxv4f64.i64(<vscale x 4 x double> [[MASKEDOFF:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
+//
+vfloat64m4_t test_vfsqrt_v_f64m4_m(vbool16_t mask, vfloat64m4_t maskedoff,
+                                   vfloat64m4_t op1, size_t vl) {
+  return vfsqrt_v_f64m4_m(mask, maskedoff, op1, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vfsqrt_v_f64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfsqrt.mask.nxv8f64.i32(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vfsqrt_v_f64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfsqrt.mask.nxv8f64.i64(<vscale x 8 x double> [[MASKEDOFF:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
+//
+vfloat64m8_t test_vfsqrt_v_f64m8_m(vbool8_t mask, vfloat64m8_t maskedoff,
+                                   vfloat64m8_t op1, size_t vl) {
+  return vfsqrt_v_f64m8_m(mask, maskedoff, op1, vl);
+}