Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)

Summary:
 This is a resurrection of work first proposed and discussed in Aug 2015:
   http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
   http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html

 The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.

 This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.

 In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
   require that the alignments for source & dest be equal.

 For example, code which used to read:
  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
  call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)

 Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.

s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g

 The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
   source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
        and those that use use MemIntrinsicInst::[get|set]Alignment() to use
        getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
        MemIntrinsicInst::[get|set]Alignment() methods.

Reviewers: pete, hfinkel, lhames, reames, bollu

Reviewed By: reames

Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits

Differential Revision: https://reviews.llvm.org/D41675

llvm-svn: 322965

6 files changed, 118 insertions, 118 deletions

diff --git a/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll b/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
index 8cabc7dae13..b40fcb3e492 100644
--- a/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
+++ b/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
@@ -1,6 +1,6 @@
 ; RUN: opt -mtriple=amdgcn-unknown-amdhsa -S -amdgpu-annotate-kernel-features < %s | FileCheck -check-prefix=HSA %s
 
-declare void @llvm.memcpy.p1i32.p4i32.i32(i32 addrspace(1)* nocapture, i32 addrspace(4)* nocapture, i32, i32, i1) #0
+declare void @llvm.memcpy.p1i32.p4i32.i32(i32 addrspace(1)* nocapture, i32 addrspace(4)* nocapture, i32, i1) #0
 
 @lds.i32 = unnamed_addr addrspace(3) global i32 undef, align 4
 @lds.arr = unnamed_addr addrspace(3) global [256 x i32] undef, align 4
@@ -68,7 +68,7 @@ define amdgpu_kernel void @cmpxchg_constant_cast_group_gv_gep_to_flat(i32 addrsp
 
 ; HSA: @memcpy_constant_cast_group_gv_gep_to_flat(i32 addrspace(1)* %out) #2
 define amdgpu_kernel void @memcpy_constant_cast_group_gv_gep_to_flat(i32 addrspace(1)* %out) #1 {
-  call void @llvm.memcpy.p1i32.p4i32.i32(i32 addrspace(1)* %out, i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 32, i32 4, i1 false)
+  call void @llvm.memcpy.p1i32.p4i32.i32(i32 addrspace(1)* align 4 %out, i32 addrspace(4)* align 4 getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 32, i1 false)
   ret void
 }
 
diff --git a/llvm/test/CodeGen/AMDGPU/lds-alignment.ll b/llvm/test/CodeGen/AMDGPU/lds-alignment.ll
index c23dea2b6b7..84c8d9b778c 100644
--- a/llvm/test/CodeGen/AMDGPU/lds-alignment.ll
+++ b/llvm/test/CodeGen/AMDGPU/lds-alignment.ll
@@ -9,16 +9,16 @@
 @lds.missing.align.0 = internal unnamed_addr addrspace(3) global [39 x i32] undef
 @lds.missing.align.1 = internal unnamed_addr addrspace(3) global [7 x i64] undef
 
-declare void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* nocapture, i8 addrspace(1)* nocapture readonly, i32, i32, i1) #0
-declare void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* nocapture, i8 addrspace(3)* nocapture readonly, i32, i32, i1) #0
+declare void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* nocapture, i8 addrspace(1)* nocapture readonly, i32, i1) #0
+declare void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* nocapture, i8 addrspace(3)* nocapture readonly, i32, i1) #0
 
 
 ; HSA-LABEL: {{^}}test_no_round_size_1:
 ; HSA: workgroup_group_segment_byte_size = 38
 define amdgpu_kernel void @test_no_round_size_1(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 4, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 4, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %lds.align16.0.bc, i8 addrspace(1)* align 4 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out, i8 addrspace(3)* align 4 %lds.align16.0.bc, i32 38, i1 false)
   ret void
 }
 
@@ -36,12 +36,12 @@ define amdgpu_kernel void @test_no_round_size_1(i8 addrspace(1)* %out, i8 addrsp
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_size_2(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 4, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 4, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %lds.align16.0.bc, i8 addrspace(1)* align 4 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out, i8 addrspace(3)* align 4 %lds.align16.0.bc, i32 38, i1 false)
 
   %lds.align16.1.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.1 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.1.bc, i8 addrspace(1)* %in, i32 38, i32 4, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.1.bc, i32 38, i32 4, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %lds.align16.1.bc, i8 addrspace(1)* align 4 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out, i8 addrspace(3)* align 4 %lds.align16.1.bc, i32 38, i1 false)
 
   ret void
 }
@@ -52,12 +52,12 @@ define amdgpu_kernel void @test_round_size_2(i8 addrspace(1)* %out, i8 addrspace
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_size_2_align_8(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align16.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align16.0.bc, i32 38, i1 false)
 
   %lds.align8.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align8.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align8.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align8.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align8.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align8.0.bc, i32 38, i1 false)
 
   ret void
 }
@@ -67,11 +67,11 @@ define amdgpu_kernel void @test_round_size_2_align_8(i8 addrspace(1)* %out, i8 a
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_local_lds_and_arg(i8 addrspace(1)* %out, i8 addrspace(1)* %in, i8 addrspace(3)* %lds.arg) #1 {
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 4, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %lds.align16.0.bc, i8 addrspace(1)* align 4 %in, i32 38, i1 false)
 
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 4, i1 false)
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.arg, i8 addrspace(1)* %in, i32 38, i32 4, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.arg, i32 38, i32 4, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out, i8 addrspace(3)* align 4 %lds.align16.0.bc, i32 38, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %lds.arg, i8 addrspace(1)* align 4 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out, i8 addrspace(3)* align 4 %lds.arg, i32 38, i1 false)
   ret void
 }
 
@@ -79,8 +79,8 @@ define amdgpu_kernel void @test_round_local_lds_and_arg(i8 addrspace(1)* %out, i
 ; HSA: workgroup_group_segment_byte_size = 0
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_lds_arg(i8 addrspace(1)* %out, i8 addrspace(1)* %in, i8 addrspace(3)* %lds.arg) #1 {
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.arg, i8 addrspace(1)* %in, i32 38, i32 4, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.arg, i32 38, i32 4, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %lds.arg, i8 addrspace(1)* align 4 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out, i8 addrspace(3)* align 4 %lds.arg, i32 38, i1 false)
   ret void
 }
 
@@ -89,8 +89,8 @@ define amdgpu_kernel void @test_round_lds_arg(i8 addrspace(1)* %out, i8 addrspac
 ; HSA: workgroup_group_segment_byte_size = 0
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_high_align_lds_arg(i8 addrspace(1)* %out, i8 addrspace(1)* %in, i8 addrspace(3)* align 64 %lds.arg) #1 {
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.arg, i8 addrspace(1)* %in, i32 38, i32 64, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.arg, i32 38, i32 64, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 64 %lds.arg, i8 addrspace(1)* align 64 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 64 %out, i8 addrspace(3)* align 64 %lds.arg, i32 38, i1 false)
   ret void
 }
 
@@ -100,12 +100,12 @@ define amdgpu_kernel void @test_high_align_lds_arg(i8 addrspace(1)* %out, i8 add
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_missing_alignment_size_2_order0(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.missing.align.0.bc = bitcast [39 x i32] addrspace(3)* @lds.missing.align.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.missing.align.0.bc, i8 addrspace(1)* %in, i32 160, i32 4, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.missing.align.0.bc, i32 160, i32 4, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %lds.missing.align.0.bc, i8 addrspace(1)* align 4 %in, i32 160, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out, i8 addrspace(3)* align 4 %lds.missing.align.0.bc, i32 160, i1 false)
 
   %lds.missing.align.1.bc = bitcast [7 x i64] addrspace(3)* @lds.missing.align.1 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.missing.align.1.bc, i8 addrspace(1)* %in, i32 56, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.missing.align.1.bc, i32 56, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.missing.align.1.bc, i8 addrspace(1)* align 8 %in, i32 56, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.missing.align.1.bc, i32 56, i1 false)
 
   ret void
 }
@@ -116,12 +116,12 @@ define amdgpu_kernel void @test_missing_alignment_size_2_order0(i8 addrspace(1)*
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_missing_alignment_size_2_order1(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.missing.align.1.bc = bitcast [7 x i64] addrspace(3)* @lds.missing.align.1 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.missing.align.1.bc, i8 addrspace(1)* %in, i32 56, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.missing.align.1.bc, i32 56, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.missing.align.1.bc, i8 addrspace(1)* align 8 %in, i32 56, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.missing.align.1.bc, i32 56, i1 false)
 
   %lds.missing.align.0.bc = bitcast [39 x i32] addrspace(3)* @lds.missing.align.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.missing.align.0.bc, i8 addrspace(1)* %in, i32 160, i32 4, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.missing.align.0.bc, i32 160, i32 4, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %lds.missing.align.0.bc, i8 addrspace(1)* align 4 %in, i32 160, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out, i8 addrspace(3)* align 4 %lds.missing.align.0.bc, i32 160, i1 false)
 
   ret void
 }
@@ -144,16 +144,16 @@ define amdgpu_kernel void @test_missing_alignment_size_2_order1(i8 addrspace(1)*
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_size_3_order0(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.align32.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align32.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align32.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align32.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align32.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align32.0.bc, i32 38, i1 false)
 
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align16.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align16.0.bc, i32 38, i1 false)
 
   %lds.align8.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align8.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align8.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align8.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align8.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align8.0.bc, i32 38, i1 false)
 
   ret void
 }
@@ -165,16 +165,16 @@ define amdgpu_kernel void @test_round_size_3_order0(i8 addrspace(1)* %out, i8 ad
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_size_3_order1(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.align32.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align32.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align32.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align32.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align32.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align32.0.bc, i32 38, i1 false)
 
   %lds.align8.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align8.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align8.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align8.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align8.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align8.0.bc, i32 38, i1 false)
 
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align16.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align16.0.bc, i32 38, i1 false)
 
   ret void
 }
@@ -186,16 +186,16 @@ define amdgpu_kernel void @test_round_size_3_order1(i8 addrspace(1)* %out, i8 ad
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_size_3_order2(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align16.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align16.0.bc, i32 38, i1 false)
 
   %lds.align32.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align32.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align32.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align32.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align32.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align32.0.bc, i32 38, i1 false)
 
   %lds.align8.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align8.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align8.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align8.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align8.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align8.0.bc, i32 38, i1 false)
 
   ret void
 }
@@ -207,16 +207,16 @@ define amdgpu_kernel void @test_round_size_3_order2(i8 addrspace(1)* %out, i8 ad
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_size_3_order3(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align16.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align16.0.bc, i32 38, i1 false)
 
   %lds.align8.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align8.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align8.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align8.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align8.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align8.0.bc, i32 38, i1 false)
 
   %lds.align32.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align32.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align32.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align32.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align32.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align32.0.bc, i32 38, i1 false)
 
   ret void
 }
@@ -228,16 +228,16 @@ define amdgpu_kernel void @test_round_size_3_order3(i8 addrspace(1)* %out, i8 ad
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_size_3_order4(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.align8.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align8.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align8.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align8.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align8.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align8.0.bc, i32 38, i1 false)
 
   %lds.align32.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align32.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align32.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align32.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align32.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align32.0.bc, i32 38, i1 false)
 
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align16.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align16.0.bc, i32 38, i1 false)
 
   ret void
 }
@@ -249,16 +249,16 @@ define amdgpu_kernel void @test_round_size_3_order4(i8 addrspace(1)* %out, i8 ad
 ; HSA: group_segment_alignment = 4
 define amdgpu_kernel void @test_round_size_3_order5(i8 addrspace(1)* %out, i8 addrspace(1)* %in) #1 {
   %lds.align8.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align8.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align8.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align8.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align8.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align8.0.bc, i32 38, i1 false)
 
   %lds.align16.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align16.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align16.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align16.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align16.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align16.0.bc, i32 38, i1 false)
 
   %lds.align32.0.bc = bitcast [38 x i8] addrspace(3)* @lds.align32.0 to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %lds.align32.0.bc, i8 addrspace(1)* %in, i32 38, i32 8, i1 false)
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out, i8 addrspace(3)* %lds.align32.0.bc, i32 38, i32 8, i1 false)
+  call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 8 %lds.align32.0.bc, i8 addrspace(1)* align 8 %in, i32 38, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 8 %out, i8 addrspace(3)* align 8 %lds.align32.0.bc, i32 38, i1 false)
 
   ret void
 }
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.memcpy.ll b/llvm/test/CodeGen/AMDGPU/llvm.memcpy.ll
index 4068c020e70..77eb4900ea5 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.memcpy.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.memcpy.ll
@@ -1,9 +1,9 @@
 ; RUN: llc -march=amdgcn -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
 ; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
 
-declare void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* nocapture, i8 addrspace(3)* nocapture, i32, i32, i1) nounwind
-declare void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* nocapture, i8 addrspace(1)* nocapture, i64, i32, i1) nounwind
-declare void @llvm.memcpy.p1i8.p2i8.i64(i8 addrspace(1)* nocapture, i8 addrspace(2)* nocapture, i64, i32, i1) nounwind
+declare void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* nocapture, i8 addrspace(3)* nocapture, i32, i1) nounwind
+declare void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* nocapture, i8 addrspace(1)* nocapture, i64, i1) nounwind
+declare void @llvm.memcpy.p1i8.p2i8.i64(i8 addrspace(1)* nocapture, i8 addrspace(2)* nocapture, i64, i1) nounwind
 
 
 ; FUNC-LABEL: {{^}}test_small_memcpy_i64_lds_to_lds_align1:
@@ -83,7 +83,7 @@ declare void @llvm.memcpy.p1i8.p2i8.i64(i8 addrspace(1)* nocapture, i8 addrspace
 define amdgpu_kernel void @test_small_memcpy_i64_lds_to_lds_align1(i64 addrspace(3)* noalias %out, i64 addrspace(3)* noalias %in) nounwind {
   %bcin = bitcast i64 addrspace(3)* %in to i8 addrspace(3)*
   %bcout = bitcast i64 addrspace(3)* %out to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* %bcout, i8 addrspace(3)* %bcin, i32 32, i32 1, i1 false) nounwind
+  call void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* %bcout, i8 addrspace(3)* %bcin, i32 32, i1 false) nounwind
   ret void
 }
 
@@ -128,7 +128,7 @@ define amdgpu_kernel void @test_small_memcpy_i64_lds_to_lds_align1(i64 addrspace
 define amdgpu_kernel void @test_small_memcpy_i64_lds_to_lds_align2(i64 addrspace(3)* noalias %out, i64 addrspace(3)* noalias %in) nounwind {
   %bcin = bitcast i64 addrspace(3)* %in to i8 addrspace(3)*
   %bcout = bitcast i64 addrspace(3)* %out to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* %bcout, i8 addrspace(3)* %bcin, i32 32, i32 2, i1 false) nounwind
+  call void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* align 2 %bcout, i8 addrspace(3)* align 2 %bcin, i32 32, i1 false) nounwind
   ret void
 }
 
@@ -147,7 +147,7 @@ define amdgpu_kernel void @test_small_memcpy_i64_lds_to_lds_align2(i64 addrspace
 define amdgpu_kernel void @test_small_memcpy_i64_lds_to_lds_align4(i64 addrspace(3)* noalias %out, i64 addrspace(3)* noalias %in) nounwind {
   %bcin = bitcast i64 addrspace(3)* %in to i8 addrspace(3)*
   %bcout = bitcast i64 addrspace(3)* %out to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* %bcout, i8 addrspace(3)* %bcin, i32 32, i32 4, i1 false) nounwind
+  call void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* align 4 %bcout, i8 addrspace(3)* align 4 %bcin, i32 32, i1 false) nounwind
   ret void
 }
 
@@ -164,7 +164,7 @@ define amdgpu_kernel void @test_small_memcpy_i64_lds_to_lds_align4(i64 addrspace
 define amdgpu_kernel void @test_small_memcpy_i64_lds_to_lds_align8(i64 addrspace(3)* noalias %out, i64 addrspace(3)* noalias %in) nounwind {
   %bcin = bitcast i64 addrspace(3)* %in to i8 addrspace(3)*
   %bcout = bitcast i64 addrspace(3)* %out to i8 addrspace(3)*
-  call void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* %bcout, i8 addrspace(3)* %bcin, i32 32, i32 8, i1 false) nounwind
+  call void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* align 8 %bcout, i8 addrspace(3)* align 8 %bcin, i32 32, i1 false) nounwind
   ret void
 }
 
@@ -241,7 +241,7 @@ define amdgpu_kernel void @test_small_memcpy_i64_lds_to_lds_align8(i64 addrspace
 define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align1(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noalias %in) nounwind {
   %bcin = bitcast i64 addrspace(1)* %in to i8 addrspace(1)*
   %bcout = bitcast i64 addrspace(1)* %out to i8 addrspace(1)*
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %bcout, i8 addrspace(1)* %bcin, i64 32, i32 1, i1 false) nounwind
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %bcout, i8 addrspace(1)* %bcin, i64 32, i1 false) nounwind
   ret void
 }
 
@@ -284,7 +284,7 @@ define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align1(i64 add
 define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align2(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noalias %in) nounwind {
   %bcin = bitcast i64 addrspace(1)* %in to i8 addrspace(1)*
   %bcout = bitcast i64 addrspace(1)* %out to i8 addrspace(1)*
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %bcout, i8 addrspace(1)* %bcin, i64 32, i32 2, i1 false) nounwind
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* align 2 %bcout, i8 addrspace(1)* align 2 %bcin, i64 32, i1 false) nounwind
   ret void
 }
 
@@ -297,7 +297,7 @@ define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align2(i64 add
 define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align4(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noalias %in) nounwind {
   %bcin = bitcast i64 addrspace(1)* %in to i8 addrspace(1)*
   %bcout = bitcast i64 addrspace(1)* %out to i8 addrspace(1)*
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %bcout, i8 addrspace(1)* %bcin, i64 32, i32 4, i1 false) nounwind
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* align 4 %bcout, i8 addrspace(1)* align 4 %bcin, i64 32, i1 false) nounwind
   ret void
 }
 
@@ -310,7 +310,7 @@ define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align4(i64 add
 define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align8(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noalias %in) nounwind {
   %bcin = bitcast i64 addrspace(1)* %in to i8 addrspace(1)*
   %bcout = bitcast i64 addrspace(1)* %out to i8 addrspace(1)*
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %bcout, i8 addrspace(1)* %bcin, i64 32, i32 8, i1 false) nounwind
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* align 8 %bcout, i8 addrspace(1)* align 8 %bcin, i64 32, i1 false) nounwind
   ret void
 }
 
@@ -323,7 +323,7 @@ define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align8(i64 add
 define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align16(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noalias %in) nounwind {
   %bcin = bitcast i64 addrspace(1)* %in to i8 addrspace(1)*
   %bcout = bitcast i64 addrspace(1)* %out to i8 addrspace(1)*
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %bcout, i8 addrspace(1)* %bcin, i64 32, i32 16, i1 false) nounwind
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* align 16 %bcout, i8 addrspace(1)* align 16 %bcin, i64 32, i1 false) nounwind
   ret void
 }
 
@@ -342,7 +342,7 @@ define amdgpu_kernel void @test_small_memcpy_i64_global_to_global_align16(i64 ad
 ; SI-DAG: buffer_store_dwordx4
 define amdgpu_kernel void @test_memcpy_const_string_align4(i8 addrspace(1)* noalias %out) nounwind {
   %str = bitcast [16 x i8] addrspace(2)* @hello.align4 to i8 addrspace(2)*
-  call void @llvm.memcpy.p1i8.p2i8.i64(i8 addrspace(1)* %out, i8 addrspace(2)* %str, i64 32, i32 4, i1 false)
+  call void @llvm.memcpy.p1i8.p2i8.i64(i8 addrspace(1)* align 4 %out, i8 addrspace(2)* align 4 %str, i64 32, i1 false)
   ret void
 }
 
@@ -367,6 +367,6 @@ define amdgpu_kernel void @test_memcpy_const_string_align4(i8 addrspace(1)* noal
 ; SI: buffer_store_byte
 define amdgpu_kernel void @test_memcpy_const_string_align1(i8 addrspace(1)* noalias %out) nounwind {
   %str = bitcast [16 x i8] addrspace(2)* @hello.align1 to i8 addrspace(2)*
-  call void @llvm.memcpy.p1i8.p2i8.i64(i8 addrspace(1)* %out, i8 addrspace(2)* %str, i64 32, i32 1, i1 false)
+  call void @llvm.memcpy.p1i8.p2i8.i64(i8 addrspace(1)* %out, i8 addrspace(2)* %str, i64 32, i1 false)
   ret void
 }
diff --git a/llvm/test/CodeGen/AMDGPU/lower-mem-intrinsics.ll b/llvm/test/CodeGen/AMDGPU/lower-mem-intrinsics.ll
index 778467207a0..498a65dc0a6 100644
--- a/llvm/test/CodeGen/AMDGPU/lower-mem-intrinsics.ll
+++ b/llvm/test/CodeGen/AMDGPU/lower-mem-intrinsics.ll
@@ -1,16 +1,16 @@
 ; RUN: opt -S -amdgpu-lower-intrinsics %s | FileCheck -check-prefix=OPT %s
 
-declare void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* nocapture, i8 addrspace(1)* nocapture readonly, i64, i32, i1) #1
-declare void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* nocapture, i8 addrspace(3)* nocapture readonly, i32, i32, i1) #1
+declare void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* nocapture, i8 addrspace(1)* nocapture readonly, i64, i1) #1
+declare void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* nocapture, i8 addrspace(3)* nocapture readonly, i32, i1) #1
 
-declare void @llvm.memmove.p1i8.p1i8.i64(i8 addrspace(1)* nocapture, i8 addrspace(1)* nocapture readonly, i64, i32, i1) #1
-declare void @llvm.memset.p1i8.i64(i8 addrspace(1)* nocapture, i8, i64, i32, i1) #1
+declare void @llvm.memmove.p1i8.p1i8.i64(i8 addrspace(1)* nocapture, i8 addrspace(1)* nocapture readonly, i64, i1) #1
+declare void @llvm.memset.p1i8.i64(i8 addrspace(1)* nocapture, i8, i64, i1) #1
 
 ; Test the upper bound for sizes to leave
 ; OPT-LABEL: @max_size_small_static_memcpy_caller0(
-; OPT: call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1024, i32 1, i1 false)
+; OPT: call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1024, i1 false)
 define amdgpu_kernel void @max_size_small_static_memcpy_caller0(i8 addrspace(1)* %dst, i8 addrspace(1)* %src) #0 {
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1024, i32 1, i1 false)
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1024, i1 false)
   ret void
 }
 
@@ -26,14 +26,14 @@ define amdgpu_kernel void @max_size_small_static_memcpy_caller0(i8 addrspace(1)*
 ; OPT-NEXT: [[T5:%[0-9]+]] = icmp ult i64 [[T4]], 1025
 ; OPT-NEXT: br i1 [[T5]], label %load-store-loop, label %memcpy-split
 define amdgpu_kernel void @min_size_large_static_memcpy_caller0(i8 addrspace(1)* %dst, i8 addrspace(1)* %src) #0 {
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1025, i32 1, i1 false)
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1025, i1 false)
   ret void
 }
 
 ; OPT-LABEL: @max_size_small_static_memmove_caller0(
-; OPT: call void @llvm.memmove.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1024, i32 1, i1 false)
+; OPT: call void @llvm.memmove.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1024, i1 false)
 define amdgpu_kernel void @max_size_small_static_memmove_caller0(i8 addrspace(1)* %dst, i8 addrspace(1)* %src) #0 {
-  call void @llvm.memmove.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1024, i32 1, i1 false)
+  call void @llvm.memmove.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1024, i1 false)
   ret void
 }
 
@@ -44,14 +44,14 @@ define amdgpu_kernel void @max_size_small_static_memmove_caller0(i8 addrspace(1)
 ; OPT: getelementptr
 ; OPT-NEXT: store i8
 define amdgpu_kernel void @min_size_large_static_memmove_caller0(i8 addrspace(1)* %dst, i8 addrspace(1)* %src) #0 {
-  call void @llvm.memmove.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1025, i32 1, i1 false)
+  call void @llvm.memmove.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 1025, i1 false)
   ret void
 }
 
 ; OPT-LABEL: @max_size_small_static_memset_caller0(
-; OPT: call void @llvm.memset.p1i8.i64(i8 addrspace(1)* %dst, i8 %val, i64 1024, i32 1, i1 false)
+; OPT: call void @llvm.memset.p1i8.i64(i8 addrspace(1)* %dst, i8 %val, i64 1024, i1 false)
 define amdgpu_kernel void @max_size_small_static_memset_caller0(i8 addrspace(1)* %dst, i8 %val) #0 {
-  call void @llvm.memset.p1i8.i64(i8 addrspace(1)* %dst, i8 %val, i64 1024, i32 1, i1 false)
+  call void @llvm.memset.p1i8.i64(i8 addrspace(1)* %dst, i8 %val, i64 1024, i1 false)
   ret void
 }
 
@@ -60,7 +60,7 @@ define amdgpu_kernel void @max_size_small_static_memset_caller0(i8 addrspace(1)*
 ; OPT: getelementptr
 ; OPT: store i8
 define amdgpu_kernel void @min_size_large_static_memset_caller0(i8 addrspace(1)* %dst, i8 %val) #0 {
-  call void @llvm.memset.p1i8.i64(i8 addrspace(1)* %dst, i8 %val, i64 1025, i32 1, i1 false)
+  call void @llvm.memset.p1i8.i64(i8 addrspace(1)* %dst, i8 %val, i64 1025, i1 false)
   ret void
 }
 
@@ -68,7 +68,7 @@ define amdgpu_kernel void @min_size_large_static_memset_caller0(i8 addrspace(1)*
 ; OPT-NOT: call
 ; OPT: phi
 define amdgpu_kernel void @variable_memcpy_caller0(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 %n) #0 {
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 %n, i32 1, i1 false)
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 %n, i1 false)
   ret void
 }
 
@@ -76,7 +76,7 @@ define amdgpu_kernel void @variable_memcpy_caller0(i8 addrspace(1)* %dst, i8 add
 ; OPT-NOT: call
 ; OPT: phi
 define amdgpu_kernel void @variable_memcpy_caller1(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 %n) #0 {
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 %n, i32 1, i1 false)
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst, i8 addrspace(1)* %src, i64 %n, i1 false)
   ret void
 }
 
@@ -87,8 +87,8 @@ define amdgpu_kernel void @variable_memcpy_caller1(i8 addrspace(1)* %dst, i8 add
 ; OPT: phi
 ; OPT-NOT: call
 define amdgpu_kernel void @memcpy_multi_use_one_function(i8 addrspace(1)* %dst0, i8 addrspace(1)* %dst1, i8 addrspace(1)* %src, i64 %n, i64 %m) #0 {
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst0, i8 addrspace(1)* %src, i64 %n, i32 1, i1 false)
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst1, i8 addrspace(1)* %src, i64 %m, i32 1, i1 false)
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst0, i8 addrspace(1)* %src, i64 %n, i1 false)
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst1, i8 addrspace(1)* %src, i64 %m, i1 false)
   ret void
 }
 
@@ -99,7 +99,7 @@ define amdgpu_kernel void @memcpy_multi_use_one_function(i8 addrspace(1)* %dst0,
 ; OPT: getelementptr inbounds i8, i8 addrspace(1)*
 ; OPT: store i8
 define amdgpu_kernel void @memcpy_alt_type(i8 addrspace(1)* %dst, i8 addrspace(3)* %src, i32 %n) #0 {
-  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %dst, i8 addrspace(3)* %src, i32 %n, i32 1, i1 false)
+  call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %dst, i8 addrspace(3)* %src, i32 %n, i1 false)
   ret void
 }
 
@@ -110,10 +110,10 @@ define amdgpu_kernel void @memcpy_alt_type(i8 addrspace(1)* %dst, i8 addrspace(3
 ; OPT: getelementptr inbounds i8, i8 addrspace(1)*
 ; OPT: store i8
 
-; OPT: call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst1, i8 addrspace(1)* %src, i64 102, i32 1, i1 false)
+; OPT: call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst1, i8 addrspace(1)* %src, i64 102, i1 false)
 define amdgpu_kernel void @memcpy_multi_use_one_function_keep_small(i8 addrspace(1)* %dst0, i8 addrspace(1)* %dst1, i8 addrspace(1)* %src, i64 %n) #0 {
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst0, i8 addrspace(1)* %src, i64 %n, i32 1, i1 false)
-  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst1, i8 addrspace(1)* %src, i64 102, i32 1, i1 false)
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst0, i8 addrspace(1)* %src, i64 %n, i1 false)
+  call void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* %dst1, i8 addrspace(1)* %src, i64 102, i1 false)
   ret void
 }
 
diff --git a/llvm/test/CodeGen/AMDGPU/promote-alloca-mem-intrinsics.ll b/llvm/test/CodeGen/AMDGPU/promote-alloca-mem-intrinsics.ll
index 7343dd6bbda..fcf64ce8016 100644
--- a/llvm/test/CodeGen/AMDGPU/promote-alloca-mem-intrinsics.ll
+++ b/llvm/test/CodeGen/AMDGPU/promote-alloca-mem-intrinsics.ll
@@ -1,52 +1,52 @@
 ; RUN: opt -S -mtriple=amdgcn-unknown-amdhsa -mcpu=kaveri -amdgpu-promote-alloca < %s | FileCheck %s
 
-declare void @llvm.memcpy.p0i8.p1i8.i32(i8* nocapture, i8 addrspace(1)* nocapture, i32, i32, i1) #0
-declare void @llvm.memcpy.p1i8.p0i8.i32(i8 addrspace(1)* nocapture, i8* nocapture, i32, i32, i1) #0
+declare void @llvm.memcpy.p0i8.p1i8.i32(i8* nocapture, i8 addrspace(1)* nocapture, i32, i1) #0
+declare void @llvm.memcpy.p1i8.p0i8.i32(i8 addrspace(1)* nocapture, i8* nocapture, i32, i1) #0
 
-declare void @llvm.memmove.p0i8.p1i8.i32(i8* nocapture, i8 addrspace(1)* nocapture, i32, i32, i1) #0
-declare void @llvm.memmove.p1i8.p0i8.i32(i8 addrspace(1)* nocapture, i8* nocapture, i32, i32, i1) #0
+declare void @llvm.memmove.p0i8.p1i8.i32(i8* nocapture, i8 addrspace(1)* nocapture, i32, i1) #0
+declare void @llvm.memmove.p1i8.p0i8.i32(i8 addrspace(1)* nocapture, i8* nocapture, i32, i1) #0
 
-declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) #0
+declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) #0
 
 declare i32 @llvm.objectsize.i32.p0i8(i8*, i1, i1) #1
 
 ; CHECK-LABEL: @promote_with_memcpy(
 ; CHECK: getelementptr inbounds [64 x [17 x i32]], [64 x [17 x i32]] addrspace(3)* @promote_with_memcpy.alloca, i32 0, i32 %{{[0-9]+}}
-; CHECK: call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* %alloca.bc, i8 addrspace(1)* %in.bc, i32 68, i32 4, i1 false)
-; CHECK: call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* %out.bc, i8 addrspace(3)* %alloca.bc, i32 68, i32 4, i1 false)
+; CHECK: call void @llvm.memcpy.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %alloca.bc, i8 addrspace(1)* align 4 %in.bc, i32 68, i1 false)
+; CHECK: call void @llvm.memcpy.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out.bc, i8 addrspace(3)* align 4 %alloca.bc, i32 68, i1 false)
 define amdgpu_kernel void @promote_with_memcpy(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0 {
   %alloca = alloca [17 x i32], align 4
   %alloca.bc = bitcast [17 x i32]* %alloca to i8*
   %in.bc = bitcast i32 addrspace(1)* %in to i8 addrspace(1)*
   %out.bc = bitcast i32 addrspace(1)* %out to i8 addrspace(1)*
-  call void @llvm.memcpy.p0i8.p1i8.i32(i8* %alloca.bc, i8 addrspace(1)* %in.bc, i32 68, i32 4, i1 false)
-  call void @llvm.memcpy.p1i8.p0i8.i32(i8 addrspace(1)* %out.bc, i8* %alloca.bc, i32 68, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p1i8.i32(i8* align 4 %alloca.bc, i8 addrspace(1)* align 4 %in.bc, i32 68, i1 false)
+  call void @llvm.memcpy.p1i8.p0i8.i32(i8 addrspace(1)* align 4 %out.bc, i8* align 4 %alloca.bc, i32 68, i1 false)
   ret void
 }
 
 ; CHECK-LABEL: @promote_with_memmove(
 ; CHECK: getelementptr inbounds [64 x [17 x i32]], [64 x [17 x i32]] addrspace(3)* @promote_with_memmove.alloca, i32 0, i32 %{{[0-9]+}}
-; CHECK: call void @llvm.memmove.p3i8.p1i8.i32(i8 addrspace(3)* %alloca.bc, i8 addrspace(1)* %in.bc, i32 68, i32 4, i1 false)
-; CHECK: call void @llvm.memmove.p1i8.p3i8.i32(i8 addrspace(1)* %out.bc, i8 addrspace(3)* %alloca.bc, i32 68, i32 4, i1 false)
+; CHECK: call void @llvm.memmove.p3i8.p1i8.i32(i8 addrspace(3)* align 4 %alloca.bc, i8 addrspace(1)* align 4 %in.bc, i32 68, i1 false)
+; CHECK: call void @llvm.memmove.p1i8.p3i8.i32(i8 addrspace(1)* align 4 %out.bc, i8 addrspace(3)* align 4 %alloca.bc, i32 68, i1 false)
 define amdgpu_kernel void @promote_with_memmove(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0 {
   %alloca = alloca [17 x i32], align 4
   %alloca.bc = bitcast [17 x i32]* %alloca to i8*
   %in.bc = bitcast i32 addrspace(1)* %in to i8 addrspace(1)*
   %out.bc = bitcast i32 addrspace(1)* %out to i8 addrspace(1)*
-  call void @llvm.memmove.p0i8.p1i8.i32(i8* %alloca.bc, i8 addrspace(1)* %in.bc, i32 68, i32 4, i1 false)
-  call void @llvm.memmove.p1i8.p0i8.i32(i8 addrspace(1)* %out.bc, i8* %alloca.bc, i32 68, i32 4, i1 false)
+  call void @llvm.memmove.p0i8.p1i8.i32(i8* align 4 %alloca.bc, i8 addrspace(1)* align 4 %in.bc, i32 68, i1 false)
+  call void @llvm.memmove.p1i8.p0i8.i32(i8 addrspace(1)* align 4 %out.bc, i8* align 4 %alloca.bc, i32 68, i1 false)
   ret void
 }
 
 ; CHECK-LABEL: @promote_with_memset(
 ; CHECK: getelementptr inbounds [64 x [17 x i32]], [64 x [17 x i32]] addrspace(3)* @promote_with_memset.alloca, i32 0, i32 %{{[0-9]+}}
-; CHECK: call void @llvm.memset.p3i8.i32(i8 addrspace(3)* %alloca.bc, i8 7, i32 68, i32 4, i1 false)
+; CHECK: call void @llvm.memset.p3i8.i32(i8 addrspace(3)* align 4 %alloca.bc, i8 7, i32 68, i1 false)
 define amdgpu_kernel void @promote_with_memset(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0 {
   %alloca = alloca [17 x i32], align 4
   %alloca.bc = bitcast [17 x i32]* %alloca to i8*
   %in.bc = bitcast i32 addrspace(1)* %in to i8 addrspace(1)*
   %out.bc = bitcast i32 addrspace(1)* %out to i8 addrspace(1)*
-  call void @llvm.memset.p0i8.i32(i8* %alloca.bc, i8 7, i32 68, i32 4, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* align 4 %alloca.bc, i8 7, i32 68, i1 false)
   ret void
 }
 
diff --git a/llvm/test/CodeGen/AMDGPU/stack-size-overflow.ll b/llvm/test/CodeGen/AMDGPU/stack-size-overflow.ll
index 45a399b058c..322e5ca6219 100644
--- a/llvm/test/CodeGen/AMDGPU/stack-size-overflow.ll
+++ b/llvm/test/CodeGen/AMDGPU/stack-size-overflow.ll
@@ -1,7 +1,7 @@
 ; RUN: not llc -march=amdgcn < %s 2>&1 | FileCheck -check-prefix=ERROR %s
 ; RUN: not llc -march=amdgcn < %s | FileCheck -check-prefix=GCN %s
 
-declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) #1
+declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) #1
 
 ; ERROR: error: stack size limit exceeded (4294967296) in stack_size_limit
 ; GCN: ; ScratchSize: 4294967296
@@ -9,6 +9,6 @@ define amdgpu_kernel void @stack_size_limit() #0 {
 entry:
   %alloca = alloca [1073741823 x i32], align 4
   %bc = bitcast [1073741823 x i32]* %alloca to i8*
-  call void @llvm.memset.p0i8.i32(i8* %bc, i8 9, i32 1073741823, i32 1, i1 true)
+  call void @llvm.memset.p0i8.i32(i8* %bc, i8 9, i32 1073741823, i1 true)
   ret void
 }