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

8 files changed, 29 insertions, 29 deletions

diff --git a/llvm/test/CodeGen/Hexagon/adjust-latency-stackST.ll b/llvm/test/CodeGen/Hexagon/adjust-latency-stackST.ll
index 915db91635f..e8d8364f4ed 100644
--- a/llvm/test/CodeGen/Hexagon/adjust-latency-stackST.ll
+++ b/llvm/test/CodeGen/Hexagon/adjust-latency-stackST.ll
@@ -26,7 +26,7 @@ b1:
   store %struct.0* %v5, %struct.0** %v2, align 4
   %v6 = bitcast %struct.0* %v5 to i8*
   %v7 = load i8*, i8** bitcast (%struct.0** @G to i8**), align 4
-  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %v6, i8* %v7, i32 48, i32 4, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %v6, i8* align 4 %v7, i32 48, i1 false)
   %v8 = getelementptr inbounds %struct.0, %struct.0* %a0, i32 0, i32 2, i32 0, i32 1
   store i32 5, i32* %v8, align 4
   %v9 = getelementptr inbounds %struct.0, %struct.0* %v5, i32 0, i32 2, i32 0, i32 1
@@ -64,14 +64,14 @@ b32:                                              ; preds = %b1
   %v33 = bitcast %struct.0* %a0 to i8**
   %v34 = load i8*, i8** %v33, align 4
   %v35 = bitcast %struct.0* %a0 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %v35, i8* %v34, i32 48, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %v35, i8* align 4 %v34, i32 48, i1 false)
   br label %b36
 
 b36:                                              ; preds = %b32, %b18
   ret i32 undef
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture writeonly, i8* nocapture readonly, i32, i32, i1) #1
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture writeonly, i8* nocapture readonly, i32, i1) #1
 
 declare i32 @f0(...) #0
 declare i32 @f1(...) #0
diff --git a/llvm/test/CodeGen/Hexagon/branchfolder-keep-impdef.ll b/llvm/test/CodeGen/Hexagon/branchfolder-keep-impdef.ll
index e09f7986621..541d9d51142 100644
--- a/llvm/test/CodeGen/Hexagon/branchfolder-keep-impdef.ll
+++ b/llvm/test/CodeGen/Hexagon/branchfolder-keep-impdef.ll
@@ -18,11 +18,11 @@ b1:                                               ; preds = %b0
 b2:                                               ; preds = %b1, %b0
   %t1 = phi i8* [ %t0, %b1 ], [ undef, %b0 ]
   %t2 = getelementptr inbounds i8, i8* %t1, i32 %p0
-  tail call void @llvm.memmove.p0i8.p0i8.i32(i8* undef, i8* %t2, i32 undef, i32 1, i1 false) #1
+  tail call void @llvm.memmove.p0i8.p0i8.i32(i8* undef, i8* %t2, i32 undef, i1 false) #1
   unreachable
 }
 
-declare void @llvm.memmove.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) #0
+declare void @llvm.memmove.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) #0
 
 attributes #0 = { argmemonly nounwind }
 attributes #1 = { nounwind }
diff --git a/llvm/test/CodeGen/Hexagon/early-if-conversion-bug1.ll b/llvm/test/CodeGen/Hexagon/early-if-conversion-bug1.ll
index 6739b03985d..804b083e84b 100644
--- a/llvm/test/CodeGen/Hexagon/early-if-conversion-bug1.ll
+++ b/llvm/test/CodeGen/Hexagon/early-if-conversion-bug1.ll
@@ -23,7 +23,7 @@ target triple = "hexagon"
 %union.anon.0 = type { i8 }
 
 ; Function Attrs: nounwind
-declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) #0
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) #0
 
 declare i32 @__gxx_personality_v0(...)
 
@@ -98,7 +98,7 @@ entry:
 if.then:                                          ; preds = %entry
   %1 = bitcast %"class.std::__1::ostreambuf_iterator"* %retval to i8*
   %2 = bitcast %"class.std::__1::ostreambuf_iterator"* %__s to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %1, i8* %2, i32 4, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %1, i8* align 4 %2, i32 4, i1 false)
   br label %return
 
 if.end:                                           ; preds = %entry
@@ -166,7 +166,7 @@ if.then12:                                        ; preds = %if.then8
   store %"class.std::__1::basic_streambuf"* null, %"class.std::__1::basic_streambuf"** %__sbuf_13, align 4
   %22 = bitcast %"class.std::__1::ostreambuf_iterator"* %retval to i8*
   %23 = bitcast %"class.std::__1::ostreambuf_iterator"* %__s to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %22, i8* %23, i32 4, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %22, i8* align 4 %23, i32 4, i1 false)
   br label %return
 
 if.end14:                                         ; preds = %if.then8
@@ -296,7 +296,7 @@ if.then22:                                        ; preds = %invoke.cont
   store %"class.std::__1::basic_streambuf"* null, %"class.std::__1::basic_streambuf"** %__sbuf_23, align 4
   %53 = bitcast %"class.std::__1::ostreambuf_iterator"* %retval to i8*
   %54 = bitcast %"class.std::__1::ostreambuf_iterator"* %__s to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %53, i8* %54, i32 4, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %53, i8* align 4 %54, i32 4, i1 false)
   store i32 1, i32* %cleanup.dest.slot
   br label %cleanup
 
@@ -361,7 +361,7 @@ if.then34:                                        ; preds = %if.then30
   store %"class.std::__1::basic_streambuf"* null, %"class.std::__1::basic_streambuf"** %__sbuf_35, align 4
   %69 = bitcast %"class.std::__1::ostreambuf_iterator"* %retval to i8*
   %70 = bitcast %"class.std::__1::ostreambuf_iterator"* %__s to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %69, i8* %70, i32 4, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %69, i8* align 4 %70, i32 4, i1 false)
   br label %return
 
 if.end36:                                         ; preds = %if.then30
@@ -381,7 +381,7 @@ if.end37:                                         ; preds = %if.end36, %if.end25
   %74 = load i32, i32* %__r.i, align 4
   %75 = bitcast %"class.std::__1::ostreambuf_iterator"* %retval to i8*
   %76 = bitcast %"class.std::__1::ostreambuf_iterator"* %__s to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %75, i8* %76, i32 4, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %75, i8* align 4 %76, i32 4, i1 false)
   br label %return
 
 return:                                           ; preds = %if.end37, %if.then34, %cleanup, %if.then12, %if.then
diff --git a/llvm/test/CodeGen/Hexagon/mem-fi-add.ll b/llvm/test/CodeGen/Hexagon/mem-fi-add.ll
index a46029fdb5e..4ec62c58170 100644
--- a/llvm/test/CodeGen/Hexagon/mem-fi-add.ll
+++ b/llvm/test/CodeGen/Hexagon/mem-fi-add.ll
@@ -13,14 +13,14 @@ define void @foo() #0 {
 entry:
   %t = alloca [4 x [2 x i32]], align 8
   %0 = bitcast [4 x [2 x i32]]* %t to i8*
-  call void @llvm.memset.p0i8.i32(i8* %0, i8 0, i32 32, i32 8, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* align 8 %0, i8 0, i32 32, i1 false)
   %arraydecay = getelementptr inbounds [4 x [2 x i32]], [4 x [2 x i32]]* %t, i32 0, i32 0
   call void @bar([2 x i32]* %arraydecay) #1
   ret void
 }
 
 ; Function Attrs: nounwind
-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
 
 declare void @bar([2 x i32]*) #2
 
diff --git a/llvm/test/CodeGen/Hexagon/memcpy-likely-aligned.ll b/llvm/test/CodeGen/Hexagon/memcpy-likely-aligned.ll
index f2677efc304..2cce3f12c1d 100644
--- a/llvm/test/CodeGen/Hexagon/memcpy-likely-aligned.ll
+++ b/llvm/test/CodeGen/Hexagon/memcpy-likely-aligned.ll
@@ -25,8 +25,8 @@ entry:
   %m = getelementptr inbounds %struct.n, %struct.n* %p, i32 0, i32 0
   %arraydecay = getelementptr inbounds [2 x %struct.l], [2 x %struct.l]* %m, i32 0, i32 0
   %3 = bitcast %struct.l* %arraydecay to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %3, i8* getelementptr inbounds ({ <{ { %struct.e, { i8, i8, i8, [5 x i8] }, %struct.e }, { %struct.e, { i8, i8, i8, [5 x i8] }, %struct.e } }> }, { <{ { %struct.e, { i8, i8, i8, [5 x i8] }, %struct.e }, { %struct.e, { i8, i8, i8, [5 x i8] }, %struct.e } }> }* @y, i32 0, i32 0, i32 0, i32 0, i32 0), i32 32, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %3, i8* align 4 getelementptr inbounds ({ <{ { %struct.e, { i8, i8, i8, [5 x i8] }, %struct.e }, { %struct.e, { i8, i8, i8, [5 x i8] }, %struct.e } }> }, { <{ { %struct.e, { i8, i8, i8, [5 x i8] }, %struct.e }, { %struct.e, { i8, i8, i8, [5 x i8] }, %struct.e } }> }* @y, i32 0, i32 0, i32 0, i32 0, i32 0), i32 32, i1 false)
   ret void
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i1) nounwind
diff --git a/llvm/test/CodeGen/Hexagon/rdf-filter-defs.ll b/llvm/test/CodeGen/Hexagon/rdf-filter-defs.ll
index 735b20e697f..5c55c500634 100644
--- a/llvm/test/CodeGen/Hexagon/rdf-filter-defs.ll
+++ b/llvm/test/CodeGen/Hexagon/rdf-filter-defs.ll
@@ -120,7 +120,7 @@ b18:
   %t45 = getelementptr inbounds i8, i8* %t42, i32 %p1
   %t46 = load i32, i32* %t0, align 4
   %t47 = sub i32 %t46, %p1
-  tail call void @llvm.memmove.p0i8.p0i8.i32(i8* %t44, i8* %t45, i32 %t47, i32 1, i1 false) #1
+  tail call void @llvm.memmove.p0i8.p0i8.i32(i8* %t44, i8* %t45, i32 %t47, i1 false) #1
   %t48 = icmp eq %type.0* %p0, %p2
   %t49 = load i32, i32* %t22, align 4
   %t50 = icmp ugt i32 %t49, 15
@@ -158,7 +158,7 @@ b25:
   %t61 = select i1 %t60, i32 %t13, i32 0
   %t62 = add i32 %t61, %p3
   %t63 = getelementptr inbounds i8, i8* %t59, i32 %t62
-  tail call void @llvm.memmove.p0i8.p0i8.i32(i8* %t55, i8* %t63, i32 %t13, i32 1, i1 false) #1
+  tail call void @llvm.memmove.p0i8.p0i8.i32(i8* %t55, i8* %t63, i32 %t13, i1 false) #1
   br label %b27
 
 b26:
@@ -171,7 +171,7 @@ b26:
   %t70 = bitcast %type.3* %t67 to i8*
   %t71 = select i1 %t66, i8* %t69, i8* %t70
   %t72 = getelementptr inbounds i8, i8* %t71, i32 %p3
-  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %t55, i8* %t72, i32 %t13, i32 1, i1 false) #1
+  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %t55, i8* %t72, i32 %t13, i1 false) #1
   br label %b27
 
 b27:
@@ -203,8 +203,8 @@ b33:
   ret %type.0* %p0
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture writeonly, i8* nocapture readonly, i32, i32, i1) #0
-declare void @llvm.memmove.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) #0
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture writeonly, i8* nocapture readonly, i32, i1) #0
+declare void @llvm.memmove.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) #0
 
 declare void @blah(%type.4*) local_unnamed_addr
 declare void @danny(%type.4*) local_unnamed_addr
diff --git a/llvm/test/CodeGen/Hexagon/store-imm-stack-object.ll b/llvm/test/CodeGen/Hexagon/store-imm-stack-object.ll
index c0eaea26cc2..5566bda7683 100644
--- a/llvm/test/CodeGen/Hexagon/store-imm-stack-object.ll
+++ b/llvm/test/CodeGen/Hexagon/store-imm-stack-object.ll
@@ -59,11 +59,11 @@ b0:
   store i32 875770417, i32* %v4, align 4
   %v11 = getelementptr inbounds [100 x i8], [100 x i8]* %v5, i32 0, i32 0
   call void @llvm.lifetime.start(i64 100, i8* %v11)
-  call void @llvm.memset.p0i8.i32(i8* %v11, i8 0, i32 100, i32 8, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* align 8 %v11, i8 0, i32 100, i1 false)
   store i8 50, i8* %v11, align 8
   %v12 = getelementptr inbounds [101 x i8], [101 x i8]* %v6, i32 0, i32 0
   call void @llvm.lifetime.start(i64 101, i8* %v12)
-  call void @llvm.memset.p0i8.i32(i8* %v12, i8 0, i32 101, i32 8, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* align 8 %v12, i8 0, i32 101, i1 false)
   store i8 49, i8* %v12, align 8
   call void @test3(i8* %v7, i8* %v8, i8* %v9, i8* %v10, i8* %v11, i8* %v12)
   call void @llvm.lifetime.end(i64 101, i8* %v12)
@@ -77,7 +77,7 @@ b0:
 
 declare void @llvm.lifetime.start(i64, i8* nocapture) #0
 declare void @llvm.lifetime.end(i64, i8* nocapture) #0
-declare void @llvm.memset.p0i8.i32(i8* nocapture writeonly, i8, i32, i32, i1) #0
+declare void @llvm.memset.p0i8.i32(i8* nocapture writeonly, i8, i32, i1) #0
 
 declare void @test3(i8*, i8*, i8*, i8*, i8*, i8*)
 declare void @test4(i8*, i8*, i8*, i8*)
diff --git a/llvm/test/CodeGen/Hexagon/tail-call-mem-intrinsics.ll b/llvm/test/CodeGen/Hexagon/tail-call-mem-intrinsics.ll
index 90fb75e5be0..7f0fb6281ff 100644
--- a/llvm/test/CodeGen/Hexagon/tail-call-mem-intrinsics.ll
+++ b/llvm/test/CodeGen/Hexagon/tail-call-mem-intrinsics.ll
@@ -4,7 +4,7 @@
 ; CHECK: jump memcpy
 define void @tail_memcpy(i8* nocapture %p, i8* nocapture readonly %q, i32 %n) #0 {
 entry:
-  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %p, i8* %q, i32 %n, i32 1, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %p, i8* %q, i32 %n, i1 false)
   ret void
 }
 
@@ -12,7 +12,7 @@ entry:
 ; CHECK: jump memmove
 define void @tail_memmove(i8* nocapture %p, i8* nocapture readonly %q, i32 %n) #0 {
 entry:
-  tail call void @llvm.memmove.p0i8.p0i8.i32(i8* %p, i8* %q, i32 %n, i32 1, i1 false)
+  tail call void @llvm.memmove.p0i8.p0i8.i32(i8* %p, i8* %q, i32 %n, i1 false)
   ret void
 }
 
@@ -20,12 +20,12 @@ entry:
 ; CHECK: jump memset
 define void @tail_memset(i8* nocapture %p, i8 %c, i32 %n) #0 {
 entry:
-  tail call void @llvm.memset.p0i8.i32(i8* %p, i8 %c, i32 %n, i32 1, i1 false)
+  tail call void @llvm.memset.p0i8.i32(i8* %p, i8 %c, i32 %n, i1 false)
   ret void
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) #0
-declare void @llvm.memmove.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) #0
-declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) #0
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) #0
+declare void @llvm.memmove.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) #0
+declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) #0
 
 attributes #0 = { nounwind }