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

26 files changed, 110 insertions, 110 deletions

diff --git a/llvm/test/CodeGen/PowerPC/2011-12-05-NoSpillDupCR.ll b/llvm/test/CodeGen/PowerPC/2011-12-05-NoSpillDupCR.ll
index e87fca07410..14a89b03495 100644
--- a/llvm/test/CodeGen/PowerPC/2011-12-05-NoSpillDupCR.ll
+++ b/llvm/test/CodeGen/PowerPC/2011-12-05-NoSpillDupCR.ll
@@ -179,7 +179,7 @@ for.end.7:                                        ; preds = %entry, %for.end.7
   br i1 %exitcond.7, label %for.end12, label %for.end.7
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i1) nounwind
 
 declare i32 @puts(i8* nocapture) nounwind
 
diff --git a/llvm/test/CodeGen/PowerPC/2011-12-06-SpillAndRestoreCR.ll b/llvm/test/CodeGen/PowerPC/2011-12-06-SpillAndRestoreCR.ll
index ca752f568e0..02da82d4029 100644
--- a/llvm/test/CodeGen/PowerPC/2011-12-06-SpillAndRestoreCR.ll
+++ b/llvm/test/CodeGen/PowerPC/2011-12-06-SpillAndRestoreCR.ll
@@ -213,7 +213,7 @@ for.end23:                                        ; preds = %for.end17
   ret i32 0
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i1) nounwind
 
 declare i32 @puts(i8* nocapture) nounwind
 
diff --git a/llvm/test/CodeGen/PowerPC/MMO-flags-assertion.ll b/llvm/test/CodeGen/PowerPC/MMO-flags-assertion.ll
index ab9f76f4609..64fa85a8195 100644
--- a/llvm/test/CodeGen/PowerPC/MMO-flags-assertion.ll
+++ b/llvm/test/CodeGen/PowerPC/MMO-flags-assertion.ll
@@ -4,7 +4,7 @@
 ; Assertion `MMO->getFlags() == getFlags() && "Flags mismatch !"' failed.
 
 declare void @_Z3fn11F(%class.F* byval align 8) local_unnamed_addr
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1)
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i1)
 declare signext i32 @_ZN1F11isGlobalRegEv(%class.F*) local_unnamed_addr
 declare void @llvm.lifetime.start.p0i8(i64, i8* nocapture)
 declare void @_Z10EmitLValuev(%class.F* sret) local_unnamed_addr
@@ -28,7 +28,7 @@ entry:
   call void @_Z10EmitLValuev(%class.F* nonnull sret %XLValue)
   %1 = bitcast %class.F* %agg.tmp1 to i8*
   call void @llvm.lifetime.start.p0i8(i64 96, i8* nonnull %1)
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull %1, i8* nonnull %0, i64 96, i32 8, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 nonnull %1, i8* align 8 nonnull %0, i64 96, i1 false)
   call void @_Z3fn11F(%class.F* byval nonnull align 8 %XLValue)
   %call.i = call signext i32 @_ZN1F11isGlobalRegEv(%class.F* nonnull %agg.tmp1)
   call void @llvm.lifetime.end.p0i8(i64 96, i8* nonnull %1)
diff --git a/llvm/test/CodeGen/PowerPC/aantidep-inline-asm-use.ll b/llvm/test/CodeGen/PowerPC/aantidep-inline-asm-use.ll
index f0c0deacf4d..d31a5553bf9 100644
--- a/llvm/test/CodeGen/PowerPC/aantidep-inline-asm-use.ll
+++ b/llvm/test/CodeGen/PowerPC/aantidep-inline-asm-use.ll
@@ -112,7 +112,7 @@ _ZN10SubProcess12SafeSyscalls5closeEi.exit22:     ; preds = %_ZN10SubProcess12Sa
   br label %.thread
 
 .thread:                                          ; preds = %45, %.thread.outer
-  call void @llvm.memset.p0i8.i64(i8* undef, i8 0, i64 56, i32 8, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* align 8 undef, i8 0, i64 56, i1 false)
   store i8* %21, i8** undef, align 8
   store i32 1073741824, i32* undef, align 8
   %22 = call { i64, i64, i64, i64, i64, i64, i64 } asm sideeffect "sc\0A\09mfcr $0", "=&{r0},=&{r3},=&{r4},=&{r5},=&{r6},=&{r7},=&{r8},{r0},{r3},{r4},{r5},~{cr0},~{ctr},~{memory},~{r11},~{r12}"(i64 342, i64 80871424, i64 undef, i64 0) #2, !srcloc !1
@@ -296,7 +296,7 @@ _ZN10SubProcess12SafeSyscalls5fcntlEiil.exit:     ; preds = %_ZN10SubProcess12Sa
 }
 
 ; Function Attrs: nounwind argmemonly
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) #1
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) #1
 
 attributes #0 = { nounwind "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="pwr8" "target-features"="+altivec,+bpermd,+crypto,+direct-move,+extdiv,+power8-vector,+vsx,-qpx" "unsafe-fp-math"="false" "use-soft-float"="false" }
 attributes #1 = { nounwind argmemonly }
diff --git a/llvm/test/CodeGen/PowerPC/ctrloop-reg.ll b/llvm/test/CodeGen/PowerPC/ctrloop-reg.ll
index 477d2aacdf4..7a7d81537bf 100644
--- a/llvm/test/CodeGen/PowerPC/ctrloop-reg.ll
+++ b/llvm/test/CodeGen/PowerPC/ctrloop-reg.ll
@@ -74,7 +74,7 @@ declare i32 @interp(%struct.ref_s.1.49.91.115.121.139.145.151.157.163.169.175.18
 
 declare i32 @dict_lookup(%struct.ref_s.1.49.91.115.121.139.145.151.157.163.169.175.181.211*, %struct.ref_s.1.49.91.115.121.139.145.151.157.163.169.175.181.211*, %struct.ref_s.1.49.91.115.121.139.145.151.157.163.169.175.181.211*, %struct.ref_s.1.49.91.115.121.139.145.151.157.163.169.175.181.211**)
 
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i1) nounwind
 
 declare i32 @obj_compare(...)
 
diff --git a/llvm/test/CodeGen/PowerPC/emptystruct.ll b/llvm/test/CodeGen/PowerPC/emptystruct.ll
index bd8a974ab12..b0e41ec29e4 100644
--- a/llvm/test/CodeGen/PowerPC/emptystruct.ll
+++ b/llvm/test/CodeGen/PowerPC/emptystruct.ll
@@ -21,7 +21,7 @@ entry:
   %0 = load %struct.empty*, %struct.empty** %a2.addr, align 8
   %1 = bitcast %struct.empty* %agg.result to i8*
   %2 = bitcast %struct.empty* %0 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %2, i64 0, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %2, i64 0, i1 false)
   ret void
 }
 
@@ -31,7 +31,7 @@ entry:
 ; CHECK-NOT: std 6,
 ; CHECK: blr
 
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i1) nounwind
 
 define void @caller(%struct.empty* noalias sret %agg.result) nounwind {
 entry:
diff --git a/llvm/test/CodeGen/PowerPC/fsl-e500mc.ll b/llvm/test/CodeGen/PowerPC/fsl-e500mc.ll
index fe3e19b72d1..b1bb09da3e1 100644
--- a/llvm/test/CodeGen/PowerPC/fsl-e500mc.ll
+++ b/llvm/test/CodeGen/PowerPC/fsl-e500mc.ll
@@ -15,8 +15,8 @@ entry:
 ; CHECK-NOT: bl memcpy
   %0 = bitcast %struct.teststruct* %agg.result to i8*
   %1 = bitcast %struct.teststruct* %in to i8*
-  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %0, i8* %1, i32 52, i32 4, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %0, i8* align 4 %1, i32 52, 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/PowerPC/fsl-e5500.ll b/llvm/test/CodeGen/PowerPC/fsl-e5500.ll
index dae47fb037f..595d91ad620 100644
--- a/llvm/test/CodeGen/PowerPC/fsl-e5500.ll
+++ b/llvm/test/CodeGen/PowerPC/fsl-e5500.ll
@@ -15,8 +15,8 @@ entry:
 ; CHECK-NOT: bl memcpy
   %0 = bitcast %struct.teststruct* %agg.result to i8*
   %1 = bitcast %struct.teststruct* %in to i8*
-  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* %1, i64 100, i32 4, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %0, i8* align 4 %1, i64 100, i1 false)
   ret void
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i1) nounwind
diff --git a/llvm/test/CodeGen/PowerPC/glob-comp-aa-crash.ll b/llvm/test/CodeGen/PowerPC/glob-comp-aa-crash.ll
index 51275f3cdae..9f0e706da32 100644
--- a/llvm/test/CodeGen/PowerPC/glob-comp-aa-crash.ll
+++ b/llvm/test/CodeGen/PowerPC/glob-comp-aa-crash.ll
@@ -35,7 +35,7 @@ entry:
 invoke.cont:                                      ; preds = %entry
   %__exception_ = getelementptr inbounds %"class.std::__1::__assoc_sub_state", %"class.std::__1::__assoc_sub_state"* %this, i64 0, i32 1
   %0 = bitcast { i64, i64 }* %tmp to i8*
-  call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 16, i32 8, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* align 8 %0, i8 0, i64 16, i1 false)
   call void @_ZNSt15__exception_ptr13exception_ptrC1EMS0_FvvE(%"class.std::__exception_ptr::exception_ptr"* %ref.tmp, { i64, i64 }* byval %tmp) #5
   %call = call zeroext i1 @_ZNSt15__exception_ptrneERKNS_13exception_ptrES2_(%"class.std::__exception_ptr::exception_ptr"* %__exception_, %"class.std::__exception_ptr::exception_ptr"* %ref.tmp) #5
   call void @_ZNSt15__exception_ptr13exception_ptrD1Ev(%"class.std::__exception_ptr::exception_ptr"* %ref.tmp) #5
@@ -120,7 +120,7 @@ declare void @_ZNSt3__15mutex6unlockEv(%"class.std::__1::mutex"*) #1
 declare void @_ZNSt3__15mutex4lockEv(%"class.std::__1::mutex"*) #0
 
 ; Function Attrs: nounwind
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) #3
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) #3
 
 attributes #0 = { optsize "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
 attributes #1 = { nounwind optsize "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
diff --git a/llvm/test/CodeGen/PowerPC/isel-rc-nox0.ll b/llvm/test/CodeGen/PowerPC/isel-rc-nox0.ll
index e3479f8586f..582778f01dd 100644
--- a/llvm/test/CodeGen/PowerPC/isel-rc-nox0.ll
+++ b/llvm/test/CodeGen/PowerPC/isel-rc-nox0.ll
@@ -21,7 +21,7 @@ crc32_gentab.exit:                                ; preds = %for.cond1.preheader
   br label %for.cond1.preheader.i2961.i
 
 for.cond1.preheader.i2961.i:                      ; preds = %for.inc44.i2977.i, %crc32_gentab.exit
-  call void @llvm.memset.p0i8.i64(i8* bitcast ([1 x [9 x i32]]* @g_62 to i8*), i8 -1, i64 36, i32 4, i1 false) #1
+  call void @llvm.memset.p0i8.i64(i8* align 4 bitcast ([1 x [9 x i32]]* @g_62 to i8*), i8 -1, i64 36, i1 false) #1
   %0 = load i32, i32* %retval.0.i.i.i, align 4
   %tobool.i2967.i = icmp eq i32 %0, 0
   br label %for.body21.i2968.i
@@ -40,7 +40,7 @@ func_80.exit2978.i:                               ; preds = %for.inc44.i2977.i
 }
 
 ; Function Attrs: nounwind
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) #1
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) #1
 
 attributes #0 = { nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "ssp-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
 attributes #1 = { nounwind }
diff --git a/llvm/test/CodeGen/PowerPC/licm-remat.ll b/llvm/test/CodeGen/PowerPC/licm-remat.ll
index f9c14052452..e72a8b0cd3e 100644
--- a/llvm/test/CodeGen/PowerPC/licm-remat.ll
+++ b/llvm/test/CodeGen/PowerPC/licm-remat.ll
@@ -13,9 +13,9 @@
 @_ZN6snappy8internalL8wordmaskE = internal unnamed_addr constant [5 x i32] [i32 0, i32 255, i32 65535, i32 16777215, i32 -1], align 4
 
 ; Function Attrs: argmemonly nounwind
-declare void @llvm.memmove.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1) #2
+declare void @llvm.memmove.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1) #2
 ; Function Attrs: argmemonly nounwind
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1) #2
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i1) #2
 
 define linkonce_odr void @ZN6snappyDecompressor_(%"class.snappy::SnappyDecompressor"* %this, %"class.snappy::SnappyIOVecWriter"* %writer) {
 ; CHECK-LABEL: ZN6snappyDecompressor_:
@@ -126,7 +126,7 @@ if.end18.i207:                                    ; preds = %if.then10.i193, %co
   %iov_base.i.i200 = getelementptr inbounds %"struct.snappy::iovec", %"struct.snappy::iovec"* %12, i64 %17, i32 0
   %18 = load i8*, i8** %iov_base.i.i200, align 8
   %add.ptr.i.i201 = getelementptr inbounds i8, i8* %18, i64 %15
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %add.ptr.i.i201, i8* %add.ptr24, i64 %.sroa.speculated.i199, i32 1, i1 false) #12
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %add.ptr.i.i201, i8* %add.ptr24, i64 %.sroa.speculated.i199, i1 false) #12
   %add30.i203 = add i64 0, %.sroa.speculated.i199
   store i64 %add30.i203, i64* null, align 8
   %.pre245 = load i64, i64* %0, align 8
@@ -164,7 +164,7 @@ cleanup102:                                       ; preds = %land.lhs.true5.i
   %iov_base.i.i = getelementptr inbounds %"struct.snappy::iovec", %"struct.snappy::iovec"* %7, i64 %8, i32 0
   %23 = load i8*, i8** %iov_base.i.i, align 8
   %add.ptr.i.i = getelementptr inbounds i8, i8* %23, i64 %9
-  call void @llvm.memmove.p0i8.p0i8.i64(i8* %add.ptr.i.i, i8* %incdec.ptr, i64 16, i32 1, i1 false) #12
+  call void @llvm.memmove.p0i8.p0i8.i64(i8* %add.ptr.i.i, i8* %incdec.ptr, i64 16, i1 false) #12
   %24 = load <2 x i64>, <2 x i64>* %1, align 8
   %25 = insertelement <2 x i64> undef, i64 %conv9, i32 0
   %26 = shufflevector <2 x i64> %25, <2 x i64> undef, <2 x i32> zeroinitializer
diff --git a/llvm/test/CodeGen/PowerPC/lxv-aligned-stack-slots.ll b/llvm/test/CodeGen/PowerPC/lxv-aligned-stack-slots.ll
index 9c5432578b4..e8b65a38075 100644
--- a/llvm/test/CodeGen/PowerPC/lxv-aligned-stack-slots.ll
+++ b/llvm/test/CodeGen/PowerPC/lxv-aligned-stack-slots.ll
@@ -24,7 +24,7 @@ define void @unaligned_slot() #0 {
   %1 = alloca %class2, align 8
   %2 = getelementptr inbounds %class2, %class2* %1, i64 0, i32 0, i32 0, i32 2
   %3 = bitcast %union.anon* %2 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull getelementptr inbounds (%class1, %class1* @ext, i64 0, i32 0, i32 1, i64 8), i8* nonnull %3, i64 16, i32 8, i1 false) #2
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 nonnull getelementptr inbounds (%class1, %class1* @ext, i64 0, i32 0, i32 1, i64 8), i8* align 8 nonnull %3, i64 16, i1 false) #2
   ret void
 }
 ; CHECK-LABEL: aligned_slot:
@@ -34,12 +34,12 @@ define void @aligned_slot() #0 {
   %1 = alloca %class2, align 16
   %2 = getelementptr inbounds %class2, %class2* %1, i64 0, i32 0, i32 0, i32 2
   %3 = bitcast %union.anon* %2 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull getelementptr inbounds (%class1, %class1* @ext, i64 0, i32 0, i32 1, i64 8), i8* nonnull %3, i64 16, i32 8, i1 false) #2
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 nonnull getelementptr inbounds (%class1, %class1* @ext, i64 0, i32 0, i32 1, i64 8), i8* align 8 nonnull %3, i64 16, i1 false) #2
   ret void
 }
 
 ; Function Attrs: argmemonly nounwind
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1) #1
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i1) #1
 
 attributes #0 = { nounwind "target-cpu"="pwr9" "target-features"="+altivec,+bpermd,+crypto,+direct-move,+extdiv,+htm,+power8-vector,+power9-vector,+vsx,-qpx" "unsafe-fp-math"="false" "use-soft-float"="false" }
 attributes #1 = { argmemonly nounwind }
diff --git a/llvm/test/CodeGen/PowerPC/memcpy-vec.ll b/llvm/test/CodeGen/PowerPC/memcpy-vec.ll
index 3046b26e76c..d97d604128e 100644
--- a/llvm/test/CodeGen/PowerPC/memcpy-vec.ll
+++ b/llvm/test/CodeGen/PowerPC/memcpy-vec.ll
@@ -9,7 +9,7 @@ define void @foo1(double* nocapture %x, double* nocapture readonly %y) #0 {
 entry:
   %0 = bitcast double* %x to i8*
   %1 = bitcast double* %y to i8*
-  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* %1, i64 32, i32 8, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %0, i8* align 8 %1, i64 32, i1 false)
   ret void
 
 ; PWR7-LABEL: @foo1
@@ -34,14 +34,14 @@ entry:
 }
 
 ; Function Attrs: nounwind
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1) #0
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1) #0
 
 ; Function Attrs: nounwind
 define void @foo2(double* nocapture %x, double* nocapture readonly %y) #0 {
 entry:
   %0 = bitcast double* %x to i8*
   %1 = bitcast double* %y to i8*
-  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* %1, i64 128, i32 8, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %0, i8* align 8 %1, i64 128, i1 false)
   ret void
 
 ; PWR7-LABEL: @foo2
@@ -64,7 +64,7 @@ entry:
 define void @bar1(double* nocapture %x) #0 {
 entry:
   %0 = bitcast double* %x to i8*
-  tail call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 128, i32 8, i1 false)
+  tail call void @llvm.memset.p0i8.i64(i8* align 8 %0, i8 0, i64 128, i1 false)
   ret void
 
 ; PWR7-LABEL: @bar1
@@ -87,7 +87,7 @@ entry:
 define void @bar2(double* nocapture %x) #0 {
 entry:
   %0 = bitcast double* %x to i8*
-  tail call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 128, i32 32, i1 false)
+  tail call void @llvm.memset.p0i8.i64(i8* align 32 %0, i8 0, i64 128, i1 false)
   ret void
 
 ; PWR7-LABEL: @bar2
@@ -107,7 +107,7 @@ entry:
 }
 
 ; Function Attrs: nounwind
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) #0
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) #0
 
 attributes #0 = { nounwind }
 
diff --git a/llvm/test/CodeGen/PowerPC/memcpy_dereferenceable.ll b/llvm/test/CodeGen/PowerPC/memcpy_dereferenceable.ll
index ed821849f09..36d37f294ee 100644
--- a/llvm/test/CodeGen/PowerPC/memcpy_dereferenceable.ll
+++ b/llvm/test/CodeGen/PowerPC/memcpy_dereferenceable.ll
@@ -22,7 +22,7 @@ entry:
 
 end:
   ; copy third element into first element by memcpy
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull %dst, i8* %src, i64 16, i32 8, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 nonnull %dst, i8* align 8 %src, i64 16, i1 false)
   ; copy third element into second element by LD/ST
   %vec2 = load <2 x i64>, <2 x i64>* %pvec2, align 8
   store <2 x i64> %vec2, <2 x i64>* %pvec1, align 8
@@ -30,7 +30,7 @@ end:
 
 dummy:
   ; to make use of %src in another BB
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %src, i8* %src, i64 0, i32 0, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %src, i8* %src, i64 0, i1 false)
   br label %end
 }
 
@@ -55,7 +55,7 @@ entry:
 
 end:
   ; copy third element into first element by memcpy
-  call void @llvm.memmove.p0i8.p0i8.i64(i8* nonnull %dst, i8* %src, i64 16, i32 8, i1 false)
+  call void @llvm.memmove.p0i8.p0i8.i64(i8* align 8 nonnull %dst, i8* align 8 %src, i64 16, i1 false)
   ; copy third element into second element by LD/ST
   %vec2 = load <2 x i64>, <2 x i64>* %pvec2, align 8
   store <2 x i64> %vec2, <2 x i64>* %pvec1, align 8
@@ -63,12 +63,12 @@ end:
 
 dummy:
   ; to make use of %src in another BB
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %src, i8* %src, i64 0, i32 0, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %src, i8* %src, i64 0, i1 false)
   br label %end
 }
 
 ; Function Attrs: argmemonly nounwind
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1) #1
-declare void @llvm.memmove.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1) #1
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i1) #1
+declare void @llvm.memmove.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i1) #1
 
 attributes #1 = { argmemonly nounwind }
diff --git a/llvm/test/CodeGen/PowerPC/memset-nc-le.ll b/llvm/test/CodeGen/PowerPC/memset-nc-le.ll
index cd6253f50e5..2924ff3d44b 100644
--- a/llvm/test/CodeGen/PowerPC/memset-nc-le.ll
+++ b/llvm/test/CodeGen/PowerPC/memset-nc-le.ll
@@ -7,7 +7,7 @@ define void @test_vsx() unnamed_addr #0 align 2 {
 entry:
   %0 = load i32, i32* undef, align 4
   %1 = trunc i32 %0 to i8
-  call void @llvm.memset.p0i8.i64(i8* null, i8 %1, i64 32, i32 1, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* null, i8 %1, i64 32, i1 false)
   ret void
 
 ; CHECK-LABEL: @test_vsx
@@ -17,7 +17,7 @@ entry:
 }
 
 ; Function Attrs: nounwind
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) #1
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) #1
 
 attributes #0 = { nounwind "target-cpu"="pwr8" }
 attributes #1 = { nounwind }
diff --git a/llvm/test/CodeGen/PowerPC/memset-nc.ll b/llvm/test/CodeGen/PowerPC/memset-nc.ll
index fd4327ef4d6..663d0cb1d67 100644
--- a/llvm/test/CodeGen/PowerPC/memset-nc.ll
+++ b/llvm/test/CodeGen/PowerPC/memset-nc.ll
@@ -8,7 +8,7 @@ define void @test_qpx() unnamed_addr #0 align 2 {
 entry:
   %0 = load i32, i32* undef, align 4
   %1 = trunc i32 %0 to i8
-  call void @llvm.memset.p0i8.i64(i8* null, i8 %1, i64 64, i32 32, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* align 32 null, i8 %1, i64 64, i1 false)
   ret void
 
 ; CHECK-LABEL: @test_qpx
@@ -22,14 +22,14 @@ entry:
 }
 
 ; Function Attrs: nounwind
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) #1
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) #1
 
 ; Function Attrs: nounwind
 define void @test_vsx() unnamed_addr #2 align 2 {
 entry:
   %0 = load i32, i32* undef, align 4
   %1 = trunc i32 %0 to i8
-  call void @llvm.memset.p0i8.i64(i8* null, i8 %1, i64 32, i32 1, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* null, i8 %1, i64 32, i1 false)
   ret void
 
 ; CHECK-LABEL: @test_vsx
diff --git a/llvm/test/CodeGen/PowerPC/merge-st-chain-op.ll b/llvm/test/CodeGen/PowerPC/merge-st-chain-op.ll
index bfb911c0115..4d5b9170ece 100644
--- a/llvm/test/CodeGen/PowerPC/merge-st-chain-op.ll
+++ b/llvm/test/CodeGen/PowerPC/merge-st-chain-op.ll
@@ -17,7 +17,7 @@ _ZN4llvm18IntrusiveRefCntPtrIN5clang13DiagnosticIDsEEC2EPS2_.exit: ; preds = %en
   store <2 x i8*> <i8* bitcast (i64* getelementptr inbounds ([0 x i64], [0 x i64]* @_ZNSs4_Rep20_S_empty_rep_storageE, i64 0, i64 3) to i8*), i8* bitcast (i64* getelementptr inbounds ([0 x i64], [0 x i64]* @_ZNSs4_Rep20_S_empty_rep_storageE, i64 0, i64 3) to i8*)>, <2 x i8*>* undef, align 8
   %IgnoreWarnings.i = getelementptr inbounds i8, i8* %call2, i64 4
   %0 = bitcast i8* %IgnoreWarnings.i to i32*
-  call void @llvm.memset.p0i8.i64(i8* null, i8 0, i64 48, i32 8, i1 false) #4
+  call void @llvm.memset.p0i8.i64(i8* align 8 null, i8 0, i64 48, i1 false) #4
   store i32 251658240, i32* %0, align 4
   store i256 37662610426935100959726589394453639584271499769928088551424, i256* null, align 8
   store i32 1, i32* %ref_cnt.i.i, align 4
@@ -31,7 +31,7 @@ return:                                           ; preds = %entry
 declare noalias i8* @_Znwm() #1
 
 ; Function Attrs: nounwind argmemonly
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) #2
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) #2
 
 attributes #0 = { nounwind "target-cpu"="pwr7" }
 attributes #1 = { nobuiltin "target-cpu"="pwr7" }
diff --git a/llvm/test/CodeGen/PowerPC/ppc-empty-fs.ll b/llvm/test/CodeGen/PowerPC/ppc-empty-fs.ll
index c89ca2ea2ab..8b600893225 100644
--- a/llvm/test/CodeGen/PowerPC/ppc-empty-fs.ll
+++ b/llvm/test/CodeGen/PowerPC/ppc-empty-fs.ll
@@ -15,14 +15,14 @@ entry:
   store i64 %x.coerce, i64* %0, align 1
   %1 = bitcast %struct.fab* %agg.result to i8*
   %2 = bitcast %struct.fab* %x to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %2, i64 8, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %1, i8* align 4 %2, i64 8, i1 false)
   ret void
 }
 
 ; CHECK: func_fab
 
 ; Function Attrs: nounwind
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1) #1
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1) #1
 
 attributes #0 = { nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "no-realign-stack" "stack-protector-buffer-size"="8" "target-features"="" "unsafe-fp-math"="false" "use-soft-float"="false" }
 attributes #1 = { nounwind }
diff --git a/llvm/test/CodeGen/PowerPC/pr27350.ll b/llvm/test/CodeGen/PowerPC/pr27350.ll
index f6ad38e2f29..7dbd5110700 100644
--- a/llvm/test/CodeGen/PowerPC/pr27350.ll
+++ b/llvm/test/CodeGen/PowerPC/pr27350.ll
@@ -1,12 +1,12 @@
 ; RUN: llc -verify-machineinstrs -mcpu=ppc64le -mtriple=powerpc64le-unknown-linux-gnu < %s
 
 ; Function Attrs: argmemonly nounwind
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1) #0
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1) #0
 
 ; Function Attrs: nounwind
 define internal fastcc void @foo() unnamed_addr #1 align 2 {
 entry:
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* undef, i8* null, i64 16, i32 8, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 undef, i8* align 8 null, i64 16, i1 false)
   %0 = load <2 x i64>, <2 x i64>* null, align 8
   %1 = extractelement <2 x i64> %0, i32 1
   %.fca.1.insert159.i = insertvalue [2 x i64] undef, i64 %1, 1
diff --git a/llvm/test/CodeGen/PowerPC/resolvefi-basereg.ll b/llvm/test/CodeGen/PowerPC/resolvefi-basereg.ll
index 9e83f0979fe..731f37d2770 100644
--- a/llvm/test/CodeGen/PowerPC/resolvefi-basereg.ll
+++ b/llvm/test/CodeGen/PowerPC/resolvefi-basereg.ll
@@ -32,9 +32,9 @@ entry:
   %agg.tmp117 = alloca %struct.S1998, align 16
   %agg.tmp118 = alloca %struct.S1998, align 16
   %agg.tmp119 = alloca %struct.S1998, align 16
-  call void @llvm.memset.p0i8.i64(i8* bitcast (%struct.S1998* @s1998 to i8*), i8 0, i64 5168, i32 16, i1 false)
-  call void @llvm.memset.p0i8.i64(i8* bitcast ([5 x %struct.S1998]* @a1998 to i8*), i8 0, i64 25840, i32 16, i1 false)
-  call void @llvm.memset.p0i8.i64(i8* bitcast (%struct.Info* @info to i8*), i8 0, i64 832, i32 8, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* align 16 bitcast (%struct.S1998* @s1998 to i8*), i8 0, i64 5168, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* align 16 bitcast ([5 x %struct.S1998]* @a1998 to i8*), i8 0, i64 25840, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* align 8 bitcast (%struct.Info* @info to i8*), i8 0, i64 832, i1 false)
   store i8* bitcast (%struct.S1998* @s1998 to i8*), i8** getelementptr inbounds (%struct.Info, %struct.Info* @info, i32 0, i32 2), align 8
   store i8* bitcast ([5 x %struct.S1998]* @a1998 to i8*), i8** getelementptr inbounds (%struct.Info, %struct.Info* @info, i32 0, i32 3), align 8
   store i8* bitcast (%struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 3) to i8*), i8** getelementptr inbounds (%struct.Info, %struct.Info* @info, i32 0, i32 4), align 8
@@ -329,32 +329,32 @@ if.end:                                           ; preds = %if.then, %entry
   %61 = load i32, i32* %j, align 4
   store i32 %61, i32* getelementptr inbounds (%struct.Info, %struct.Info* @info, i32 0, i32 1), align 4
   %62 = bitcast %struct.S1998* %agg.tmp111 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %62, i8* bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %62, i8* align 16 bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i1 false)
   %63 = bitcast %struct.S1998* %agg.tmp112 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %63, i8* bitcast (%struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 2) to i8*), i64 5168, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %63, i8* align 16 bitcast (%struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 2) to i8*), i64 5168, i1 false)
   call void @check1998(%struct.S1998* sret %agg.tmp, %struct.S1998* byval align 16 %agg.tmp111, %struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 1), %struct.S1998* byval align 16 %agg.tmp112)
   call void @checkx1998(%struct.S1998* byval align 16 %agg.tmp)
   %64 = bitcast %struct.S1998* %agg.tmp113 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %64, i8* bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %64, i8* align 16 bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i1 false)
   %65 = bitcast %struct.S1998* %agg.tmp114 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %65, i8* bitcast (%struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 2) to i8*), i64 5168, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %65, i8* align 16 bitcast (%struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 2) to i8*), i64 5168, i1 false)
   %66 = bitcast %struct.S1998* %agg.tmp115 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %66, i8* bitcast (%struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 2) to i8*), i64 5168, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %66, i8* align 16 bitcast (%struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 2) to i8*), i64 5168, i1 false)
   call void (i32, ...) @check1998va(i32 signext 1, double 1.000000e+00, %struct.S1998* byval align 16 %agg.tmp113, i64 2, %struct.S1998* byval align 16 %agg.tmp114, %struct.S1998* byval align 16 %agg.tmp115)
   %67 = bitcast %struct.S1998* %agg.tmp116 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %67, i8* bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %67, i8* align 16 bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i1 false)
   %68 = bitcast %struct.S1998* %agg.tmp117 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %68, i8* bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %68, i8* align 16 bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i1 false)
   %69 = bitcast %struct.S1998* %agg.tmp118 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %69, i8* bitcast (%struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 2) to i8*), i64 5168, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %69, i8* align 16 bitcast (%struct.S1998* getelementptr inbounds ([5 x %struct.S1998], [5 x %struct.S1998]* @a1998, i32 0, i64 2) to i8*), i64 5168, i1 false)
   %70 = bitcast %struct.S1998* %agg.tmp119 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %70, i8* bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %70, i8* align 16 bitcast (%struct.S1998* @s1998 to i8*), i64 5168, i1 false)
   call void (i32, ...) @check1998va(i32 signext 2, %struct.S1998* byval align 16 %agg.tmp116, %struct.S1998* byval align 16 %agg.tmp117, ppc_fp128 0xM40000000000000000000000000000000, %struct.S1998* byval align 16 %agg.tmp118, %struct.S1998* byval align 16 %agg.tmp119)
   ret void
 }
 
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1)
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1)
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1)
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1)
 
 declare void @check1998(%struct.S1998* sret, %struct.S1998* byval align 16, %struct.S1998*, %struct.S1998* byval align 16)
 declare void @check1998va(i32 signext, ...)
diff --git a/llvm/test/CodeGen/PowerPC/resolvefi-disp.ll b/llvm/test/CodeGen/PowerPC/resolvefi-disp.ll
index 72755df0eb9..68a31278e77 100644
--- a/llvm/test/CodeGen/PowerPC/resolvefi-disp.ll
+++ b/llvm/test/CodeGen/PowerPC/resolvefi-disp.ll
@@ -30,17 +30,17 @@ entry:
   %b2 = alloca %struct.S2760, align 32
   %2 = bitcast %struct.S2760* %arg0 to i8*
   %3 = bitcast %struct.S2760* %0 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* %3, i64 11104, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %2, i8* align 16 %3, i64 11104, i1 false)
   %4 = bitcast %struct.S2760* %arg2 to i8*
   %5 = bitcast %struct.S2760* %1 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %4, i8* %5, i64 11104, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %4, i8* align 16 %5, i64 11104, i1 false)
   store %struct.S2760* %arg1, %struct.S2760** %arg1.addr, align 8
   %6 = bitcast %struct.S2760* %ret to i8*
-  call void @llvm.memset.p0i8.i64(i8* %6, i8 0, i64 11104, i32 32, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* align 32 %6, i8 0, i64 11104, i1 false)
   %7 = bitcast %struct.S2760* %b1 to i8*
-  call void @llvm.memset.p0i8.i64(i8* %7, i8 0, i64 11104, i32 32, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* align 32 %7, i8 0, i64 11104, i1 false)
   %8 = bitcast %struct.S2760* %b2 to i8*
-  call void @llvm.memset.p0i8.i64(i8* %8, i8 0, i64 11104, i32 32, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* align 32 %8, i8 0, i64 11104, i1 false)
   %b = getelementptr inbounds %struct.S2760, %struct.S2760* %arg0, i32 0, i32 1
   %g = getelementptr inbounds %struct.anon, %struct.anon* %b, i32 0, i32 1
   %9 = load i64, i64* %g, align 8
@@ -61,11 +61,11 @@ if.end:                                           ; preds = %if.then, %entry
   store i64 %12, i64* %g4, align 8
   %13 = bitcast %struct.S2760* %agg.result to i8*
   %14 = bitcast %struct.S2760* %ret to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %13, i8* %14, i64 11104, i32 32, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 32 %13, i8* align 32 %14, i64 11104, i1 false)
   ret void
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1)
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1)
 
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1)
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1)
 
diff --git a/llvm/test/CodeGen/PowerPC/structsinmem.ll b/llvm/test/CodeGen/PowerPC/structsinmem.ll
index 01b0848e707..c8ea3be7cce 100644
--- a/llvm/test/CodeGen/PowerPC/structsinmem.ll
+++ b/llvm/test/CodeGen/PowerPC/structsinmem.ll
@@ -43,19 +43,19 @@ entry:
   %p6 = alloca %struct.s6, align 4
   %p7 = alloca %struct.s7, align 4
   %0 = bitcast %struct.s1* %p1 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* getelementptr inbounds (%struct.s1, %struct.s1* @caller1.p1, i32 0, i32 0), i64 1, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* getelementptr inbounds (%struct.s1, %struct.s1* @caller1.p1, i32 0, i32 0), i64 1, i1 false)
   %1 = bitcast %struct.s2* %p2 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* bitcast (%struct.s2* @caller1.p2 to i8*), i64 2, i32 2, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 2 %1, i8* align 2 bitcast (%struct.s2* @caller1.p2 to i8*), i64 2, i1 false)
   %2 = bitcast %struct.s3* %p3 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* bitcast ({ i16, i8, i8 }* @caller1.p3 to i8*), i64 4, i32 2, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 2 %2, i8* align 2 bitcast ({ i16, i8, i8 }* @caller1.p3 to i8*), i64 4, i1 false)
   %3 = bitcast %struct.s4* %p4 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %3, i8* bitcast (%struct.s4* @caller1.p4 to i8*), i64 4, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %3, i8* align 4 bitcast (%struct.s4* @caller1.p4 to i8*), i64 4, i1 false)
   %4 = bitcast %struct.s5* %p5 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %4, i8* bitcast ({ i32, i8, [3 x i8] }* @caller1.p5 to i8*), i64 8, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %4, i8* align 4 bitcast ({ i32, i8, [3 x i8] }* @caller1.p5 to i8*), i64 8, i1 false)
   %5 = bitcast %struct.s6* %p6 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %5, i8* bitcast ({ i32, i16, [2 x i8] }* @caller1.p6 to i8*), i64 8, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %5, i8* align 4 bitcast ({ i32, i16, [2 x i8] }* @caller1.p6 to i8*), i64 8, i1 false)
   %6 = bitcast %struct.s7* %p7 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %6, i8* bitcast ({ i32, i16, i8, i8 }* @caller1.p7 to i8*), i64 8, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %6, i8* align 4 bitcast ({ i32, i16, i8, i8 }* @caller1.p7 to i8*), i64 8, i1 false)
   %call = call i32 @callee1(i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, %struct.s1* byval %p1, %struct.s2* byval %p2, %struct.s3* byval %p3, %struct.s4* byval %p4, %struct.s5* byval %p5, %struct.s6* byval %p6, %struct.s7* byval %p7)
   ret i32 %call
 
@@ -68,7 +68,7 @@ entry:
 ; CHECK: std {{[0-9]+}}, 160(1)
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i1) nounwind
 
 define internal i32 @callee1(i32 %z1, i32 %z2, i32 %z3, i32 %z4, i32 %z5, i32 %z6, i32 %z7, i32 %z8, %struct.s1* byval %v1, %struct.s2* byval %v2, %struct.s3* byval %v3, %struct.s4* byval %v4, %struct.s5* byval %v5, %struct.s6* byval %v6, %struct.s7* byval %v7) nounwind {
 entry:
@@ -132,19 +132,19 @@ entry:
   %p6 = alloca %struct.t6, align 1
   %p7 = alloca %struct.t7, align 1
   %0 = bitcast %struct.t1* %p1 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* getelementptr inbounds (%struct.t1, %struct.t1* @caller2.p1, i32 0, i32 0), i64 1, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* getelementptr inbounds (%struct.t1, %struct.t1* @caller2.p1, i32 0, i32 0), i64 1, i1 false)
   %1 = bitcast %struct.t2* %p2 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* bitcast ({ i16 }* @caller2.p2 to i8*), i64 2, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* bitcast ({ i16 }* @caller2.p2 to i8*), i64 2, i1 false)
   %2 = bitcast %struct.t3* %p3 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* bitcast (%struct.t3* @caller2.p3 to i8*), i64 3, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* bitcast (%struct.t3* @caller2.p3 to i8*), i64 3, i1 false)
   %3 = bitcast %struct.t4* %p4 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %3, i8* bitcast ({ i32 }* @caller2.p4 to i8*), i64 4, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %3, i8* bitcast ({ i32 }* @caller2.p4 to i8*), i64 4, i1 false)
   %4 = bitcast %struct.t5* %p5 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %4, i8* bitcast (%struct.t5* @caller2.p5 to i8*), i64 5, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %4, i8* bitcast (%struct.t5* @caller2.p5 to i8*), i64 5, i1 false)
   %5 = bitcast %struct.t6* %p6 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %5, i8* bitcast (%struct.t6* @caller2.p6 to i8*), i64 6, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %5, i8* bitcast (%struct.t6* @caller2.p6 to i8*), i64 6, i1 false)
   %6 = bitcast %struct.t7* %p7 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %6, i8* bitcast (%struct.t7* @caller2.p7 to i8*), i64 7, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %6, i8* bitcast (%struct.t7* @caller2.p7 to i8*), i64 7, i1 false)
   %call = call i32 @callee2(i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, %struct.t1* byval %p1, %struct.t2* byval %p2, %struct.t3* byval %p3, %struct.t4* byval %p4, %struct.t5* byval %p5, %struct.t6* byval %p6, %struct.t7* byval %p7)
   ret i32 %call
 
diff --git a/llvm/test/CodeGen/PowerPC/structsinregs.ll b/llvm/test/CodeGen/PowerPC/structsinregs.ll
index 54679f259e9..d8afc8f8559 100644
--- a/llvm/test/CodeGen/PowerPC/structsinregs.ll
+++ b/llvm/test/CodeGen/PowerPC/structsinregs.ll
@@ -43,19 +43,19 @@ entry:
   %p6 = alloca %struct.s6, align 4
   %p7 = alloca %struct.s7, align 4
   %0 = bitcast %struct.s1* %p1 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* getelementptr inbounds (%struct.s1, %struct.s1* @caller1.p1, i32 0, i32 0), i64 1, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* getelementptr inbounds (%struct.s1, %struct.s1* @caller1.p1, i32 0, i32 0), i64 1, i1 false)
   %1 = bitcast %struct.s2* %p2 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* bitcast (%struct.s2* @caller1.p2 to i8*), i64 2, i32 2, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 2 %1, i8* align 2 bitcast (%struct.s2* @caller1.p2 to i8*), i64 2, i1 false)
   %2 = bitcast %struct.s3* %p3 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* bitcast ({ i16, i8, i8 }* @caller1.p3 to i8*), i64 4, i32 2, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 2 %2, i8* align 2 bitcast ({ i16, i8, i8 }* @caller1.p3 to i8*), i64 4, i1 false)
   %3 = bitcast %struct.s4* %p4 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %3, i8* bitcast (%struct.s4* @caller1.p4 to i8*), i64 4, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %3, i8* align 4 bitcast (%struct.s4* @caller1.p4 to i8*), i64 4, i1 false)
   %4 = bitcast %struct.s5* %p5 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %4, i8* bitcast ({ i32, i8, [3 x i8] }* @caller1.p5 to i8*), i64 8, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %4, i8* align 4 bitcast ({ i32, i8, [3 x i8] }* @caller1.p5 to i8*), i64 8, i1 false)
   %5 = bitcast %struct.s6* %p6 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %5, i8* bitcast ({ i32, i16, [2 x i8] }* @caller1.p6 to i8*), i64 8, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %5, i8* align 4 bitcast ({ i32, i16, [2 x i8] }* @caller1.p6 to i8*), i64 8, i1 false)
   %6 = bitcast %struct.s7* %p7 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %6, i8* bitcast ({ i32, i16, i8, i8 }* @caller1.p7 to i8*), i64 8, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %6, i8* align 4 bitcast ({ i32, i16, i8, i8 }* @caller1.p7 to i8*), i64 8, i1 false)
   %call = call i32 @callee1(%struct.s1* byval %p1, %struct.s2* byval %p2, %struct.s3* byval %p3, %struct.s4* byval %p4, %struct.s5* byval %p5, %struct.s6* byval %p6, %struct.s7* byval %p7)
   ret i32 %call
 
@@ -69,7 +69,7 @@ entry:
 ; CHECK: lbz 3, 160(31)
 }
 
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i1) nounwind
 
 define internal i32 @callee1(%struct.s1* byval %v1, %struct.s2* byval %v2, %struct.s3* byval %v3, %struct.s4* byval %v4, %struct.s5* byval %v5, %struct.s6* byval %v6, %struct.s7* byval %v7) nounwind {
 entry:
@@ -125,19 +125,19 @@ entry:
   %p6 = alloca %struct.t6, align 1
   %p7 = alloca %struct.t7, align 1
   %0 = bitcast %struct.t1* %p1 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* getelementptr inbounds (%struct.t1, %struct.t1* @caller2.p1, i32 0, i32 0), i64 1, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* getelementptr inbounds (%struct.t1, %struct.t1* @caller2.p1, i32 0, i32 0), i64 1, i1 false)
   %1 = bitcast %struct.t2* %p2 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* bitcast ({ i16 }* @caller2.p2 to i8*), i64 2, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* bitcast ({ i16 }* @caller2.p2 to i8*), i64 2, i1 false)
   %2 = bitcast %struct.t3* %p3 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* bitcast (%struct.t3* @caller2.p3 to i8*), i64 3, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* bitcast (%struct.t3* @caller2.p3 to i8*), i64 3, i1 false)
   %3 = bitcast %struct.t4* %p4 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %3, i8* bitcast ({ i32 }* @caller2.p4 to i8*), i64 4, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %3, i8* bitcast ({ i32 }* @caller2.p4 to i8*), i64 4, i1 false)
   %4 = bitcast %struct.t5* %p5 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %4, i8* bitcast (%struct.t5* @caller2.p5 to i8*), i64 5, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %4, i8* bitcast (%struct.t5* @caller2.p5 to i8*), i64 5, i1 false)
   %5 = bitcast %struct.t6* %p6 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %5, i8* bitcast (%struct.t6* @caller2.p6 to i8*), i64 6, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %5, i8* bitcast (%struct.t6* @caller2.p6 to i8*), i64 6, i1 false)
   %6 = bitcast %struct.t7* %p7 to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %6, i8* bitcast (%struct.t7* @caller2.p7 to i8*), i64 7, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %6, i8* bitcast (%struct.t7* @caller2.p7 to i8*), i64 7, i1 false)
   %call = call i32 @callee2(%struct.t1* byval %p1, %struct.t2* byval %p2, %struct.t3* byval %p3, %struct.t4* byval %p4, %struct.t5* byval %p5, %struct.t6* byval %p6, %struct.t7* byval %p7)
   ret i32 %call
 
diff --git a/llvm/test/CodeGen/PowerPC/stwu8.ll b/llvm/test/CodeGen/PowerPC/stwu8.ll
index 4dfef4b0323..f6d7ec9334c 100644
--- a/llvm/test/CodeGen/PowerPC/stwu8.ll
+++ b/llvm/test/CodeGen/PowerPC/stwu8.ll
@@ -13,7 +13,7 @@ define void @test1(%class.spell_checker.21.103.513.538* %this) unnamed_addr alig
 entry:
   %_M_header.i.i.i.i.i.i = getelementptr inbounds %class.spell_checker.21.103.513.538, %class.spell_checker.21.103.513.538* %this, i64 0, i32 0, i32 0, i32 0, i32 1
   %0 = bitcast %"struct.std::_Rb_tree_node_base.17.99.509.534"* %_M_header.i.i.i.i.i.i to i8*
-  call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 40, i32 4, i1 false) nounwind
+  call void @llvm.memset.p0i8.i64(i8* align 4 %0, i8 0, i64 40, i1 false) nounwind
   store %"struct.std::_Rb_tree_node_base.17.99.509.534"* %_M_header.i.i.i.i.i.i, %"struct.std::_Rb_tree_node_base.17.99.509.534"** undef, align 8
   unreachable
 }
@@ -21,4 +21,4 @@ entry:
 ; CHECK: @test1
 ; CHECK: stwu
 
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) nounwind
diff --git a/llvm/test/CodeGen/PowerPC/tailcall-string-rvo.ll b/llvm/test/CodeGen/PowerPC/tailcall-string-rvo.ll
index cf365dfa2a6..c48ee467031 100644
--- a/llvm/test/CodeGen/PowerPC/tailcall-string-rvo.ll
+++ b/llvm/test/CodeGen/PowerPC/tailcall-string-rvo.ll
@@ -32,7 +32,7 @@ bb:
   %tmp1 = bitcast %class.basic_string.11.42.73* %arg to %union.anon.8.39.70**
   store %union.anon.8.39.70* %tmp, %union.anon.8.39.70** %tmp1, align 8
   %tmp2 = bitcast %union.anon.8.39.70* %tmp to i8*
-  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp2, i8* nonnull undef, i64 13, i32 1, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp2, i8* nonnull undef, i64 13, i1 false)
   %tmp3 = getelementptr inbounds %class.basic_string.11.42.73, %class.basic_string.11.42.73* %arg, i64 0, i32 0, i32 0, i32 1
   store i64 13, i64* %tmp3, align 8
   %tmp4 = getelementptr inbounds %class.basic_string.11.42.73, %class.basic_string.11.42.73* %arg, i64 0, i32 0, i32 0, i32 2, i32 1, i64 5
@@ -42,6 +42,6 @@ bb:
 }
 
 ; Function Attrs: argmemonly nounwind
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1) #0
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1) #0
 
 attributes #0 = { argmemonly nounwind }
diff --git a/llvm/test/CodeGen/PowerPC/toc-load-sched-bug.ll b/llvm/test/CodeGen/PowerPC/toc-load-sched-bug.ll
index 21ccbf6f1ea..8e2aadf87fa 100644
--- a/llvm/test/CodeGen/PowerPC/toc-load-sched-bug.ll
+++ b/llvm/test/CodeGen/PowerPC/toc-load-sched-bug.ll
@@ -222,7 +222,7 @@ if.then:                                          ; preds = %_ZNK4llvm7ErrorOrIS
   %Filename.i = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %ref.tmp, i64 0, i32 2
   %10 = getelementptr inbounds %"class.std::allocator", %"class.std::allocator"* %ref.tmp.i.i2.i, i64 0, i32 0
   %11 = bitcast %"class.llvm::SMDiagnostic"* %ref.tmp to i8*
-  call void @llvm.memset.p0i8.i64(i8* %11, i8 0, i64 16, i32 8, i1 false) #3
+  call void @llvm.memset.p0i8.i64(i8* align 8 %11, i8 0, i64 16, i1 false) #3
   call void @llvm.lifetime.start.p0i8(i64 1, i8* %10) #3
   %tobool.i.i4.i = icmp eq i8* %4, null
   br i1 %tobool.i.i4.i, label %if.then.i.i6.i, label %if.end.i.i8.i
@@ -265,7 +265,7 @@ _ZN4llvm12SMDiagnosticC2ENS_9StringRefENS_9SourceMgr8DiagKindES1_.exit: ; preds
   store i8* bitcast (i64* getelementptr inbounds ([0 x i64], [0 x i64]* @_ZNSs4_Rep20_S_empty_rep_storageE, i64 0, i64 3) to i8*), i8** %_M_p.i.i.i.i.i, align 8, !tbaa !13
   %Ranges.i = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %ref.tmp, i64 0, i32 8
   %13 = bitcast %"class.std::vector.79"* %Ranges.i to i8*
-  call void @llvm.memset.p0i8.i64(i8* %13, i8 0, i64 24, i32 8, i1 false) #3
+  call void @llvm.memset.p0i8.i64(i8* align 8 %13, i8 0, i64 24, i1 false) #3
   %14 = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %ref.tmp, i64 0, i32 9, i32 0, i32 0, i32 0, i32 1, i32 0, i32 0, i64 0
   %BeginX.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %ref.tmp, i64 0, i32 9, i32 0, i32 0, i32 0, i32 0, i32 0
   store i8* %14, i8** %BeginX.i.i.i.i.i.i, align 8, !tbaa !23
@@ -275,13 +275,13 @@ _ZN4llvm12SMDiagnosticC2ENS_9StringRefENS_9SourceMgr8DiagKindES1_.exit: ; preds
   %add.ptr.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %ref.tmp, i64 0, i32 9, i32 0, i32 0, i32 0, i32 1, i32 0, i32 0, i64 96
   store i8* %add.ptr.i.i.i.i.i.i, i8** %CapacityX.i.i.i.i.i.i, align 8, !tbaa !26
   %15 = bitcast %"class.llvm::SMDiagnostic"* %Err to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %15, i8* %11, i64 16, i32 8, i1 false) #3
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %15, i8* align 8 %11, i64 16, i1 false) #3
   %Filename.i38 = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %Err, i64 0, i32 2
   call void @_ZNSs4swapERSs(%"class.std::basic_string"* %Filename.i38, %"class.std::basic_string"* dereferenceable(8) %Filename.i) #3
   %LineNo.i39 = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %Err, i64 0, i32 3
   %16 = bitcast i32* %LineNo.i39 to i8*
   %17 = bitcast i32* %LineNo.i to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %16, i8* %17, i64 12, i32 4, i1 false) #3
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %16, i8* align 4 %17, i64 12, i1 false) #3
   %Message.i40 = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %Err, i64 0, i32 6
   call void @_ZNSs4swapERSs(%"class.std::basic_string"* %Message.i40, %"class.std::basic_string"* dereferenceable(8) %Message.i) #3
   %LineContents.i = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %Err, i64 0, i32 7
@@ -294,7 +294,7 @@ _ZN4llvm12SMDiagnosticC2ENS_9StringRefENS_9SourceMgr8DiagKindES1_.exit: ; preds
   %_M_end_of_storage.i11.i.i.i = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %Err, i64 0, i32 8, i32 0, i32 0, i32 2
   %_M_start2.i.i.i.i = getelementptr inbounds %"class.llvm::SMDiagnostic", %"class.llvm::SMDiagnostic"* %ref.tmp, i64 0, i32 8, i32 0, i32 0, i32 0
   %19 = bitcast %"class.std::vector.79"* %Ranges.i41 to i8*
-  call void @llvm.memset.p0i8.i64(i8* %19, i8 0, i64 16, i32 8, i1 false) #3
+  call void @llvm.memset.p0i8.i64(i8* align 8 %19, i8 0, i64 16, i1 false) #3
   %20 = load %"struct.std::pair"*, %"struct.std::pair"** %_M_start2.i.i.i.i, align 8, !tbaa !27
   store %"struct.std::pair"* %20, %"struct.std::pair"** %_M_start.i7.i.i.i, align 8, !tbaa !27
   store %"struct.std::pair"* null, %"struct.std::pair"** %_M_start2.i.i.i.i, align 8, !tbaa !27
@@ -449,7 +449,7 @@ declare void @__assert_fail(i8*, i8*, i32 zeroext, i8*) #4
 declare dereferenceable(8) %"class.std::basic_string"* @_ZNSs6insertEmPKcm(%"class.std::basic_string"*, i64, i8*, i64) #1
 
 ; Function Attrs: nounwind
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1) #3
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1) #3
 
 ; Function Attrs: nounwind
 declare void @_ZNSs4_Rep10_M_destroyERKSaIcE(%"struct.std::basic_string<char, std::char_traits<char>, std::allocator<char> >::_Rep"*, %"class.std::allocator"* dereferenceable(1)) #0
@@ -471,7 +471,7 @@ declare %"class.llvm::Module"* @_ZN4llvm7ParseIREPNS_12MemoryBufferERNS_12SMDiag
 declare void @_ZNSs4swapERSs(%"class.std::basic_string"*, %"class.std::basic_string"* dereferenceable(8)) #1
 
 ; Function Attrs: nounwind
-declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) #3
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) #3
 
 attributes #0 = { nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
 attributes #1 = { "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }