diff options
| author | Daniel Neilson <dneilson@azul.com> | 2018-01-19 17:13:12 +0000 |
|---|---|---|
| committer | Daniel Neilson <dneilson@azul.com> | 2018-01-19 17:13:12 +0000 |
| commit | 1e68724d24ba38de7c7cdb2e1939d78c8b37cc0d (patch) | |
| tree | ef2e22d141b391e512da3c2df5c65f906eb7b98f /llvm/test/CodeGen/X86 | |
| parent | 6e938effaaf2016eb76e1b73aba2aa38a245cb70 (diff) | |
| download | bcm5719-llvm-1e68724d24ba38de7c7cdb2e1939d78c8b37cc0d.tar.gz bcm5719-llvm-1e68724d24ba38de7c7cdb2e1939d78c8b37cc0d.zip | |
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
Diffstat (limited to 'llvm/test/CodeGen/X86')
55 files changed, 168 insertions, 168 deletions
diff --git a/llvm/test/CodeGen/X86/2007-10-15-CoalescerCrash.ll b/llvm/test/CodeGen/X86/2007-10-15-CoalescerCrash.ll index 65e5ed76213..5cdf4dec3c5 100644 --- a/llvm/test/CodeGen/X86/2007-10-15-CoalescerCrash.ll +++ b/llvm/test/CodeGen/X86/2007-10-15-CoalescerCrash.ll @@ -224,7 +224,7 @@ declare void @fancy_abort(i8*, i32, i8*) declare i8* @pool_alloc(%struct.alloc_pool_def*) -declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) +declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) declare void @link_block(%struct.basic_block_def*, %struct.basic_block_def*) diff --git a/llvm/test/CodeGen/X86/2009-01-25-NoSSE.ll b/llvm/test/CodeGen/X86/2009-01-25-NoSSE.ll index 58bce75fc73..6d390b71114 100644 --- a/llvm/test/CodeGen/X86/2009-01-25-NoSSE.ll +++ b/llvm/test/CodeGen/X86/2009-01-25-NoSSE.ll @@ -12,11 +12,11 @@ define void @foo() nounwind { entry: %termios = alloca %struct.ktermios, align 8 %termios1 = bitcast %struct.ktermios* %termios to i8* - call void @llvm.memset.p0i8.i64(i8* %termios1, i8 0, i64 44, i32 8, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 8 %termios1, i8 0, i64 44, i1 false) call void @bar(%struct.ktermios* %termios) nounwind ret void } declare void @bar(%struct.ktermios*) -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/X86/2009-11-16-UnfoldMemOpBug.ll b/llvm/test/CodeGen/X86/2009-11-16-UnfoldMemOpBug.ll index 1cfd108db65..b58ee5be82e 100644 --- a/llvm/test/CodeGen/X86/2009-11-16-UnfoldMemOpBug.ll +++ b/llvm/test/CodeGen/X86/2009-11-16-UnfoldMemOpBug.ll @@ -17,7 +17,7 @@ bb1: ; CHECK: movups %xmm0, 12(%rsp) ; CHECK: movaps %xmm1, (%rsp) %tmp2 = phi i32 [ %tmp3, %bb1 ], [ 0, %entry ] - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp1, i8* getelementptr inbounds ([28 x i8], [28 x i8]* @str, i64 0, i64 0), i64 28, i32 1, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp1, i8* getelementptr inbounds ([28 x i8], [28 x i8]* @str, i64 0, i64 0), i64 28, i1 false) %tmp3 = add i32 %tmp2, 1 %tmp4 = icmp eq i32 %tmp3, %count br i1 %tmp4, label %bb2, label %bb1 @@ -26,4 +26,4 @@ bb2: 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/X86/2010-04-08-CoalescerBug.ll b/llvm/test/CodeGen/X86/2010-04-08-CoalescerBug.ll index c3dfbfc15ec..6c9c743eed7 100644 --- a/llvm/test/CodeGen/X86/2010-04-08-CoalescerBug.ll +++ b/llvm/test/CodeGen/X86/2010-04-08-CoalescerBug.ll @@ -19,8 +19,8 @@ entry: %tmp4 = getelementptr inbounds %struct.FC, %struct.FC* %tmp3, i64 0, i32 1, i64 0 %tmp5 = bitcast [32 x i32]* %BitValueArray to i8* %tmp6 = bitcast i32* %tmp4 to i8* - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp5, i8* %tmp6, i64 128, i32 4, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %tmp5, i8* align 4 %tmp6, i64 128, i1 false) unreachable } -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/X86/2010-04-21-CoalescerBug.ll b/llvm/test/CodeGen/X86/2010-04-21-CoalescerBug.ll index d5987645cfc..46dedb48ff1 100644 --- a/llvm/test/CodeGen/X86/2010-04-21-CoalescerBug.ll +++ b/llvm/test/CodeGen/X86/2010-04-21-CoalescerBug.ll @@ -8,8 +8,8 @@ define void @t(%struct.CMTimeMapping* noalias nocapture sret %agg.result) nounwind optsize ssp { entry: %agg.result1 = bitcast %struct.CMTimeMapping* %agg.result to i8* ; <i8*> [#uses=1] - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %agg.result1, i8* null, i64 96, i32 4, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %agg.result1, i8* align 4 null, i64 96, 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/X86/2010-06-25-CoalescerSubRegDefDead.ll b/llvm/test/CodeGen/X86/2010-06-25-CoalescerSubRegDefDead.ll index ffb51572a30..dd7c3fa571c 100644 --- a/llvm/test/CodeGen/X86/2010-06-25-CoalescerSubRegDefDead.ll +++ b/llvm/test/CodeGen/X86/2010-06-25-CoalescerSubRegDefDead.ll @@ -26,7 +26,7 @@ bb: ; CHECK: rep;stosl %tmp5 = bitcast i32* %tmp4 to i8* - call void @llvm.memset.p0i8.i64(i8* %tmp5, i8 0, i64 84, i32 4, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 4 %tmp5, i8 0, i64 84, i1 false) %tmp6 = getelementptr inbounds %struct.type, %struct.type* %s, i32 0, i32 62 store i32* null, i32** %tmp6, align 8 br label %bb1 @@ -36,4 +36,4 @@ bb1: ret i32 42 } -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/X86/2010-09-17-SideEffectsInChain.ll b/llvm/test/CodeGen/X86/2010-09-17-SideEffectsInChain.ll index 9e33d2bf6ac..3a5942513e8 100644 --- a/llvm/test/CodeGen/X86/2010-09-17-SideEffectsInChain.ll +++ b/llvm/test/CodeGen/X86/2010-09-17-SideEffectsInChain.ll @@ -2,7 +2,7 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64" target triple = "x86_64-apple-darwin10.4" -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 fastcc i32 @cli_magic_scandesc(i8* %in) nounwind ssp { entry: @@ -12,7 +12,7 @@ entry: %d = load i8, i8* %b, align 8 %e = load i8, i8* %c, align 8 %f = bitcast [64 x i8]* %a to i8* - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %f, i8* %in, i64 64, i32 8, i1 false) nounwind + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %f, i8* align 8 %in, i64 64, i1 false) nounwind store i8 %d, i8* %b, align 8 store i8 %e, i8* %c, align 8 ret i32 0 diff --git a/llvm/test/CodeGen/X86/2012-01-10-UndefExceptionEdge.ll b/llvm/test/CodeGen/X86/2012-01-10-UndefExceptionEdge.ll index 20615afdfa1..97a33893fa0 100644 --- a/llvm/test/CodeGen/X86/2012-01-10-UndefExceptionEdge.ll +++ b/llvm/test/CodeGen/X86/2012-01-10-UndefExceptionEdge.ll @@ -14,7 +14,7 @@ target triple = "i386-apple-macosx10.7" @Exception = external unnamed_addr constant { i8*, i8* } -declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind +declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) nounwind define void @f(i32* nocapture %arg, i32* nocapture %arg1, i32* nocapture %arg2, i32* nocapture %arg3, i32 %arg4, i32 %arg5) optsize ssp personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) { bb: @@ -85,7 +85,7 @@ bb41: ; preds = %bb38 to label %bb42 unwind label %bb20 bb42: ; preds = %bb41 - tail call void @llvm.memset.p0i8.i32(i8* %tmp32, i8 0, i32 %tmp9, i32 1, i1 false) nounwind + tail call void @llvm.memset.p0i8.i32(i8* %tmp32, i8 0, i32 %tmp9, i1 false) nounwind br i1 %tmp35, label %bb43, label %bb45 bb43: ; preds = %bb42 @@ -101,7 +101,7 @@ bb45: ; preds = %bb57, %bb42 br i1 %tmp47, label %bb48, label %bb59 bb48: ; preds = %bb45 - tail call void @llvm.memset.p0i8.i32(i8* %tmp32, i8 0, i32 %tmp9, i32 1, i1 false) nounwind + tail call void @llvm.memset.p0i8.i32(i8* %tmp32, i8 0, i32 %tmp9, i1 false) nounwind br i1 %tmp36, label %bb49, label %bb57 bb49: ; preds = %bb49, %bb48 @@ -120,7 +120,7 @@ bb57: ; preds = %bb49, %bb48 bb59: ; preds = %bb45 %tmp60 = ashr i32 %tmp46, 31 - tail call void @llvm.memset.p0i8.i32(i8* null, i8 0, i32 %tmp37, i32 1, i1 false) nounwind + tail call void @llvm.memset.p0i8.i32(i8* null, i8 0, i32 %tmp37, i1 false) nounwind br i1 %tmp36, label %bb61, label %bb67 bb61: ; preds = %bb61, %bb59 diff --git a/llvm/test/CodeGen/X86/alignment-2.ll b/llvm/test/CodeGen/X86/alignment-2.ll index a38a3626702..b191b986232 100644 --- a/llvm/test/CodeGen/X86/alignment-2.ll +++ b/llvm/test/CodeGen/X86/alignment-2.ll @@ -23,8 +23,8 @@ bb: ; CHECK-NOT: movaps {{[0-9]*}}(%{{[a-z]*}}), {{%xmm[0-9]}} %myopt = alloca %struct.printQueryOpt, align 4 %tmp = bitcast %struct.printQueryOpt* %myopt to i8* - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp, i8* bitcast (%struct.printQueryOpt* getelementptr inbounds (%struct._psqlSettings, %struct._psqlSettings* @pset, i32 0, i32 4) to i8*), i32 76, i32 4, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %tmp, i8* align 4 bitcast (%struct.printQueryOpt* getelementptr inbounds (%struct._psqlSettings, %struct._psqlSettings* @pset, i32 0, i32 4) to i8*), i32 76, i1 false) ret i8 0 } -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/X86/bug26810.ll b/llvm/test/CodeGen/X86/bug26810.ll index 816bc8224d8..263008131e7 100644 --- a/llvm/test/CodeGen/X86/bug26810.ll +++ b/llvm/test/CodeGen/X86/bug26810.ll @@ -114,7 +114,7 @@ loop.exit: ; preds = %for.body.i define void @init() local_unnamed_addr #1 { entry: - call void @llvm.memset.p0i8.i32(i8* bitcast ([8 x <2 x double>]* @"\01?v@@3PAU__m128d@@A" to i8*), i8 0, i32 128, i32 16, i1 false) + call void @llvm.memset.p0i8.i32(i8* align 16 bitcast ([8 x <2 x double>]* @"\01?v@@3PAU__m128d@@A" to i8*), i8 0, i32 128, i1 false) %call.i = tail call i64 @_time64(i64* null) %conv = trunc i64 %call.i to i32 tail call void @srand(i32 %conv) @@ -284,7 +284,7 @@ declare i32 @fclose(%struct._iobuf* nocapture) local_unnamed_addr #5 declare i64 @_time64(i64*) local_unnamed_addr #4 ; Function Attrs: argmemonly nounwind -declare void @llvm.memset.p0i8.i32(i8* nocapture writeonly, i8, i32, i32, i1) #6 +declare void @llvm.memset.p0i8.i32(i8* nocapture writeonly, i8, i32, i1) #6 attributes #0 = { norecurse "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="pentium4" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" } attributes #1 = { "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="pentium4" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" } diff --git a/llvm/test/CodeGen/X86/darwin-bzero.ll b/llvm/test/CodeGen/X86/darwin-bzero.ll index 3d03ec677a0..60032665255 100644 --- a/llvm/test/CodeGen/X86/darwin-bzero.ll +++ b/llvm/test/CodeGen/X86/darwin-bzero.ll @@ -3,12 +3,12 @@ ; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck -check-prefixes=CHECK,NOBZERO %s ; RUN: llc < %s -mtriple=x86_64-apple-ios10.0-simulator | FileCheck -check-prefixes=CHECK,NOBZERO %s -declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind +declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) nounwind ; CHECK-LABEL: foo: ; BZERO: {{calll|callq}} ___bzero ; NOBZERO-NOT: bzero define void @foo(i8* %p, i32 %len) { - call void @llvm.memset.p0i8.i32(i8* %p, i8 0, i32 %len, i32 1, i1 false) + call void @llvm.memset.p0i8.i32(i8* %p, i8 0, i32 %len, i1 false) ret void } diff --git a/llvm/test/CodeGen/X86/fast-isel-call.ll b/llvm/test/CodeGen/X86/fast-isel-call.ll index 3f394514e2c..2f3f4151424 100644 --- a/llvm/test/CodeGen/X86/fast-isel-call.ll +++ b/llvm/test/CodeGen/X86/fast-isel-call.ll @@ -31,10 +31,10 @@ define void @test2(%struct.s* %d) nounwind { ; CHECK: movl %eax, 8(%esp) } -declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind +declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) nounwind define void @test3(i8* %a) { - call void @llvm.memset.p0i8.i32(i8* %a, i8 0, i32 100, i32 1, i1 false) + call void @llvm.memset.p0i8.i32(i8* %a, i8 0, i32 100, i1 false) ret void ; CHECK-LABEL: test3: ; CHECK: movl {{.*}}, (%esp) @@ -43,10 +43,10 @@ define void @test3(i8* %a) { ; CHECK: calll {{.*}}memset } -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 define void @test4(i8* %a, i8* %b) { - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a, i8* %b, i32 100, i32 1, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a, i8* %b, i32 100, i1 false) ret void ; CHECK-LABEL: test4: ; CHECK: movl {{.*}}, (%esp) diff --git a/llvm/test/CodeGen/X86/fast-isel-deadcode.ll b/llvm/test/CodeGen/X86/fast-isel-deadcode.ll index 5381dc4858a..b7eab1a45c6 100644 --- a/llvm/test/CodeGen/X86/fast-isel-deadcode.ll +++ b/llvm/test/CodeGen/X86/fast-isel-deadcode.ll @@ -131,10 +131,10 @@ func.exit: ; preds = %if.then.i, %if.else.i, %if.end.5.i store { <2 x float>, float } %.fca.1.insert.i, { <2 x float>, float }* %tmp, align 8 %2 = bitcast { <2 x float>, float }* %tmp to i8* %3 = bitcast %struct.FVector* %ref.tmp to i8* - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %3, i8* %2, i64 12, i32 4, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %3, i8* align 4 %2, i64 12, i1 false) %4 = bitcast %struct.FVector* %v to i8* %5 = bitcast %struct.FVector* %ref.tmp to i8* - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %4, i8* %5, i64 12, i32 4, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %4, i8* align 4 %5, i64 12, i1 false) %6 = bitcast %struct.FVector* %v to i8* call void @llvm.lifetime.end.p0i8(i64 12, i8* %6) nounwind ret i32 0 @@ -142,6 +142,6 @@ func.exit: ; preds = %if.then.i, %if.else.i, %if.end.5.i declare void @llvm.lifetime.start.p0i8(i64, i8* nocapture) argmemonly nounwind -declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1) argmemonly nounwind +declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1) argmemonly nounwind declare void @llvm.lifetime.end.p0i8(i64, i8* nocapture) argmemonly nounwind diff --git a/llvm/test/CodeGen/X86/fast-isel-x86-64.ll b/llvm/test/CodeGen/X86/fast-isel-x86-64.ll index c87353ed1f5..7fb2670e6d1 100644 --- a/llvm/test/CodeGen/X86/fast-isel-x86-64.ll +++ b/llvm/test/CodeGen/X86/fast-isel-x86-64.ll @@ -172,11 +172,11 @@ entry: ; CHECK: callq } -declare void @llvm.memcpy.p0i8.p0i8.i64(i8*, i8*, i64, i32, i1) +declare void @llvm.memcpy.p0i8.p0i8.i64(i8*, i8*, i64, i1) ; rdar://9289488 - fast-isel shouldn't bail out on llvm.memcpy define void @test15(i8* %a, i8* %b) nounwind { - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 4, i32 4, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %a, i8* align 4 %b, i64 4, i1 false) ret void ; CHECK-LABEL: test15: ; CHECK-NEXT: movl (%rsi), %eax diff --git a/llvm/test/CodeGen/X86/force-align-stack-alloca.ll b/llvm/test/CodeGen/X86/force-align-stack-alloca.ll index 8d42680e199..e9f38e9af62 100644 --- a/llvm/test/CodeGen/X86/force-align-stack-alloca.ll +++ b/llvm/test/CodeGen/X86/force-align-stack-alloca.ll @@ -67,10 +67,10 @@ entry: if.then: %0 = alloca i8, i32 %i - call void @llvm.memset.p0i8.i32(i8* %0, i8 0, i32 %i, i32 1, i1 false) + call void @llvm.memset.p0i8.i32(i8* %0, i8 0, i32 %i, i1 false) %call = call i32 @f(i8* %0) %conv = sext i32 %call to i64 ret i64 %conv } -declare void @llvm.memset.p0i8.i32(i8*, i8, i32, i32, i1) nounwind +declare void @llvm.memset.p0i8.i32(i8*, i8, i32, i1) nounwind diff --git a/llvm/test/CodeGen/X86/immediate_merging.ll b/llvm/test/CodeGen/X86/immediate_merging.ll index e1c29191498..a6e36c73467 100644 --- a/llvm/test/CodeGen/X86/immediate_merging.ll +++ b/llvm/test/CodeGen/X86/immediate_merging.ll @@ -94,7 +94,7 @@ entry: ret i32 0 } -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 @AA = common global [100 x i8] zeroinitializer, align 1 @@ -121,6 +121,6 @@ define void @foomemset() optsize { ; X64-NEXT: movq %rax, {{.*}}(%rip) ; X64-NEXT: retq entry: - call void @llvm.memset.p0i8.i32(i8* getelementptr inbounds ([100 x i8], [100 x i8]* @AA, i32 0, i32 0), i8 33, i32 24, i32 1, i1 false) + call void @llvm.memset.p0i8.i32(i8* getelementptr inbounds ([100 x i8], [100 x i8]* @AA, i32 0, i32 0), i8 33, i32 24, i1 false) ret void } diff --git a/llvm/test/CodeGen/X86/immediate_merging64.ll b/llvm/test/CodeGen/X86/immediate_merging64.ll index 57f5b3b79d9..12be8bdff83 100644 --- a/llvm/test/CodeGen/X86/immediate_merging64.ll +++ b/llvm/test/CodeGen/X86/immediate_merging64.ll @@ -19,7 +19,7 @@ define i1 @imm_multiple_users(i64 %a, i64* %b) optsize { ret i1 %cmp } -declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) +declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) ; Inlined memsets requiring multiple same-sized stores should be lowered using ; the register, rather than immediate, form of stores when optimizing for @@ -31,6 +31,6 @@ define void @memset_zero(i8* noalias nocapture %D) optsize { ; CHECK-NEXT: movq %rax, 7(%rdi) ; CHECK-NEXT: movq %rax, (%rdi) ; CHECK-NEXT: retq - tail call void @llvm.memset.p0i8.i64(i8* %D, i8 0, i64 15, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i64(i8* %D, i8 0, i64 15, i1 false) ret void } diff --git a/llvm/test/CodeGen/X86/lea-opt-memop-check-1.ll b/llvm/test/CodeGen/X86/lea-opt-memop-check-1.ll index 6ad55d42868..00d47fae25a 100644 --- a/llvm/test/CodeGen/X86/lea-opt-memop-check-1.ll +++ b/llvm/test/CodeGen/X86/lea-opt-memop-check-1.ll @@ -6,7 +6,7 @@ target datalayout = "e-m:x-p:32:32-i64:64-f80:32-n8:16:32-a:0:32-S32" target triple = "i686-pc-windows-msvc" -declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) argmemonly nounwind +declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) argmemonly nounwind declare <2 x i64> @_mm_xor_si128(<2 x i64>, <2 x i64>) optsize declare <2 x i64> @llvm.x86.pclmulqdq(<2 x i64>, <2 x i64>, i8) nounwind readnone declare <4 x float> @_mm_castsi128_ps(<2 x i64>) optsize @@ -15,7 +15,7 @@ declare <4 x float> @_mm_castsi128_ps(<2 x i64>) optsize define void @test1(i8* nocapture readonly %src, i32 %len) #0 { %parts = alloca [4 x i32], align 4 %part0 = bitcast [4 x i32]* %parts to i8* - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %part0, i8* %src, i32 %len, i32 1, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* %part0, i8* %src, i32 %len, i1 false) %call0 = tail call <2 x i64> @_mm_xor_si128(<2 x i64> undef, <2 x i64> <i64 -9187201950435737472, i64 -9187201950435737472>) %tmp0 = tail call <2 x i64> @llvm.x86.pclmulqdq(<2 x i64> undef, <2 x i64> <i64 7631803798, i64 5708721108>, i8 16) %call1 = tail call <4 x float> @_mm_castsi128_ps(<2 x i64> %tmp0) diff --git a/llvm/test/CodeGen/X86/load-slice.ll b/llvm/test/CodeGen/X86/load-slice.ll index 8803512eec0..3cbb70bd70d 100644 --- a/llvm/test/CodeGen/X86/load-slice.ll +++ b/llvm/test/CodeGen/X86/load-slice.ll @@ -70,7 +70,7 @@ entry: } ; 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 ; Function Attrs: nounwind declare void @llvm.lifetime.start.p0i8(i64, i8* nocapture) diff --git a/llvm/test/CodeGen/X86/lsr-normalization.ll b/llvm/test/CodeGen/X86/lsr-normalization.ll index a8e3ab1ae99..f56256aa468 100644 --- a/llvm/test/CodeGen/X86/lsr-normalization.ll +++ b/llvm/test/CodeGen/X86/lsr-normalization.ll @@ -21,7 +21,7 @@ define i32 @main(i32 %arg, i8** nocapture %arg1) nounwind { bb: %tmp = alloca %0, align 8 ; <%0*> [#uses=11] %tmp2 = bitcast %0* %tmp to i8* ; <i8*> [#uses=1] - call void @llvm.memset.p0i8.i64(i8* %tmp2, i8 0, i64 16, i32 8, i1 false) nounwind + call void @llvm.memset.p0i8.i64(i8* align 8 %tmp2, i8 0, i64 16, i1 false) nounwind %tmp3 = getelementptr inbounds %0, %0* %tmp, i64 0, i32 0 ; <%0**> [#uses=3] store %0* %tmp, %0** %tmp3 %tmp4 = getelementptr inbounds %0, %0* %tmp, i64 0, i32 1 ; <%0**> [#uses=1] @@ -98,7 +98,7 @@ declare void @_ZNSt15_List_node_base4hookEPS_(%0*, %0*) declare noalias i8* @_Znwm(i64) -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 declare void @_ZdlPv(i8*) nounwind diff --git a/llvm/test/CodeGen/X86/mcu-abi.ll b/llvm/test/CodeGen/X86/mcu-abi.ll index 1cc277c863f..baafea26c22 100644 --- a/llvm/test/CodeGen/X86/mcu-abi.ll +++ b/llvm/test/CodeGen/X86/mcu-abi.ll @@ -70,7 +70,7 @@ define void @ret_large_struct(%struct.st12_t* noalias nocapture sret %agg.result entry: %0 = bitcast %struct.st12_t* %agg.result to i8* %1 = bitcast %struct.st12_t* %r to i8* - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %0, i8* %1, i32 48, i32 1, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* %0, i8* %1, i32 48, i1 false) ret void } @@ -104,7 +104,7 @@ define i32 @test_fp128(fp128* %ptr) #0 { ret i32 %ret } -declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) #1 +declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) #1 ; CHECK-LABEL: test_alignment_d: ; CHECK-NOT: andl {{.+}}, %esp diff --git a/llvm/test/CodeGen/X86/mem-intrin-base-reg.ll b/llvm/test/CodeGen/X86/mem-intrin-base-reg.ll index 59a2207b470..1a45ed57b26 100644 --- a/llvm/test/CodeGen/X86/mem-intrin-base-reg.ll +++ b/llvm/test/CodeGen/X86/mem-intrin-base-reg.ll @@ -9,12 +9,12 @@ target triple = "i686-pc-windows-msvc" ; which all of the X86 string instructions use. declare void @escape_vla_and_icmp(i8*, i1 zeroext) -declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) -declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) +declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) +declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) define i32 @memcpy_novla_vector(<4 x i32>* %vp0, i8* %a, i8* %b, i32 %n, i1 zeroext %cond) { %foo = alloca <4 x i32>, align 16 - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a, i8* %b, i32 128, i32 4, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %a, i8* align 4 %b, i32 128, i1 false) br i1 %cond, label %spill_vectors, label %no_vectors no_vectors: @@ -40,7 +40,7 @@ spill_vectors: define i32 @memcpy_vla_vector(<4 x i32>* %vp0, i8* %a, i8* %b, i32 %n, i1 zeroext %cond) { %foo = alloca <4 x i32>, align 16 - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a, i8* %b, i32 128, i32 4, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %a, i8* align 4 %b, i32 128, i1 false) br i1 %cond, label %spill_vectors, label %no_vectors no_vectors: @@ -69,7 +69,7 @@ spill_vectors: define i32 @memset_vla_vector(<4 x i32>* %vp0, i8* %a, i32 %n, i1 zeroext %cond) { %foo = alloca <4 x i32>, align 16 - call void @llvm.memset.p0i8.i32(i8* %a, i8 42, i32 128, i32 4, i1 false) + call void @llvm.memset.p0i8.i32(i8* align 4 %a, i8 42, i32 128, i1 false) br i1 %cond, label %spill_vectors, label %no_vectors no_vectors: diff --git a/llvm/test/CodeGen/X86/memcpy-2.ll b/llvm/test/CodeGen/X86/memcpy-2.ll index 040dd153d64..6deeaa69853 100644 --- a/llvm/test/CodeGen/X86/memcpy-2.ll +++ b/llvm/test/CodeGen/X86/memcpy-2.ll @@ -48,7 +48,7 @@ entry: ; X86-64: movq $0 %tmp1 = alloca [25 x i8] %tmp2 = bitcast [25 x i8]* %tmp1 to i8* - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* getelementptr inbounds ([25 x i8], [25 x i8]* @.str, i32 0, i32 0), i32 25, i32 1, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 1 %tmp2, i8* align 1 getelementptr inbounds ([25 x i8], [25 x i8]* @.str, i32 0, i32 0), i32 25, i1 false) unreachable } @@ -86,7 +86,7 @@ entry: ; X86-64: movaps %xmm0, (%rdi) %tmp2 = bitcast %struct.s0* %a to i8* ; <i8*> [#uses=1] %tmp3 = bitcast %struct.s0* %b to i8* ; <i8*> [#uses=1] - tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* %tmp3, i32 16, i32 16, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 16 %tmp2, i8* align 16 %tmp3, i32 16, i1 false) ret void } @@ -135,7 +135,7 @@ entry: ; X86-64: movq %rax, (%rdi) %tmp2 = bitcast %struct.s0* %a to i8* ; <i8*> [#uses=1] %tmp3 = bitcast %struct.s0* %b to i8* ; <i8*> [#uses=1] - tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* %tmp3, i32 16, i32 8, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 8 %tmp2, i8* align 8 %tmp3, i32 16, i1 false) ret void } @@ -202,8 +202,8 @@ entry: %tmp1 = alloca [30 x i8] %tmp2 = bitcast [30 x i8]* %tmp1 to i8* - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* getelementptr inbounds ([30 x i8], [30 x i8]* @.str2, i32 0, i32 0), i32 30, i32 1, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 1 %tmp2, i8* align 1 getelementptr inbounds ([30 x i8], [30 x i8]* @.str2, i32 0, i32 0), i32 30, i1 false) unreachable } -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/X86/memcpy-from-string.ll b/llvm/test/CodeGen/X86/memcpy-from-string.ll index d62d9e20254..8e2444ebe0e 100644 --- a/llvm/test/CodeGen/X86/memcpy-from-string.ll +++ b/llvm/test/CodeGen/X86/memcpy-from-string.ll @@ -17,8 +17,8 @@ target triple = "x86_64-unknown-linux-gnu" ; CHECK: movw $15212, 4(%rdi) ; CHECK: movl $1802117222, (%rdi) define void @foo(i8* %tmp2) { - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp2, i8* getelementptr inbounds ([10 x i8], [10 x i8]* @0, i64 0, i64 3), i64 7, i32 1, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp2, i8* getelementptr inbounds ([10 x i8], [10 x i8]* @0, i64 0, i64 3), i64 7, 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) diff --git a/llvm/test/CodeGen/X86/memcpy.ll b/llvm/test/CodeGen/X86/memcpy.ll index 4351014192b..87e350a9039 100644 --- a/llvm/test/CodeGen/X86/memcpy.ll +++ b/llvm/test/CodeGen/X86/memcpy.ll @@ -1,14 +1,14 @@ ; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=core2 | FileCheck %s -check-prefix=LINUX ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core2 | FileCheck %s -check-prefix=DARWIN -declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind -declare void @llvm.memcpy.p256i8.p256i8.i64(i8 addrspace(256)* nocapture, i8 addrspace(256)* nocapture, i64, i32, i1) nounwind +declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i1) nounwind +declare void @llvm.memcpy.p256i8.p256i8.i64(i8 addrspace(256)* nocapture, i8 addrspace(256)* nocapture, i64, i1) nounwind ; Variable memcpy's should lower to calls. define i8* @test1(i8* %a, i8* %b, i64 %n) nounwind { entry: - tail call void @llvm.memcpy.p0i8.p0i8.i64( i8* %a, i8* %b, i64 %n, i32 1, i1 0 ) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 %n, i1 0 ) ret i8* %a ; LINUX-LABEL: test1: @@ -20,7 +20,7 @@ define i8* @test2(i64* %a, i64* %b, i64 %n) nounwind { entry: %tmp14 = bitcast i64* %a to i8* %tmp25 = bitcast i64* %b to i8* - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp14, i8* %tmp25, i64 %n, i32 8, i1 0 ) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %tmp14, i8* align 8 %tmp25, i64 %n, i1 0 ) ret i8* %tmp14 ; LINUX-LABEL: test2: @@ -35,7 +35,7 @@ entry: ; rdar://8821501 define void @test3(i8* nocapture %A, i8* nocapture %B) nounwind optsize noredzone { entry: - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i1 false) ret void ; LINUX-LABEL: test3: ; LINUX: memcpy @@ -61,7 +61,7 @@ entry: } define void @test3_minsize(i8* nocapture %A, i8* nocapture %B) nounwind minsize noredzone { - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i1 false) ret void ; LINUX-LABEL: test3_minsize: ; LINUX: memcpy @@ -71,7 +71,7 @@ define void @test3_minsize(i8* nocapture %A, i8* nocapture %B) nounwind minsize } define void @test3_minsize_optsize(i8* nocapture %A, i8* nocapture %B) nounwind optsize minsize noredzone { - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i1 false) ret void ; LINUX-LABEL: test3_minsize_optsize: ; LINUX: memcpy @@ -83,7 +83,7 @@ define void @test3_minsize_optsize(i8* nocapture %A, i8* nocapture %B) nounwind ; Large constant memcpy's should be inlined when not optimizing for size. define void @test4(i8* nocapture %A, i8* nocapture %B) nounwind noredzone { entry: - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i1 false) ret void ; LINUX-LABEL: test4: ; LINUX: movq @@ -105,7 +105,7 @@ entry: define void @test5(i8* nocapture %C) nounwind uwtable ssp { entry: - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %C, i8* getelementptr inbounds ([30 x i8], [30 x i8]* @.str, i64 0, i64 0), i64 16, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %C, i8* getelementptr inbounds ([30 x i8], [30 x i8]* @.str, i64 0, i64 0), i64 16, i1 false) ret void ; DARWIN-LABEL: test5: @@ -122,7 +122,7 @@ entry: ; DARWIN: test6 ; DARWIN: movw $0, 8 ; DARWIN: movq $120, 0 - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* null, i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str2, i64 0, i64 0), i64 10, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* null, i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str2, i64 0, i64 0), i64 10, i1 false) ret void } @@ -136,14 +136,14 @@ define void @PR15348(i8* %a, i8* %b) { ; LINUX: movq ; LINUX: movq ; LINUX: movq - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 17, i32 0, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 17, i1 false) ret void } ; Memcpys from / to address space 256 should be lowered to appropriate loads / ; stores if small enough. define void @addrspace256(i8 addrspace(256)* %a, i8 addrspace(256)* %b) nounwind { - tail call void @llvm.memcpy.p256i8.p256i8.i64(i8 addrspace(256)* %a, i8 addrspace(256)* %b, i64 16, i32 8, i1 false) + tail call void @llvm.memcpy.p256i8.p256i8.i64(i8 addrspace(256)* align 8 %a, i8 addrspace(256)* align 8 %b, i64 16, i1 false) ret void ; LINUX-LABEL: addrspace256: ; LINUX: movq %gs: diff --git a/llvm/test/CodeGen/X86/memset-2.ll b/llvm/test/CodeGen/X86/memset-2.ll index e94432884b1..a0511f2804a 100644 --- a/llvm/test/CodeGen/X86/memset-2.ll +++ b/llvm/test/CodeGen/X86/memset-2.ll @@ -11,7 +11,7 @@ define fastcc void @t1() nounwind { ; CHECK-NEXT: calll _memset ; CHECK-NEXT: addl $16, %esp entry: - call void @llvm.memset.p0i8.i32(i8* null, i8 0, i32 188, i32 1, i1 false) + call void @llvm.memset.p0i8.i32(i8* null, i8 0, i32 188, i1 false) unreachable } @@ -23,11 +23,11 @@ define fastcc void @t2(i8 signext %c) nounwind { ; CHECK-NEXT: movl $76, {{[0-9]+}}(%esp) ; CHECK-NEXT: calll _memset entry: - call void @llvm.memset.p0i8.i32(i8* undef, i8 %c, i32 76, i32 1, i1 false) + call void @llvm.memset.p0i8.i32(i8* undef, i8 %c, i32 76, i1 false) unreachable } -declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind +declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) nounwind define void @t3(i8* nocapture %s, i8 %a) nounwind { ; CHECK-LABEL: t3: @@ -39,7 +39,7 @@ define void @t3(i8* nocapture %s, i8 %a) nounwind { ; CHECK-NEXT: movl %ecx, (%eax) ; CHECK-NEXT: retl entry: - tail call void @llvm.memset.p0i8.i32(i8* %s, i8 %a, i32 8, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i32(i8* %s, i8 %a, i32 8, i1 false) ret void } @@ -56,6 +56,6 @@ define void @t4(i8* nocapture %s, i8 %a) nounwind { ; CHECK-NEXT: movb %cl, 14(%eax) ; CHECK-NEXT: retl entry: - tail call void @llvm.memset.p0i8.i32(i8* %s, i8 %a, i32 15, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i32(i8* %s, i8 %a, i32 15, i1 false) ret void } diff --git a/llvm/test/CodeGen/X86/memset-3.ll b/llvm/test/CodeGen/X86/memset-3.ll index 455e6756013..47c7ab99d29 100644 --- a/llvm/test/CodeGen/X86/memset-3.ll +++ b/llvm/test/CodeGen/X86/memset-3.ll @@ -5,8 +5,8 @@ define void @t() nounwind ssp { entry: %buf = alloca [512 x i8], align 1 %ptr = getelementptr inbounds [512 x i8], [512 x i8]* %buf, i32 0, i32 0 - call void @llvm.memset.p0i8.i32(i8* %ptr, i8 undef, i32 512, i32 1, i1 false) + call void @llvm.memset.p0i8.i32(i8* %ptr, i8 undef, i32 512, i1 false) unreachable } -declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind +declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) nounwind diff --git a/llvm/test/CodeGen/X86/memset-nonzero.ll b/llvm/test/CodeGen/X86/memset-nonzero.ll index cc434bf18ab..37b98b40192 100644 --- a/llvm/test/CodeGen/X86/memset-nonzero.ll +++ b/llvm/test/CodeGen/X86/memset-nonzero.ll @@ -225,7 +225,7 @@ define void @memset_16_nonconst_bytes(i8* %x, i8 %c) { ; AVX2-NEXT: vpbroadcastb %xmm0, %xmm0 ; AVX2-NEXT: vmovdqu %xmm0, (%rdi) ; AVX2-NEXT: retq - tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 16, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 16, i1 false) ret void } @@ -268,7 +268,7 @@ define void @memset_32_nonconst_bytes(i8* %x, i8 %c) { ; AVX2-NEXT: vmovdqu %ymm0, (%rdi) ; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq - tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 32, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 32, i1 false) ret void } @@ -319,7 +319,7 @@ define void @memset_64_nonconst_bytes(i8* %x, i8 %c) { ; AVX2-NEXT: vmovdqu %ymm0, (%rdi) ; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq - tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 64, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 64, i1 false) ret void } @@ -386,7 +386,7 @@ define void @memset_128_nonconst_bytes(i8* %x, i8 %c) { ; AVX2-NEXT: vmovdqu %ymm0, (%rdi) ; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq - tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 128, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 128, i1 false) ret void } @@ -451,9 +451,9 @@ define void @memset_256_nonconst_bytes(i8* %x, i8 %c) { ; AVX2-NEXT: vmovdqu %ymm0, (%rdi) ; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq - tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 256, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i64(i8* %x, i8 %c, i64 256, i1 false) ret void } -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 diff --git a/llvm/test/CodeGen/X86/memset-sse-stack-realignment.ll b/llvm/test/CodeGen/X86/memset-sse-stack-realignment.ll index d77a7ed3816..68fa15e3398 100644 --- a/llvm/test/CodeGen/X86/memset-sse-stack-realignment.ll +++ b/llvm/test/CodeGen/X86/memset-sse-stack-realignment.ll @@ -9,7 +9,7 @@ define void @test1(i32 %t) nounwind { %tmp1210 = alloca i8, i32 32, align 4 - call void @llvm.memset.p0i8.i64(i8* %tmp1210, i8 0, i64 32, i32 4, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 4 %tmp1210, i8 0, i64 32, i1 false) %x = alloca i8, i32 %t call void @dummy(i8* %x) ret void @@ -42,7 +42,7 @@ define void @test1(i32 %t) nounwind { define void @test2(i32 %t) nounwind { %tmp1210 = alloca i8, i32 16, align 4 - call void @llvm.memset.p0i8.i64(i8* %tmp1210, i8 0, i64 16, i32 4, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 4 %tmp1210, i8 0, i64 16, i1 false) %x = alloca i8, i32 %t call void @dummy(i8* %x) ret void @@ -74,4 +74,4 @@ define void @test2(i32 %t) nounwind { declare void @dummy(i8*) -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/X86/memset.ll b/llvm/test/CodeGen/X86/memset.ll index c9d8fbd58aa..6d5c4cd0f8a 100644 --- a/llvm/test/CodeGen/X86/memset.ll +++ b/llvm/test/CodeGen/X86/memset.ll @@ -58,14 +58,14 @@ entry: %up_mvd116 = getelementptr [8 x %struct.x], [8 x %struct.x]* %up_mvd, i32 0, i32 0 ; <%struct.x*> [#uses=1] %tmp110117 = bitcast [8 x %struct.x]* %up_mvd to i8* ; <i8*> [#uses=1] - call void @llvm.memset.p0i8.i64(i8* %tmp110117, i8 0, i64 32, i32 8, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 8 %tmp110117, i8 0, i64 32, i1 false) call void @foo( %struct.x* %up_mvd116 ) nounwind ret void } declare void @foo(%struct.x*) -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 ; Ensure that alignment of '0' in an @llvm.memset intrinsic results in ; unaligned loads and stores. @@ -97,6 +97,6 @@ define void @PR15348(i8* %a) { ; YMM-NEXT: vmovups %xmm0, (%eax) ; YMM-NEXT: movb $0, 16(%eax) ; YMM-NEXT: retl - call void @llvm.memset.p0i8.i64(i8* %a, i8 0, i64 17, i32 0, i1 false) + call void @llvm.memset.p0i8.i64(i8* %a, i8 0, i64 17, i1 false) ret void } diff --git a/llvm/test/CodeGen/X86/memset64-on-x86-32.ll b/llvm/test/CodeGen/X86/memset64-on-x86-32.ll index 0fc21920409..f9707c66863 100644 --- a/llvm/test/CodeGen/X86/memset64-on-x86-32.ll +++ b/llvm/test/CodeGen/X86/memset64-on-x86-32.ll @@ -51,9 +51,9 @@ define void @bork() nounwind { ; SLOW_64-NEXT: movq $0, 8 ; SLOW_64-NEXT: movq $0, 0 ; SLOW_64-NEXT: retq - call void @llvm.memset.p0i8.i64(i8* null, i8 0, i64 80, i32 4, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 4 null, i8 0, i64 80, i1 false) ret void } -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/X86/misaligned-memset.ll b/llvm/test/CodeGen/X86/misaligned-memset.ll index ef8e0e81ad7..f7a6d577c08 100644 --- a/llvm/test/CodeGen/X86/misaligned-memset.ll +++ b/llvm/test/CodeGen/X86/misaligned-memset.ll @@ -7,9 +7,9 @@ define i32 @main() nounwind ssp { entry: %retval = alloca i32, align 4 store i32 0, i32* %retval - call void @llvm.memset.p0i8.i64(i8* bitcast (i64* getelementptr inbounds ([3 x i64], [3 x i64]* @a, i32 0, i64 1) to i8*), i8 0, i64 16, i32 1, i1 false) + call void @llvm.memset.p0i8.i64(i8* bitcast (i64* getelementptr inbounds ([3 x i64], [3 x i64]* @a, i32 0, i64 1) to i8*), i8 0, i64 16, i1 false) %0 = load i32, i32* %retval ret i32 %0 } -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/X86/misched-new.ll b/llvm/test/CodeGen/X86/misched-new.ll index 4e42c931454..5a93577a214 100644 --- a/llvm/test/CodeGen/X86/misched-new.ll +++ b/llvm/test/CodeGen/X86/misched-new.ll @@ -11,7 +11,7 @@ ; FIXME: There should be an assert in the coalescer that we're not rematting ; "not-quite-dead" copies, but that breaks a lot of tests <rdar://problem/11148682>. -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 ; From oggenc. ; After coalescing, we have a dead superreg (RAX) definition. @@ -24,7 +24,7 @@ entry: br i1 undef, label %for.cond.preheader, label %if.end for.cond.preheader: ; preds = %entry - call void @llvm.memcpy.p0i8.p0i8.i64(i8* undef, i8* null, i64 128, i32 4, i1 false) nounwind + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 undef, i8* align 4 null, i64 128, i1 false) nounwind unreachable if.end: ; preds = %entry diff --git a/llvm/test/CodeGen/X86/negate-add-zero.ll b/llvm/test/CodeGen/X86/negate-add-zero.ll index 64f20a6f81b..beb87e3e903 100644 --- a/llvm/test/CodeGen/X86/negate-add-zero.ll +++ b/llvm/test/CodeGen/X86/negate-add-zero.ll @@ -1133,4 +1133,4 @@ declare %"struct.std::basic_ostream<char,std::char_traits<char> >"* @_ZlsIdLi5EL declare %"struct.std::basic_ostream<char,std::char_traits<char> >"* @_ZlsIdLi5ELi6EERSoS0_RK15FixedMatrixBaseIT_XT0_EXT1_EE(%"struct.std::basic_ostream<char,std::char_traits<char> >"*, %"struct.FixedMatrixBase<double,5,6>"*) -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/X86/optimize-max-0.ll b/llvm/test/CodeGen/X86/optimize-max-0.ll index 2dde95738d1..b5e8627a88b 100644 --- a/llvm/test/CodeGen/X86/optimize-max-0.ll +++ b/llvm/test/CodeGen/X86/optimize-max-0.ll @@ -173,7 +173,7 @@ bb23: ; preds = %bb24, %bb.nph %47 = mul i32 %y.21, %w %.sum5 = add i32 %47, %.sum3 %48 = getelementptr i8, i8* %j, i32 %.sum5 - tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %48, i8* %46, i32 %w, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %48, i8* %46, i32 %w, i1 false) br label %bb24 bb24: ; preds = %bb23 @@ -190,7 +190,7 @@ bb26: ; preds = %bb24.bb26_crit_edge %50 = getelementptr i8, i8* %j, i32 %.sum4 %51 = mul i32 %x, %w %52 = sdiv i32 %51, 2 - tail call void @llvm.memset.p0i8.i32(i8* %50, i8 -128, i32 %52, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i32(i8* %50, i8 -128, i32 %52, i1 false) ret void bb29: ; preds = %bb20, %entry @@ -208,7 +208,7 @@ bb30: ; preds = %bb31, %bb.nph11 %57 = getelementptr i8, i8* %r, i32 %56 %58 = mul i32 %y.310, %w %59 = getelementptr i8, i8* %j, i32 %58 - tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %59, i8* %57, i32 %w, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %59, i8* %57, i32 %w, i1 false) br label %bb31 bb31: ; preds = %bb30 @@ -224,7 +224,7 @@ bb33: ; preds = %bb31.bb33_crit_edge %61 = getelementptr i8, i8* %j, i32 %60 %62 = mul i32 %x, %w %63 = sdiv i32 %62, 2 - tail call void @llvm.memset.p0i8.i32(i8* %61, i8 -128, i32 %63, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i32(i8* %61, i8 -128, i32 %63, i1 false) ret void return: ; preds = %bb20 @@ -398,7 +398,7 @@ bb23: ; preds = %bb24, %bb.nph %47 = mul i32 %y.21, %w %.sum5 = add i32 %47, %.sum3 %48 = getelementptr i8, i8* %j, i32 %.sum5 - tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %48, i8* %46, i32 %w, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %48, i8* %46, i32 %w, i1 false) br label %bb24 bb24: ; preds = %bb23 @@ -415,7 +415,7 @@ bb26: ; preds = %bb24.bb26_crit_edge %50 = getelementptr i8, i8* %j, i32 %.sum4 %51 = mul i32 %x, %w %52 = udiv i32 %51, 2 - tail call void @llvm.memset.p0i8.i32(i8* %50, i8 -128, i32 %52, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i32(i8* %50, i8 -128, i32 %52, i1 false) ret void bb29: ; preds = %bb20, %entry @@ -433,7 +433,7 @@ bb30: ; preds = %bb31, %bb.nph11 %57 = getelementptr i8, i8* %r, i32 %56 %58 = mul i32 %y.310, %w %59 = getelementptr i8, i8* %j, i32 %58 - tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %59, i8* %57, i32 %w, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %59, i8* %57, i32 %w, i1 false) br label %bb31 bb31: ; preds = %bb30 @@ -449,13 +449,13 @@ bb33: ; preds = %bb31.bb33_crit_edge %61 = getelementptr i8, i8* %j, i32 %60 %62 = mul i32 %x, %w %63 = udiv i32 %62, 2 - tail call void @llvm.memset.p0i8.i32(i8* %61, i8 -128, i32 %63, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i32(i8* %61, i8 -128, i32 %63, i1 false) ret void return: ; preds = %bb20 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 -declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind +declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) nounwind diff --git a/llvm/test/CodeGen/X86/pr11985.ll b/llvm/test/CodeGen/X86/pr11985.ll index 94b37215f63..99084d61140 100644 --- a/llvm/test/CodeGen/X86/pr11985.ll +++ b/llvm/test/CodeGen/X86/pr11985.ll @@ -24,7 +24,7 @@ define float @foo(i8* nocapture %buf, float %a, float %b) nounwind uwtable { ; NEHALEM-NEXT: movups %xmm2, (%rdi) entry: - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %buf, i8* blockaddress(@foo, %out), i64 22, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %buf, i8* blockaddress(@foo, %out), i64 22, i1 false) br label %out out: ; preds = %entry @@ -32,4 +32,4 @@ out: ; preds = %entry ret float %add } -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/X86/pr14333.ll b/llvm/test/CodeGen/X86/pr14333.ll index 89779302d7f..8298ca5a7d6 100644 --- a/llvm/test/CodeGen/X86/pr14333.ll +++ b/llvm/test/CodeGen/X86/pr14333.ll @@ -6,7 +6,7 @@ define void @bar(%foo* %zed) { %tmp2 = getelementptr inbounds %foo, %foo* %zed, i64 0, i32 1 store i64 0, i64* %tmp2, align 8 %tmp3 = bitcast %foo* %zed to i8* - call void @llvm.memset.p0i8.i64(i8* %tmp3, i8 0, i64 16, i32 8, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 8 %tmp3, i8 0, i64 16, i1 false) ret void } -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/X86/pr34088.ll b/llvm/test/CodeGen/X86/pr34088.ll index 4fa24a50648..2fb000f3538 100644 --- a/llvm/test/CodeGen/X86/pr34088.ll +++ b/llvm/test/CodeGen/X86/pr34088.ll @@ -31,13 +31,13 @@ define i32 @pr34088() local_unnamed_addr { entry: %foo = alloca %struct.Foo, align 4 %0 = bitcast %struct.Foo* %foo to i8* - call void @llvm.memset.p0i8.i32(i8* nonnull %0, i8 0, i32 20, i32 4, i1 false) + call void @llvm.memset.p0i8.i32(i8* align 4 nonnull %0, i8 0, i32 20, i1 false) %buffer1 = getelementptr inbounds %struct.Foo, %struct.Foo* %foo, i32 0, i32 1, i32 1 %1 = bitcast %struct.Buffer* %buffer1 to i64* %2 = load i64, i64* %1, align 4 - call void @llvm.memset.p0i8.i32(i8* nonnull %0, i8 -51, i32 20, i32 4, i1 false) + call void @llvm.memset.p0i8.i32(i8* align 4 nonnull %0, i8 -51, i32 20, i1 false) store i64 %2, i64* %1, align 4 ret i32 0 } -declare void @llvm.memset.p0i8.i32(i8* nocapture writeonly, i8, i32, i32, i1) +declare void @llvm.memset.p0i8.i32(i8* nocapture writeonly, i8, i32, i1) diff --git a/llvm/test/CodeGen/X86/ragreedy-hoist-spill.ll b/llvm/test/CodeGen/X86/ragreedy-hoist-spill.ll index 0178c9ec1c9..a6d4c6e97bc 100644 --- a/llvm/test/CodeGen/X86/ragreedy-hoist-spill.ll +++ b/llvm/test/CodeGen/X86/ragreedy-hoist-spill.ll @@ -63,7 +63,7 @@ SyTime.exit2720: br i1 %cmp293427, label %for.body.lr.ph, label %while.body.preheader for.body.lr.ph: - call void @llvm.memset.p0i8.i64(i8* undef, i8 32, i64 512, i32 16, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 16 undef, i8 32, i64 512, i1 false) br label %while.body.preheader while.body.preheader: @@ -377,7 +377,7 @@ cleanup: declare i32 @fileno(%struct.TMP.2* nocapture) declare i64 @"\01_write"(i32, i8*, i64) declare i32 @__maskrune(i32, i64) -declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) +declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) !llvm.ident = !{!0} diff --git a/llvm/test/CodeGen/X86/regparm.ll b/llvm/test/CodeGen/X86/regparm.ll index f427010edc5..01a734f9f47 100644 --- a/llvm/test/CodeGen/X86/regparm.ll +++ b/llvm/test/CodeGen/X86/regparm.ll @@ -9,7 +9,7 @@ target datalayout = "e-m:e-p:32:32-f64:32:64-f80:32-n8:16:32-S128" target triple = "i386-unknown-linux-gnu" ; Function Attrs: argmemonly nounwind -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 define void @use_memset(i8* inreg nocapture %dest, i8 inreg %c, i32 inreg %n) local_unnamed_addr #0 { entry: @@ -30,12 +30,12 @@ entry: ;FASTWIN: movzbl %dl, %edx ;FASTWIN-NEXT: calll _memset ;FASTWIN-NEXT: retl - tail call void @llvm.memset.p0i8.i32(i8* %dest, i8 %c, i32 %n, i32 1, i1 false) + tail call void @llvm.memset.p0i8.i32(i8* %dest, i8 %c, i32 %n, i1 false) ret void } ; Function Attrs: argmemonly nounwind -declare void @llvm.memset.p0i8.i32(i8* nocapture writeonly, i8, i32, i32, i1) #1 +declare void @llvm.memset.p0i8.i32(i8* nocapture writeonly, i8, i32, i1) #1 attributes #0 = { nounwind "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="pentium4" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" } diff --git a/llvm/test/CodeGen/X86/remat-fold-load.ll b/llvm/test/CodeGen/X86/remat-fold-load.ll index 3478033bfbf..e640974bdd2 100644 --- a/llvm/test/CodeGen/X86/remat-fold-load.ll +++ b/llvm/test/CodeGen/X86/remat-fold-load.ll @@ -16,7 +16,7 @@ target triple = "i386-unknown-linux-gnu" %type_d = type { i64 } %type_e = type { %type_c, i64 } -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 define linkonce_odr void @test() nounwind { entry: @@ -41,7 +41,7 @@ if.then.i.i.i.i71: ; preds = %while.body12 %tmp1 = getelementptr inbounds %type_a, %type_a* %tmp, i32 0, i32 1, i32 0, i32 1 %buf_6.i.i.i.i70 = bitcast %type_d* %tmp1 to i8** %tmp2 = load i8*, i8** %buf_6.i.i.i.i70, align 4 - call void @llvm.memcpy.p0i8.p0i8.i32(i8* undef, i8* %tmp2, i32 undef, i32 1, i1 false) nounwind + call void @llvm.memcpy.p0i8.p0i8.i32(i8* undef, i8* %tmp2, i32 undef, i1 false) nounwind unreachable if.else.i.i.i.i74: ; preds = %while.body12 @@ -69,7 +69,7 @@ if.then.i.i.i.i92: ; preds = %if.else.i.i.i.i74 %tmp12 = getelementptr inbounds %type_e, %type_e* %tmp9, i32 0, i32 0, i32 1 %buf_6.i.i.i.i91 = bitcast %type_d* %tmp12 to i8** %tmp13 = load i8*, i8** %buf_6.i.i.i.i91, align 4 - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %call4.i.i.i.i89, i8* %tmp13, i32 %tmp10, i32 1, i1 false) nounwind + call void @llvm.memcpy.p0i8.p0i8.i32(i8* %call4.i.i.i.i89, i8* %tmp13, i32 %tmp10, i1 false) nounwind br label %A if.else.i.i.i.i95: ; preds = %if.else.i.i.i.i74 diff --git a/llvm/test/CodeGen/X86/slow-unaligned-mem.ll b/llvm/test/CodeGen/X86/slow-unaligned-mem.ll index a3a21892339..54c248f3b04 100644 --- a/llvm/test/CodeGen/X86/slow-unaligned-mem.ll +++ b/llvm/test/CodeGen/X86/slow-unaligned-mem.ll @@ -88,9 +88,9 @@ define void @store_zeros(i8* %a) { ; FAST: # %bb.0: ; FAST-NEXT: movl {{[0-9]+}}(%esp), %eax ; FAST-NOT: movl - call void @llvm.memset.p0i8.i64(i8* %a, i8 0, i64 64, i32 1, i1 false) + call void @llvm.memset.p0i8.i64(i8* %a, i8 0, i64 64, i1 false) ret void } -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/X86/small-byval-memcpy.ll b/llvm/test/CodeGen/X86/small-byval-memcpy.ll index 3c03750199c..c5c9a3d8416 100644 --- a/llvm/test/CodeGen/X86/small-byval-memcpy.ll +++ b/llvm/test/CodeGen/X86/small-byval-memcpy.ll @@ -2,10 +2,10 @@ ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=nehalem | FileCheck %s --check-prefix=NEHALEM ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=btver2 | FileCheck %s --check-prefix=BTVER2 -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) define void @copy16bytes(i8* nocapture %a, i8* nocapture readonly %b) { - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 16, i32 1, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 16, i1 false) ret void ; CHECK-LABEL: copy16bytes diff --git a/llvm/test/CodeGen/X86/stack-align.ll b/llvm/test/CodeGen/X86/stack-align.ll index 192306462d1..338ced0ebf1 100644 --- a/llvm/test/CodeGen/X86/stack-align.ll +++ b/llvm/test/CodeGen/X86/stack-align.ll @@ -71,7 +71,7 @@ define x86_stdcallcc void @test5(%struct.sixteen* byval nocapture readonly align %1 = getelementptr inbounds [16 x i8], [16 x i8]* %d.sroa.0, i32 0, i32 0 call void @llvm.lifetime.start.p0i8(i64 16, i8* %1) %2 = getelementptr inbounds %struct.sixteen, %struct.sixteen* %s, i32 0, i32 0, i32 0 - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %1, i8* %2, i32 16, i32 1, i1 true) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* %1, i8* %2, i32 16, i1 true) call void @llvm.lifetime.end.p0i8(i64 16, i8* %1) ret void ; CHECK-LABEL: test5: @@ -84,7 +84,7 @@ define x86_stdcallcc void @test5(%struct.sixteen* byval nocapture readonly align declare void @llvm.lifetime.start.p0i8(i64, i8* nocapture) argmemonly nounwind -declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) argmemonly nounwind +declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) argmemonly nounwind declare void @llvm.lifetime.end.p0i8(i64, i8* nocapture) argmemonly nounwind diff --git a/llvm/test/CodeGen/X86/stack-protector.ll b/llvm/test/CodeGen/X86/stack-protector.ll index 5166ed5b02a..d4eee18244f 100644 --- a/llvm/test/CodeGen/X86/stack-protector.ll +++ b/llvm/test/CodeGen/X86/stack-protector.ll @@ -3768,7 +3768,7 @@ entry: %test.coerce = alloca { i64, i8 } %0 = bitcast { i64, i8 }* %test.coerce to i8* %1 = bitcast %struct.small_char* %test to i8* - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* %1, i64 12, i32 0, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* %1, i64 12, i1 false) %2 = getelementptr { i64, i8 }, { i64, i8 }* %test.coerce, i32 0, i32 0 %3 = load i64, i64* %2, align 1 %4 = getelementptr { i64, i8 }, { i64, i8 }* %test.coerce, i32 0, i32 1 @@ -3806,7 +3806,7 @@ entry: %test.coerce = alloca { i64, i8 } %0 = bitcast { i64, i8 }* %test.coerce to i8* %1 = bitcast %struct.small_char* %test to i8* - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* %1, i64 12, i32 0, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* %1, i64 12, i1 false) %2 = getelementptr { i64, i8 }, { i64, i8 }* %test.coerce, i32 0, i32 0 %3 = load i64, i64* %2, align 1 %4 = getelementptr { i64, i8 }, { i64, i8 }* %test.coerce, i32 0, i32 1 @@ -3922,7 +3922,7 @@ declare void @_Z3exceptPi(i32*) declare i32 @__gxx_personality_v0(...) declare i32* @getp() declare i32 @dummy(...) -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) attributes #0 = { ssp } attributes #1 = { sspstrong } diff --git a/llvm/test/CodeGen/X86/tail-dup-merge-loop-headers.ll b/llvm/test/CodeGen/X86/tail-dup-merge-loop-headers.ll index 197fd72586a..736a6d8500d 100644 --- a/llvm/test/CodeGen/X86/tail-dup-merge-loop-headers.ll +++ b/llvm/test/CodeGen/X86/tail-dup-merge-loop-headers.ll @@ -97,7 +97,7 @@ if.end19: ; preds = %entry br i1 %or.cond203, label %cleanup, label %if.end50 if.end50: ; preds = %if.end19 - tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull %call, i8* undef, i64 %conv, i32 1, i1 false) + tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull %call, i8* undef, i64 %conv, i1 false) %cmp1.i.i = icmp ugt i32 %mul, 3 br i1 %cmp1.i.i, label %shared_preheader, label %wunpsect.exit.thread.loopexit391 @@ -185,6 +185,6 @@ declare void @cli_dbgmsg(i8*, ...) local_unnamed_addr #0 declare i8* @cli_calloc(i64, i64) local_unnamed_addr #0 ; 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 } attributes #1 = { argmemonly nounwind } diff --git a/llvm/test/CodeGen/X86/tailcall-mem-intrinsics.ll b/llvm/test/CodeGen/X86/tailcall-mem-intrinsics.ll index 7491ea659ba..ee3489701df 100644 --- a/llvm/test/CodeGen/X86/tailcall-mem-intrinsics.ll +++ b/llvm/test/CodeGen/X86/tailcall-mem-intrinsics.ll @@ -4,7 +4,7 @@ ; CHECK: jmp 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: jmp 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 } @@ -20,7 +20,7 @@ entry: ; CHECK: jmp 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 } @@ -28,7 +28,7 @@ entry: ; CHECK: jmp memcpy define i8* @tail_memcpy_ret(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 i8* %p } @@ -36,7 +36,7 @@ entry: ; CHECK: jmp memmove define i8* @tail_memmove_ret(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 i8* %p } @@ -44,12 +44,12 @@ entry: ; CHECK: jmp memset define i8* @tail_memset_ret(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 i8* %p } -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 } diff --git a/llvm/test/CodeGen/X86/tlv-1.ll b/llvm/test/CodeGen/X86/tlv-1.ll index 5f017d31dbb..0dbd00c55eb 100644 --- a/llvm/test/CodeGen/X86/tlv-1.ll +++ b/llvm/test/CodeGen/X86/tlv-1.ll @@ -7,7 +7,7 @@ define void @main() nounwind ssp { ; CHECK-LABEL: main: entry: - call void @llvm.memset.p0i8.i64(i8* getelementptr inbounds (%struct.A, %struct.A* @c, i32 0, i32 0, i32 0), i8 0, i64 60, i32 1, i1 false) + call void @llvm.memset.p0i8.i64(i8* getelementptr inbounds (%struct.A, %struct.A* @c, i32 0, i32 0, i32 0), i8 0, i64 60, i1 false) unreachable ; CHECK: movq _c@TLVP(%rip), %rdi ; CHECK-NEXT: callq *(%rdi) @@ -31,7 +31,7 @@ entry: ret i32 %sub } -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 @a = thread_local global i32 0 ; <i32*> [#uses=0] @b = thread_local global i32 0 ; <i32*> [#uses=0] diff --git a/llvm/test/CodeGen/X86/unaligned-load.ll b/llvm/test/CodeGen/X86/unaligned-load.ll index 644a3644730..1ceca2cd06d 100644 --- a/llvm/test/CodeGen/X86/unaligned-load.ll +++ b/llvm/test/CodeGen/X86/unaligned-load.ll @@ -12,14 +12,14 @@ entry: bb: ; preds = %bb, %entry %String2Loc9 = getelementptr inbounds [31 x i8], [31 x i8]* %String2Loc, i64 0, i64 0 - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %String2Loc9, i8* getelementptr inbounds ([31 x i8], [31 x i8]* @.str3, i64 0, i64 0), i64 31, i32 1, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* %String2Loc9, i8* getelementptr inbounds ([31 x i8], [31 x i8]* @.str3, i64 0, i64 0), i64 31, i1 false) br label %bb return: ; No predecessors! 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 ; I386: calll {{_?}}memcpy diff --git a/llvm/test/CodeGen/X86/unused_stackslots.ll b/llvm/test/CodeGen/X86/unused_stackslots.ll index 82fd3db1ccb..dca01275ca7 100644 --- a/llvm/test/CodeGen/X86/unused_stackslots.ll +++ b/llvm/test/CodeGen/X86/unused_stackslots.ll @@ -202,14 +202,14 @@ land.lhs.true54: ; preds = %for.end50 br i1 %tobool56, label %for.inc73, label %for.body61.preheader for.body61.preheader: ; preds = %land.lhs.true54 - call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 4, i64 0) to i8*), i8* %tmp1, i64 32, i32 16, i1 false) - call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 6, i64 0) to i8*), i8* %tmp2, i64 32, i32 16, i1 false) - call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 8, i64 0) to i8*), i8* %tmp3, i64 32, i32 16, i1 false) - call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 10, i64 0) to i8*), i8* %tmp4, i64 32, i32 16, i1 false) - call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 12, i64 0) to i8*), i8* %tmp5, i64 32, i32 16, i1 false) - call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 14, i64 0) to i8*), i8* %tmp6, i64 32, i32 16, i1 false) - call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 16, i64 0) to i8*), i8* %tmp7, i64 32, i32 16, i1 false) - call void @llvm.memcpy.p0i8.p0i8.i64(i8* nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 18, i64 0) to i8*), i8* %tmp8, i64 32, i32 16, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 4, i64 0) to i8*), i8* align 16 %tmp1, i64 32, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 6, i64 0) to i8*), i8* align 16 %tmp2, i64 32, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 8, i64 0) to i8*), i8* align 16 %tmp3, i64 32, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 10, i64 0) to i8*), i8* align 16 %tmp4, i64 32, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 12, i64 0) to i8*), i8* align 16 %tmp5, i64 32, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 14, i64 0) to i8*), i8* align 16 %tmp6, i64 32, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 16, i64 0) to i8*), i8* align 16 %tmp7, i64 32, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 nonnull bitcast (i32* getelementptr ([4 x i32], [4 x i32]* @e, i64 18, i64 0) to i8*), i8* align 16 %tmp8, i64 32, i1 false) %call70 = tail call i32 @distortion4x4(i32* nonnull getelementptr inbounds ([4 x i32], [4 x i32]* @e, i64 0, i64 0)) #3 %add71 = add nsw i32 %call70, %m.3.lcssa.lcssa br label %for.inc73 @@ -234,7 +234,7 @@ declare void @LumaPrediction4x4(i32, i32, i32, i32, i32, i16 signext, i16 signex declare i32 @distortion4x4(i32*) #2 ; Function Attrs: argmemonly 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 ; Function Attrs: argmemonly nounwind declare void @llvm.lifetime.end.p0i8(i64, i8* nocapture) #1 diff --git a/llvm/test/CodeGen/X86/unwindraise.ll b/llvm/test/CodeGen/X86/unwindraise.ll index db39f4ed455..2da07fb1753 100644 --- a/llvm/test/CodeGen/X86/unwindraise.ll +++ b/llvm/test/CodeGen/X86/unwindraise.ll @@ -34,7 +34,7 @@ entry: call fastcc void @uw_init_context_1(%struct._Unwind_Context* %this_context, i8* %0, i8* %1) %2 = bitcast %struct._Unwind_Context* %cur_context to i8* %3 = bitcast %struct._Unwind_Context* %this_context to i8* - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* %3, i64 240, i32 8, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %2, i8* align 8 %3, i64 240, i1 false) %personality = getelementptr inbounds %struct._Unwind_FrameState, %struct._Unwind_FrameState* %fs, i64 0, i32 6 %retaddr_column.i = getelementptr inbounds %struct._Unwind_FrameState, %struct._Unwind_FrameState* %fs, i64 0, i32 9 %flags.i.i.i.i = getelementptr inbounds %struct._Unwind_Context, %struct._Unwind_Context* %cur_context, i64 0, i32 5 @@ -121,7 +121,7 @@ while.end: ; preds = %if.then4 %16 = ptrtoint i8* %15 to i64 %private_2 = getelementptr inbounds %struct._Unwind_Exception, %struct._Unwind_Exception* %exc, i64 0, i32 3 store i64 %16, i64* %private_2, align 8 - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* %3, i64 240, i32 8, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %2, i8* align 8 %3, i64 240, i1 false) %17 = bitcast %struct._Unwind_FrameState* %fs.i to i8* call void @llvm.lifetime.start.p0i8(i64 -1, i8* %17) %personality.i = getelementptr inbounds %struct._Unwind_FrameState, %struct._Unwind_FrameState* %fs.i, i64 0, i32 6 @@ -234,7 +234,7 @@ declare i8* @llvm.eh.dwarf.cfa(i32) nounwind declare i8* @llvm.returnaddress(i32) nounwind readnone -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 fastcc i64 @uw_install_context_1(%struct._Unwind_Context*, %struct._Unwind_Context*) uwtable diff --git a/llvm/test/CodeGen/X86/variable-sized-darwin-bzero.ll b/llvm/test/CodeGen/X86/variable-sized-darwin-bzero.ll index 3971190f02c..55c5dd39d12 100644 --- a/llvm/test/CodeGen/X86/variable-sized-darwin-bzero.ll +++ b/llvm/test/CodeGen/X86/variable-sized-darwin-bzero.ll @@ -1,8 +1,8 @@ ; RUN: llc < %s -mtriple=i686-apple-darwin10 | grep __bzero define void @foo(i8* %p, i64 %n) { - call void @llvm.memset.p0i8.i64(i8* %p, i8 0, i64 %n, i32 4, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 4 %p, i8 0, i64 %n, i1 false) ret void } -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/X86/vectorcall.ll b/llvm/test/CodeGen/X86/vectorcall.ll index 598a339ee2f..9914780e04c 100644 --- a/llvm/test/CodeGen/X86/vectorcall.ll +++ b/llvm/test/CodeGen/X86/vectorcall.ll @@ -157,7 +157,7 @@ entry: %retval = alloca %struct.HVA4, align 16 %0 = bitcast %struct.HVA4* %retval to i8* %1 = bitcast %struct.HVA4* %b to i8* - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %0, i8* %1, i32 64, i32 16, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 16 %0, i8* align 16 %1, i32 64, i1 false) %2 = load %struct.HVA4, %struct.HVA4* %retval, align 16 ret %struct.HVA4 %2 } @@ -168,18 +168,18 @@ entry: ; CHECK: movaps 48(%{{[re]}}sp), %xmm3 ; CHECK: ret{{[ql]}} -declare void @llvm.memset.p0i8.i64(i8* nocapture writeonly, i8, i64, i32, i1) -declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1) -declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture writeonly, i8* nocapture readonly, i32, i32, i1) +declare void @llvm.memset.p0i8.i64(i8* nocapture writeonly, i8, i64, i1) +declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i1) +declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture writeonly, i8* nocapture readonly, i32, i1) define x86_vectorcallcc void @test_mixed_7(%struct.HVA5* noalias sret %agg.result) { entry: %a = alloca %struct.HVA5, align 16 %0 = bitcast %struct.HVA5* %a to i8* - call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 80, i32 16, i1 false) + call void @llvm.memset.p0i8.i64(i8* align 16 %0, i8 0, i64 80, i1 false) %1 = bitcast %struct.HVA5* %agg.result to i8* %2 = bitcast %struct.HVA5* %a to i8* - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %2, i64 80, i32 16, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 %1, i8* align 16 %2, i64 80, i1 false) ret void } ; CHECK-LABEL: test_mixed_7 diff --git a/llvm/test/CodeGen/X86/x86-64-static-relo-movl.ll b/llvm/test/CodeGen/X86/x86-64-static-relo-movl.ll index 5da3a470503..658187e2204 100644 --- a/llvm/test/CodeGen/X86/x86-64-static-relo-movl.ll +++ b/llvm/test/CodeGen/X86/x86-64-static-relo-movl.ll @@ -10,7 +10,7 @@ define void @setup() { %pending = alloca %struct.MatchInfo, align 8 %t = bitcast %struct.MatchInfo* %pending to i8* - call void @llvm.memcpy.p0i8.p0i8.i64(i8* %t, i8* bitcast (%struct.MatchInfo* @NO_MATCH to i8*), i64 512, i32 8, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %t, i8* align 8 bitcast (%struct.MatchInfo* @NO_MATCH to i8*), i64 512, i1 false) %u = getelementptr inbounds %struct.MatchInfo, %struct.MatchInfo* %pending, i32 0, i32 2 %v = load i64, i64* %u, align 8 br label %done @@ -21,4 +21,4 @@ done: } ; Function Attrs: nounwind -declare void @llvm.memcpy.p0i8.p0i8.i64(i8*, i8*, i64, i32, i1) +declare void @llvm.memcpy.p0i8.p0i8.i64(i8*, i8*, i64, i1) diff --git a/llvm/test/CodeGen/X86/x86-repmov-copy-eflags.ll b/llvm/test/CodeGen/X86/x86-repmov-copy-eflags.ll index ad398885728..49afb39b6d4 100644 --- a/llvm/test/CodeGen/X86/x86-repmov-copy-eflags.ll +++ b/llvm/test/CodeGen/X86/x86-repmov-copy-eflags.ll @@ -10,7 +10,7 @@ entry: %g = alloca %struct.T, align 8 %r = alloca i32, align 8 store i32 0, i32* %r, align 4 - call void @llvm.memcpy.p0i8.p0i8.i32(i8* %p, i8* %q, i32 24, i32 8, i1 false) + call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 8 %p, i8* align 8 %q, i32 24, i1 false) br label %while.body while.body: ; preds = %while.body, %entry @@ -26,7 +26,7 @@ while.end: ; preds = %while.body } ; Function Attrs: argmemonly nounwind -declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) #1 +declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i1) #1 declare void @g(%struct.T*) |
