llvm-svn: 36527

previous clients.  This fixes MallocBench/gs

llvm-svn: 36525

llvm-svn: 36523

llvm-svn: 36521

for unrelated casts.

llvm-svn: 36511

llvm-svn: 36475

llvm-svn: 36444

llvm-svn: 36441

copies from a constant global, then we can change the reads to read from the
global instead of from the alloca.  This eliminates the alloca and the memcpy,
and promotes secondary optimizations (because the loads are now loads from
a constant global).

This is important for a common C idiom:

void foo() {
   int A[] = {1,2,3,4,5,6,7,8,9...};
   ... only reads of A ...
}

For some reason, people forget to mark the array static or const.

This triggers on these multisource benchmarks:
JM/ldecode: block_pos, [3 x [4 x [4 x i32]]]
FreeBench/mason: m, [18 x i32], inlined 4 times
MiBench/office-stringsearch: search_strings, [1332 x i8*]
MiBench/office-stringsearch: find_strings, [1333 x i8*]
Prolangs-C++/city: dirs, [9 x i8*], inlined 4 places

and these spec benchmarks:
177.mesa: message, [8 x [32 x i8]]
186.crafty: bias_rl45, [64 x i32]
186.crafty: diag_sq, [64 x i32]
186.crafty: empty, [9 x i8]
186.crafty: xlate, [15 x i8]
186.crafty: status, [13 x i8]
186.crafty: bdinfo, [25 x i8]
445.gobmk: routines, [16 x i8*]
458.sjeng: piece_rep, [14 x i8*]
458.sjeng: t, [13 x i32], inlined 4 places.
464.h264ref: block8x8_idx, [3 x [4 x [4 x i32]]]
464.h264ref: block_pos, [3 x [4 x [4 x i32]]]
464.h264ref: j_off_tab, [12 x i32]

This implements Transforms/ScalarRepl/memcpy-from-global.ll

llvm-svn: 36429

llvm-svn: 36426

this approach is no longer appropriate.

llvm-svn: 36421

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20070423/048376.html

llvm-svn: 36417

my approach to this, so hopefully I'll find a way to do this without making this slower.

llvm-svn: 36392

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20070423/048333.html

llvm-svn: 36380

llvm-svn: 36300

llvm-svn: 36299

llvm-svn: 36271

llvm-svn: 36261

llvm-svn: 36260

llvm-svn: 36255

llvm-svn: 36254

llvm-svn: 36252

llvm-svn: 36249

gets called.

llvm-svn: 36208

Fix
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20070416/047888.html

llvm-svn: 36182

target for tabs checking.

llvm-svn: 36146

Remove ImmediateDominator analysis.  The same information can be obtained from DomTree.  A lot of code for
constructing ImmediateDominator is now folded into DomTree construction.

This is part of the ongoing work for PR217.

llvm-svn: 36063

This sinks the two stores in this example into a single store in cond_next.  In this
case, it allows elimination of the load as well:

        store double 0.000000e+00, double* @s.3060
        %tmp3 = fcmp ogt double %tmp1, 5.000000e-01             ; <i1> [#uses=1]
        br i1 %tmp3, label %cond_true, label %cond_next
cond_true:              ; preds = %entry
        store double 1.000000e+00, double* @s.3060
        br label %cond_next
cond_next:              ; preds = %entry, %cond_true
        %tmp6 = load double* @s.3060            ; <double> [#uses=1]

This implements Transforms/InstCombine/store-merge.ll:test2

llvm-svn: 36040

llvm-svn: 36037

llvm-svn: 36031

define i32 @test(float %f) {
        %tmp7 = insertelement <4 x float> undef, float %f, i32 0
        %tmp17 = bitcast <4 x float> %tmp7 to <4 x i32>
        %tmp19 = extractelement <4 x i32> %tmp17, i32 0
        ret i32 %tmp19
}

into:

define i32 @test(float %f) {
        %tmp19 = bitcast float %f to i32                ; <i32> [#uses=1]
        ret i32 %tmp19
}

On PPC, this is the difference between:

_test:
        mfspr r2, 256
        oris r3, r2, 8192
        mtspr 256, r3
        stfs f1, -16(r1)
        addi r3, r1, -16
        addi r4, r1, -32
        lvx v2, 0, r3
        stvx v2, 0, r4
        lwz r3, -32(r1)
        mtspr 256, r2
        blr

and:

_test:
        stfs f1, -4(r1)
        nop
        nop
        nop
        lwz r3, -4(r1)
        blr

llvm-svn: 36025

unsigned test(float f) {
 return _mm_cvtsi128_si32( (__m128i) _mm_set_ss( f*f ));
}

into:

_test:
        movss 4(%esp), %xmm0
        mulss %xmm0, %xmm0
        movd %xmm0, %eax
        ret

instead of:

_test:
        movss 4(%esp), %xmm0
        mulss %xmm0, %xmm0
        xorps %xmm1, %xmm1
        movss %xmm0, %xmm1
        movd %xmm1, %eax
        ret

GCC gets:

_test:
        subl    $28, %esp
        movss   32(%esp), %xmm0
        mulss   %xmm0, %xmm0
        xorps   %xmm1, %xmm1
        movss   %xmm0, %xmm1
        movaps  %xmm1, %xmm0
        movd    %xmm0, 12(%esp)
        movl    12(%esp), %eax
        addl    $28, %esp
        ret

llvm-svn: 36020

llvm-svn: 35981

llvm-svn: 35979

out to do! :)

This fixes a problem where LSR would insert a bunch of code into each MBB
that uses a particular subexpression (e.g. IV+base+C).  The problem is that
this code cannot be CSE'd back together if inserted into different blocks.

This patch changes LSR to attempt to insert a single copy of this code and
share it, allowing codegenprepare to duplicate the code if it can be sunk
into various addressing modes.  On CodeGen/ARM/lsr-code-insertion.ll,
for example, this gives us code like:

        add r8, r0, r5
        str r6, [r8, #+4]
..
        ble LBB1_4      @cond_next
LBB1_3: @cond_true
        str r10, [r8, #+4]
LBB1_4: @cond_next
...
LBB1_5: @cond_true55
        ldr r6, LCPI1_1
        str r6, [r8, #+4]

instead of:

        add r10, r0, r6
        str r8, [r10, #+4]
...
        ble LBB1_4      @cond_next
LBB1_3: @cond_true
        add r8, r0, r6
        str r10, [r8, #+4]
LBB1_4: @cond_next
...
LBB1_5: @cond_true55
        add r8, r0, r6
        ldr r10, LCPI1_1
        str r10, [r8, #+4]

Besides being smaller and more efficient, this makes it immediately
obvious that it is profitable to predicate LBB1_3 now :)

llvm-svn: 35972

this fixes problems where codegenprepare would sink expressions into load/stores
that are not valid, and fixes cases where it would miss important valid ones.

This fixes several serious codesize and perf issues, particularly on targets
with complex addressing modes like arm and x86.  For example, now we compile
CodeGen/X86/isel-sink.ll to:

_test:
        movl 8(%esp), %eax
        movl 4(%esp), %ecx
        cmpl $1233, %eax
        ja LBB1_2       #F
LBB1_1: #T
        movl $4, (%ecx,%eax,4)
        movl $141, %eax
        ret
LBB1_2: #F
        movl (%ecx,%eax,4), %eax
        ret

instead of:

_test:
        movl 8(%esp), %eax
        leal (,%eax,4), %ecx
        addl 4(%esp), %ecx
        cmpl $1233, %eax
        ja LBB1_2       #F
LBB1_1: #T
        movl $4, (%ecx)
        movl $141, %eax
        ret
LBB1_2: #F
        movl (%ecx), %eax
        ret

llvm-svn: 35970

llvm-svn: 35906

        icmp slt i32 %X, 0              ; <i1>:0 [#uses=1]
        sext i1 %0 to i32               ; <i32>:1 [#uses=1]

into:

        %X.lobit = ashr i32 %X, 31              ; <i32> [#uses=1]

This implements InstCombine/icmp.ll:test[34]

llvm-svn: 35891

Transforms/InstCombine/icmp.ll

llvm-svn: 35890

llvm-svn: 35886