RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-30 15:32:52 +01:00
Code Issues Projects Releases Wiki Activity
20,390 Commits 4 Branches 5 Tags 704 MiB
9c6b86dcb4
Commit Graph

1840 Commits

Author SHA1 Message Date
Chris Lattner
a103a2e9c6 Fix a regression on X86, where FP values can be promoted too. 
llvm-svn: 22822
 2005-08-17 06:06:25 +00:00
Jim Laskey
59b9ee0529 Added generic code expansion for [signed|unsigned] i32 to [f32|f64] casts in the 
legalizer.  PowerPC now uses this expansion instead of ISel version.

Example:

// signed integer to double conversion
double f1(signed x) {
  return (double)x;
}

// unsigned integer to double conversion
double f2(unsigned x) {
  return (double)x;
}

// signed integer to float conversion
float f3(signed x) {
  return (float)x;
}

// unsigned integer to float conversion
float f4(unsigned x) {
  return (float)x;
}


Byte Code:

internal fastcc double %_Z2f1i(int %x) {
entry:
        %tmp.1 = cast int %x to double          ; <double> [#uses=1]
        ret double %tmp.1
}

internal fastcc double %_Z2f2j(uint %x) {
entry:
        %tmp.1 = cast uint %x to double         ; <double> [#uses=1]
        ret double %tmp.1
}

internal fastcc float %_Z2f3i(int %x) {
entry:
        %tmp.1 = cast int %x to float           ; <float> [#uses=1]
        ret float %tmp.1
}

internal fastcc float %_Z2f4j(uint %x) {
entry:
        %tmp.1 = cast uint %x to float          ; <float> [#uses=1]
        ret float %tmp.1
}

internal fastcc double %_Z2g1i(int %x) {
entry:
        %buffer = alloca [2 x uint]             ; <[2 x uint]*> [#uses=3]
        %tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0                ; <uint*> [#uses=1]
        store uint 1127219200, uint* %tmp.0
        %tmp.2 = cast int %x to uint            ; <uint> [#uses=1]
        %tmp.3 = xor uint %tmp.2, 2147483648            ; <uint> [#uses=1]
        %tmp.5 = getelementptr [2 x uint]* %buffer, int 0, int 1                ; <uint*> [#uses=1]
        store uint %tmp.3, uint* %tmp.5
        %tmp.9 = cast [2 x uint]* %buffer to double*            ; <double*> [#uses=1]
        %tmp.10 = load double* %tmp.9           ; <double> [#uses=1]
        %tmp.13 = load double* cast (long* %signed_bias to double*)             ; <double> [#uses=1]
        %tmp.14 = sub double %tmp.10, %tmp.13           ; <double> [#uses=1]
        ret double %tmp.14
}

internal fastcc double %_Z2g2j(uint %x) {
entry:
        %buffer = alloca [2 x uint]             ; <[2 x uint]*> [#uses=3]
        %tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0                ; <uint*> [#uses=1]
        store uint 1127219200, uint* %tmp.0
        %tmp.1 = getelementptr [2 x uint]* %buffer, int 0, int 1                ; <uint*> [#uses=1]
        store uint %x, uint* %tmp.1
        %tmp.4 = cast [2 x uint]* %buffer to double*            ; <double*> [#uses=1]
        %tmp.5 = load double* %tmp.4            ; <double> [#uses=1]
        %tmp.8 = load double* cast (long* %unsigned_bias to double*)            ; <double> [#uses=1]
        %tmp.9 = sub double %tmp.5, %tmp.8              ; <double> [#uses=1]
        ret double %tmp.9
}

internal fastcc float %_Z2g3i(int %x) {
entry:
        %buffer = alloca [2 x uint]             ; <[2 x uint]*> [#uses=3]
        %tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0                ; <uint*> [#uses=1]
        store uint 1127219200, uint* %tmp.0
        %tmp.2 = cast int %x to uint            ; <uint> [#uses=1]
        %tmp.3 = xor uint %tmp.2, 2147483648            ; <uint> [#uses=1]
        %tmp.5 = getelementptr [2 x uint]* %buffer, int 0, int 1                ; <uint*> [#uses=1]
        store uint %tmp.3, uint* %tmp.5
        %tmp.9 = cast [2 x uint]* %buffer to double*            ; <double*> [#uses=1]
        %tmp.10 = load double* %tmp.9           ; <double> [#uses=1]
        %tmp.13 = load double* cast (long* %signed_bias to double*)             ; <double> [#uses=1]
        %tmp.14 = sub double %tmp.10, %tmp.13           ; <double> [#uses=1]
        %tmp.16 = cast double %tmp.14 to float          ; <float> [#uses=1]
        ret float %tmp.16
}

internal fastcc float %_Z2g4j(uint %x) {
entry:
        %buffer = alloca [2 x uint]             ; <[2 x uint]*> [#uses=3]
        %tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0                ; <uint*> [#uses=1]
        store uint 1127219200, uint* %tmp.0
        %tmp.1 = getelementptr [2 x uint]* %buffer, int 0, int 1                ; <uint*> [#uses=1]
        store uint %x, uint* %tmp.1
        %tmp.4 = cast [2 x uint]* %buffer to double*            ; <double*> [#uses=1]
        %tmp.5 = load double* %tmp.4            ; <double> [#uses=1]
        %tmp.8 = load double* cast (long* %unsigned_bias to double*)            ; <double> [#uses=1]
        %tmp.9 = sub double %tmp.5, %tmp.8              ; <double> [#uses=1]
        %tmp.11 = cast double %tmp.9 to float           ; <float> [#uses=1]
        ret float %tmp.11
}


PowerPC Code:

        .machine ppc970


        .const
        .align  2
.CPIl1__Z2f1i_0:                                        ; float 0x4330000080000000
        .long   1501560836      ; float 4.5036e+15
        .text
        .align  2
        .globl  l1__Z2f1i
l1__Z2f1i:
.LBBl1__Z2f1i_0:        ; entry
        xoris r2, r3, 32768
        stw r2, -4(r1)
        lis r2, 17200
        stw r2, -8(r1)
        lfd f0, -8(r1)
        lis r2, ha16(.CPIl1__Z2f1i_0)
        lfs f1, lo16(.CPIl1__Z2f1i_0)(r2)
        fsub f1, f0, f1
        blr


        .const
        .align  2
.CPIl2__Z2f2j_0:                                        ; float 0x4330000000000000
        .long   1501560832      ; float 4.5036e+15
        .text
        .align  2
        .globl  l2__Z2f2j
l2__Z2f2j:
.LBBl2__Z2f2j_0:        ; entry
        stw r3, -4(r1)
        lis r2, 17200
        stw r2, -8(r1)
        lfd f0, -8(r1)
        lis r2, ha16(.CPIl2__Z2f2j_0)
        lfs f1, lo16(.CPIl2__Z2f2j_0)(r2)
        fsub f1, f0, f1
        blr


        .const
        .align  2
.CPIl3__Z2f3i_0:                                        ; float 0x4330000080000000
        .long   1501560836      ; float 4.5036e+15
        .text
        .align  2
        .globl  l3__Z2f3i
l3__Z2f3i:
.LBBl3__Z2f3i_0:        ; entry
        xoris r2, r3, 32768
        stw r2, -4(r1)
        lis r2, 17200
        stw r2, -8(r1)
        lfd f0, -8(r1)
        lis r2, ha16(.CPIl3__Z2f3i_0)
        lfs f1, lo16(.CPIl3__Z2f3i_0)(r2)
        fsub f0, f0, f1
        frsp f1, f0
        blr


        .const
        .align  2
.CPIl4__Z2f4j_0:                                        ; float 0x4330000000000000
        .long   1501560832      ; float 4.5036e+15
        .text
        .align  2
        .globl  l4__Z2f4j
l4__Z2f4j:
.LBBl4__Z2f4j_0:        ; entry
        stw r3, -4(r1)
        lis r2, 17200
        stw r2, -8(r1)
        lfd f0, -8(r1)
        lis r2, ha16(.CPIl4__Z2f4j_0)
        lfs f1, lo16(.CPIl4__Z2f4j_0)(r2)
        fsub f0, f0, f1
        frsp f1, f0
        blr

llvm-svn: 22814
 2005-08-17 00:39:29 +00:00
Chris Lattner
bd8cbd4951 add a new TargetConstant node 
llvm-svn: 22813
 2005-08-17 00:34:06 +00:00
Chris Lattner
3b7e157005 Eliminate the RegSDNode class, which 3 nodes (CopyFromReg/CopyToReg/ImplicitDef) 
used to tack a register number onto the node.

Instead of doing this, make a new node, RegisterSDNode, which is a leaf
containing a register number.  These three operations just become normal
DAG nodes now, instead of requiring special handling.

Note that with this change, it is no longer correct to make illegal
CopyFromReg/CopyToReg nodes.  The legalizer will not touch them, and this
is bad, so don't do it. :)

llvm-svn: 22806
 2005-08-16 21:55:35 +00:00
Nate Begeman
f6b6378f23 Implement BR_CC and BRTWOWAY_CC. This allows the removal of a rather nasty 
fixme from the PowerPC backend.  Emit slightly better code for legalizing
select_cc.

llvm-svn: 22805
 2005-08-16 19:49:35 +00:00
Chris Lattner
65b9983515 Allow passing a dag into dump and getOperationName. If one is available 
when printing a node, use it to render target operations with their
target instruction name instead of "<<unknown>>".

llvm-svn: 22804
 2005-08-16 18:33:07 +00:00
Chris Lattner
1b07a165e0 Use a extant helper to do this. 
llvm-svn: 22802
 2005-08-16 18:31:23 +00:00
Chris Lattner
73348d1e89 Add some methods for dag->dag isel. 
Split RemoveNodeFromCSEMaps out of DeleteNodesIfDead to do it.

llvm-svn: 22801
 2005-08-16 18:17:10 +00:00
Nate Begeman
54423e60c6 Fix last night's PPC32 regressions by 
1. Not selecting the false value of a select_cc in the false arm, which
   isn't legal for nested selects.
2. Actually returning the node we created and Legalized in the FP_TO_UINT
   Expander.

llvm-svn: 22789
 2005-08-14 18:38:32 +00:00
Nate Begeman
9be6a214ff Teach the legalizer how to legalize FP_TO_UINT. 
Teach the legalizer to promote FP_TO_UINT to FP_TO_SINT if the wider
  FP_TO_UINT is also illegal.  This allows us on PPC to codegen
  unsigned short foo(float a) { return a; }

as:
_foo:
.LBB_foo_0:     ; entry
        fctiwz f0, f1
        stfd f0, -8(r1)
        lwz r2, -4(r1)
        rlwinm r3, r2, 0, 16, 31
        blr

instead of:
_foo:
.LBB_foo_0:     ; entry
        fctiwz f0, f1
        stfd f0, -8(r1)
        lwz r2, -4(r1)
        lis r3, ha16(.CPI_foo_0)
        lfs f0, lo16(.CPI_foo_0)(r3)
        fcmpu cr0, f1, f0
        blt .LBB_foo_2  ; entry
.LBB_foo_1:     ; entry
        fsubs f0, f1, f0
        fctiwz f0, f0
        stfd f0, -16(r1)
        lwz r2, -12(r1)
        xoris r2, r2, 32768
.LBB_foo_2:     ; entry
        rlwinm r3, r2, 0, 16, 31
        blr

llvm-svn: 22785
 2005-08-14 01:20:53 +00:00
Nate Begeman
021a5b3fe1 Remove an unncessary argument to SimplifySelectCC and add an additional 
assert when creating a select_cc node.

llvm-svn: 22780
 2005-08-13 06:14:17 +00:00
Nate Begeman
4e8f777256 Fix the fabs regression on x86 by abstracting the select_cc optimization 
out into SimplifySelectCC.  This allows both ISD::SELECT and ISD::SELECT_CC
to use the same set of simplifying folds.

llvm-svn: 22779
 2005-08-13 06:00:21 +00:00
Chris Lattner
e06d2c3760 implement a couple of simple shift foldings. 
e.g.  (X & 7) >> 3   -> 0

llvm-svn: 22774
 2005-08-12 23:54:58 +00:00
Nate Begeman
09c56e0432 Add a select_cc optimization for recognizing abs(int). This speeds up an 
integer MPEG encoding loop by a factor of two.

llvm-svn: 22758
 2005-08-11 02:18:13 +00:00
Nate Begeman
206e850add Some SELECT_CC cleanups: 
1. move assertions for node creation to getNode()
2. legalize the values returned in ExpandOp immediately
3. Move select_cc optimizations from SELECT's getNode() to SELECT_CC's,
   allowing them to be cleaned up significantly.

This paves the way to pick up additional optimizations on SELECT_CC, such
as sum-of-absolute-differences.

llvm-svn: 22757
 2005-08-11 01:12:20 +00:00
Nate Begeman
eddc9d4856 Add new node, SELECT_CC. This node is for targets that don't natively 
implement SELECT.

llvm-svn: 22755
 2005-08-10 20:51:12 +00:00
Chris Lattner
51cf9fd316 Fix an oversight that may be causing PR617. 
llvm-svn: 22753
 2005-08-10 17:37:53 +00:00
Chris Lattner
3179a74493 Fix spelling, fix some broken canonicalizations by my last patch 
llvm-svn: 22734
 2005-08-09 23:09:05 +00:00
Chris Lattner
3290ca9983 add cc nodes to the AllNodes list so they show up in Graphviz output 
llvm-svn: 22731
 2005-08-09 20:40:02 +00:00
Chris Lattner
0fa4402b59 Eliminate the SetCCSDNode in favor of a CondCodeSDNode class. This pulls the 
CC out of the SetCC operation, making SETCC a standard ternary operation and
CC's a standard DAG leaf.  This will make it possible for other node to use
CC's as operands in the future...

llvm-svn: 22728
 2005-08-09 20:20:18 +00:00
Chris Lattner
e7f14fb39d Handle 64-bit constant exprs on 64-bit targets. 
llvm-svn: 22696
 2005-08-08 04:26:32 +00:00
Chris Lattner
fdb467b18d add a small simplification that can be exposed after promotion/expansion 
llvm-svn: 22691
 2005-08-07 05:00:44 +00:00
Chris Lattner
d3a8084e5b Change FindEarliestCallSeqEnd (used by libcall insertion) to use a set to 
avoid revisiting nodes more than once.  This eliminates a source of
potentially exponential behavior.  For a small function in 191.fma3d
(hexah_stress_divergence_), this speeds up isel from taking > 20mins to
taking 0.07s.

llvm-svn: 22680
 2005-08-05 18:10:27 +00:00
Chris Lattner
c7a67abac2 Fix a use-of-dangling-pointer bug, from the introduction of SrcValue's. 
llvm-svn: 22679
 2005-08-05 16:55:31 +00:00
Chris Lattner
644edfb51e Fix a latent bug in the libcall inserter that was exposed by Nate's patch 
yesterday.  This fixes whetstone and a bunch of programs in the External tests.

llvm-svn: 22678
 2005-08-05 16:23:57 +00:00
Nate Begeman
348caa49b3 Fix a fixme in LegalizeDAG 
llvm-svn: 22661
 2005-08-04 21:43:28 +00:00
Misha Brukman
8b8272b648 * Unbreak release build 
* Add comments to #endif pragmas for readability

llvm-svn: 22647
 2005-08-04 14:22:41 +00:00
Chris Lattner
d124203207 Fix PR611, codegen'ing SREM of FP operands to fmod or fmodf instead of 
the sequence used for integer ops

llvm-svn: 22629
 2005-08-03 20:31:37 +00:00
Chris Lattner
cc8ae687e1 Update to use the new MathExtras.h support for log2 computation. 
Patch contributed by Jim Laskey!

llvm-svn: 22594
 2005-08-02 19:26:06 +00:00
Chris Lattner
83f0262a2c Fix casts from long to sbyte on ppc 
llvm-svn: 22570
 2005-08-01 18:16:37 +00:00
Jeff Cohen
019104459d Keep tabs and trailing spaces out. 
llvm-svn: 22565
 2005-07-30 18:33:25 +00:00
Chris Lattner
d742a80e9e fix float->long conversions on x86 
llvm-svn: 22563
 2005-07-30 01:40:57 +00:00
Chris Lattner
e0b705ba00 Allow targets to have custom expanders for FP_TO_*INT conversions where 
both the src and dest values are legal

llvm-svn: 22555
 2005-07-30 00:04:12 +00:00
Chris Lattner
8d48aef4e3 Allow targets to define custom expanders for FP_TO_*INT 
llvm-svn: 22548
 2005-07-29 00:33:32 +00:00
Chris Lattner
f355c0f6ea allow a target to request that unknown FP_TO_*INT conversion be promoted to 
a larger integer destination.

llvm-svn: 22547
 2005-07-29 00:11:56 +00:00
Chris Lattner
6b4f386826 instead of having all conversions be handled by one case value, and then have 
subcases inside, break things out earlier.

llvm-svn: 22546
 2005-07-28 23:31:12 +00:00
Andrew Lenharth
f623af9b64 new is not a valid default anywhere, so make this pure virtual 
llvm-svn: 22542
 2005-07-28 18:13:59 +00:00
Chris Lattner
b0658628c1 Fix debug info to not print out recently freed memory. 
llvm-svn: 22529
 2005-07-27 23:11:25 +00:00
Chris Lattner
1a3a4c7791 Print symbolic register names in debug dumps 
llvm-svn: 22528
 2005-07-27 23:03:38 +00:00
Jeff Cohen
bd51ec7461 Eliminate all remaining tabs and trailing spaces. 
llvm-svn: 22523
 2005-07-27 06:12:32 +00:00
Nate Begeman
a25a2010e3 Remove unnecessary FP_EXTEND. This causes worse codegen for SSE. 
llvm-svn: 22469
 2005-07-19 16:50:03 +00:00
Chris Lattner
d4f9ab3809 The assertion was wrong: the code only worked for i64. While we're at it, 
expand the code to work for all integer datatypes.  This should unbreak
alpha.

llvm-svn: 22464
 2005-07-18 04:31:14 +00:00
Chris Lattner
07d79f8aa7 Only get the .bss and .data sections when needed instead of unconditionally. 
This allows is to not emit empty sections when .data or .bss is not used.

llvm-svn: 22457
 2005-07-16 17:41:06 +00:00
Chris Lattner
60bcec0238 Refactor getSection() method to make it easier to use. 
llvm-svn: 22455
 2005-07-16 17:36:04 +00:00
Chris Lattner
40fbf63df8 Major refactor of the ELFWriter code. Instead of building up one big 
vector that represents the .o file at once, build up a vector for each
section of the .o file.  This is needed because the .o file writer needs
to be able to switch between sections as it emits them (e.g. switch
between the .text section and the .rel section when emitting code).

This patch has no functionality change.

llvm-svn: 22453
 2005-07-16 08:01:13 +00:00
Nate Begeman
160c12d896 Teach the legalizer how to promote SINT_TO_FP to a wider SINT_TO_FP that 
the target natively supports.  This eliminates some special-case code from
the x86 backend and generates better code as well.

For an i8 to f64 conversion, before & after:

_x87 before:
        subl $2, %esp
        movb 6(%esp), %al
        movsbw %al, %ax
        movw %ax, (%esp)
        filds (%esp)
        addl $2, %esp
        ret

_x87 after:
        subl $2, %esp
        movsbw 6(%esp), %ax
        movw %ax, (%esp)
        filds (%esp)
        addl $2, %esp
        ret

_sse before:
        subl $12, %esp
        movb 16(%esp), %al
        movsbl %al, %eax
        cvtsi2sd %eax, %xmm0
        addl $12, %esp
        ret

_sse after:
        subl $12, %esp
        movsbl 16(%esp), %eax
        cvtsi2sd %eax, %xmm0
        addl $12, %esp
        ret

llvm-svn: 22452
 2005-07-16 02:02:34 +00:00
Chris Lattner
10da57bfed Break the code for expanding UINT_TO_FP operations out into its own 
SelectionDAGLegalize::ExpandLegalUINT_TO_FP method.

Add a new method, PromoteLegalUINT_TO_FP, which allows targets to request
that UINT_TO_FP operations be promoted to a larger input type.  This is
useful for targets that have some UINT_TO_FP or SINT_TO_FP operations but
not all of them (like X86).

The same should be done with SINT_TO_FP, but this patch does not do that
yet.

llvm-svn: 22447
 2005-07-16 00:19:57 +00:00
Chris Lattner
94e486c56e You can't use config options without config.h 
llvm-svn: 22446
 2005-07-15 22:48:31 +00:00
Chris Lattner
d8eb6ea6da Make this use the new autoconf support for finding the executables for 
gv and Graphviz.

llvm-svn: 22434
 2005-07-14 05:33:13 +00:00
Chris Lattner
d9f1a60c61 As discussed on IRC, this stuff is just for debugging. 
llvm-svn: 22432
 2005-07-14 05:17:43 +00:00
Chris Lattner
61b33e0bc4 If the Graphviz program is available, use it to visualize dot graphs. 
llvm-svn: 22429
 2005-07-14 01:10:55 +00:00
Chris Lattner
aeae45b371 Fix Alpha/2005-07-12-TwoMallocCalls.ll and PR593. 
It is not safe to call LegalizeOp on something that has already been legalized.
Instead, just force another iteration of legalization.

This could affect all platforms but X86, as this codepath is dynamically
dead on X86 (ISD::MEMSET and friends are legal).

llvm-svn: 22419
 2005-07-13 02:00:04 +00:00
Chris Lattner
628a248ff9 Fix test/Regression/CodeGen/Generic/2005-07-12-memcpy-i64-length.ll 
llvm-svn: 22417
 2005-07-13 01:42:45 +00:00
Chris Lattner
bec12eb953 Add support for 64-bit elf files 
llvm-svn: 22400
 2005-07-12 06:57:52 +00:00
Jeff Cohen
7bc4266cf1 VC++ demands that the function returns a value 
llvm-svn: 22393
 2005-07-12 02:53:33 +00:00
Chris Lattner
8dd11b0f9c Clean up code, no functionality changes. 
llvm-svn: 22382
 2005-07-11 06:34:30 +00:00
Chris Lattner
d710b0a025 Emit a symbol table entry for each function we output to the ELF file. This 
allows objdump to know which function we are emitting to:

00000000 <foo>:     <----
   0:   b8 01 00 00 00          mov    $0x1,%eax
   5:   03 44 24 04             add    0x4(%esp,1),%eax
   9:   c3                      ret

... and allows .o files to be useful for linking :)

llvm-svn: 22378
 2005-07-11 06:17:35 +00:00
Chris Lattner
34d2a2ae23 add code to emit the .text section to the section header. 
Add a *VERY INITIAL* machine code emitter class.  This is enough to take
this C function:
int foo(int X) { return X +1; }

and make objdump produce the following:

$ objdump -d t-llvm.o

t-llvm.o:     file format elf32-i386

Disassembly of section .text:

00000000 <.text>:
   0:   b8 01 00 00 00          mov    $0x1,%eax
   5:   03 44 24 04             add    0x4(%esp,1),%eax
   9:   c3                      ret


Anything using branches or refering to the constant pool or requiring
relocations will not work yet.

llvm-svn: 22375
 2005-07-11 05:17:18 +00:00
Chris Lattner
8c10fbf3cc Use a name mangler object to uniquify names and remove nonstandard 
characters from them.

llvm-svn: 22371
 2005-07-11 03:11:47 +00:00
Chris Lattner
6e49696ba6 Change *EXTLOAD to use an VTSDNode operand instead of being an MVTSDNode. 
This is the last MVTSDNode.

This allows us to eliminate a bunch of special case code for handling
MVTSDNodes.

llvm-svn: 22367
 2005-07-10 01:55:33 +00:00
Chris Lattner
273b81e0c0 Change TRUNCSTORE to use a VTSDNode operand instead of being an MVTSTDNode 
llvm-svn: 22366
 2005-07-10 00:29:18 +00:00
Chris Lattner
c355896290 Introduce a new VTSDNode class with the ultimate goal of eliminating the 
MVTSDNode class.  This class is used to provide an operand to operators
that require an extra type.  We start by converting FP_ROUND_INREG and
SIGN_EXTEND_INREG over to using it.

llvm-svn: 22364
 2005-07-10 00:07:11 +00:00
Chris Lattner
de44e16474 Add support for emitting a .data section and .bss section. 
Add support for emitting external and .bss symbols.

llvm-svn: 22358
 2005-07-08 05:47:00 +00:00
Chris Lattner
efccb190b5 Add support for emitting the symbol table (and its string table) of the 
module to the ELF file.  Test it by adding support for emitting common
symbols.  This allows us to compile this:

%X = weak global int 0
%Y = weak global int 0
%Z = weak global int 0

to an elf file that 'readelf's this:

Symbol table '.symtab' contains 4 entries:
   Num:    Value  Size Type    Bind   Vis      Ndx Name
     0: 00000000     0 NOTYPE  LOCAL  DEFAULT  UND
     1: 00000004     4 OBJECT  GLOBAL DEFAULT  COM X
     2: 00000004     4 OBJECT  GLOBAL DEFAULT  COM Y
     3: 00000004     4 OBJECT  GLOBAL DEFAULT  COM Z

llvm-svn: 22343
 2005-07-07 07:02:20 +00:00
Chris Lattner
bf100c8bdb Make several cleanups to Andrews varargs change: 
1. Pass Value*'s into lowering methods so that the proper pointers can be
   added to load/stores from the valist
2. Intrinsics that return void should only return a token chain, not a token
   chain/retval pair.
3. Rename LowerVAArgNext -> LowerVAArg, because VANext is long gone.

llvm-svn: 22338
 2005-07-05 19:57:53 +00:00
Andrew Lenharth
3543e3b3a9 2 fixes: 
1: Legalize operand in UINT_TO_FP expanision

2: SRA x, const i8 was not promoting the constant to shift amount type.
llvm-svn: 22337
 2005-07-05 19:52:39 +00:00
Andrew Lenharth
c9903eb2cc I really didn't think this was necessary. But, Legalize wasn't running again 
and legalizing the extload.  Strange.  Should fix most alpha regressions.

llvm-svn: 22329
 2005-07-02 20:58:53 +00:00
Andrew Lenharth
b8c48ce74e oops 
llvm-svn: 22320
 2005-06-30 19:32:57 +00:00
Andrew Lenharth
04aa18bd2a FP EXTLOAD is not support on all archs, expand to LOAD and FP_EXTEND 
llvm-svn: 22319
 2005-06-30 19:22:37 +00:00
Andrew Lenharth
898efb338a restore old srcValueNode behavior and try to to work around it 
llvm-svn: 22315
 2005-06-29 18:54:02 +00:00
Andrew Lenharth
edccb834bb tracking the instructions causing loads and stores provides more information than just the pointer being loaded or stored 
llvm-svn: 22311
 2005-06-29 15:57:19 +00:00
Andrew Lenharth
d534c5cb2a Adapt the code for handling uint -> fp conversion for the 32 bit case to 
handling it in the 64 bit case.  The two code paths should probably be merged.

llvm-svn: 22302
 2005-06-27 23:28:32 +00:00
Chris Lattner
8be1980730 iniital checkin of ELFWriter implementation 
For now, the elf writer is only capable of emitting an empty elf file, with
a section table and a section table string table.  This will be enhanced
in the future :)

llvm-svn: 22291
 2005-06-27 06:29:00 +00:00
Andrew Lenharth
4fd2bde906 If we support structs as va_list, we must pass pointers to them to va_copy 
See last commit for LangRef, this implements it on all targets.

llvm-svn: 22273
 2005-06-22 21:04:42 +00:00
Andrew Lenharth
a9214fec08 core changes for varargs 
llvm-svn: 22254
 2005-06-18 18:34:52 +00:00
Nate Begeman
ed49a51836 Fix bug 537 test 2, which checks to make sure that we fold A+(B-A) -> B for 
integer types.  Add a couple checks to not perform these kinds of transform
on floating point values.

llvm-svn: 22228
 2005-06-16 07:06:03 +00:00
Duraid Madina
f4ca48c89e aCC and STLport complained about this, because they're like that 
llvm-svn: 22053
 2005-05-15 13:05:48 +00:00
Chris Lattner
811dc49f55 Add some simplifications for MULH[SU]. This allows us to compile this: 
long %bar(long %X) {
  %Y = mul long %X, 4294967297
  ret long %Y
}

to this:

l1_bar:
        mov %EAX, DWORD PTR [%ESP + 4]
        mov %EDX, %EAX
        add %EDX, DWORD PTR [%ESP + 8]
        ret

instead of:

l1_bar:
        mov %ECX, DWORD PTR [%ESP + 4]
        mov %EDX, 1
        mov %EAX, %ECX
        mul %EDX
        add %EDX, %ECX
        add %EDX, DWORD PTR [%ESP + 8]
        mov %EAX, %ECX
        ret

llvm-svn: 22044
 2005-05-15 05:39:08 +00:00
Chris Lattner
54bf1cc313 When inserting callee-save register reloads, make sure to skip over any 
terminator instructions before the 'ret' in case the target has a
multi-instruction return sequence.

llvm-svn: 22041
 2005-05-15 03:09:58 +00:00
Chris Lattner
46de5c99bd Fix construction of ioport intrinsics, fixing X86/io.llx and io-port.llx 
llvm-svn: 22026
 2005-05-14 13:56:55 +00:00
Chris Lattner
052759b78c allow token chain at start or end of node 
llvm-svn: 22020
 2005-05-14 08:34:53 +00:00
Chris Lattner
d9e36f94bb remove special case hacks for readport/readio from the binary operator 
codepath

llvm-svn: 22019
 2005-05-14 07:45:46 +00:00
Chris Lattner
d1d8fbee2d Implement fixme's by memoizing nodes. 
llvm-svn: 22018
 2005-05-14 07:42:29 +00:00
Chris Lattner
ac7d55f114 Turn this into a wrapper for a simpler version of getNode. 
llvm-svn: 22016
 2005-05-14 07:32:14 +00:00
Chris Lattner
a035798c4b Eliminate special purpose hacks for dynamic_stack_alloc. 
llvm-svn: 22015
 2005-05-14 07:29:57 +00:00
Chris Lattner
b94e243d14 Use the general mechanism for creating multi-value nodes instead of using 
special case hacks.

llvm-svn: 22014
 2005-05-14 07:25:05 +00:00
Chris Lattner
ad411081fb Wrap long line, actually add node to the graph. 
llvm-svn: 22011
 2005-05-14 06:42:57 +00:00
Chris Lattner
6f7b63c7d7 legalize target-specific operations 
llvm-svn: 22010
 2005-05-14 06:34:48 +00:00
Chris Lattner
1afb5ae575 add a getNode() version that allows construction of any node type. 
llvm-svn: 22009
 2005-05-14 06:20:26 +00:00
Chris Lattner
6e81a4090f LowerOperation takes a dag 
llvm-svn: 22004
 2005-05-14 05:50:48 +00:00
Chris Lattner
369427683d Print the symbolic register name in a register allocator debug dump. 
llvm-svn: 22002
 2005-05-14 05:34:15 +00:00
Chris Lattner
1202c26d6e Allow targets to have a custom int64->fp expander if desired 
llvm-svn: 22001
 2005-05-14 05:33:54 +00:00
Chris Lattner
2163eeaa67 Align doubles on 8-byte boundaries if possible. 
llvm-svn: 21993
 2005-05-13 23:14:17 +00:00
Chris Lattner
d9dd5bb798 print stack object alignment in -print-machineinstr dumps 
llvm-svn: 21992
 2005-05-13 22:54:44 +00:00
Chris Lattner
a3e57bd844 Tolerate instrs with extra args 
llvm-svn: 21982
 2005-05-13 21:07:15 +00:00
Chris Lattner
9d788e93a6 Add an isTailCall flag to LowerCallTo 
llvm-svn: 21958
 2005-05-13 18:50:42 +00:00
Chris Lattner
3a76f85d43 Handle TAILCALL node 
llvm-svn: 21957
 2005-05-13 18:43:43 +00:00
Chris Lattner
01eba53a10 Emit function entry code after lowering hte arguments. 
llvm-svn: 21931
 2005-05-13 07:33:32 +00:00
Chris Lattner
fdc4816996 Allow targets to emit code into the entry block of each function 
llvm-svn: 21930
 2005-05-13 07:23:21 +00:00
Chris Lattner
5cd760d12d allow a virtual register to be associated with live-in values. 
llvm-svn: 21927
 2005-05-13 07:08:07 +00:00
Chris Lattner
670c7f516c Fix a problem that nate reduced for me. 
llvm-svn: 21923
 2005-05-13 05:17:00 +00:00
Chris Lattner
59bb0edb45 rename variables and functions to match renamed DAG nodes. Bonus feature: 
I can actually remember which one is which now!

llvm-svn: 21922
 2005-05-13 05:09:11 +00:00
Chris Lattner
c7013ec3a9 do not call expandop on the same value more than once. This fixes 
X86/2004-02-22-Casts.llx

llvm-svn: 21919
 2005-05-13 04:45:13 +00:00
Chris Lattner
51de10e0c6 fix a bad typeo 
llvm-svn: 21917
 2005-05-12 23:51:40 +00:00
Chris Lattner
00d2fb482f update comment 
llvm-svn: 21916
 2005-05-12 23:24:44 +00:00
Chris Lattner
094bbfcebb rename the ADJCALLSTACKDOWN/ADJCALLSTACKUP nodes to be CALLSEQ_START/BEGIN. 
llvm-svn: 21915
 2005-05-12 23:24:06 +00:00
Chris Lattner
dd2700de99 Pass calling convention to use into lower call to 
llvm-svn: 21900
 2005-05-12 19:56:57 +00:00
Chris Lattner
ad48ef0a7d fix expansion of ct[lt]z nodes 
llvm-svn: 21896
 2005-05-12 19:27:51 +00:00
Chris Lattner
6b5bacbc0b Expand 64-bit ctlz/cttz nodes for 32-bit targets 
llvm-svn: 21895
 2005-05-12 19:05:01 +00:00
Chris Lattner
3677432d39 Fix uint->fp casts on PPC, allowing UnitTests/2005-05-12-Int64ToFP to 
work on it.

llvm-svn: 21894
 2005-05-12 18:52:34 +00:00
Chris Lattner
dbcdac1ebf Allow something to be legalized multiple times. This can be used to reduce 
legalization iteration

llvm-svn: 21892
 2005-05-12 16:53:42 +00:00
Chris Lattner
a9a41e8856 Oops, don't do this after we figure out where to insert the call chains. 
llvm-svn: 21890
 2005-05-12 07:00:44 +00:00
Chris Lattner
b58308e6d4 Make sure to expand all nodes, avoiding unintentional node duplication. 
llvm-svn: 21889
 2005-05-12 06:54:21 +00:00
Chris Lattner
9f40cfa0a1 handle a common case generated by the uint64 -> FP code path better 
llvm-svn: 21888
 2005-05-12 06:27:02 +00:00
Chris Lattner
1c248e7462 add fixme 
llvm-svn: 21887
 2005-05-12 06:04:14 +00:00
Chris Lattner
1196356365 Fix a problem where early legalization can cause token chain problems. 
llvm-svn: 21885
 2005-05-12 04:49:08 +00:00
Chris Lattner
b38ffd7fbf Make legalize a bit more efficient, and canonicalize sub X, C -> add X, -C 
llvm-svn: 21882
 2005-05-12 00:17:04 +00:00
Nate Begeman
e84f776b5d Necessary changes to codegen cttz efficiently on PowerPC 
1. Teach LegalizeDAG how to better legalize CTTZ if the target doesn't have
   CTPOP, but does have CTLZ
2. Teach PPC32 how to do sub x, const -> add x, -const for valid consts
3. Teach PPC32 how to do and (xor a, -1) b -> andc b, a
4. Teach PPC32 that ISD::CTLZ -> PPC::CNTLZW

llvm-svn: 21880
 2005-05-11 23:43:56 +00:00
Chris Lattner
8f2d079b36 Fix lowering of ctlz, so now UnitTests/2005-05-11-Popcount-ffs-fls passes 
with the CBE

llvm-svn: 21875
 2005-05-11 20:24:12 +00:00
Chris Lattner
303ac68c80 Fix lowering of cttz to work with signed values 
llvm-svn: 21874
 2005-05-11 20:02:14 +00:00
Chris Lattner
330f44f3b6 fix and concisify intinsic lowering for ctpop. Unfortunately, this code 
looks completely untested. :(

llvm-svn: 21873
 2005-05-11 19:42:05 +00:00
Chris Lattner
eeeaf45bba Fix the last remaining bug preventing us from switching the X86 BE over 
from the simple isel to the pattern isel.  This forces inserted libcalls
to serialize against other function calls, which was breaking
UnitTests/2005-05-12-Int64ToFP.  Hopefully this will fix issues on other
targets as well.

llvm-svn: 21872
 2005-05-11 19:02:11 +00:00
Chris Lattner
296754995e Do not memoize ADJCALLSTACKDOWN nodes, provide a method to hack on them. 
llvm-svn: 21871
 2005-05-11 18:57:39 +00:00
Chris Lattner
74763db128 wrap long line 
llvm-svn: 21870
 2005-05-11 18:57:06 +00:00
Chris Lattner
d76582b540 Make sure to legalize generated ctpop nodes, convert tabs to spaces 
llvm-svn: 21868
 2005-05-11 18:35:21 +00:00
Duraid Madina
8ad9786fcd expand count-leading/trailing-zeros; the test 2005-05-11-Popcount-ffs-fls.c 
should now pass (the "LLVM" and "REF" results should be identical)

llvm-svn: 21866
 2005-05-11 08:45:08 +00:00
Chris Lattner
b452b5aa42 Add some notes for expanding clz/ctz 
llvm-svn: 21862
 2005-05-11 05:27:09 +00:00
Chris Lattner
4f05136f61 Simplify this code, use the proper shift amount 
llvm-svn: 21861
 2005-05-11 05:21:31 +00:00
Chris Lattner
3edc8ecb53 Legalize this correctly 
llvm-svn: 21859
 2005-05-11 05:09:47 +00:00
Chris Lattner
457996c4a6 implement expansion of ctpop nodes, implementing CodeGen/Generic/llvm-ct-intrinsics.ll 
llvm-svn: 21856
 2005-05-11 04:51:16 +00:00
Chris Lattner
ce84b90a3d Print bit count nodes correctly 
llvm-svn: 21855
 2005-05-11 04:50:30 +00:00
Jeff Cohen
afc58006b7 Silence some VC++ warnings 
llvm-svn: 21838
 2005-05-10 02:22:38 +00:00
Chris Lattner
5edb4c4af6 The semantics of cast X to bool are a comparison against zero, not a truncation! 
llvm-svn: 21833
 2005-05-09 22:17:13 +00:00
Chris Lattner
95c836384b legalize readio/writeio into a load/store if requested 
llvm-svn: 21827
 2005-05-09 20:36:57 +00:00
Chris Lattner
7cc8edfc30 legalize READPORT, WRITEPORT, READIO, WRITEIO, at least in the basic cases 
where they are directly supported by the architecture.  Wrap a bunch of
long lines :(

llvm-svn: 21826
 2005-05-09 20:23:03 +00:00
Chris Lattner
af6bde0db6 Add support for matching the READPORT, WRITEPORT, READIO, WRITEIO intrinsics 
llvm-svn: 21825
 2005-05-09 20:22:36 +00:00
Chris Lattner
eee649df34 Add support for READPORT, WRITEPORT, READIO, WRITEIO 
llvm-svn: 21824
 2005-05-09 20:22:17 +00:00
Chris Lattner
c3fa88e7c8 Fold shifts into subsequent SHL's. These shifts often arise due to addrses 
arithmetic lowering.

llvm-svn: 21818
 2005-05-09 17:06:45 +00:00
Chris Lattner
a1e633ef7a Don't use the load/store instruction as the source pointer, use the pointer 
being stored/loaded through!

llvm-svn: 21806
 2005-05-09 04:28:51 +00:00
Chris Lattner
bfbefe0837 memoize all nodes, even null Value* nodes. Do not add two token chain outputs 
llvm-svn: 21805
 2005-05-09 04:14:13 +00:00
Chris Lattner
b85030373d wrap long lines 
llvm-svn: 21804
 2005-05-09 04:08:33 +00:00
Chris Lattner
6ffae1a3ec Print SrcValue nodes correctly 
llvm-svn: 21803
 2005-05-09 04:08:27 +00:00
Chris Lattner
6d85b91b24 Wrap long lines. Fix "warning: conflicting types for built-in function 'memset'" 
warning from the CBE+GCC.

llvm-svn: 21779
 2005-05-08 19:46:29 +00:00
Misha Brukman
1996bf6ea5 * Order #includes alphabetically 
* Remove commented-out debug printouts

llvm-svn: 21707
 2005-05-05 23:45:17 +00:00
Chris Lattner
6e8167d1c2 When hitting an unsupported intrinsic, actually print it 
Lower debug info to noops.

llvm-svn: 21698
 2005-05-05 17:55:17 +00:00
Andrew Lenharth
09c3c4add4 ctpop lowering in legalize 
llvm-svn: 21697
 2005-05-05 15:55:21 +00:00
Andrew Lenharth
9282d00d4f Make promoteOp work for CT* 
Proof?

ubyte %bar(ubyte %x) {
entry:
        %tmp.1 = call ubyte %llvm.ctlz( ubyte %x )
        ret ubyte %tmp.1
}

==>

zapnot $16,1,$0
CTLZ $0,$0
subq $0,56,$0
zapnot $0,1,$0
ret $31,($26),1

llvm-svn: 21691
 2005-05-04 19:11:05 +00:00
Andrew Lenharth
8b64bd0fd5 Implement count leading zeros (ctlz), count trailing zeros (cttz), and count 
population (ctpop).  Generic lowering is implemented, however only promotion
is implemented for SelectionDAG at the moment.

More coming soon.

llvm-svn: 21676
 2005-05-03 17:19:30 +00:00
Alkis Evlogimenos
66f1632de8 Do not use deprecated APIs 
llvm-svn: 21639
 2005-04-30 07:13:31 +00:00
Chris Lattner
fe72cdf838 Codegen and legalize sin/cos/llvm.sqrt as FSIN/FCOS/FSQRT calls. This patch 
was contributed by Morten Ofstad, with some minor tweaks and bug fixes added
by me.

llvm-svn: 21636
 2005-04-30 04:43:14 +00:00
Chris Lattner
0366e4c0d3 Lower llvm.sqrt -> fsqrt/sqrt 
llvm-svn: 21629
 2005-04-30 04:07:50 +00:00
Chris Lattner
6ec8bb9e8d Legalize FSQRT, FSIN, FCOS nodes, patch contributed by Morten Ofstad 
llvm-svn: 21606
 2005-04-28 21:44:33 +00:00
Chris Lattner
4678a790e6 Add FSQRT, FSIN, FCOS nodes, patch contributed by Morten Ofstad 
llvm-svn: 21605
 2005-04-28 21:44:03 +00:00
Andrew Lenharth
2a00530fa7 Implement Value* tracking for loads and stores in the selection DAG. This enables one to use alias analysis in the backends. 
(TRUNK)Stores and (EXT|ZEXT|SEXT)Loads have an extra SDOperand which is a SrcValueSDNode which contains the Value*.  Note that if the operation is introduced by the backend, it will still have the operand, but the value* will be null.

llvm-svn: 21599
 2005-04-27 20:10:01 +00:00
Chris Lattner
15bcc5273b Fold (X > -1) | (Y > -1) --> (X&Y > -1) 
llvm-svn: 21552
 2005-04-26 01:18:33 +00:00
Chris Lattner
d8ac4da793 implement some more logical compares with constants, so that: 
int foo1(int x, int y) {
  int t1 = x >= 0;
  int t2 = y >= 0;
  return t1 & t2;
}
int foo2(int x, int y) {
  int t1 = x == -1;
  int t2 = y == -1;
  return t1 & t2;
}

produces:

_foo1:
        or r2, r4, r3
        srwi r2, r2, 31
        xori r3, r2, 1
        blr
_foo2:
        and r2, r4, r3
        addic r2, r2, 1
        li r2, 0
        addze r3, r2
        blr

instead of:

_foo1:
        srwi r2, r4, 31
        xori r2, r2, 1
        srwi r3, r3, 31
        xori r3, r3, 1
        and r3, r2, r3
        blr
_foo2:
        addic r2, r4, 1
        li r2, 0
        addze r2, r2
        addic r3, r3, 1
        li r3, 0
        addze r3, r3
        and r3, r2, r3
        blr

llvm-svn: 21547
 2005-04-25 21:20:28 +00:00
Chris Lattner
7931b75a81 Codegen x < 0 | y < 0 as (x|y) < 0. This allows us to compile this to: 
_foo:
        or r2, r4, r3
        srwi r3, r2, 31
        blr

instead of:

_foo:
        srwi r2, r4, 31
        srwi r3, r3, 31
        or r3, r2, r3
        blr

llvm-svn: 21544
 2005-04-25 21:03:25 +00:00
Misha Brukman
a9a1982a44 Convert tabs to spaces 
llvm-svn: 21439
 2005-04-22 04:01:18 +00:00
Misha Brukman
774e55c446 Remove trailing whitespace 
llvm-svn: 21420
 2005-04-21 22:36:52 +00:00
Chris Lattner
87fbc1c554 Improve and elimination. On PPC, for: 
bool %test(int %X) {
        %Y = and int %X, 8
        %Z = setne int %Y, 0
        ret bool %Z
}

we now generate this:

        rlwinm r2, r3, 0, 28, 28
        srwi r3, r2, 3

instead of this:

        rlwinm r2, r3, 0, 28, 28
        srwi r2, r2, 3
        rlwinm r3, r2, 0, 31, 31

I'll leave it to Nate to get it down to one instruction. :)

---------------------------------------------------------------------

llvm-svn: 21391
 2005-04-21 06:28:15 +00:00
Chris Lattner
d0a2fda2c6 Fold (x & 8) != 0 and (x & 8) == 8 into (x & 8) >> 3. 
This turns this PPC code:

        rlwinm r2, r3, 0, 28, 28
        cmpwi cr7, r2, 8
        mfcr r2
        rlwinm r3, r2, 31, 31, 31

into this:

        rlwinm r2, r3, 0, 28, 28
        srwi r2, r2, 3
        rlwinm r3, r2, 0, 31, 31

Next up, nuking the extra and.

llvm-svn: 21390
 2005-04-21 06:12:41 +00:00
Chris Lattner
188ecaab1d Fold setcc of MVT::i1 operands into logical operations 
llvm-svn: 21319
 2005-04-18 04:48:12 +00:00
Chris Lattner
72aca1b758 Another minor simplification: handle setcc (zero_extend x), c -> setcc(x, c') 
llvm-svn: 21318
 2005-04-18 04:30:45 +00:00
Chris Lattner
e6117e5d4f Another simple xform 
llvm-svn: 21317
 2005-04-18 04:11:19 +00:00
Chris Lattner
f6f5b23a00 Fold: 
// (X != 0) | (Y != 0) -> (X|Y != 0)
        // (X == 0) & (Y == 0) -> (X|Y == 0)

Compiling this:

int %bar(int %a, int %b) {
        entry:
        %tmp.1 = setne int %a, 0
        %tmp.2 = setne int %b, 0
        %tmp.3 = or bool %tmp.1, %tmp.2
        %retval = cast bool %tmp.3 to int
        ret int %retval
        }

to this:

_bar:
        or r2, r3, r4
        addic r3, r2, -1
        subfe r3, r3, r2
        blr

instead of:

_bar:
        addic r2, r3, -1
        subfe r2, r2, r3
        addic r3, r4, -1
        subfe r3, r3, r4
        or r3, r2, r3
        blr

llvm-svn: 21316
 2005-04-18 03:59:53 +00:00
Chris Lattner
a32c50520c Make the AND elimination operation recursive and significantly more powerful, 
eliminating an and for Nate's testcase:

int %bar(int %a, int %b) {
        entry:
        %tmp.1 = setne int %a, 0
        %tmp.2 = setne int %b, 0
        %tmp.3 = or bool %tmp.1, %tmp.2
        %retval = cast bool %tmp.3 to int
        ret int %retval
        }

generating:

_bar:
        addic r2, r3, -1
        subfe r2, r2, r3
        addic r3, r4, -1
        subfe r3, r3, r4
        or r3, r2, r3
        blr

instead of:

_bar:
        addic r2, r3, -1
        subfe r2, r2, r3
        addic r3, r4, -1
        subfe r3, r3, r4
        or r2, r2, r3
        rlwinm r3, r2, 0, 31, 31
        blr

llvm-svn: 21315
 2005-04-18 03:48:41 +00:00
Nate Begeman
ce63e383b8 Add a couple missing transforms in getSetCC that were triggering assertions 
in the PPC Pattern ISel

llvm-svn: 21297
 2005-04-14 08:56:52 +00:00
Nate Begeman
20b3399465 Disbale the broken fold of shift + sz[ext] for now 
Move the transform for select (a < 0) ? b : 0 into the dag from ppc isel
Enable the dag to fold and (setcc, 1) -> setcc for targets where setcc
  always produces zero or one.

llvm-svn: 21291
 2005-04-13 21:23:31 +00:00
Chris Lattner
89f7e115a4 fix an infinite loop 
llvm-svn: 21289
 2005-04-13 20:06:29 +00:00
Chris Lattner
475fe85ddf fix some serious miscompiles on ia64, alpha, and ppc 
llvm-svn: 21288
 2005-04-13 19:53:40 +00:00
Chris Lattner
03d675414e avoid work when possible, perhaps fix the problem nate and andrew are seeing 
with != 0 comparisons vanishing.

llvm-svn: 21287
 2005-04-13 19:41:05 +00:00
Chris Lattner
9540cf8c7e Implement expansion of unsigned i64 -> FP. 
Note that this probably only works for little endian targets, but is enough
to get siod working :)

llvm-svn: 21280
 2005-04-13 05:09:42 +00:00
Chris Lattner
1a6247ff51 Make expansion of uint->fp cast assert out instead of infinitely recurse. 
llvm-svn: 21275
 2005-04-13 03:42:14 +00:00
Chris Lattner
63450e87d9 add back the optimization that Nate added for shl X, (zext_inreg y) 
llvm-svn: 21273
 2005-04-13 02:58:13 +00:00
Chris Lattner
759afe07d7 Oops, remove these too. 
llvm-svn: 21272
 2005-04-13 02:47:57 +00:00
Chris Lattner
4f188f949c Instead of making ZERO_EXTEND_INREG nodes, use the helper method in 
SelectionDAG to do the job with AND.  Don't legalize Z_E_I anymore as
it is gone

llvm-svn: 21266
 2005-04-13 02:38:47 +00:00
Chris Lattner
bce0030a88 Remove all foldings of ZERO_EXTEND_INREG, moving them to work for AND nodes 
instead.  OVerall, this increases the amount of folding we can do.

llvm-svn: 21265
 2005-04-13 02:38:18 +00:00
Nate Begeman
38d8248a9e Fold shift x, [sz]ext(y) -> shift x, y 
llvm-svn: 21262
 2005-04-12 23:32:28 +00:00
Nate Begeman
a56527ea5f Fold shift by size larger than type size to undef 
Make llvm undef values generate ISD::UNDEF nodes

llvm-svn: 21261
 2005-04-12 23:12:17 +00:00
Chris Lattner
58f72ab722 promote extload i1 -> extload i8 
llvm-svn: 21258
 2005-04-12 20:30:10 +00:00
Chris Lattner
cfc7093ca6 Remove some redundant checks, add a couple of new ones. This allows us to 
compile this:

int foo (unsigned long a, unsigned long long g) {
  return a >= g;
}

To:

foo:
        movl 8(%esp), %eax
        cmpl %eax, 4(%esp)
        setae %al
        cmpl $0, 12(%esp)
        sete %cl
        andb %al, %cl
        movzbl %cl, %eax
        ret

instead of:

foo:
        movl 8(%esp), %eax
        cmpl %eax, 4(%esp)
        setae %al
        movzbw %al, %cx
        movl 12(%esp), %edx
        cmpl $0, %edx
        sete %al
        movzbw %al, %ax
        cmpl $0, %edx
        cmove %cx, %ax
        movzbl %al, %eax
        ret

llvm-svn: 21244
 2005-04-12 02:54:39 +00:00
Chris Lattner
61f353dbdc Emit comparisons against the sign bit better. Codegen this: 
bool %test1(long %X) {
        %A = setlt long %X, 0
        ret bool %A
}

like this:

test1:
        cmpl $0, 8(%esp)
        setl %al
        movzbl %al, %eax
        ret

instead of:

test1:
        movl 8(%esp), %ecx
        cmpl $0, %ecx
        setl %al
        movzbw %al, %ax
        cmpl $0, 4(%esp)
        setb %dl
        movzbw %dl, %dx
        cmpl $0, %ecx
        cmove %dx, %ax
        movzbl %al, %eax
        ret

llvm-svn: 21243
 2005-04-12 02:19:10 +00:00
Chris Lattner
6cbbb55967 Emit long comparison against -1 better. Instead of this (x86): 
test2:
        movl 8(%esp), %eax
        notl %eax
        movl 4(%esp), %ecx
        notl %ecx
        orl %eax, %ecx
        cmpl $0, %ecx
        sete %al
        movzbl %al, %eax
        ret

or this (PPC):

_test2:
        nor r2, r4, r4
        nor r3, r3, r3
        or r2, r2, r3
        cntlzw r2, r2
        srwi r3, r2, 5
        blr

Emit this:

test2:
        movl 8(%esp), %eax
        andl 4(%esp), %eax
        cmpl $-1, %eax
        sete %al
        movzbl %al, %eax
        ret

or this:

_test2:
.LBB_test2_0:   ;
        and r2, r4, r3
        cmpwi cr0, r2, -1
        li r3, 1
        li r2, 0
        beq .LBB_test2_2        ;
.LBB_test2_1:   ;
        or r3, r2, r2
.LBB_test2_2:   ;
        blr

it seems like the PPC isel could do better for R32 == -1 case.

llvm-svn: 21242
 2005-04-12 01:46:05 +00:00
Chris Lattner
37534d43d0 canonicalize x <u 1 -> x == 0. On this testcase: 
unsigned long long g;
unsigned long foo (unsigned long a) {
  return (a >= g) ? 1 : 0;
}

It changes the ppc code from:

_foo:
.LBB_foo_0:     ; entry
        mflr r11
        stw r11, 8(r1)
        bl "L00000$pb"
"L00000$pb":
        mflr r2
        addis r2, r2, ha16(L_g$non_lazy_ptr-"L00000$pb")
        lwz r2, lo16(L_g$non_lazy_ptr-"L00000$pb")(r2)
        lwz r4, 0(r2)
        lwz r2, 4(r2)
        cmplw cr0, r3, r2
        li r2, 1
        li r3, 0
        bge .LBB_foo_2  ; entry
.LBB_foo_1:     ; entry
        or r2, r3, r3
.LBB_foo_2:     ; entry
        cmplwi cr0, r4, 1
        li r3, 1
        li r5, 0
        blt .LBB_foo_4  ; entry
.LBB_foo_3:     ; entry
        or r3, r5, r5
.LBB_foo_4:     ; entry
        cmpwi cr0, r4, 0
        beq .LBB_foo_6  ; entry
.LBB_foo_5:     ; entry
        or r2, r3, r3
.LBB_foo_6:     ; entry
        rlwinm r3, r2, 0, 31, 31
        lwz r11, 8(r1)
        mtlr r11
        blr


to:

_foo:
.LBB_foo_0:     ; entry
        mflr r11
        stw r11, 8(r1)
        bl "L00000$pb"
"L00000$pb":
        mflr r2
        addis r2, r2, ha16(L_g$non_lazy_ptr-"L00000$pb")
        lwz r2, lo16(L_g$non_lazy_ptr-"L00000$pb")(r2)
        lwz r4, 0(r2)
        lwz r2, 4(r2)
        cmplw cr0, r3, r2
        li r2, 1
        li r3, 0
        bge .LBB_foo_2  ; entry
.LBB_foo_1:     ; entry
        or r2, r3, r3
.LBB_foo_2:     ; entry
        cntlzw r3, r4
        srwi r3, r3, 5
        cmpwi cr0, r4, 0
        beq .LBB_foo_4  ; entry
.LBB_foo_3:     ; entry
        or r2, r3, r3
.LBB_foo_4:     ; entry
        rlwinm r3, r2, 0, 31, 31
        lwz r11, 8(r1)
        mtlr r11
        blr

llvm-svn: 21241
 2005-04-12 00:28:49 +00:00
Chris Lattner
7f0f0854fa Teach the dag mechanism that this: 
long long test2(unsigned A, unsigned B) {
        return ((unsigned long long)A << 32) + B;
}

is equivalent to this:

long long test1(unsigned A, unsigned B) {
        return ((unsigned long long)A << 32) | B;
}

Now they are both codegen'd to this on ppc:

_test2:
        blr

or this on x86:

test2:
        movl 4(%esp), %edx
        movl 8(%esp), %eax
        ret

llvm-svn: 21231
 2005-04-11 20:29:59 +00:00
Chris Lattner
71f3d4ce57 Fix expansion of shifts by exactly NVT bits on arch's (like X86) that have 
masking shifts.

This fixes the miscompilation of this:

long long test1(unsigned A, unsigned B) {
        return ((unsigned long long)A << 32) | B;
}

into this:

test1:
        movl 4(%esp), %edx
        movl %edx, %eax
        orl 8(%esp), %eax
        ret

allowing us to generate this instead:

test1:
        movl 4(%esp), %edx
        movl 8(%esp), %eax
        ret

llvm-svn: 21230
 2005-04-11 20:08:52 +00:00
Nate Begeman
32163963cb Fix libcall code to not pass a NULL Chain to LowerCallTo 
Fix libcall code to not crash or assert looking for an ADJCALLSTACKUP node
  when it is known that there is no ADJCALLSTACKDOWN to match.
Expand i64 multiply when ISD::MULHU is legal for the target.

llvm-svn: 21214
 2005-04-11 03:01:51 +00:00
Chris Lattner
4f26677dc9 Don't bother sign/zext_inreg'ing the result of an and operation if we know 
the result does change as a result of the extend.

This improves codegen for Alpha on this testcase:

int %a(ushort* %i) {
        %tmp.1 = load ushort* %i
        %tmp.2 = cast ushort %tmp.1 to int
        %tmp.4 = and int %tmp.2, 1
        ret int %tmp.4
}

Generating:

a:
        ldgp $29, 0($27)
        ldwu $0,0($16)
        and $0,1,$0
        ret $31,($26),1

instead of:

a:
        ldgp $29, 0($27)
        ldwu $0,0($16)
        and $0,1,$0
        addl $0,0,$0
        ret $31,($26),1

btw, alpha really should switch to livein/outs for args :)

llvm-svn: 21213
 2005-04-10 23:37:16 +00:00
Chris Lattner
c730ea00e2 Teach legalize to deal with targets that don't support some SEXTLOAD/ZEXTLOADs 
llvm-svn: 21212
 2005-04-10 22:54:25 +00:00
Chris Lattner
1b9e1e26cb don't zextload fp values! 
llvm-svn: 21209
 2005-04-10 17:40:35 +00:00
Chris Lattner
0c089eae41 Until we have a dag combiner, promote using zextload's instead of extloads. 
This gives the optimizer a bit of information about the top-part of the
value.

llvm-svn: 21205
 2005-04-10 04:33:47 +00:00
Chris Lattner
9d13d0b958 Fold zext_inreg(zextload), likewise for sext's 
llvm-svn: 21204
 2005-04-10 04:33:08 +00:00
Chris Lattner
9c8fe594e5 add a simple xform 
llvm-svn: 21203
 2005-04-10 04:04:49 +00:00
Chris Lattner
b3518a838c Fix a thinko. If the operand is promoted, pass the promoted value into 
the new zero extend, not the original operand.  This fixes cast bool -> long
on ppc.

Add an unrelated fixme

llvm-svn: 21196
 2005-04-10 01:13:15 +00:00
Chris Lattner
034716de24 add a little peephole optimization. This allows us to codegen: 
int a(short i) {
        return i & 1;
}

as

_a:
        andi. r3, r3, 1
        blr

instead of:

_a:
        rlwinm r2, r3, 0, 16, 31
        andi. r3, r2, 1
        blr

on ppc.  It should also help the other risc targets.

llvm-svn: 21189
 2005-04-09 21:43:54 +00:00
Chris Lattner
77ab286605 there is no need to remove this instruction, linscan does it already as it 
removes noop moves.

llvm-svn: 21183
 2005-04-09 16:24:20 +00:00
Chris Lattner
f408e9a07b Adjust live intervals to support a livein set 
llvm-svn: 21182
 2005-04-09 16:17:50 +00:00
Chris Lattner
1a9c8fc64a Consider the livein/out set for a function, allowing targets to not have to 
use ugly imp_def/imp_uses for arguments and return values.

llvm-svn: 21180
 2005-04-09 15:23:25 +00:00
Chris Lattner
afa0001d54 recognize some patterns as fabs operations, so that fabs at the source level 
is deconstructed then reconstructed here.  This catches 19 fabs's in 177.mesa
9 in 168.wupwise, 5 in 171.swim, 3 in 172.mgrid, and 14 in 173.applu out of
specfp2000.

This allows the X86 code generator to make MUCH better code than before for
each of these and saves one instr on ppc.

This depends on the previous CFE patch to expose these correctly.

llvm-svn: 21171
 2005-04-09 05:15:53 +00:00
Chris Lattner
8e6eafa8e1 Emit BRCONDTWOWAY when possible. 
llvm-svn: 21167
 2005-04-09 03:30:29 +00:00
Chris Lattner
55b73bda6c Legalize BRCONDTWOWAY into a BRCOND/BR pair if a target doesn't support it. 
llvm-svn: 21166
 2005-04-09 03:30:19 +00:00
Chris Lattner
da902bdf1b print and fold BRCONDTWOWAY correctly 
llvm-svn: 21165
 2005-04-09 03:27:28 +00:00
Chris Lattner
31170cd2ec canonicalize a bunch of operations involving fneg 
llvm-svn: 21160
 2005-04-09 03:02:46 +00:00
Chris Lattner
9a56ef5693 If a target zero or sign extends the result of its setcc, allow folding of 
this into sign/zero extension instructions later.

On PPC, for example, this testcase:

%G = external global sbyte
implementation
void %test(int %X, int %Y) {
  %C = setlt int %X, %Y
  %D = cast bool %C to sbyte
  store sbyte %D, sbyte* %G
  ret void
}

Now codegens to:

        cmpw cr0, r3, r4
        li r3, 1
        li r4, 0
        blt .LBB_test_2 ;
.LBB_test_1:    ;
        or r3, r4, r4
.LBB_test_2:    ;
        addis r2, r2, ha16(L_G$non_lazy_ptr-"L00000$pb")
        lwz r2, lo16(L_G$non_lazy_ptr-"L00000$pb")(r2)
        stb r3, 0(r2)

instead of:

        cmpw cr0, r3, r4
        li r3, 1
        li r4, 0
        blt .LBB_test_2 ;
.LBB_test_1:    ;
        or r3, r4, r4
.LBB_test_2:    ;
***     rlwinm r3, r3, 0, 31, 31
        addis r2, r2, ha16(L_G$non_lazy_ptr-"L00000$pb")
        lwz r2, lo16(L_G$non_lazy_ptr-"L00000$pb")(r2)
        stb r3, 0(r2)

llvm-svn: 21148
 2005-04-07 19:43:53 +00:00
Chris Lattner
bbe0e9e9db Remove somethign I had for testing 
llvm-svn: 21144
 2005-04-07 18:58:54 +00:00
Chris Lattner
ee836c7b32 This patch does two things. First, it canonicalizes 'X >= C' -> 'X > C-1' 
(likewise for <= >=u >=u).

Second, it implements a special case hack to turn 'X gtu SINTMAX' -> 'X lt 0'

On powerpc, for example, this changes this:

        lis r2, 32767
        ori r2, r2, 65535
        cmplw cr0, r3, r2
        bgt .LBB_test_2

into:

        cmpwi cr0, r3, 0
        blt .LBB_test_2

llvm-svn: 21142
 2005-04-07 18:14:58 +00:00
Chris Lattner
22bbc2351e Fix a really scary bug that Nate found where we weren't deleting the right 
elements auto of the autoCSE maps.

llvm-svn: 21128
 2005-04-07 00:30:13 +00:00
Nate Begeman
7898fc8cc8 Teach ExpandShift how to handle shifts by a constant. This allows targets 
like PowerPC to codegen long shifts in many fewer instructions.

llvm-svn: 21122
 2005-04-06 21:13:14 +00:00
Nate Begeman
4457b4994c Expand SREM and UREM for targets that claim not to have them, like PowerPC 
llvm-svn: 21103
 2005-04-06 00:23:54 +00:00
Nate Begeman
12af81407b Add MULHU and MULHS nodes for the high part of an (un)signed 32x32=64b 
multiply.

llvm-svn: 21102
 2005-04-05 22:36:56 +00:00
Chris Lattner
f81edb57b6 Make sure to notice that explicit physregs are used in the function 
llvm-svn: 21084
 2005-04-04 21:35:34 +00:00
Nate Begeman
a8be5b976f Handle expanding arguments to ISD::TRUNCATE. This happens on PowerPC when 
you have something like i16 = truncate i64.  This fixes Regression/C/casts

llvm-svn: 21073
 2005-04-04 00:57:08 +00:00
Chris Lattner
a8bccb73cd Fix sign_extend and zero_extend of promoted value types to expanded value 
types.  This occurs when casting short to long on PPC for example.

llvm-svn: 21072
 2005-04-03 23:41:52 +00:00
Duraid Madina
3a10f491f0 add support for prefix/suffix strings to go around GlobalValue(s) 
(which may or be function pointers) in the asmprinter. For the moment,
this changes nothing, except the IA64 backend which can use this to write:

  data8.ua  @fptr(blah__blah__mangled_function_name)

  (by setting FunctionAddrPrefix/Suffix to "@fptr(" / ")")

llvm-svn: 21024
 2005-04-02 12:21:51 +00:00
Chris Lattner
1a15f58a92 transform fabs/fabsf calls into FABS nodes. 
llvm-svn: 21014
 2005-04-02 05:26:53 +00:00
Chris Lattner
206a694a7b Expand fabs into fneg 
llvm-svn: 21013
 2005-04-02 05:26:37 +00:00
Chris Lattner
fcf6ee0a8b Turn -0.0 - X -> fneg 
llvm-svn: 21011
 2005-04-02 05:04:50 +00:00
Chris Lattner
8644181cd6 Several changes mixed up here. First when legalizing a DAG with pcmarker, 
dont' regen the whole dag if unneccesary.  Second, fix and ugly bug with
the _PARTS nodes that caused legalize to produce multiples of them.
Finally, implement initial support for FABS and FNEG.  Currently FNEG is
the only one to be trusted though.

llvm-svn: 21009
 2005-04-02 05:00:07 +00:00
Chris Lattner
c8f36868e6 print fneg/fabs 
llvm-svn: 21008
 2005-04-02 04:58:41 +00:00
Chris Lattner
8be5696874 fix some bugs in the implementation of SHL_PARTS and friends. 
llvm-svn: 21004
 2005-04-02 04:00:59 +00:00
Chris Lattner
964ab5d408 Turn expanded shift operations into (e.g.) SHL_PARTS if the target supports it. 
llvm-svn: 21002
 2005-04-02 03:38:53 +00:00
Chris Lattner
33ca1ce8e0 Print some new nodes 
llvm-svn: 21001
 2005-04-02 03:30:42 +00:00
Chris Lattner
20027c6b30 Fix a bug when inserting a libcall into a function with no other calls. 
llvm-svn: 20999
 2005-04-02 03:22:40 +00:00
Nate Begeman
893f5729ce Fix a warning about an unhandled switch case 
llvm-svn: 20994
 2005-04-02 00:41:14 +00:00
Nate Begeman
4034852ba9 Add ISD::UNDEF node 
Teach the SelectionDAG code how to expand and promote it
Have PPC32 LowerCallTo generate ISD::UNDEF for int arg regs used up by fp
  arguments, but not shadowing their value.  This allows us to do the right
  thing with both fixed and vararg floating point arguments.

llvm-svn: 20988
 2005-04-01 22:34:39 +00:00
Chris Lattner
c81870e4e6 print the machine CFG in the -print-machineinstrs dump 
llvm-svn: 20976
 2005-04-01 06:48:38 +00:00
Andrew Lenharth
7db3834ecf PCMarker support for DAG and Alpha 
llvm-svn: 20965
 2005-03-31 21:24:06 +00:00
Chris Lattner
abb59a3c21 Instead of setting up the CFG edges at selectiondag construction time, set 
them up after the code has been emitted.  This allows targets to select one
mbb as multiple mbb's as needed.

llvm-svn: 20937
 2005-03-30 01:10:47 +00:00
Chris Lattner
02a4d3bd9b Fix a bug that andrew noticed where we do not correctly sign/zero extend 
returned integer values all of the way to 64-bits (we only did it to 32-bits
leaving the top bits undefined).  This causes problems for targets like alpha
whose ABI's define the top bits too.

llvm-svn: 20926
 2005-03-29 19:09:56 +00:00
Chris Lattner
185e7e2c22 implement legalization of build_pair for nate 
llvm-svn: 20901
 2005-03-28 22:03:13 +00:00
Andrew Lenharth
c287cd1e4e First step in adding pcmarker intrinsic. Second step (soon) is adding backend support. 
llvm-svn: 20900
 2005-03-28 20:05:49 +00:00
Nate Begeman
f821401825 Change interface to LowerCallTo to take a boolean isVarArg argument. 
llvm-svn: 20842
 2005-03-26 01:29:23 +00:00
Chris Lattner
c9a3ea81bf Fix the missing symbols problem Bill was hitting. Patch contributed by 
Bill Wendling!!

llvm-svn: 20649
 2005-03-17 15:38:16 +00:00
Chris Lattner
4b688a1c70 This mega patch converts us from using Function::a{iterator|begin|end} to 
using Function::arg_{iterator|begin|end}.  Likewise Module::g* -> Module::global_*.

This patch is contributed by Gabor Greif, thanks!

llvm-svn: 20597
 2005-03-15 04:54:21 +00:00
Chris Lattner
4422ffd421 I didn't mean to check this in. :( 
llvm-svn: 20555
 2005-03-10 20:59:51 +00:00
Chris Lattner
fa9e43b38c Fix a bug where we would incorrectly do a sign ext instead of a zero ext 
because we were checking the wrong thing.  Thanks to andrew for pointing
this out!

llvm-svn: 20554
 2005-03-10 20:55:51 +00:00
Chris Lattner
ea2e61b83a Allow the live interval analysis pass to be a bit more aggressive about 
numbering values in live ranges for physical registers.

The alpha backend currently generates code that looks like this:

  vreg = preg
...
  preg = vreg
  use preg
...
  preg = vreg
  use preg

etc.  Because vreg contains the value of preg coming in, each of the
copies back into preg contain that initial value as well.

In the case of the Alpha, this allows this testcase:

void "foo"(int %blah) {
        store int 5, int *%MyVar
        store int 12, int* %MyVar2
        ret void
}

to compile to:

foo:
        ldgp $29, 0($27)
        ldiq $0,5
        stl $0,MyVar
        ldiq $0,12
        stl $0,MyVar2
        ret $31,($26),1

instead of:

foo:
        ldgp $29, 0($27)
        bis $29,$29,$0
        ldiq $1,5
        bis $0,$0,$29
        stl $1,MyVar
        ldiq $1,12
        bis $0,$0,$29
        stl $1,MyVar2
        ret $31,($26),1

This does not seem to have any noticable effect on X86 code.

This fixes PR535.

llvm-svn: 20536
 2005-03-09 23:05:19 +00:00
Chris Lattner
e0d0c64c8a constant fold FP_ROUND_INREG, ZERO_EXTEND_INREG, and SIGN_EXTEND_INREG 
This allows the alpha backend to compile:

bool %test(uint %P) {
        %c = seteq uint %P, 0
        ret bool %c
}

into:

test:
        ldgp $29, 0($27)
        ZAP $16,240,$0
        CMPEQ $0,0,$0
        AND $0,1,$0
        ret $31,($26),1

instead of:

test:
        ldgp $29, 0($27)
        ZAP $16,240,$0
        ldiq $1,0
        ZAP $1,240,$1
        CMPEQ $0,$1,$0
        AND $0,1,$0
        ret $31,($26),1

... and fixes PR534.

llvm-svn: 20534
 2005-03-09 18:37:12 +00:00
Alkis Evlogimenos
422af394b6 Lower llvm.isunordered(a, b) into a != a | b != b. 
llvm-svn: 20382
 2005-03-01 02:07:58 +00:00
Chris Lattner
9ccfcab3db Lower prefetch to a noop, patch contributed by Justin Wick! 
llvm-svn: 20375
 2005-02-28 19:27:23 +00:00
Chris Lattner
4ba91f5168 Fix a bug in the 'store fpimm, ptr' -> 'store intimm, ptr' handling code. 
Changing 'op' here caused us to not enter the store into a map, causing
reemission of the code!!  In practice, a simple loop like this:

no_exit:                ; preds = %no_exit, %entry
        %indvar = phi uint [ %indvar.next, %no_exit ], [ 0, %entry ]            ; <uint> [#uses=3]
        %tmp.4 = getelementptr "complex long double"* %P, uint %indvar, uint 0          ; <double*> [#uses=1]
        store double 0.000000e+00, double* %tmp.4
        %indvar.next = add uint %indvar, 1              ; <uint> [#uses=2]
        %exitcond = seteq uint %indvar.next, %N         ; <bool> [#uses=1]
        br bool %exitcond, label %return, label %no_exit

was being code gen'd to:

.LBBtest_1:     # no_exit
        movl %edx, %esi
        shll $4, %esi
        movl $0, 4(%eax,%esi)
        movl $0, (%eax,%esi)
        incl %edx
        movl $0, (%eax,%esi)
        movl $0, 4(%eax,%esi)
        cmpl %ecx, %edx
        jne .LBBtest_1  # no_exit

Note that we are doing 4 32-bit stores instead of 2.  Now we generate:

.LBBtest_1:     # no_exit
        movl %edx, %esi
        incl %esi
        shll $4, %edx
        movl $0, (%eax,%edx)
        movl $0, 4(%eax,%edx)
        cmpl %ecx, %esi
        movl %esi, %edx
        jne .LBBtest_1  # no_exit

This is much happier, though it would be even better if the increment of ESI
was scheduled after the compare :-/

llvm-svn: 20265
 2005-02-22 07:23:39 +00:00
Misha Brukman
381d248dc6 Fix compilation errors with VS 2005, patch by Aaron Gray. 
llvm-svn: 20231
 2005-02-17 21:39:27 +00:00
Chris Lattner
89105cec43 Don't rely on doubles comparing identical to each other, which doesn't work 
for 0.0 and -0.0.

llvm-svn: 20230
 2005-02-17 20:17:32 +00:00
Chris Lattner
0de03b45ab Don't sink argument loads into loops or other bad places. This disables folding of argument loads with instructions that are not in the entry block. 
llvm-svn: 20228
 2005-02-17 19:40:32 +00:00
Chris Lattner
43b14db4d9 Print GEP offsets as signed values instead of unsigned values. On X86, this 
prints:

getelementptr (int* %A, int -1)

as: "(A) - 4" instead of "(A) + 18446744073709551612", which makes the
assembler much happier.

This fixes test/Regression/CodeGen/X86/2005-02-14-IllegalAssembler.ll,
and Benchmarks/Prolangs-C/cdecl with LLC on X86.

llvm-svn: 20183
 2005-02-14 21:40:26 +00:00
Chris Lattner
c808a143af Fix a case where were incorrectly compiled cast from short to int on 64-bit 
targets.

llvm-svn: 20030
 2005-02-04 18:39:19 +00:00
Andrew Lenharth
d2d24eee40 fix constant pointer outputing on 64 bit machines 
llvm-svn: 20026
 2005-02-04 13:47:16 +00:00
Chris Lattner
c3f476e9c2 Fix yet another memset issue. 
llvm-svn: 19986
 2005-02-02 03:44:41 +00:00
Chris Lattner
9cf60e3459 Fix some bugs andrew noticed legalizing memset for alpha 
llvm-svn: 19969
 2005-02-01 18:38:28 +00:00
Chris Lattner
382abe80a0 Improve conformance with the Misha spelling benchmark suite 
llvm-svn: 19930
 2005-01-30 00:09:23 +00:00