Commit Graph

1107 Commits

Author SHA1 Message Date
Chris Lattner
707ebc5dd6 update the backends to work with the new CopyFromReg/CopyToReg/ImplicitDef nodes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22807 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-16 21:56:37 +00:00
Nate Begeman
7cbd525ba8 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.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22805 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-16 19:49:35 +00:00
Nate Begeman
889f2c1acf Fix last night's X86 regressions by putting code for SSE in the if(SSE)
block.  nur.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22788 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-14 18:37:02 +00:00
Nate Begeman
875f360e22 Fix FP_TO_UINT with Scalar SSE2 now that the legalizer can handle it. We
now generate the relatively good code sequences:
unsigned short foo(float a) { return a; }
_foo:
        movss 4(%esp), %xmm0
        cvttss2si %xmm0, %eax
        movzwl %ax, %eax
        ret

and
unsigned bar(float a) { return a; }
_bar:
        movss .CPI_bar_0, %xmm0
        movss 4(%esp), %xmm1
        movapd %xmm1, %xmm2
        subss %xmm0, %xmm2
        cvttss2si %xmm2, %eax
        xorl $-2147483648, %eax
        cvttss2si %xmm1, %ecx
        ucomiss %xmm0, %xmm1
        cmovb %ecx, %eax
        ret


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22786 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-14 04:36:51 +00:00
Chris Lattner
88ac32ca26 Update the targets to the new SETCC/CondCodeSDNode interfaces.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22729 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-09 20:21:10 +00:00
Chris Lattner
8c4a8735ec adjust to change in getSubtarget() api
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22687 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-05 21:54:27 +00:00
Nate Begeman
8c00f8cdc7 Add Subtarget support to PowerPC. Next up, using it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22644 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-04 07:12:09 +00:00
Nate Begeman
1c73c7be9d Scalar SSE: load +0.0 -> xorps/xorpd
Scalar SSE: a < b ? c : 0.0 -> cmpss, andps
Scalar SSE: float -> i16 needs to be promoted


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22637 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-03 23:26:28 +00:00
Chris Lattner
0561b3ff9f Update to use the new MathExtras.h support for log2 computation.
Patch contributed by Jim Laskey!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22594 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-02 19:26:06 +00:00
Jeff Cohen
d29b6aa608 Keep tabs and trailing spaces out.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22565 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-30 18:33:25 +00:00
Chris Lattner
a88da08c6e fix a typeo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22561 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-30 00:43:00 +00:00
Chris Lattner
a35e1dfd12 Change the fp to integer code to not perform 2-byte stores followed by
1 byte loads and other operations.  This is bad for store-forwarding on
common CPUs.  We now do this:

fnstcw WORD PTR [%ESP]
mov %AX, WORD PTR [%ESP]

instead of:

fnstcw WORD PTR [%ESP]
mov %AL, BYTE PTR [%ESP + 1]


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22559 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-30 00:17:52 +00:00
Chris Lattner
01546c5d50 Use a custom expander for all FP to int conversions, as the X86 only has
FP-to-int-in-memory: this exposes the load from the stored slot to the
selection dag, allowing it to be folded into other operaions.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22556 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-30 00:05:54 +00:00
Andrew Lenharth
422f3d58a8 turn off GOT on archs that didn't use it (not that it appeard to harm them much with it on)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22553 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-29 23:32:02 +00:00
Chris Lattner
5d06b8caae Implement a FIXME: move a bunch of cruft for handling FP_TO_*INT operations
that the X86 does not support to the legalizer.  This allows it to be better
optimized, etc, and will help with SSE support.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22551 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-29 01:00:29 +00:00
Chris Lattner
f7443dace5 Don't forget to diddle with the control word when performing an FISTP64.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22550 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-29 00:54:34 +00:00
Chris Lattner
745d538119 Use a custom expander to compile this:
long %test4(double %X) {
        %tmp.1 = cast double %X to long         ; <long> [#uses=1]
        ret long %tmp.1
}

to this:

_test4:
        sub %ESP, 12
        fld QWORD PTR [%ESP + 16]
        fistp QWORD PTR [%ESP]
        mov %EDX, DWORD PTR [%ESP + 4]
        mov %EAX, DWORD PTR [%ESP]
        add %ESP, 12
        ret

instead of this:

_test4:
        sub %ESP, 28
        fld QWORD PTR [%ESP + 32]
        fstp QWORD PTR [%ESP]
        call ___fixdfdi
        add %ESP, 28
        ret


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22549 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-29 00:40:01 +00:00
Jeff Cohen
00b16889ab Eliminate all remaining tabs and trailing spaces.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22523 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-27 06:12:32 +00:00
Jeff Cohen
9eb59ec548 Eliminate tabs and trailing spaces.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22520 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-27 05:53:44 +00:00
Andrew Lenharth
908bc862d5 update interface
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22498 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-22 20:49:37 +00:00
Reid Spencer
a0f5bf306c For: memory operations -> stores
This is the first incremental patch to implement this feature. It adds no
functionality to LLVM but setup up the information needed from targets in
order to implement the optimization correctly. Each target needs to specify
the maximum number of store operations for conversion of the llvm.memset,
llvm.memcpy, and llvm.memmove intrinsics into a sequence of store operations.
The limit needs to be chosen at the threshold of performance for such an
optimization (generally smallish). The target also needs to specify whether
the target can support unaligned stores for multi-byte store operations.
This helps ensure the optimization doesn't generate code that will trap on
an alignment errors.
More patches to follow.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22468 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-19 04:52:44 +00:00
Nate Begeman
5a8441ea3f 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


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22452 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-16 02:02:34 +00:00
Nate Begeman
11cefd926a Teach the register allocator that movaps is also a move instruction
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22451 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-16 02:00:20 +00:00
Nate Begeman
9035b99abf A couple more darwinisms
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22450 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-16 01:59:47 +00:00
Chris Lattner
a28381cac2 Remove all knowledge of UINT_TO_FP from the X86 backend, relying on the
legalizer to eliminate them.  With this comes the expected code quality
improvements, such as, for this:

double foo(unsigned short X) { return X; }

we now generate this:

_foo:
        subl $4, %esp
        movzwl 8(%esp), %eax
        movl %eax, (%esp)
        fildl (%esp)
        addl $4, %esp
        ret

instead of this:

_foo:
        subl $4, %esp
        movw 8(%esp), %ax
        movzwl %ax, %eax   ;; Load not folded into this.
        movl %eax, (%esp)
        fildl (%esp)
        addl $4, %esp
        ret

-Chris


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22449 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-16 00:28:20 +00:00
Nate Begeman
16b04f3d5e Get closer to fully working scalar FP in SSE regs. This gets singlesource
working, and Olden/power.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22441 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-15 00:38:55 +00:00
Nate Begeman
6c7cb29038 Add support for printing the sse scalar comparison instruction mnemonics.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22440 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-14 22:52:25 +00:00
Nate Begeman
d3a490a08e Check in the last of the darwin-specific code necessary to get shootout
working before modifying the asm printer to use the subtarget info.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22408 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-12 18:34:58 +00:00
Nate Begeman
0e219eb9b8 Clean up the TargetSubtarget class a bit, removing an unnecessary argument
to the constructor.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22392 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-12 02:41:19 +00:00
Chris Lattner
b151acadc8 Minor changes to improve comments and fix the build on _WIN32 systems.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22391 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-12 02:36:10 +00:00
Chris Lattner
3249bfc828 Add a note
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22390 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-12 02:35:36 +00:00
Nate Begeman
fb5792f416 Implement Subtarget support
Implement the X86 Subtarget.

This consolidates the checks for target triple, and setting options based
on target triple into one place.  This allows us to convert the asm printer
and isel over from being littered with "forDarwin", "forCygwin", etc. into
just having the appropriate flags for each subtarget feature controlling
the code for that feature.

This patch also implements indirect external and weak references in the
X86 pattern isel, for darwin.  Next up is to convert over the asm printers
to use this new interface.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22389 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-12 01:41:54 +00:00
Nate Begeman
73213f6c07 Commit some pending darwin changes before subtarget support.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22388 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-12 01:37:28 +00:00
Chris Lattner
c569d79a2a Output .size directives to tell the assembler the size of each function.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22381 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-11 06:29:14 +00:00
Chris Lattner
b12c9fa565 Fix crazy indentation
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22380 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-11 06:25:47 +00:00
Chris Lattner
81b6ed7ed1 Refactor things a bit to allow the ELF code emitter to run the X86 machine code emitter
after itself.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22376 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-11 05:17:48 +00:00
Chris Lattner
efb4bf0de6 Remove prototype for non-existant function
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22372 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-11 04:20:55 +00:00
Chris Lattner
bce81ae51e 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.

Also, remove some uses of dyn_cast that should really be cast (which is
cheaper in a release build).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22368 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-10 01:56:13 +00:00
Chris Lattner
9fadb4c1c0 Change TRUNCSTORE to use a VTSDNode operand instead of being an MVTSTDNode
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22366 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-10 00:29:18 +00:00
Nate Begeman
72b286b0a0 Add support for assembling .s files on mac os x for intel
Add support for running bugpoint on mac os x for intel


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22351 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-08 00:23:26 +00:00
Chris Lattner
8d4b9eddc0 Restore some code that was accidentally removed by Nate's patch yesterday.
This fixes the regressions from last night.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22344 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-07 17:12:53 +00:00
Nate Begeman
b8aa3ac678 Fix a typo in my checkin today that caused regressions. Oops!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22341 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-07 06:32:01 +00:00
Nate Begeman
f63be7d395 First round of support for doing scalar FP using the SSE2 ISA extension and
XMM registers.  There are many known deficiencies and fixmes, which will be
addressed ASAP.  The major benefit of this work is that it will allow the
LLVM register allocator to allocate FP registers across basic blocks.

The x86 backend will still default to x87 style FP.  To enable this work,
you must pass -enable-sse-scalar-fp and either -sse2 or -sse3 to llc.

An example before and after would be for:
double foo(double *P) { double Sum = 0; int i; for (i = 0; i < 1000; ++i)
                        Sum += P[i]; return Sum; }

The inner loop looks like the following:
x87:
.LBB_foo_1:     # no_exit
        fldl (%esp)
        faddl (%eax,%ecx,8)
        fstpl (%esp)
        incl %ecx
        cmpl $1000, %ecx
        #FP_REG_KILL
        jne .LBB_foo_1  # no_exit

SSE2:
        addsd (%eax,%ecx,8), %xmm0
        incl %ecx
        cmpl $1000, %ecx
        #FP_REG_KILL
        jne .LBB_foo_1  # no_exit


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22340 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-06 18:59:04 +00:00
Chris Lattner
e0fe225e89 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.
4. Now that we have Value*'s available in the lowering methods, pass them
   into any load/stores from the valist that are emitted


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22339 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-05 19:58:54 +00:00
Chris Lattner
08568cfe25 Fit to 80 columns
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22336 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-05 17:50:16 +00:00
Chris Lattner
6a709a418b Percolate the call up to the right superclass
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22330 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-03 17:34:39 +00:00
Nate Begeman
4eb74ba602 The statistic needs to be in the correct namespace.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22327 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-01 23:56:38 +00:00
Chris Lattner
b36cbd0286 Refactor X86AsmPrinter.cpp into multiple files. Patch contributed
by Aaron Gray, cleaned up by me.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22324 91177308-0d34-0410-b5e6-96231b3b80d8
2005-07-01 22:44:09 +00:00
Nate Begeman
9d19eb459c Make the x86 asm printer darwin-aware. This mostly entails doing the same
thing as cygwin most of the time, and printing our alignments in log2
rather than number of bytes.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22316 91177308-0d34-0410-b5e6-96231b3b80d8
2005-06-30 00:53:20 +00:00
Nate Begeman
f1702ac589 Initial set of .td file changes necessary to get scalar fp in xmm registers
working.  The instruction selector changes will hopefully be coming later
this week once they are debugged.  This is necessary to support the darwin
x86 FP model, and is recommended by intel as the replacement for x87.  As
a bonus, the register allocator knows how to deal with these registers
across basic blocks, unliky the FP stackifier.  This leads to significantly
better codegen in several cases.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22300 91177308-0d34-0410-b5e6-96231b3b80d8
2005-06-27 21:20:31 +00:00