fix boolean XOR (which fixes up comparisons..)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25462 91177308-0d34-0410-b5e6-96231b3b80d8

BOOM!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25460 91177308-0d34-0410-b5e6-96231b3b80d8

click click

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25459 91177308-0d34-0410-b5e6-96231b3b80d8

Use pattern information to determine whether the use expects this
instruction to produce a result. e.g MUL8m, the instruction does not
produce a explicit result. However it produces an implicit result in
AL which would be copied to a temp. The root operator of the matching
pattern is a mul so the use would expect it to produce a result.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25458 91177308-0d34-0410-b5e6-96231b3b80d8

Avoid generating a redundant setcc.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25457 91177308-0d34-0410-b5e6-96231b3b80d8

For PR696:
Don't do floor->floorf conversion if floorf is not available. This checks
the compiler's host, not its target, which is incorrect for cross-compilers
Not sure that's important as we don't build many cross-compilers.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25456 91177308-0d34-0410-b5e6-96231b3b80d8

fix calls that return f32

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25455 91177308-0d34-0410-b5e6-96231b3b80d8

Make sure that libm is used during config tests so that ceil, floor, and
friends are actually detected.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25454 91177308-0d34-0410-b5e6-96231b3b80d8

For PR696:
Add checks for ceil, ceilf, floor, and floorf

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25453 91177308-0d34-0410-b5e6-96231b3b80d8

Implement casts.ll:test26: a cast from float -> double -> integer, doesn't
need the float->double part.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25452 91177308-0d34-0410-b5e6-96231b3b80d8

Add a new testcase for an eliminable cast

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25451 91177308-0d34-0410-b5e6-96231b3b80d8

implement support for f32 arguments past the first 6 words

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25450 91177308-0d34-0410-b5e6-96231b3b80d8

This test passes now.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25449 91177308-0d34-0410-b5e6-96231b3b80d8

1. Identify bytecode modules that have upgraded intrinsics by setting a
   boolean flag if we read a function prototype that needs upgrading.
2. Don't upgrade the CallInst instruction until after its been inserted
   into the basic block, and only if we know that we have seen an
   upgraded intrinsic function.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25448 91177308-0d34-0410-b5e6-96231b3b80d8

Don't forget about casting the result of an upgraded call to an intrinsic
in the case where it needs to cast back to a signed type.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25447 91177308-0d34-0410-b5e6-96231b3b80d8

1. Documentation upgrade.
2. Have UpgradeInstrinicCall return an Instruction* instead of a CallInst*
   and return the needed CastInst* if the result of the upgraded call needs
   to be casted back to a signed type.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25446 91177308-0d34-0410-b5e6-96231b3b80d8

Add a flag to identify bytecode files that have intrinsic functions that
need to be upgraded.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25445 91177308-0d34-0410-b5e6-96231b3b80d8

adc and sbb need an incoming flag to ensure it reads the carry flag
from add / sub.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25444 91177308-0d34-0410-b5e6-96231b3b80d8

make this work for Reid, who renamed his CVS module dir :)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25443 91177308-0d34-0410-b5e6-96231b3b80d8

new testcase the autoupgrader in the bc reader fails to handle

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25442 91177308-0d34-0410-b5e6-96231b3b80d8

Make get_suffix faster by using a switch on getTypeID rather than a series
of comparisons on the various type objects.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25441 91177308-0d34-0410-b5e6-96231b3b80d8

Another typo

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25440 91177308-0d34-0410-b5e6-96231b3b80d8

add a note

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25439 91177308-0d34-0410-b5e6-96231b3b80d8

Two peepholes:
(or (x >> c) | (y << (32 - c))) ==> (shrd x, y, c)
(or (x << c) | (y >> (32 - c))) ==> (shld x, y, c)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25438 91177308-0d34-0410-b5e6-96231b3b80d8

Prevent unnecessary CopyToReg when the same HW register appears in two spots
in the pattern.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25437 91177308-0d34-0410-b5e6-96231b3b80d8

Didn't mean to check that in.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25436 91177308-0d34-0410-b5e6-96231b3b80d8

A obvious typo

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25435 91177308-0d34-0410-b5e6-96231b3b80d8

Make sure intrinsic auto-upgrade is invoked correctly.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25434 91177308-0d34-0410-b5e6-96231b3b80d8

Don't accept the ctpop, cttz, or ctlz intrinsics with signed arguments. The
interface requires unsigned arguments.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25433 91177308-0d34-0410-b5e6-96231b3b80d8

Add a new interface function to AutoUpgrade for simultaneously upgrading
the Function and the CallInst: UpgradeCallsToIntrinsic(Function*). Also,
re-factor the AutoUpgrade implementation to eliminate some duplication of
code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25432 91177308-0d34-0410-b5e6-96231b3b80d8

Expand test to cover the non-overloaded variants of the arithmetic and
floating point intrinsics.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25431 91177308-0d34-0410-b5e6-96231b3b80d8

If not internalizing, don't mark llvm.global[cd]tors const, as a fix for a
hypothetical future boog.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25430 91177308-0d34-0410-b5e6-96231b3b80d8

Don't internalize llvm.global[cd]tor unless there are uses of it.  This
unbreaks front-ends that don't use __main (like the new CFE).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25429 91177308-0d34-0410-b5e6-96231b3b80d8

was ignoring the legalized chain in this case, fixed SPASS on alpha

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25428 91177308-0d34-0410-b5e6-96231b3b80d8

a reduced test case, the GlobalAddress is never legalized

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25427 91177308-0d34-0410-b5e6-96231b3b80d8

Get rid of code in the DAGCombiner that is duplicated in SelectionDAG.cpp
Now all constant folding in the code generator is in one place.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25426 91177308-0d34-0410-b5e6-96231b3b80d8

Temporary work around for a libcall insertion bug: If a target doesn't
support FSIN/FCOS nodes, do not lower sin/cos to them.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25425 91177308-0d34-0410-b5e6-96231b3b80d8

Make sure that cloning a module clones its target triple and dependent
library list as well.  This should help bugpoint.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25424 91177308-0d34-0410-b5e6-96231b3b80d8

Don't assert on 'select_cc SETUO'

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25423 91177308-0d34-0410-b5e6-96231b3b80d8

fix out of date comment

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25422 91177308-0d34-0410-b5e6-96231b3b80d8

Fix Regression/CodeGen/PowerPC/2006-01-18-InvalidBranchOpcodeAssert.ll

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25421 91177308-0d34-0410-b5e6-96231b3b80d8

new testcase

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25420 91177308-0d34-0410-b5e6-96231b3b80d8

Fix a backwards conditional that caused an inf loop in some cases.  This
fixes: test/Regression/CodeGen/Generic/2005-01-18-SetUO-InfLoop.ll

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25419 91177308-0d34-0410-b5e6-96231b3b80d8

this testcase sent the dag combiner spinning into an infinite loop

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25418 91177308-0d34-0410-b5e6-96231b3b80d8

Added minimum Dwarf aranges.  Cleaned up some section headers.  Line number
support now works in gdb.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25417 91177308-0d34-0410-b5e6-96231b3b80d8

SRA shift amount must be in i8

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25416 91177308-0d34-0410-b5e6-96231b3b80d8

If a call return type is i1, insert a truncate from X86::AL to i1.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25415 91177308-0d34-0410-b5e6-96231b3b80d8

Fixed style of curly brace.  No functionality changes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25414 91177308-0d34-0410-b5e6-96231b3b80d8

Fix lowering of calls which return f32 values.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25413 91177308-0d34-0410-b5e6-96231b3b80d8

Pastos.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25412 91177308-0d34-0410-b5e6-96231b3b80d8

fix a broken comment

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25411 91177308-0d34-0410-b5e6-96231b3b80d8

Add frame work for additional dwarf sections.  Comments will improve as code
is added.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25410 91177308-0d34-0410-b5e6-96231b3b80d8

Test case for the insertelement operation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25409 91177308-0d34-0410-b5e6-96231b3b80d8

VMCore support for the insertelement operation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25408 91177308-0d34-0410-b5e6-96231b3b80d8

Constant folding support for the insertelement operation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25407 91177308-0d34-0410-b5e6-96231b3b80d8

Lowerpacked and SCCP support for the insertelement operation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25406 91177308-0d34-0410-b5e6-96231b3b80d8

Support for the insertelement operation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25405 91177308-0d34-0410-b5e6-96231b3b80d8

Reader support for the insertelement operation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25404 91177308-0d34-0410-b5e6-96231b3b80d8

Lexer and parser support for the insertelement operation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25403 91177308-0d34-0410-b5e6-96231b3b80d8

Instruction and constant expression definitions for the insertelement
operation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25402 91177308-0d34-0410-b5e6-96231b3b80d8

Bug fix: missing LegalizeOp() on newly created nodes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25401 91177308-0d34-0410-b5e6-96231b3b80d8

Suppress "no newline at end of file" warnings.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25400 91177308-0d34-0410-b5e6-96231b3b80d8

Misc. errors.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25399 91177308-0d34-0410-b5e6-96231b3b80d8

Clean up the FFS optimization code, and make it correctly create the appropriate
unsigned llvm.cttz.* intrinsic, fixing the 2005-05-11-Popcount-ffs-fls regression
last night.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25398 91177308-0d34-0410-b5e6-96231b3b80d8

Adding basic support for Dwarf line number debug information.

I promise to keep future commits smaller.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25396 91177308-0d34-0410-b5e6-96231b3b80d8

Adding UniqueVector.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25395 91177308-0d34-0410-b5e6-96231b3b80d8

Regenerated the Lex and Yacc output files on Linux.  It seems that our
Linux machines don't like the source code generated on MacOS X for some
reason.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25394 91177308-0d34-0410-b5e6-96231b3b80d8

Reduce memory consumption and force (somewhat) access to entries via ID.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25393 91177308-0d34-0410-b5e6-96231b3b80d8

Emit a type matching check for ComplexPatterns.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25392 91177308-0d34-0410-b5e6-96231b3b80d8

Zero extending load from i1 to i8.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25391 91177308-0d34-0410-b5e6-96231b3b80d8

Ensure that the CFE always emits intrinsics in their unsigned version, not with signed operands

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25390 91177308-0d34-0410-b5e6-96231b3b80d8

Visual Studio still has issues with being left out.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25389 91177308-0d34-0410-b5e6-96231b3b80d8

oops, this shouldn't have gotten in

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25388 91177308-0d34-0410-b5e6-96231b3b80d8

SSE does not support i64 SINT_TO_FP (FP stack doesn't either, but we custom
expand it), so ask legalizer to expand i32 UINT_TO_FP.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25386 91177308-0d34-0410-b5e6-96231b3b80d8

use proper (82-bit) spills/fills when spilling FP regs, so that
divides don't get broken. this fixes obsequi, smg2000, and probably
a bunch of other stuff (tm)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25385 91177308-0d34-0410-b5e6-96231b3b80d8

fixing divides

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25383 91177308-0d34-0410-b5e6-96231b3b80d8

Revert this, I didn't mean to commit it

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25382 91177308-0d34-0410-b5e6-96231b3b80d8

Added a FIXME comment about why FST is currently flagged to fpGETRESULT.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25381 91177308-0d34-0410-b5e6-96231b3b80d8

Yet another getTargetNode() variant. I promise one of these days I'll fix
tblgen so this is unnecessary.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25380 91177308-0d34-0410-b5e6-96231b3b80d8

Add support for programs with a null argv[0]

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25379 91177308-0d34-0410-b5e6-96231b3b80d8

Bug fixes: fpGETRESULT should produces a flag result and X86ISD::FST should
read a flag.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25378 91177308-0d34-0410-b5e6-96231b3b80d8

Redundant inline keyword.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25377 91177308-0d34-0410-b5e6-96231b3b80d8

UniqueVector template provides a means of enumerating objects uniquely.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25376 91177308-0d34-0410-b5e6-96231b3b80d8

More typo's

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25375 91177308-0d34-0410-b5e6-96231b3b80d8

Some typo's

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25374 91177308-0d34-0410-b5e6-96231b3b80d8

like bswap, the ct* intrinsics require unsigned operands

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25373 91177308-0d34-0410-b5e6-96231b3b80d8

Consistency

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25372 91177308-0d34-0410-b5e6-96231b3b80d8

fix short immediate loads

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25371 91177308-0d34-0410-b5e6-96231b3b80d8

this is fixed

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25370 91177308-0d34-0410-b5e6-96231b3b80d8

stack and rpcc

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25369 91177308-0d34-0410-b5e6-96231b3b80d8

Fix FP_TO_INT**_IN_MEM lowering.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25368 91177308-0d34-0410-b5e6-96231b3b80d8

There is no llvm.bswap.i8, need at least 2 bytes to swap.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25367 91177308-0d34-0410-b5e6-96231b3b80d8

For PR411:
This patch is an incremental step towards supporting a flat symbol table.
It de-overloads the intrinsic functions by providing type-specific intrinsics
and arranging for automatically upgrading from the old overloaded name to
the new non-overloaded name. Specifically:
  llvm.isunordered -> llvm.isunordered.f32, llvm.isunordered.f64
  llvm.sqrt -> llvm.sqrt.f32, llvm.sqrt.f64
  llvm.ctpop -> llvm.ctpop.i8, llvm.ctpop.i16, llvm.ctpop.i32, llvm.ctpop.i64
  llvm.ctlz -> llvm.ctlz.i8, llvm.ctlz.i16, llvm.ctlz.i32, llvm.ctlz.i64
  llvm.cttz -> llvm.cttz.i8, llvm.cttz.i16, llvm.cttz.i32, llvm.cttz.i64
New code should not use the overloaded intrinsic names. Warnings will be
emitted if they are used.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25366 91177308-0d34-0410-b5e6-96231b3b80d8

For PR411:
This file makes the helper functions for auto-upgrade of llvm assembly and
bytecode more accessible. This is part of de-overloading of intrinsic
functions to support the flat symbol table (no type planes).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25365 91177308-0d34-0410-b5e6-96231b3b80d8

Friendly names

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25364 91177308-0d34-0410-b5e6-96231b3b80d8

fix a crash due to missing parens

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25363 91177308-0d34-0410-b5e6-96231b3b80d8

add notes from my *other* email acct.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25362 91177308-0d34-0410-b5e6-96231b3b80d8

transfer some notes from my email to somewhere useful.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25361 91177308-0d34-0410-b5e6-96231b3b80d8

Fix these testcases  :(
Apparently Andrew hasn't implemented ReadCycleCounter in the new isel and
renamed ornot to eqv

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25360 91177308-0d34-0410-b5e6-96231b3b80d8

Add a couple very minor tweaks

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25359 91177308-0d34-0410-b5e6-96231b3b80d8