Commit Graph

36 Commits

Author SHA1 Message Date
Robert Bocchino ca27f0320b VMCore support for the insertelement operation.
llvm-svn: 25408
2006-01-17 20:07:22 +00:00
Robert Bocchino de7f1c9a63 Added constant folding support for the extractelement operation.
llvm-svn: 25187
2006-01-10 20:03:46 +00:00
Chris Lattner fed8ceb56a Implement a few symbolic constant folding things. X ? Y : Y is Y.
Fold:
seteq ({ short }* cast (int 1 to { short }*), { short }* null)
setlt ({ short }* cast (int 1 to { short }*), { short }* cast (int 2 to { short }*))

to false/true.  These last two commonly occur in the output of compilers that
tag integers, like cozmic's scheme compiler.

Tested by Regression/Assembler/ConstantExprFold.llx

llvm-svn: 25112
2006-01-05 07:49:30 +00:00
Chris Lattner 0f7e9f5180 fix some formatting problems
llvm-svn: 25110
2006-01-05 07:19:51 +00:00
Chris Lattner 6b52be6a89 implement constant folding of ==/!= on constant packed, simplify some code.
llvm-svn: 25074
2006-01-04 02:20:54 +00:00
Chris Lattner f0f4068196 implement constant folding for the element-wise binary operations
llvm-svn: 25073
2006-01-04 02:15:02 +00:00
Chris Lattner 1171d95d88 don't crash when trying to constant fold packed expressions.
llvm-svn: 25072
2006-01-04 02:03:29 +00:00
Jeff Cohen 4e3aedeaa6 Use ANSI-approved way of getting the value infinity (otherwise VC++ won't compile it)
llvm-svn: 21662
2005-05-03 03:13:01 +00:00
Andrew Lenharth c73e633a41 fold fp div by 0 to inf, the way gcc does. This is legal according to the FP spec
llvm-svn: 21655
2005-05-02 21:25:47 +00:00
Reid Spencer 9c47b25868 Shut GCC 4.0 up about classes with virtual functions but no virtual
destructor.

llvm-svn: 21510
2005-04-24 22:27:20 +00:00
Misha Brukman b1c9317bb4 Remove trailing whitespace
llvm-svn: 21427
2005-04-21 23:48:37 +00:00
Chris Lattner feaf92f7ad Fix a nasty thinko in my previous commit.
llvm-svn: 19881
2005-01-28 23:17:27 +00:00
Chris Lattner 60c47267a9 Fix ConstProp/2005-01-28-SetCCGEP.ll: indexing over zero sized elements does
not change the address.

llvm-svn: 19874
2005-01-28 19:09:51 +00:00
Chris Lattner cd4003e228 No need to pessimize current code for future possibilities.
llvm-svn: 19311
2005-01-06 16:26:38 +00:00
Chris Lattner 9e90720658 Fix test/Regression/CFrontend/2003-11-01-EmptyStructCrash.c
llvm-svn: 18115
2004-11-22 19:15:27 +00:00
Chris Lattner ba18b9a0da Generalize this code to turn any cast-to-first-element-of into a gep constexpr
llvm-svn: 17914
2004-11-17 17:59:35 +00:00
Chris Lattner f60137501f fold gep undef, ... -> undef
This comes up many times in perlbmk and probably others.

llvm-svn: 17100
2004-10-17 21:54:55 +00:00
Chris Lattner 192eaccc1a Fix constant folding relational operators with undef operands.
llvm-svn: 17077
2004-10-17 04:01:51 +00:00
Chris Lattner fd7bf724d3 Implement constant folding of undef values.
llvm-svn: 17070
2004-10-16 23:31:32 +00:00
Chris Lattner 13128ab8fb Allow creation of GEP constantexprs with a vector of value* operands as
well as a vector of constant*'s.  It turns out that this is more efficient
and all of the clients want to do that, so we should cater to them.

llvm-svn: 16923
2004-10-11 22:52:25 +00:00
Chris Lattner b2b7f90caf If we are trying to create a ConstantExpr cast that is really a GEP to the
first element of an array, return a GEP instead of a cast.  This allows us
to transparently fold this:

int* getelementptr (int* cast ([100 x int]* %Gbody to int*), int 40)

into this:

int* getelementptr ([100 x int]* %Gbody, int 0, int 40)

llvm-svn: 16911
2004-10-11 03:57:30 +00:00
Reid Spencer accd7c708d bug 122:
- Replace ConstantPointerRef usage with GlobalValue usage
- Minimize redundant isa<GlobalValue> usage
- Correct isa<Constant> for GlobalValue subclass

llvm-svn: 14927
2004-07-17 23:47:01 +00:00
Chris Lattner 4bbd409749 Implement folding of expressions like 'uint cast (int* getelementptr (int*
null, uint 1) to uint)' to a constant integer.  We can only do this with
primitive LLVM types, because other types have target-specific sizes.

llvm-svn: 14837
2004-07-15 01:16:59 +00:00
Chris Lattner 71068a0462 When folding constant expr gep's, don't force the use of long indices.
llvm-svn: 14658
2004-07-07 04:45:13 +00:00
Chris Lattner 6b7275996c Rename Type::PrimitiveID to TypeId and ::getPrimitiveID() to ::getTypeID()
llvm-svn: 14201
2004-06-17 18:19:28 +00:00
Reid Spencer 8eb06df5ab Fix a bug that Chris asserts emphatically is a bug. The changed clause
would always return false because the Type of a type value is always
Type::TypeTY and can never be a floating point type.

llvm-svn: 13902
2004-05-30 01:19:48 +00:00
Chris Lattner 192e326cd3 Add a missing break, which caused a crash in an obscure situation
llvm-svn: 12825
2004-04-11 01:29:30 +00:00
Chris Lattner 69193f93b6 Support getelementptr instructions which use uint's to index into structure
types and can have arbitrary 32- and 64-bit integer types indexing into
sequential types.

llvm-svn: 12653
2004-04-05 01:30:19 +00:00
Chris Lattner 6ea4b52903 Teach the constant folder how to do select instructions
llvm-svn: 12321
2004-03-12 05:53:32 +00:00
Chris Lattner ea0789ca7b Implement test/Regression/Assembler/2004-03-07-FunctionAddressAlignment.llx
llvm-svn: 12218
2004-03-08 06:17:35 +00:00
Chris Lattner a9eddae213 Get all instruction definitions
llvm-svn: 11706
2004-02-22 06:25:38 +00:00
Chris Lattner 04b60fe5ad Move the folding of gep null, 0, 0, 0 to a place where it can be shared and
enjoyed by all, fixing a fixme.  Add an assert

llvm-svn: 11505
2004-02-16 20:46:13 +00:00
Chris Lattner 125ed54fdc Fix a crasher bug in my constant folding rewrite
llvm-svn: 11044
2004-02-01 01:23:19 +00:00
Chris Lattner 061da2f059 Implement a bunch of symbolic constant folding opportunities. This implements
testcase test/Regression/Assembler/ConstantExprFold.llx

Note that these kinds of things only rarely show up in source code, but are
exceedingly common in the intermediate stages of algorithms like SCCP.  By
folding things (especially relational operators) that use symbolic constants,
we are able to speculatively fold more conditional branches, which can
lead to some big simplifications.

It would be easy to add a lot more special cases here, so if you notice
SCCP missing anything "obvious", you know what to make smarter.  :)

llvm-svn: 10812
2004-01-13 05:51:55 +00:00
Chris Lattner 1dd054c41b Rearrange and comment code better. No functionality changes
llvm-svn: 10808
2004-01-12 22:07:24 +00:00
Chris Lattner 5a945e3616 Rename ConstantHandling.* -> ConstantFolding.*
Move a bunch of (now) private stuff from ConstantFolding.h into
ConstantFolding.cpp.

This _finally_ gets us to a place where we have a sane constant folder.  The
rules are:

1. LLVM clients now use ConstantExpr::get* methods to fold constants.  If they
   cannot be folded, a constantexpr is created, so these methods always return
   valid Constant*'s.
2. The implementation of ConstantExpr::get* uses the functions exposed by
   ConstantFolding.h to try to fold constants.  If they cannot be folded,
   they should return a null pointer.
3. The implementation of ConstantFolding can do whatever it wants, and only
   has one client (Constants.cpp)

This cuts down on the wierd dependencies, and eliminates the two interfaces.
The old constanthandling interface was especially bad for clients to use
because almost none of them took the failure condition into consideration,
thus leading to obscure problems.

llvm-svn: 10807
2004-01-12 21:13:12 +00:00