Commit Graph

218 Commits

Author SHA1 Message Date
Reid Spencer 9a6bb5fab7 Remove some more dead code resulting from adding setTypeName().
llvm-svn: 13862
2004-05-28 00:21:06 +00:00
Reid Spencer 3ae29942bf Remove an assertion that uses Type::TypeTy that is never hit and will
break when Type::TypeTy goes away. Also remove a dead block of code and
dead comments.

llvm-svn: 13861
2004-05-27 22:05:50 +00:00
Reid Spencer 5b4413cb03 Provide the correct patch for bug 345. The solution is to add a setTypeName
function to llvmAsmParser.y and then use it in the one place in the grammar
that needs it. Also had to make Type::setName public because setTypeName
needs it in order to retain compatibility with setValueName.

llvm-svn: 13795
2004-05-26 21:48:31 +00:00
Chris Lattner 1f08746f72 A quick and ugly hack to fix PR345. I used TypeTy specifically to make
Reid cringe :)

llvm-svn: 13788
2004-05-26 17:08:25 +00:00
Reid Spencer ea41b4fb83 Changed to use SymbolTable's new lookup interface.
llvm-svn: 13758
2004-05-25 17:29:21 +00:00
Chris Lattner ca96cee67b Fix a memory leak. We leaked the vector holding the entries in switch tables.
llvm-svn: 13023
2004-04-17 23:49:15 +00:00
Chris Lattner fd9fbe187d 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.

Auto-upgrade .ll files that use ubytes to index into structures to use uint's.

llvm-svn: 12652
2004-04-05 01:30:04 +00:00
Chris Lattner c0ba90e35c Avoid TRUE and FALSE which apparently conflict with some macros on OSX
llvm-svn: 12566
2004-03-31 03:49:47 +00:00
Chris Lattner deab5e212d Fix bug: Assembler/2004-03-30-UnclosedFunctionCrash.llx
llvm-svn: 12551
2004-03-30 20:58:25 +00:00
Chris Lattner 6536f0c35f Allow parsing select instruction and constant expr
llvm-svn: 12313
2004-03-12 05:51:36 +00:00
Chris Lattner e2b27080e7 Insert functions into the module promptly, not lazily. This fixes a bug
I introduced last night.  Note to self: test the *correct* tree...

llvm-svn: 12220
2004-03-08 16:14:19 +00:00
Chris Lattner 75998c0885 Eliminate a REALLY HORRIBLE API: mutateReferences, which is gross gross gross.
llvm-svn: 12212
2004-03-08 06:09:57 +00:00
Chris Lattner 7289e62874 It turns out that the two dimensional vectors were causing big slowdowns
in this for programs with lots of types (like the testcase in PR224).
The problem was that the type ID that the outer vector was using was not
very dense (as many types are getting resolved), so the vector is large
and gets reallocated a lot.

Since there are a lot of values in the program (the .ll file is 10M),
each reallocation has to copy the subvectors, which is also quite slow
(this wouldn't be a problem if C++ supported move semantics, but it
doesn't, at least not yet :(

Changing the outer data structure to a map speeds a release build of
llvm-as up from 11.21s to 5.13s on the testcase in PR224.

llvm-svn: 11244
2004-02-09 21:03:38 +00:00
Chris Lattner edd4500cf0 When resolving upreferences, if multiple uprefs will be resolved to the same
type at the same time, resolve the upreferences to each other before resolving
it to the outer type.  This shaves off some time from the testcase in PR224, from
25.41s -> 21.72s.

llvm-svn: 11241
2004-02-09 18:53:54 +00:00
Chris Lattner ac6db755c3 Adjust to the changed StructType interface. In particular, getElementTypes() is gone.
llvm-svn: 11228
2004-02-09 04:37:31 +00:00
Chris Lattner fa829be4d3 Start using the new and improve interface to FunctionType arguments
llvm-svn: 11224
2004-02-09 04:14:01 +00:00
Chris Lattner a9634a0ee3 This #include is not needed, it should have been removed with the last patch
llvm-svn: 11222
2004-02-09 03:22:32 +00:00
Chris Lattner 3cb4167b26 Instead of searching the entire type graph for a type to determine if it
contains the type we are looking for, just search the immediately used types.
We can only do this because we keep the "current" type in the nesting level
as we decrement upreferences.

This change speeds up the testcase in PR224 from 50.4s to 22.08s, not
too shabby.

llvm-svn: 11221
2004-02-09 03:19:29 +00:00
Chris Lattner a263b85fe6 Upreferences are always OpaqueTypes, meaning that it is impossible for a non-abstract
type from containing one.  This speeds up the asmparser on the testcase in PR224 from
61->50s.

llvm-svn: 11220
2004-02-09 03:03:10 +00:00
Chris Lattner 17235712fb Change the 'exception' destination to the 'unwind' destination. We will always
allow 'except' instead of 'unwind' here though.

llvm-svn: 11203
2004-02-08 21:48:25 +00:00
Chris Lattner 1b7d4d7b63 Don't use ConstantExpr::getShift anymore
llvm-svn: 10791
2004-01-12 19:08:43 +00:00
Chris Lattner fd2a514199 Minor code cleanups. The only bugfix is to the UR_DEBUG stuff which didn't
compile when enabled.

llvm-svn: 10657
2003-12-31 02:18:11 +00:00
Chris Lattner 87c593b882 Right, fix the problem with invoke instructions, not just call instructions
llvm-svn: 10599
2003-12-23 22:18:36 +00:00
Chris Lattner 601415ed84 Do not delete the type holder until after the call instruction has been
constructed!

llvm-svn: 10598
2003-12-23 20:39:17 +00:00
Chris Lattner e970301580 Minor cleanups, plug a minor memory leak
llvm-svn: 10596
2003-12-23 20:05:15 +00:00
Chris Lattner 6789a0b1d4 To not barf when an error occurs.
llvm-svn: 10236
2003-11-26 07:24:58 +00:00
Chris Lattner 2a18c640df Fix PR147
llvm-svn: 10204
2003-11-25 03:54:16 +00:00
Chris Lattner 3ae303cc8c Check return types of functions
llvm-svn: 10146
2003-11-21 22:32:23 +00:00
Chris Lattner 199bb859a9 Don't crash on bogus source value
llvm-svn: 10132
2003-11-21 20:27:35 +00:00
Chris Lattner ebcf5a5dba Fix bug PR107, patch contributed by Reid Spencer!
llvm-svn: 9911
2003-11-12 04:40:30 +00:00
Brian Gaeke 960707c335 Put all LLVM code into the llvm namespace, as per bug 109.
llvm-svn: 9903
2003-11-11 22:41:34 +00:00
Chris Lattner 0fc43a6048 Make sure that PHI node operands are first class types
llvm-svn: 9607
2003-10-30 01:38:18 +00:00
John Criswell 29265fe981 Added LLVM copyright header.
llvm-svn: 9321
2003-10-21 15:17:13 +00:00
Chris Lattner b94550e537 Change the Opcode enum for PHI nodes from "Instruction::PHINode" to "Instruction::PHI" to be more consistent with the other instructions.
llvm-svn: 9269
2003-10-19 21:34:28 +00:00
Chris Lattner 0079e7dd6d New revised variable argument handling support
llvm-svn: 9219
2003-10-18 05:53:13 +00:00
Chris Lattner 0aa8eaf7bf Tighten up handling of checks for shift instructions
llvm-svn: 9191
2003-10-17 05:11:44 +00:00
Chris Lattner 6187774380 Minor cleanups
llvm-svn: 9177
2003-10-16 20:12:13 +00:00
Chris Lattner 068ad84038 Add support for 'weak' linkage.
llvm-svn: 9171
2003-10-16 18:29:00 +00:00
Chris Lattner f95d9b99b3 Decrease usage of use_size()
llvm-svn: 9135
2003-10-15 16:48:29 +00:00
Chris Lattner 44d2c3514a Regularize header file comments
llvm-svn: 9071
2003-10-13 03:32:08 +00:00
Chris Lattner 8ad183194b Reserve space for PHI operands
llvm-svn: 9007
2003-10-10 16:34:58 +00:00
Chris Lattner f4120d9a2a Accept 'weak' as a linkage type. For now, just turn it into linkonce linkage
llvm-svn: 8998
2003-10-10 04:54:02 +00:00
Chris Lattner 6f8cd63693 Add better checking
llvm-svn: 8996
2003-10-10 03:56:01 +00:00
Chris Lattner 5c02edbf5c The objects mapped are really PATypeHolders, not PATypeHandles
llvm-svn: 8822
2003-10-02 19:00:34 +00:00
Chris Lattner 9eea08caa2 Oops, look at the VOLATILE marker, not the opcode
llvm-svn: 8413
2003-09-08 20:29:46 +00:00
Chris Lattner 9c58cf6d03 Add support for the unwind instruction
llvm-svn: 8408
2003-09-08 18:54:55 +00:00
Chris Lattner 8b1680eadb Parse volatile loads/stores
llvm-svn: 8402
2003-09-08 18:20:29 +00:00
Chris Lattner 667ded9bf8 Remove gross old hacky code that was in there for backwards compatibility
1 year is plenty of migration time!

llvm-svn: 8282
2003-09-01 16:31:28 +00:00
Chris Lattner 6a2b859191 Rename SwitchInst::dest_push_back -> addCase
llvm-svn: 8089
2003-08-23 23:14:52 +00:00
Chris Lattner 9eb287583b Accept double quoted strings everwhere we accept a %ABC variable name.
This introduces one more innoculous shift-reduce conflict, but will REALLY
help the type names generated by the C++ frontend, which wants to use all
kinds of crazy stuff.

llvm-svn: 8050
2003-08-22 05:42:10 +00:00
Misha Brukman 7eb05a170a Spell `necessary' correctly.
llvm-svn: 7944
2003-08-18 14:43:39 +00:00
Chris Lattner f26a8ee580 Remove redundant const qualifiers from cast<> expressions
llvm-svn: 7253
2003-07-23 15:30:06 +00:00
Misha Brukman 7fdaab4f68 The word `separate' only has one `e'.
llvm-svn: 7173
2003-07-14 17:20:40 +00:00
Chris Lattner 7969401e16 Add support for a new zeroinitializer token which can be used to get rid of
huge arrays of zero initialized values

llvm-svn: 6937
2003-06-28 20:01:34 +00:00
Chris Lattner 8826647168 Detemplatize the PATypeHandle class, which was only really instantiated on 'Type'.
llvm-svn: 6774
2003-06-18 19:22:36 +00:00
Chris Lattner d515967513 Fix bugs:
Assembler/2003-05-21-MalformedShiftCrash.llx
  Assembler/2003-05-21-ConstantShiftExpr.ll

llvm-svn: 6258
2003-05-21 17:48:56 +00:00
Chris Lattner 5bb2315a88 Fix bugs:
Assembler/2003-05-21-EmptyStructTest.ll
  Assembler/2003-05-21-MalformedStructCrash.llx

llvm-svn: 6255
2003-05-21 16:06:56 +00:00
Chris Lattner 9f3648bf02 Fix bug: Assembler/2003-05-15-SwitchBug.ll
llvm-svn: 6239
2003-05-15 21:30:00 +00:00
Chris Lattner f70da10697 Add support for the new va_arg instruction
llvm-svn: 6029
2003-05-08 02:44:12 +00:00
Chris Lattner 1fcfaf5bc0 Fix bug: Assembler/2003-04-25-UnresolvedGlobalReference.ll
llvm-svn: 5948
2003-04-25 21:47:33 +00:00
Chris Lattner 2012613040 Allow information about the target to be specified in the .ll file
llvm-svn: 5843
2003-04-22 19:07:06 +00:00
Chris Lattner 89da8a3b31 Namespacify
llvm-svn: 5840
2003-04-22 18:42:41 +00:00
Chris Lattner b672b5607f Preserve module source information in the ModuleID
llvm-svn: 5836
2003-04-22 18:02:52 +00:00
Chris Lattner 379a8d2d1c Add new linkage types to support a real frontend
llvm-svn: 5786
2003-04-16 20:28:45 +00:00
Chris Lattner fcfb1761c7 Don't allow declaring an external internal variable
llvm-svn: 5781
2003-04-16 18:13:57 +00:00
Chris Lattner 4b1d10604e Fix bug: Assember2003-04-15-ConstantInitAssertion.llx
llvm-svn: 5777
2003-04-15 16:09:31 +00:00
Chris Lattner 8ce4e7949b Fix bug: Assembler/2003-03-03-DuplicateConstant.ll
llvm-svn: 5692
2003-03-03 23:28:55 +00:00
Chris Lattner b90181e348 Fix bug: Assembler/2003-02-02-ConstGlobal.ll
llvm-svn: 5467
2003-02-02 16:40:20 +00:00
Chris Lattner 63a25242dc Fix bug: Assembler/2003-01-30-UnsignedString.ll
llvm-svn: 5445
2003-01-30 22:24:26 +00:00
Chris Lattner 98cf1f5d64 - Eliminated the deferred symbol table stuff in Module & Function, it really
wasn't an optimization and it was causing lots of bugs.

llvm-svn: 4779
2002-11-20 18:36:02 +00:00
Chris Lattner 1961329140 - Fix bug: Assembler/2002-10-15-NameClash.ll
- Clean up code a bit, s/Method/Function/

llvm-svn: 4188
2002-10-15 21:41:14 +00:00
Chris Lattner 8022a0f57c This loop executed exactly one time, turn it into straightline code
llvm-svn: 4159
2002-10-14 03:28:42 +00:00
Chris Lattner 149376dee5 - Change Function's so that their argument list is populated when they are
constructed.  Before, external functions would have an empty argument list,
    now a Function ALWAYS has a populated argument list.

llvm-svn: 4149
2002-10-13 20:57:00 +00:00
Chris Lattner fa01a52149 Minor, non-functionality changing, formatting fix
llvm-svn: 4091
2002-10-09 00:25:32 +00:00
Chris Lattner e2dd92be96 The parser now accepts "external" global variables, in addition to the backwards
compatible "uninitialized" global variables

llvm-svn: 4050
2002-10-06 22:45:09 +00:00
Chris Lattner 322bf4f3a9 Change the MallocInst & AllocaInst ctors to take the allocated type, not the
pointer type returned.

llvm-svn: 3710
2002-09-13 22:28:45 +00:00
Chris Lattner bb4fe2c3c4 Recently changed getelementptr to use 'long' indexes for sequential types
instead of uints.  This adds a translation scheme to be backwards compatible
with old .ll files.

llvm-svn: 3679
2002-09-11 01:17:27 +00:00
Chris Lattner a6834c1331 Use explicit ctor
llvm-svn: 3663
2002-09-10 22:37:46 +00:00
Chris Lattner c0686feca2 Tighten up error checking in parser, disallowing instructions that f.e.,
add pointers together.

llvm-svn: 3660
2002-09-10 19:57:26 +00:00
Chris Lattner 030effa42c Load & StoreInst no longer derive from MemAccessInst, so we don't have
to handle indexing anymore

llvm-svn: 3484
2002-08-22 22:48:55 +00:00
Chris Lattner 319c47adec - Fix asmparser and bytecode reader to not generate loads/stores with idxs
Now an obnoxious warning is emitted to discourage usage.  Eventually
    support will be removed.

llvm-svn: 3435
2002-08-21 23:51:21 +00:00
Chris Lattner 7f325cd0bc - Remove unused STRING token from lexer & parser
- Changed parser to always use parenthesis on ConstExprs to be consistent
  - Parser now passes TRUE and FALSE tokens as a special case of the ConstExpr
    machinery instead of a special case of constant int stuff
  - Fix the AsmParser to use ValueRef ::= ConstExpr, and remove
    ResolvedVal ::= ConstExpr this allows constexprs to be used in PHI nodes

llvm-svn: 3362
2002-08-16 21:14:40 +00:00
Chris Lattner d8ecff793a Parse const expr cast instruction the same way the instruction is formed
llvm-svn: 3353
2002-08-15 19:37:11 +00:00
Chris Lattner 7f1c98d060 Move ConstExpr production to unify ConstVal stuff
llvm-svn: 3351
2002-08-15 18:17:28 +00:00
Chris Lattner 61643a01c3 Fix bug: test/Regression/Assembler/2002-08-15-UnresolvedGlobalReference.ll
llvm-svn: 3350
2002-08-15 17:58:33 +00:00
Chris Lattner 330b7ac76c Remove support for Not ConstantExpr. This simplifies the unary case to only
have to support the cast instruction, so the function is renamed to getCast.

llvm-svn: 3328
2002-08-14 18:24:09 +00:00
Chris Lattner db3b202be8 AsmParser turns 'not' instructions into 'xor' instructions now.
llvm-svn: 3309
2002-08-14 17:12:33 +00:00
Chris Lattner 3cd8c56cbb Cleanup ConstantExpr handling:
* Correctly delete TypeHandles in AsmParser.  In addition to not leaking
   memory, this prevents a bug that could have occurred when a type got
   resolved that the constexpr was using
 * Check for errors in the AsmParser instead of hitting assertion failures
   deep in the code
 * Simplify the interface to the ConstantExpr class, removing unneccesary
   parameters to the ::get* methods.
 * Rename the 'getelementptr' version of ConstantExpr::get to
   ConstantExpr::getGetElementPtr

llvm-svn: 3161
2002-07-30 18:54:25 +00:00
Chris Lattner 91e0832181 Add check to fix bug:
test/Regression/Assembler/2002-07-25-ParserAssertionFailure.llx

llvm-svn: 3101
2002-07-25 20:52:56 +00:00
Chris Lattner e680139c34 GCC 3.1 changes
llvm-svn: 3074
2002-07-25 06:17:42 +00:00
Chris Lattner 8eedb941d3 * Code cleanups
* Fix a REALLY misleading error message

llvm-svn: 2960
2002-07-18 05:18:37 +00:00
Chris Lattner 980ddf5854 ConstExpr::getelementptr now takes a vector of Constants not Values
llvm-svn: 2948
2002-07-18 00:14:27 +00:00
Chris Lattner 8445b41ddd Fix bug: test/Regression/Assembler/2002-07-14-InternalLossage.llx
llvm-svn: 2907
2002-07-15 00:10:33 +00:00
Vikram S. Adve 7064eaf80b Added support for parsing expressions constructed from constant values.
(Major new non-terminal is ConstExpr.)

Add YYERROR_VERBOSE and print additional information on errors.

llvm-svn: 2894
2002-07-14 22:59:28 +00:00
Chris Lattner 113f4f4609 MEGAPATCH checkin.
For details, See: docs/2002-06-25-MegaPatchInfo.txt

llvm-svn: 2779
2002-06-25 16:13:24 +00:00
Chris Lattner 8cb1dfbdc9 * Be more typesafe: cast<x> now no longer discards constness
* The newly installed version of Bison now really likes there to be ;'s
  after productions.  Add them to avoid tons of warnings.

llvm-svn: 2757
2002-06-04 21:58:56 +00:00
Chris Lattner f4c7dd9c22 Support function declarations with either %XXX or "XXX" style for now
"XXX" style should be considered deprecated, and will hopefully be removed
in the future.

llvm-svn: 2727
2002-05-22 22:33:00 +00:00
Chris Lattner 181cc32365 Replace all usages of Type::isPointerType with isa<PointerType>
llvm-svn: 2486
2002-05-06 16:15:30 +00:00
Chris Lattner 638fa9385d Remove dead code
llvm-svn: 2473
2002-05-06 03:03:09 +00:00
Chris Lattner 1e1a9b4166 Allow open/closing curly braces instead of begin/end to define the body of a function
llvm-svn: 2451
2002-05-03 18:23:48 +00:00