Commit Graph

965 Commits

Author SHA1 Message Date
Chris Lattner 3d903f0c0b Add VMCore and code generator support for debugging intrinsics. By default
code generators completely ignore them.

llvm-svn: 10691
2004-01-05 05:36:30 +00:00
Chris Lattner 09bd1a06c1 Add some comments, add new getGlobalVariable method
llvm-svn: 10671
2003-12-31 08:43:01 +00:00
Chris Lattner be3596c8ca * Make Module::getTypeName const
* Add new Module::getTypeByName method
* Group methods in Module.cpp better

llvm-svn: 10668
2003-12-31 07:09:33 +00:00
Chris Lattner efeb39e3c0 Make the lookup method const.
llvm-svn: 10667
2003-12-31 07:08:19 +00:00
Chris Lattner eb537b2401 Fix PR198
llvm-svn: 10659
2003-12-31 03:19:37 +00:00
Chris Lattner af3f8f3b42 Add comments
llvm-svn: 10658
2003-12-31 02:50:02 +00:00
Chris Lattner 5d236005b0 Clean up a lot of the code I added yesterday by exposing the IntrinsicLowering
implementation from the TargetMachine directly.

llvm-svn: 10636
2003-12-28 21:23:38 +00:00
Chris Lattner 24650d2b5e Move into the VMCore library
llvm-svn: 10623
2003-12-28 08:30:20 +00:00
Chris Lattner d1ea9cd345 The recalclulate method was a nasty hack that was once used by the -cee pass,
which never worked itself.  The cee pass still doesn't work, but it doesn't use
this method anymore anyway, so eliminate the method.

llvm-svn: 10302
2003-12-07 00:55:32 +00:00
Chris Lattner 00f5167693 Completely rewrite domset, idom, and domtree implementation. Now it is based
on the algorithm for directly computing immediate dominators presented in this
paper:

  A Fast Algorithm for Finding Dominators in a Flowgraph
  T. Lengauer & R. Tarjan, ACM TOPLAS July 1979, pgs 121-141.

This _substantially_ speeds up construction of all dominator related information.
Post-dominators to follow.

llvm-svn: 10301
2003-12-07 00:38:08 +00:00
Chris Lattner f5569bb9a5 Fix const correctness
llvm-svn: 10229
2003-11-25 21:28:00 +00:00
Chris Lattner ad70d4a15b Relax constrains on GEP type indexes
llvm-svn: 10228
2003-11-25 21:21:46 +00:00
Chris Lattner 3ae303cc8c Check return types of functions
llvm-svn: 10146
2003-11-21 22:32:23 +00:00
Chris Lattner dcdc371522 Fix problem Reid was having with external globals and nonexternal linkage
llvm-svn: 10133
2003-11-21 20:33:27 +00:00
Chris Lattner 189d19fb04 Finegrainify namespacification
llvm-svn: 10131
2003-11-21 20:23:48 +00:00
Chris Lattner 7e5e456d96 Fix some problems with assertions printing
llvm-svn: 10129
2003-11-21 17:35:51 +00:00
Chris Lattner 9d72c2f229 improve error message
llvm-svn: 10128
2003-11-21 17:06:29 +00:00
Chris Lattner a29600046d Minor cleanups and simplifications
llvm-svn: 10127
2003-11-21 16:52:05 +00:00
Chris Lattner 2af517281d Start using the nicer terminator auto-insertion API
llvm-svn: 10111
2003-11-20 18:25:24 +00:00
Chris Lattner 9ab47d1625 Somehow I forgot poor little UnwindInst
*sniff*

llvm-svn: 10108
2003-11-20 18:11:56 +00:00
Chris Lattner 0e03ab67c3 * Finegrainify namespacification
* Add new constructors to allow insertion of terminator instructions at the
  end of basic blocks.
* Move a ReturnInst method out-of-line, so that the vtable and type info don't
  need to be emitted to every translation unit that uses the class.

llvm-svn: 10107
2003-11-20 17:45:12 +00:00
Chris Lattner 2447ef59c4 Don't crash if we are printing an orphaned basic block!
llvm-svn: 10100
2003-11-20 00:09:43 +00:00
Chris Lattner 45cfe54683 Really, there was a reason for me factoring that out
llvm-svn: 10089
2003-11-19 19:20:20 +00:00
Chris Lattner 1a02ca932c *** Significantly speed up type resultion
This change speeds up type resolution by checking to see if a type is
  recursive, and if it's not, using a more efficient algorithm.

  This dramatically reduces bytecode loading time of kc++, reducing time-to-jit
  kc++ --version to 17s from 33s

llvm-svn: 10088
2003-11-19 19:10:23 +00:00
Chris Lattner a11f78167b Finegrainify namespacification
llvm-svn: 10080
2003-11-19 06:14:38 +00:00
Chris Lattner 74c55dd4f9 Fix PR130, and testcase test/Regression/Linker/2003-11-18-TypeResolution.ll
llvm-svn: 10075
2003-11-19 01:02:52 +00:00
Chris Lattner dc2e39191a Implement == and != correctly. Before they would incorrectly return !=
for some constant exprs when they could really be the same value

llvm-svn: 10058
2003-11-17 20:19:35 +00:00
Chris Lattner e120a73166 The ConstantPointer class is now gone.
llvm-svn: 10057
2003-11-17 19:47:21 +00:00
Chris Lattner 77f20dc13f No REALLY, get rid of ConstantPointer references
llvm-svn: 10054
2003-11-17 19:21:04 +00:00
Chris Lattner 4b6addf98a Eliminate use of the ConstantPointer class
llvm-svn: 10053
2003-11-17 19:19:32 +00:00
Chris Lattner 9d9cbcf246 Constant folding shalt not be built on annotations
llvm-svn: 10052
2003-11-17 19:05:17 +00:00
Chris Lattner c96e96b26d Be a bit more tolerant of broken code
llvm-svn: 10050
2003-11-17 01:17:04 +00:00
Brian Gaeke fda1f1835b When you hand WriteAsOperand a type, it now prints out its symbolic name.
llvm-svn: 10042
2003-11-16 23:08:27 +00:00
Brian Gaeke 9f47927b28 When failing a Verifier assertion, use WriteAsOperand to print out the
offending types and/or values.

llvm-svn: 10041
2003-11-16 23:07:42 +00:00
Chris Lattner 00211f1d80 Don't print 'No predecessors!' on the entry block
llvm-svn: 10038
2003-11-16 22:59:57 +00:00
Chris Lattner ef901291c6 Expose the verifier to opt
llvm-svn: 9984
2003-11-13 19:47:29 +00:00
Chris Lattner 3b93c91836 Detect problems with PHI node operands!
llvm-svn: 9916
2003-11-12 07:13:37 +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 a570f7b681 Fix PR95. I'm checking this patch in for Reid Spencer, who figured it out
and wrote it up.  Thanks!!

llvm-svn: 9832
2003-11-09 19:39:48 +00:00
Chris Lattner da8571b464 Add an assertion
llvm-svn: 9831
2003-11-09 15:51:07 +00:00
Chris Lattner b36e8a1a2e Add support for constant expr shifts
llvm-svn: 9735
2003-11-05 19:53:03 +00:00
Chris Lattner e225507193 If we're not checking, don't check!
llvm-svn: 9732
2003-11-05 19:09:40 +00:00
Chris Lattner 41baa98478 Make code gcc 3.4 clean
llvm-svn: 9719
2003-11-05 05:15:42 +00:00
Chris Lattner 7c18b18c46 Fix InstCombine/2003-11-03-VarargsCallBug.ll
llvm-svn: 9681
2003-11-03 18:44:58 +00:00
Chris Lattner 9157ec025b Did I mention that I _HATE_ CPRs?
llvm-svn: 9639
2003-10-31 18:38:06 +00:00
Chris Lattner f6925da295 Constant pointer refs are causing these to fail unnecessarily, which is causing
a lot of code to be pessimized.  I hate CPRs.  :(

llvm-svn: 9635
2003-10-31 17:51:16 +00:00
Alkis Evlogimenos 6029d4032c Use already existing classes for post order iterator.
llvm-svn: 9629
2003-10-31 03:02:34 +00:00
Chris Lattner 8339f7d379 Make use of the new AssemblyAnnotationWriter interface
llvm-svn: 9619
2003-10-30 23:41:03 +00:00
Alkis Evlogimenos 5a320efd3b Output types in reverse postorder. This will allow the ByteCode/Reader
to create the minimum number of opaque types for each type with a
cycle in its type graph.

llvm-svn: 9615
2003-10-30 21:04:44 +00:00
Chris Lattner f14ead9210 Print the names of more opaque types
llvm-svn: 9598
2003-10-30 00:22:33 +00:00
Chris Lattner 92d60539c8 Make sure to print opaque types names if they are available.
llvm-svn: 9597
2003-10-30 00:12:51 +00:00
Alkis Evlogimenos d478d3d19c Add more debugging info to help tracing the SlotCalculator.
llvm-svn: 9575
2003-10-29 03:12:12 +00:00
Chris Lattner 8a55262c76 Fix bug: instcombine/2003-10-23-InstcombineNullFail.ll
llvm-svn: 9403
2003-10-23 05:21:48 +00:00
Chris Lattner 2c35b1c94c Make sure to add a type for va_next operands, which may not other wise
occur in the module

llvm-svn: 9336
2003-10-21 17:40:54 +00:00
Chris Lattner 7267b352fb Fix problem which effected linking apache
llvm-svn: 9335
2003-10-21 17:39:59 +00:00
John Criswell 29265fe981 Added LLVM copyright header.
llvm-svn: 9321
2003-10-21 15:17:13 +00:00
John Criswell 4436c49787 Added LLVM copyright notice to Makefiles.
llvm-svn: 9312
2003-10-20 22:26:57 +00:00
John Criswell 482202a601 Added LLVM project notice to the top of every C++ source file.
Header files will be on the way.

llvm-svn: 9298
2003-10-20 19:43:21 +00:00
Chris Lattner d77dd78c52 "Function"[0] == 'F', not 'M'
llvm-svn: 9297
2003-10-20 19:10:06 +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 5b33748497 Add support for the new varargs intrinsics and instructions
llvm-svn: 9226
2003-10-18 05:57:43 +00:00
Misha Brukman e5a2565b91 Chris says it's better to assert that TheModule is valid than silently return 0.
llvm-svn: 9212
2003-10-17 21:59:57 +00:00
Misha Brukman de360f88c8 Return the Module that we just materialized.
llvm-svn: 9201
2003-10-17 18:27:13 +00:00
Alkis Evlogimenos 8faf8d93d5 Eliminate some extraneous code in SlotCalculator::insertVal().
Rename SlotCalculator::getValSlot() to SlotCalculator::getSlot(),
       SlotCalculator::insertValue() to SlotCalculator::getOrCreateSlot(),
       SlotCalculator::insertVal() to SlotCalculator::insertValue(), and
       SlotCalculator::doInsertVal() to SlotCalculator::doInsertValue().

llvm-svn: 9190
2003-10-17 02:02:40 +00:00
Chris Lattner 068ad84038 Add support for 'weak' linkage.
llvm-svn: 9171
2003-10-16 18:29:00 +00:00
Chris Lattner 079edeb919 Completely rewrite support for the Value::use_* list. Now, all operations on
this list (except use_size()) are constant time.  Before the killUse method
(used whenever something stopped using a value) was linear time, and thus
very very slow for large programs.

This speeds GCCAS up _substantially_ on large programs: almost 2x for 176.gcc:

176.gcc:     77.07s -> 37.38s
177.mesa:     7.59s ->  5.57s
252.eon:     21.02s -> 19.52s (*)
253.perlbmk: 11.40s -> 13.05s
254.gap:      7.25s -> 7.42s

252.eon would speed up a whole lot more, but optimization time is being
dominated by the inlining pass, which needs to be fixed.

llvm-svn: 9160
2003-10-16 16:53:07 +00:00
Misha Brukman 56a8642a42 Enabling incremental bytecode loading in the JIT:
* FunctionPassManager ctor now takes in a ModuleProvider
* run() materializes function before running passes on it

llvm-svn: 9126
2003-10-14 21:38:42 +00:00
Chris Lattner 0c068c097a Stop using "reverse depth first" order
llvm-svn: 9081
2003-10-13 15:21:58 +00:00
Chris Lattner e859403772 Speed up TypesEqual by specializing it for all of the derived types, avoiding
a lot of virtual method dispatch overhead.

llvm-svn: 9078
2003-10-13 14:55:56 +00:00
Chris Lattner 1dc3f3fdc0 Avoid creating lots of pointless opaque types, with short lifetimes
llvm-svn: 9076
2003-10-13 14:03:36 +00:00
Chris Lattner c7bc3bb503 This is a header file, hear me roar
llvm-svn: 9075
2003-10-13 05:34:24 +00:00
Chris Lattner ab16a659d7 * Remove extraneous #includes
* Fix a nasty initializer ordering bug.  Any only-CFG passes which registered
  themselves before the CFGOnlyAnalysis vector initialized got forgotten and
  thus got invalidated and recomputed.

In particular, in my compiled version of gccas, the Loop information pass was
being recomputed unnecessarily.

llvm-svn: 9074
2003-10-13 05:33:01 +00:00
Chris Lattner 44d2c3514a Regularize header file comments
llvm-svn: 9071
2003-10-13 03:32:08 +00:00
Chris Lattner dd99f5bf8f Rename method
llvm-svn: 9055
2003-10-12 18:52:12 +00:00
Brian Gaeke cc31fddf13 Don't include Config/stdio.h or <stdio.h>.
llvm-svn: 9031
2003-10-10 18:46:29 +00:00
Misha Brukman fa100531e1 Fix spelling/grammar.
llvm-svn: 9023
2003-10-10 17:54:14 +00:00
Chris Lattner 815ae2bfe3 Add a sanity check for constant expression casts
llvm-svn: 8951
2003-10-07 22:19:19 +00:00
Chris Lattner 794a7490f8 Users can never be null
llvm-svn: 8895
2003-10-06 17:33:39 +00:00
Chris Lattner df9779c5c2 Be more careful handling PHI nodes, which might be of potentially high degree.
This reduces the time to verify a function from eon with a large number of
large PHI nodes from 22996s (6.38 hours) to 10.5499s

llvm-svn: 8866
2003-10-05 17:44:18 +00:00
Chris Lattner b50d135710 Type tables are now AbstractTypeUsers. This allows them to merge together
constants as necessary due to type resolution.  With this change, the
following spec benchmarks now link: 176.gcc, 177.mesa, 252.eon,
253.perlbmk, & 300.twolf.  IOW, all SPEC INT and FP benchmarks now link.

llvm-svn: 8853
2003-10-05 00:17:43 +00:00
Chris Lattner cc593fb6c6 Rename AbstractModuleProvider -> ModuleProvider, to match the header file name,
and because, while the class used by the interface is abstract, the actual
concept is not.

llvm-svn: 8850
2003-10-04 20:14:59 +00:00
Chris Lattner 3e88ef9f93 Minor random speedup: make Constant::getNullValue only call ConstantFOO::get
once!

llvm-svn: 8845
2003-10-03 19:34:51 +00:00
Chris Lattner 2c6cecf321 Explicit copy ctors are no longer needed
llvm-svn: 8844
2003-10-03 19:02:41 +00:00
Chris Lattner 7f4cfda3ba Minor cleanups
llvm-svn: 8843
2003-10-03 18:57:54 +00:00
Chris Lattner 0f21ab75fa This checkin basically amounts to a complete rewrite of the type-resolution
machinery.  This dramatically simplifies how things works, removes irritating
little corner cases, and overall improves speed and reliability.

Highlights of this change are:

1. The exponential algorithm built into the code is now gone.  For example
   the time to disassemble one bytecode file from the mesa benchmark went
   from taking 12.5s to taking 0.16s.
2. The linker bugs should be dramatically reduced.  The one remaining bug
   has to do with constant handling, which I actually introduced in
   "union-find" checkins.
3. The code is much easier to follow, as a result of fewer special cases.
   It's probably also smaller.  yaay.

llvm-svn: 8842
2003-10-03 18:46:24 +00:00
Chris Lattner 29dc65a324 These methods are dead, remove them. Because the bodies will soon be
ressurected, just ifdef them out for now.

llvm-svn: 8840
2003-10-03 18:39:57 +00:00
Chris Lattner a72b78967f Dramatically simplify DerivedType::refineAbstractTypeToInternal
This makes use of the new PATypeHolder's to keep types from being deleted
prematurely, instead of the wierd "self reference" garbage.  This is easier
to understand and more efficient as well.

llvm-svn: 8834
2003-10-03 04:48:21 +00:00
Chris Lattner 016418b8c2 Make the PATypeHolder use a simple union-find implementation to handle
merging of types.  This makes it MUCH more efficient than before, also
making things simpler.

llvm-svn: 8833
2003-10-02 23:35:57 +00:00
Chris Lattner 22e4ca8863 There is no reason for Value to be an AbstractTypeUser. This just makes things
significantly more complete.  Instead, just make DerivedType's AbstractTypeUser's,
and make Value contain a PATypeHolder.

llvm-svn: 8828
2003-10-02 19:44:40 +00:00
Chris Lattner 64d88bc788 Do not use BasicBlock::succ_iterator
llvm-svn: 8700
2003-09-24 22:03:22 +00:00
Misha Brukman cd1d0e31f5 * No longer initialize the module, a subclass is more suitable to do this
because it can add a module ID which we do not have at this time.
* Check to see if the module has been initialized when materializing it.

llvm-svn: 8674
2003-09-22 23:44:13 +00:00
Chris Lattner 5dac64f629 Rename Function::getEntryNode -> getEntryBlock
llvm-svn: 8625
2003-09-20 14:39:18 +00:00
Chris Lattner 3284ed725a Reject "llvm." as a function name
llvm-svn: 8608
2003-09-19 19:31:41 +00:00
Misha Brukman 3d72faddeb Minimal implementation of the abstract ModuleProvider interface.
llvm-svn: 8594
2003-09-18 16:17:06 +00:00
Chris Lattner c1b1651210 Change the semancics of the dropallrefs method
llvm-svn: 8572
2003-09-17 04:58:59 +00:00
Chris Lattner bb9d03b9c0 Renamed DominatorTree::Node::getNode() -> getBlock()
llvm-svn: 8469
2003-09-11 16:26:13 +00:00
Chris Lattner 630d2954be Rework dominator interfaces to handle changes in the post-dominance
construction.  Now there may be multiple root blocks, and null is a
special node used to mark the "virtual" exit node of a CFG.

llvm-svn: 8461
2003-09-10 20:37:51 +00:00
Brian Gaeke ff713ae819 Make createVerifierPass return a FunctionPass *.
llvm-svn: 8449
2003-09-10 19:37:04 +00:00
Chris Lattner 216c7b8aed Fix spell-o's
llvm-svn: 8431
2003-09-10 05:29:43 +00:00
John Criswell f711d34e98 Reverting back to the previous revision.
We want to check for length 5 because we might get the "llvm." string as the
name.  That string is in the LLVM namespace and should be checked as such.
We also don't have to worry about garbage data because (I believe) the string
class will return a valid value.  So, the switch statement will work and we
don't have to worry about the code wandering into segfault land.

llvm-svn: 8419
2003-09-09 16:58:41 +00:00
John Criswell 8d575cb2e1 Modified the check for intrinsic functions. All intrinsic functions have to
be at least 6 characters, since something must follow the "llvm." string in the
function name.
This seems to fix an assertion failure with the SingleSource tests, too.

llvm-svn: 8418
2003-09-09 16:36:09 +00:00
Chris Lattner 04ecefe232 Eliminate support for the llvm.unwind intrinisic, using the Unwind instruction instead
llvm-svn: 8411
2003-09-08 19:44:26 +00:00
Chris Lattner 66d5f575ba Add support for unwind
llvm-svn: 8407
2003-09-08 18:54:36 +00:00
Chris Lattner 504f9242c3 Add support for volatile loads/stores
llvm-svn: 8393
2003-09-08 17:45:59 +00:00
Misha Brukman 730e047678 Spell `occurrence' correctly.
llvm-svn: 8388
2003-09-08 03:08:43 +00:00
Chris Lattner 20b98d2c93 Add asserts, move code around. This gets the dropAllTypeUses partially implemented
llvm-svn: 8361
2003-09-05 05:10:04 +00:00
Chris Lattner d15598bfe0 Move functionality preserving changes.
Inline containsEquivalent
Remove some //'s

llvm-svn: 8358
2003-09-05 02:39:52 +00:00
Chris Lattner 5ac07d2221 Refactor what has effectively become copy-and-pasted code
llvm-svn: 8357
2003-09-05 02:30:47 +00:00
Chris Lattner 3f695ff02e Check in a bunch of minor fixes, plus a whole lot of #if 0'd out code, which will hopefully be enabled in the near future
This does not make any functionality changes

llvm-svn: 8355
2003-09-05 02:21:39 +00:00
Chris Lattner e4fdce4235 containsEquivalent never returns its argument
llvm-svn: 8353
2003-09-04 23:49:53 +00:00
Chris Lattner cd27235b23 Remove needless usage of getDescription()
llvm-svn: 8351
2003-09-04 23:46:03 +00:00
Chris Lattner 59be7695ef Inline method
llvm-svn: 8350
2003-09-04 23:43:40 +00:00
Chris Lattner b8b10b5c54 Don't try to be sneaky, breaking recursive types in the process
llvm-svn: 8348
2003-09-04 23:41:03 +00:00
Chris Lattner 50745187ef This is now unnecessary
llvm-svn: 8347
2003-09-04 23:38:22 +00:00
Chris Lattner 13651f0c46 Fix bug where we couldn't print a function without a name
llvm-svn: 8341
2003-09-03 17:56:43 +00:00
Chris Lattner b89d2afab1 No need to rescan types when they are created.
llvm-svn: 8339
2003-09-03 14:44:53 +00:00
Chris Lattner cc61a136f7 Another small speedup, this one to: 2.42s
llvm-svn: 8335
2003-09-02 22:52:49 +00:00
Chris Lattner 188afa5feb Don't return bogus references, and don't add an entry to the Concrete map
unless it's not empty!

llvm-svn: 8334
2003-09-02 22:50:02 +00:00
Chris Lattner 39e5095c8e Another optimization, speed up the testcase to 2.7s
llvm-svn: 8333
2003-09-02 22:15:15 +00:00
Chris Lattner 835499277b Removal of explicit stack, which requires the method to be a member (so it can
call setAbstract).  Now that we just compute abstractness we can also return
the computed value by value instead of as an argument.

llvm-svn: 8332
2003-09-02 21:56:34 +00:00
Chris Lattner 9a61bbf1d3 Remove the "recursive bit", not only is it unused by anyone, it was also
not correctly calculated, and calculating it wrong for fun seems rather
pointless.  This also speeds up my favorite testcase by .25 seconds.

llvm-svn: 8330
2003-09-02 21:41:05 +00:00
Chris Lattner ce35f14050 Don't bother doing an exhaustive recursive walk if we are at the limit of what
we need to know anyway.  This reduces the 2002-07-08-HugePerformanceProblem.llx
down to 3.210u:0.010s, which is back in the acceptable range again

llvm-svn: 8323
2003-09-02 20:06:29 +00:00
Chris Lattner 99b9ddead7 Ugh, fix bugs. Ok, so the last fix wasn't as great as I thought it was.
Now we're back to: 131.730u, 0.330s, which is still quite an improvement,
but still quite unacceptable

llvm-svn: 8322
2003-09-02 19:14:12 +00:00
Chris Lattner 2a36482e74 Further simplifications
llvm-svn: 8321
2003-09-02 16:46:41 +00:00
Chris Lattner e0703ff161 Change the code to no longer compute the "type description" immediately when
the type is analyzed.  Instead, only compute it when requested (with
getDescription), and cached for reuse later.

This dramatically speeds up LLVM in general because these descriptions almost
_never_ need to be constructed.  The only time they are used is when a type is
<<'d.  Printing of modules by themselves uses other code to print symbolic
types when possible, so these descriptions are really only used for debugging.

Also, this fixes the particularly bad case when lots of types get resolved to
each other, such as during linking of large programs.  In these cases, the type
descriptions would be repeatedly recomputed and discarded even though: A. noone
reads the description before it gets resolved, and B. many many resolutions
happen at intermediate steps, causing a HUGE waste of time.

Overall, this makes the getTypeDesc function much more light-weight, and fixes
bug: Assembler/2002-07-08-HugePerformanceProblem.llx, which went from taking
1048.770u/19.150s (which is 17.5 MINUTES, on apoc), to taking 0.020u/0.000s,
which is a nice little speedup.  :)

llvm-svn: 8320
2003-09-02 16:35:17 +00:00
Brian Gaeke e248195b22 Remove assertion which is never reached.
llvm-svn: 8318
2003-09-02 06:45:34 +00:00
Chris Lattner bd717d897f Implement new method
llvm-svn: 8238
2003-08-31 00:19:28 +00:00
Chris Lattner a38adaa88f Fix bug: Linker/2003-08-28-TypeResolvesGlobal3.ll
Use new replaceAllUsesWith stuff in a way that works even though types have not yet been propagated

llvm-svn: 8210
2003-08-29 05:37:22 +00:00
Chris Lattner c27038d7b5 Implement "unsafe" replaceAllUsesWWith stuff for use during type resolution.
llvm-svn: 8209
2003-08-29 05:36:46 +00:00
Chris Lattner 6097d94295 Fix bug: Linker/2003-08-28-TypeResolvesGlobal2.ll
llvm-svn: 8206
2003-08-29 05:11:34 +00:00
Chris Lattner 9f15812e88 Refactor code to make it useful outside of Constants.cpp
llvm-svn: 8205
2003-08-29 05:09:37 +00:00
Chris Lattner ae77242011 Fix bug: Linker/2003-08-28-TypeResolvesGlobal.ll
llvm-svn: 8202
2003-08-29 04:49:54 +00:00
Chris Lattner ea9891311a As it turns out, things will be simpler than I first expected. We no longer
need any exception handling intrinsics beyond llvm.unwind. (yaay)

llvm-svn: 8145
2003-08-25 22:35:01 +00:00
Chris Lattner 8068e0cd17 Support new 'any' support for pointer size and endianness
llvm-svn: 8119
2003-08-24 13:48:48 +00:00
Chris Lattner 7da3571af9 rethrow is really the language independent primitive here. "throw" can be written
in terms of it and llvm.exc.setcurrent.

Rework the intrinsics.

llvm-svn: 8110
2003-08-24 12:24:08 +00:00
Chris Lattner ca3e569269 Fix copy-and-paste-o
llvm-svn: 8104
2003-08-24 06:57:53 +00:00
Chris Lattner ade9410c99 Initial support for recognizing LLVM exception handling intrinsics
llvm-svn: 8102
2003-08-24 05:30:29 +00:00
Chris Lattner 6d5693566f Allow specifying the name for the newly split basic block
llvm-svn: 8097
2003-08-24 03:41:39 +00:00
Chris Lattner 180e14cb87 Rename SwitchInst::dest_push_back -> addCase
Add new removeCase method

llvm-svn: 8088
2003-08-23 23:14:37 +00:00
Chris Lattner 1c3430491e If an "LLVM name" has wierd characters in it, print it out in double quotes instead of prefixing it with %
llvm-svn: 8049
2003-08-22 05:40:38 +00:00
Misha Brukman be372b91e1 The word `dependent' has no `a'.
llvm-svn: 8030
2003-08-21 22:14:26 +00:00
Chris Lattner 941b06bf3c Implement Assembler/2003-08-21-ConstantExprCast-Fold.llx
llvm-svn: 8015
2003-08-21 19:45:55 +00:00
Chris Lattner ddab239262 Implement testcase: Assembler/2003-08-20-ConstantExprGEP-Fold.llx
llvm-svn: 7983
2003-08-20 16:11:27 +00:00
Chris Lattner f8a77578a2 Do not even attempt to compute dominator information for unreachable blocks
llvm-svn: 7963
2003-08-18 22:11:16 +00:00
Chris Lattner ab69ecc6d5 Fix linking bug in xboard
llvm-svn: 7952
2003-08-18 16:54:48 +00:00
Chris Lattner 748e9e143b Add intrinsics for the llvm.sig(set|long)jmp functions
llvm-svn: 7949
2003-08-18 15:41:24 +00:00
Misha Brukman 7eb05a170a Spell `necessary' correctly.
llvm-svn: 7944
2003-08-18 14:43:39 +00:00
Brian Gaeke 2cc4b9dd2c Add new method to FunctionPassManager to add ImmutablePasses.
llvm-svn: 7838
2003-08-14 06:07:57 +00:00
Chris Lattner e7e094467d Dramatically simplify how -time-passes works.
This also enables -time-passes for FunctionPassManagers, which allows it to
be used for the JIT

llvm-svn: 7834
2003-08-14 05:20:28 +00:00
Chris Lattner cea4d8c2f5 Implement constant folding of casts from boolean constants to other values
llvm-svn: 7812
2003-08-13 15:52:25 +00:00
Brian Gaeke b7a8388b8c Add FunctionPassManager - it's like a PassManager, but it only deals in
FunctionPasses.

llvm-svn: 7778
2003-08-12 17:22:39 +00:00
Chris Lattner 4159fdaef2 Add a bunch of new Alpha Intrinsics for Rahul Joshi
llvm-svn: 7646
2003-08-06 20:08:25 +00:00
Chris Lattner 463d6a5e7c Implement TODO: print out short form of Invoke if possible
llvm-svn: 7595
2003-08-05 15:34:45 +00:00
Tanya Lattner f8c563f76c *** empty log message ***
llvm-svn: 7447
2003-07-31 05:06:09 +00:00
Tanya Lattner a93c7aeb7e Added function to determine if an Instruction may trap.
llvm-svn: 7442
2003-07-31 04:05:50 +00:00
Chris Lattner 3a5c2826d5 Fix a bug that brian reported
llvm-svn: 7428
2003-07-30 19:04:37 +00:00
Chris Lattner 60104f00c7 Add support for Alpha intrinsics, contributed by Rahul Joshi
llvm-svn: 7373
2003-07-28 21:20:57 +00:00
Chris Lattner 4b4dacd838 Make Module::getNamedFunction prefer non-external functions if there is more than
one function of the same name

llvm-svn: 7274
2003-07-23 20:21:30 +00:00
Chris Lattner f26a8ee580 Remove redundant const qualifiers from cast<> expressions
llvm-svn: 7253
2003-07-23 15:30:06 +00:00
Chris Lattner 6077c3195f Simplify code by using ConstantInt::getRawValue instead of checking to see
whether the constant is signed or unsigned, then casting

llvm-svn: 7252
2003-07-23 15:22:26 +00:00
Misha Brukman 7fdaab4f68 The word `separate' only has one `e'.
llvm-svn: 7173
2003-07-14 17:20:40 +00:00
John Criswell 3ef61afb76 Merged in autoconf branch. This provides configuration via the autoconf
system.

llvm-svn: 7014
2003-06-30 21:59:07 +00:00
Chris Lattner d401f39872 Avoid printing out huge structures or arrays if they are just filled with zeros
llvm-svn: 6938
2003-06-28 20:08:24 +00:00
Chris Lattner 4ede64e304 Implement more aggressive folding of constant GEP instructions
llvm-svn: 6913
2003-06-26 05:22:45 +00:00
Chris Lattner 8bd8bc8399 Make assertion more descriptive
llvm-svn: 6889
2003-06-24 22:20:19 +00:00
Chris Lattner 4c9cd8262a Add support for GCC 3.3
llvm-svn: 6865
2003-06-23 19:16:20 +00:00
Chris Lattner 22ced56476 Fix bug: InstCombine/2003-06-22-ConstantExprCrash.ll
llvm-svn: 6857
2003-06-22 20:48:30 +00:00
Chris Lattner 0cfcaf0cc3 Remove usage of sys/unistd.h
llvm-svn: 6788
2003-06-19 18:42:09 +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
Brian Gaeke 87b4f0765c lib/CWriter/Writer.cpp: Copy AsmWriter's ConstantFP checking code here
into a new function FPCSafeToPrint(), and use it in printConstant()
and printFunction() to decide whether we should output ConstantFPs as
floating-point constants or as references to stack-allocated variables.

lib/VMCore/AsmWriter.cpp: Fix an apparent typo in the code mentioned above.

llvm-svn: 6762
2003-06-17 23:55:35 +00:00
Chris Lattner cc7d6ff656 Fix bug: Linker/2003-06-02-TypeResolveProblem2.ll
llvm-svn: 6699
2003-06-16 12:11:33 +00:00
Chris Lattner d295d99664 Fix (bogus) possibly uninitialized warning
llvm-svn: 6634
2003-06-05 21:01:26 +00:00
Chris Lattner 93c8f14906 * Make assertion message useful
* Kill dead conditional

llvm-svn: 6544
2003-06-02 17:42:47 +00:00
Chris Lattner 7fa6783a01 Fix bug: Linker/2003-06-02-TypeResolveProblem.ll
llvm-svn: 6542
2003-06-02 17:25:46 +00:00
Chris Lattner 39adbfca5b Don't print out unique identifier for opaque types
llvm-svn: 6511
2003-06-01 03:45:51 +00:00
Chris Lattner 8acf346e20 Fix constant folding to ALWAYS work.
llvm-svn: 6355
2003-05-27 19:16:07 +00:00
Chris Lattner 20ec7bc0db Fix problem with last checkin.
llvm-svn: 6328
2003-05-25 16:15:32 +00:00
Chris Lattner 98fa07b1d5 Fix Bug: Linker/2003-04-26-NullPtrLinkProblem.ll
This was a problem with constants having their types resolved to some new type,
but there was already a constant of the new type created.  Before, these types
were never merged together, now they are.

llvm-svn: 6314
2003-05-23 20:03:32 +00:00
Chris Lattner f31a257bfb Remove using declarations and extraneous #includes
llvm-svn: 6303
2003-05-22 21:47:17 +00:00
Chris Lattner ef76b27a19 Fix static constructor ordering problem
llvm-svn: 6302
2003-05-22 21:31:52 +00:00
Chris Lattner 5eef9ffc27 Remove using declarations
llvm-svn: 6300
2003-05-22 21:21:43 +00:00
Chris Lattner 38a9bcde76 Add support for shift constant expressions
llvm-svn: 6260
2003-05-21 17:49:25 +00:00
Misha Brukman e5838c4d72 s/convertable/convertible/g
llvm-svn: 6248
2003-05-20 18:45:36 +00:00
Chris Lattner 192623ec93 Add support for setjmp/longjmp primitives
Patch checked in for Bill Wendling :)

llvm-svn: 6241
2003-05-17 22:26:33 +00:00
Chris Lattner cb4d26fb95 Fix bug: Assembler/2003-05-15-AssemblerProblem.llx
llvm-svn: 6234
2003-05-15 19:37:21 +00:00
Chris Lattner 55ed6561a1 Remove unnecessary casts
llvm-svn: 6201
2003-05-14 17:51:05 +00:00
Chris Lattner 15285abcbe Print opaque types
llvm-svn: 6200
2003-05-14 17:50:47 +00:00
Chris Lattner 69f6af1604 Fix bug
llvm-svn: 6184
2003-05-14 02:47:13 +00:00
Chris Lattner c2ceed1b38 Implement another getelementptr folding opportunity that arises when
linking stuff with appending linkage

llvm-svn: 6180
2003-05-13 21:50:52 +00:00
Chris Lattner 2b383d2ef2 Eliminate using declarations
llvm-svn: 6179
2003-05-13 21:37:02 +00:00
Chris Lattner 4dc3fcaff3 Fix bug: Dominators/2003-05-12-UnreachableCode.ll
llvm-svn: 6158
2003-05-12 22:35:13 +00:00
Chris Lattner 26891626fc Fix Bug: ConstProp/2003-05-12-DivideError.ll
llvm-svn: 6125
2003-05-12 15:26:25 +00:00
Chris Lattner bad4b4a52b llvm.va_start only takes on argument now
Verify that va_start doesn't occur in functions with no varargs

llvm-svn: 6038
2003-05-08 15:55:31 +00:00
Chris Lattner bb346d07b5 Add more support for intrinsic functions and for varargs stuff
llvm-svn: 6035
2003-05-08 03:47:33 +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 ab7d1ab806 Remove using declarations
llvm-svn: 6022
2003-05-08 02:08:14 +00:00
Chris Lattner ef8c8332b1 Fix a bug that occurred when removing the last predecessor INTO an
infinite loop

llvm-svn: 5953
2003-04-25 23:14:19 +00:00
Chris Lattner e17c1c5ad8 ** CORRECTLY, conservatively handle constant expressions
llvm-svn: 5924
2003-04-25 02:52:06 +00:00
Chris Lattner 60430423c9 Fix a nasty bug where the ConstantMerge pass was invalidating the TargetData pass
even though it was immutable.  Immutable passes should never end up in CurrentAnalyses!

llvm-svn: 5906
2003-04-24 20:07:38 +00:00
Chris Lattner 889f620841 Remove unnecesary &*'s
llvm-svn: 5872
2003-04-23 16:37:45 +00:00
Chris Lattner c840aa64e2 Emit information about target
llvm-svn: 5844
2003-04-22 19:07:19 +00:00
Chris Lattner c3f6e00011 Add support for tracking whether a module is 64/32 bit and big/little endian
Also add a moduleID field which can be used for diagnostics

llvm-svn: 5834
2003-04-22 18:02:04 +00:00
Chris Lattner 534952b0d5 Remove arbitrary limitation
llvm-svn: 5818
2003-04-19 00:50:37 +00:00
Chris Lattner acdbe7158d Don't build constantexprs that could be folded
llvm-svn: 5801
2003-04-17 19:24:48 +00:00
Chris Lattner 1f0049c79c Allow constant folding of GEP instructions, even if we don't do a whole lot yet.
Fold ConstExpr casts better
castToPointer shouldn't be forced to return a constantpointer

llvm-svn: 5800
2003-04-17 19:24:18 +00:00
Chris Lattner 46b3d30bf6 Change the interface to constant expressions to allow automatic folding
llvm-svn: 5793
2003-04-16 22:40:51 +00:00
Chris Lattner 3ac483bc89 Add code to verify correctly linkages
llvm-svn: 5788
2003-04-16 20:42:40 +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 deccfafddb Improve the efficiency and cleanup writing a bit
llvm-svn: 5782
2003-04-16 20:20:02 +00:00
Chris Lattner 9ad7757e7c Add helper method
llvm-svn: 5753
2003-03-21 21:41:02 +00:00
Chris Lattner faca62c74a * Change the order that globals and constants are processed in
* Add support for implicit zero initializers

llvm-svn: 5750
2003-03-19 20:57:22 +00:00
Chris Lattner 83e5d39566 Fix ConstantUInt::isAllOnesValue
llvm-svn: 5734
2003-03-10 22:39:02 +00:00
Chris Lattner b825722ac0 Clean up cruft
llvm-svn: 5720
2003-03-06 23:23:32 +00:00
Chris Lattner c33ae67237 Remove #include
Extend getNullValue to work with struct and array types

llvm-svn: 5718
2003-03-06 21:02:18 +00:00
Chris Lattner fbb7739b13 Simplify some of the PHI node interfaces
llvm-svn: 5700
2003-03-05 21:15:12 +00:00
Chris Lattner d4ed5c67e9 Allow ImmutablePass's to require other immutable passes and to be initialized
llvm-svn: 5630
2003-02-26 19:10:57 +00:00
Chris Lattner ade85ecf77 Fix a misunderstanding of the standard associative containers
llvm-svn: 5565
2003-02-14 05:34:36 +00:00
Chris Lattner 3789369330 This speeds up processing LLVM a _lot_, 17% in the case of loading and destroying "vortex"
llvm-svn: 5553
2003-02-13 19:46:22 +00:00
Chris Lattner 34a5bb762d Simplify assertions
llvm-svn: 5455
2003-02-01 03:33:22 +00:00
Joel Stanley 4b27ed9759 Added implementation of alternate CallInst constructors (one ctor is
for no actual parameters, and one ctor is for one actual parameter).

llvm-svn: 5452
2003-02-01 00:39:58 +00:00
Chris Lattner 6a61d257d6 Fix bug Regression/Verifier/2002-11-05-GetelementptrPointers.ll
llvm-svn: 5273
2003-01-14 22:19:44 +00:00
Chris Lattner edac414cac Make type resolution a bit more efficient
llvm-svn: 5265
2003-01-14 19:42:39 +00:00
Chris Lattner 76bff3438b Fix bug: Assembler/2002-12-15-GlobalResolve.ll
llvm-svn: 5039
2002-12-15 16:41:52 +00:00
Chris Lattner d53daec3e1 planes is not spelled with an O
llvm-svn: 5037
2002-12-15 16:20:23 +00:00
Chris Lattner 903a25d225 User defined operators are not supposed to live beyond the lifetime of the
pass.  Detect and flag them.

llvm-svn: 4801
2002-11-21 16:54:22 +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 b251c3d727 Eliminate the concept of a deferred symbol table. The optimization really isn't,
and it causes obscure bugs to show up in passes.

llvm-svn: 4777
2002-11-20 18:07:48 +00:00
Chris Lattner b2e46c0714 Add a new Module::getNamedFunction method
llvm-svn: 4758
2002-11-19 18:41:44 +00:00
Chris Lattner 1f985e0d3c Add a method "getMainFunction()" that efficiently locates 'main' in a module
llvm-svn: 4629
2002-11-08 20:34:02 +00:00
Chris Lattner 4d2997f25d Fix nasty bug in ::isNeg()
llvm-svn: 4470
2002-10-31 17:09:06 +00:00
Chris Lattner f6cfaedcff Can simplify code now with the isCommutative() method.
llvm-svn: 4461
2002-10-31 04:24:23 +00:00
Chris Lattner cab6c33811 New isAssociative/isCommutative inspection methods, graciously contributed by
Casey Carter.

llvm-svn: 4459
2002-10-31 04:14:01 +00:00
Misha Brukman 632df28deb Fix spelling of `propagate'.
llvm-svn: 4423
2002-10-29 23:06:16 +00:00
Chris Lattner 820d971233 - Rename AnalysisUsage::preservesAll to getPreservesAll & preservesCFG to
setPreservesCFG to be less confusing.

llvm-svn: 4255
2002-10-21 20:00:28 +00:00
Chris Lattner 33100eddd2 - Eliminate SymbolTable::ParentSymTab, ST::localLookup, and
Function::ParentSymTab.  These aren't needed at all.

llvm-svn: 4186
2002-10-15 21:26:29 +00:00
Chris Lattner 163b890dfb - Dramatically simplify the Constant::mutateReferences implementation,
allowing it to be called on all constant types (structures/arrays)

llvm-svn: 4160
2002-10-14 03:30:23 +00:00
Chris Lattner 251a7e6231 Minor cleanups, no changes to functionality
llvm-svn: 4157
2002-10-14 00:48:57 +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 69ce8674b5 - Rename Instruction::First*Op to *OpsBegin, and Num*Ops to *OpsEnd to
reflect the fact that it's a range being defined.

llvm-svn: 4147
2002-10-13 19:39:16 +00:00
Chris Lattner 2c6abeac43 - Make Value::replaceAllUsesWith work with constants correctly. This fixes
bug FuncResolve/2002-08-19-ResolveGlobalVars.ll and gzip looks better.

llvm-svn: 4103
2002-10-09 23:12:59 +00:00
Chris Lattner b1dd9bbc68 - Add new Constant::replaceUsesOfWithOnConstant which has an end result
similar to User::replaceUsesOfWith but actually does the right thing for
    constants.

llvm-svn: 4102
2002-10-09 23:12:25 +00:00
Chris Lattner 6023f638a1 - Remove Value::use_remove
llvm-svn: 4090
2002-10-09 00:25:05 +00:00
Chris Lattner 3ea0c188b7 Fix NASTY N^2 behavior that was causing the gzip benchmark to take forever to
assemble.  Now we scan the use-list from the back when removing users instead
of from the front.

llvm-svn: 4086
2002-10-08 23:46:55 +00:00
Chris Lattner 0d779712e7 Fix bug: Assembler/2002-10-08-LargeArrayPerformance.ll by using
std::vector::reserve when possible

llvm-svn: 4085
2002-10-08 23:33:52 +00:00
Chris Lattner caf5b504a3 Changes to support PHINode::removeIncoming changes
llvm-svn: 4080
2002-10-08 21:36:34 +00:00
Chris Lattner fae98a2d28 - Change PHINode::removeIncomingValue to delete the phi node if the last
incoming value is removed!

llvm-svn: 4078
2002-10-08 21:34:58 +00:00
Chris Lattner acb038e5c7 Expose new "recalculate" method from dominatorset
llvm-svn: 4074
2002-10-08 19:12:08 +00:00
Chris Lattner d131986b0f Global variables are now external if they don't have initializers, not
"uninitialized"

llvm-svn: 4052
2002-10-06 22:48:09 +00:00
Chris Lattner 78bc0fa29f Check that we don't have external varaibles with internal linkage
llvm-svn: 4051
2002-10-06 22:47:32 +00:00
Chris Lattner 4cd9df88a7 PHI nodes are not allowed to exist with zero incoming values, check that
there aren't any like this.

llvm-svn: 4044
2002-10-06 21:00:31 +00:00
Chris Lattner 2cedf1ce82 Fix a nasty problem with dominance calculation for unreachable blocks.
If we had a CFG that look like Entry -> B, Unreachable -> B, then we would
not correctly determine that Entry dominated B, because Entry did not
apparently dominate "unreachable".  This patch fixes this by making the entry
node dominate all blocks, including unreachable ones.

llvm-svn: 4037
2002-10-04 14:45:48 +00:00
Chris Lattner 58185f2df9 - Print the predecessors of a basic block instead of the number of uses of
the block in the AsmWriter

llvm-svn: 4019
2002-10-02 19:38:55 +00:00
Chris Lattner 52db271374 Make sure to use the TimerGroup that we created!
llvm-svn: 3995
2002-10-01 20:12:06 +00:00
Chris Lattner d5fc9023a8 Make sure not to count the PassManager wrapers
llvm-svn: 3994
2002-10-01 20:08:11 +00:00
Chris Lattner af751b8e72 Factor timer code out of PassManager implementation, into a generic interface
exposed by Support/Timer.h.

llvm-svn: 3993
2002-10-01 19:54:07 +00:00
Chris Lattner 6b6f540959 Fix a problem that was caused by stale analyses being in CurrentAnalyses
llvm-svn: 3981
2002-09-29 22:50:22 +00:00
Chris Lattner ae3ac7e1c9 Improve printing of dominator sets
llvm-svn: 3976
2002-09-29 21:42:42 +00:00
Chris Lattner 98ae0f1897 - Add methods to ImmediateDominators & DominatorTree to allow updates
- Make DominatorTree::Node not inherit from std::vector

llvm-svn: 3939
2002-09-26 16:14:41 +00:00
Chris Lattner 4dfede8070 - Add new ctor to BasicBlock to allow insertion before any BB, not just at
the end of the function.

llvm-svn: 3934
2002-09-26 05:03:22 +00:00
Chris Lattner 480b37d479 * Fix ugly bug in previous checkin where I reused the name 'i' one too many times
* Print out immutable passes in the -debug-pass=Structure report.

llvm-svn: 3923
2002-09-25 22:26:52 +00:00
Chris Lattner ee0788d42d Add support for ImmutablePasses, which are not run, and cannot be
invalidated.

llvm-svn: 3921
2002-09-25 21:59:11 +00:00
Chris Lattner a450397247 Make sure that we abort if an error happens as early as neccesary. Before
it was possible for the passmanager to continue running passes after the
verifier even if the module was not well formed.

llvm-svn: 3820
2002-09-19 16:12:19 +00:00
Vikram S. Adve 0267d0c62e Remove specification of argument default from cpp file.
llvm-svn: 3772
2002-09-17 01:17:57 +00:00
Chris Lattner fc91ee91e8 Change the MallocInst & AllocaInst ctors to take the allocated type, not the
pointer type returned.

llvm-svn: 3711
2002-09-13 22:28:50 +00:00
Chris Lattner 426752738b include/llvm/DataTypes.h includes this #define so we don't need it anymore
llvm-svn: 3708
2002-09-13 22:24:57 +00:00
Chris Lattner 548002f3d5 Another portability fix provided via Casey Carter:
ISSUE: getTimeRecord in lib/VMCore/Pass.cpp uses timeval and
gettimeofday() without including sys/time.h.

ACTION: Include sys/time.h.
llvm-svn: 3699
2002-09-13 14:47:12 +00:00
Chris Lattner bae3c67050 Add a new BasicBlockPass::doInitialization/Finalization(Function &) pair of
methods that may be useful for BasicBlockPasses.

llvm-svn: 3688
2002-09-12 17:06:40 +00:00
Chris Lattner 236c129844 Give better assertion message for error conditions
llvm-svn: 3680
2002-09-11 01:21:04 +00:00
Chris Lattner 8678d4e5e8 Allocation insts always have one operand
llvm-svn: 3677
2002-09-11 00:22:39 +00:00
Chris Lattner 5f45ca443e Add convenience ctor to BranchInst
llvm-svn: 3676
2002-09-10 23:54:58 +00:00
Chris Lattner 66d3a46afc Tighten up assertion checking for binary operators, not allowing invalid
instructions to _even be created_.

llvm-svn: 3661
2002-09-10 19:57:53 +00:00
Chris Lattner 7d734801d1 Add support for printing constpointerrefs more nicely
llvm-svn: 3654
2002-09-10 15:53:49 +00:00
Chris Lattner 3c787446c7 Add capability to insert an instruction into a basic block immediately after
it is created, as part of the ctor call.

Eliminate the GenericBinaryInst class

llvm-svn: 3653
2002-09-10 15:45:53 +00:00
Chris Lattner 80a2568f9b Fix typeo in assertion message
llvm-svn: 3641
2002-09-10 04:52:59 +00:00
Chris Lattner 1f41925300 Do not allow adds of pointer types, or many other operations on pointer types.
llvm-svn: 3638
2002-09-09 20:26:04 +00:00
Chris Lattner ed6504b1eb Enable "garbage detection" of LLVM objects. Check for leaks after running
every pass.

llvm-svn: 3621
2002-09-08 19:00:07 +00:00
Chris Lattner 184b298edc Enable "garbage detection" of LLVM objects. Now users should be obnoxious
warnings.  If they accidentally leak LLVM Value's.

llvm-svn: 3620
2002-09-08 18:59:35 +00:00
Chris Lattner 9ed7aef651 Move code out of header files into .cpp files to make future changes easier
llvm-svn: 3605
2002-09-06 21:33:15 +00:00
Chris Lattner 6213ae0028 * Remove extraneous #includes
* Add extra argument to Function ctor to allow automatic insertion into module

llvm-svn: 3600
2002-09-06 20:46:32 +00:00
Chris Lattner 07507a4ccb Implement setcc for booleans. Fixes bug:
test/Regression/Transforms/ConstProp/2002-09-03-SetCC-Bools.ll

llvm-svn: 3576
2002-09-03 20:09:49 +00:00
Chris Lattner b0b412e66e - Renamed Type::isIntegral() to Type::isInteger()
- Added new method Type::isIntegral() that is the same as isInteger, but
    also accepts bool.

llvm-svn: 3574
2002-09-03 01:08:28 +00:00
Chris Lattner b33e4630ff - Exposed SetCondInst::getInverseCondition & ::getSwappedCondition better
llvm-svn: 3562
2002-09-01 19:46:40 +00:00
Chris Lattner 216edd4ea0 - Use Pass::AnalysisImpls instead of findAnalysisGroupMemeber
- -debug-pass=Details now prints implementation class names for analysis
    groups
  - Rework how AnalysisGroups are resolved to be simpler and better.

llvm-svn: 3535
2002-08-30 20:25:01 +00:00
Chris Lattner b3708e267e - PassManager prints analysis sets nicer
- Use Pass::AnalysisImpls instead of findAnalysisGroupMemeber

llvm-svn: 3534
2002-08-30 20:23:45 +00:00
Chris Lattner 81fabb05e5 Add support for turning an array of characters into a string.
llvm-svn: 3509
2002-08-26 17:53:56 +00:00
Chris Lattner dfb3a2cd07 Eliminated the MemAccessInst class, folding contents into GEP class.
llvm-svn: 3487
2002-08-22 23:37:20 +00:00
Chris Lattner cd709cbf53 Load & StoreInst no longer derive from MemAccessInst, so we don't have
to handle indexing anymore

llvm-svn: 3485
2002-08-22 22:49:05 +00:00
Chris Lattner 4011c62a8d Load and Store now no longer derive from MemAccessInst. Indexing a load or
store is not possible anymore.

llvm-svn: 3482
2002-08-22 22:47:47 +00:00
Chris Lattner e46f5528a3 Fix bug: test/Regression/Assembler/2002-08-22-DominanceProblem.ll
llvm-svn: 3474
2002-08-22 20:39:29 +00:00
Chris Lattner 675e7a9e3d - Fighting with linking problem due to removing the ::ID elements. Now the
implementation .cpp files for analyses are not being included into gccas
    and friends because it is linking to the .a file and there is no explicit
    symbol reference to bring in the .o file.  The new IncludeFile hack is the
    result.

llvm-svn: 3436
2002-08-21 23:51:51 +00:00
Chris Lattner 10d7b85d5e Remove debugging code accidentally checked in.
llvm-svn: 3430
2002-08-21 22:49:31 +00:00
Chris Lattner 6e041bdb12 - Implement the new AnalysisGroup feature, neccesary for Value#ing and pointer analysis
llvm-svn: 3426
2002-08-21 22:17:09 +00:00
Chris Lattner 1230f23da8 - Do not expose ::ID from any of the analyses anymore.
llvm-svn: 3415
2002-08-21 17:09:15 +00:00
Chris Lattner 4b16963ca2 - Eliminate the need for analyses to expose an ::ID member.
llvm-svn: 3414
2002-08-21 17:08:37 +00:00
Chris Lattner e821d78432 Sort -time-passes report first by user+system, then by Wall clock time.
llvm-svn: 3407
2002-08-20 18:47:53 +00:00
Chris Lattner dcf53552ce Add new SetCondInst::getInverseCondition() method.
llvm-svn: 3405
2002-08-20 18:17:12 +00:00
Chris Lattner ca5afe74ab Add missing space
llvm-svn: 3387
2002-08-19 20:42:12 +00:00
Chris Lattner 5ec216b5fe Fix the -time-passes option to not print NaN when there is zero execution time
llvm-svn: 3382
2002-08-19 15:43:33 +00:00
Chris Lattner 0c6e0b9042 - ConstantPointerRefs are now automatically removed from the module table
when they are destroyed, which makes Constant::destroyConstant an actually
    useful external interface.  Expose these methods publicly.
  - Implement destroyConstant on ConstPointerNull so that destroyConstant can
    be used on any derived type constant safely.

llvm-svn: 3379
2002-08-18 00:40:04 +00:00
Chris Lattner e0f6af9bb2 Add Module::dump() method
llvm-svn: 3375
2002-08-17 23:32:47 +00:00
Chris Lattner cfe8f53a4b Changed parser to always use parenthesis on ConstExprs to be consistent
llvm-svn: 3363
2002-08-16 21:17:11 +00:00
Chris Lattner 83b396b830 Write constexpr casts using the cast X to Y notation, not using the implicit
type result

llvm-svn: 3354
2002-08-15 19:37:43 +00:00
Chris Lattner 7f8e677715 Simplify the code
llvm-svn: 3348
2002-08-15 16:15:36 +00:00
Vikram S. Adve 243f187791 Added more helper functions for binary instructions emulating
unary Neg and Not: isNeg, isNot, getNegArgument, and getNotArgument.

llvm-svn: 3341
2002-08-15 14:15:48 +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 8fd96fcdaf Remove support for unary operators.
llvm-svn: 3326
2002-08-14 18:19:46 +00:00
Chris Lattner b193ff8f70 Remove support for NOT instruction
llvm-svn: 3323
2002-08-14 18:18:02 +00:00
Chris Lattner 3733c7374e Implement two constructor functions for "unary" instructions
llvm-svn: 3317
2002-08-14 17:52:27 +00:00
Chris Lattner 4ec79c3e6f Remove support forNOT instruction
llvm-svn: 3316
2002-08-14 17:52:06 +00:00
Chris Lattner 31408f7a80 Return null on failure, instead of aborting.
llvm-svn: 3308
2002-08-14 17:12:13 +00:00
Chris Lattner b1585a90d9 - Rename ConstantGenericIntegral -> ConstantIntegral
- Add new methods to ConstantIntegral: getMaxValue, getMinValue,
   getAllOnesValue

llvm-svn: 3299
2002-08-13 17:50:20 +00:00
Chris Lattner 41e99a07f1 - Added a new superclass of ConstantBool/ConstantInt: ConstantGenericIntegral
- Moved InstCombine: isMaxValue, isMinValue, isAllOnesValue to Constants.h

llvm-svn: 3293
2002-08-12 21:21:21 +00:00
Chris Lattner 68fe621771 Remove some extraneous #includes
llvm-svn: 3289
2002-08-12 20:24:46 +00:00
Chris Lattner 7d9a14d5d9 Changes to make GlobalValueRefMap not derive from std::map.
llvm-svn: 3288
2002-08-12 20:23:29 +00:00
Chris Lattner a09b01b2c1 UnaryOperator::create should take a name just like BinaryOperator::create
llvm-svn: 3279
2002-08-09 21:39:38 +00:00
Chris Lattner 40eb9dafed - Cleaned up the interface to AnalysisUsage to take analysis class names
instead of ::ID's.
 - Pass::getAnalysis<> now no longer takes an optional argument

llvm-svn: 3264
2002-08-08 19:01:28 +00:00
Chris Lattner 8e72d6f6e9 Implement dominator checking in the verifier, so that we check that all
defintiions dominate their uses

llvm-svn: 3214
2002-08-02 17:37:08 +00:00
Chris Lattner 6d149e0215 Fix bug: test/Regression/Other/2002-08-02-DomSetProblem.ll
llvm-svn: 3213
2002-08-02 16:51:27 +00:00
Chris Lattner d43023a85d Split dominance calculation and post dominance calculation stuff
Dominance calculation goes to VMCore library to be used by Verifier.

llvm-svn: 3210
2002-08-02 16:43:03 +00:00
Chris Lattner 6a33d6f0bb Enhance -time-passes to be more detailed and useful
llvm-svn: 3199
2002-08-01 19:33:09 +00:00
Chris Lattner e5fd3867d3 Fix bug: test/Regression/Assembler/2002-07-31-SlashInString.llx
llvm-svn: 3195
2002-07-31 23:56:44 +00:00
Chris Lattner be4826e627 *** empty log message ***
llvm-svn: 3185
2002-07-31 19:32:01 +00:00
Chris Lattner 395c27ae39 We need iostream in PassManagerT.h to use cerr
llvm-svn: 3183
2002-07-31 18:04:17 +00:00
Chris Lattner 1e4867fed2 Implement new -debug-pass=Arguments option that causes PassManager to
print out the command line options for the optimizations it is running.

llvm-svn: 3165
2002-07-30 19:51:02 +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 31aa7e72f2 Declare that these passes only depend on the CFG of the function
llvm-svn: 3157
2002-07-30 16:27:52 +00:00
Chris Lattner 198cf42178 * Move some code from Pass.cpp to PassManagerT.h
* Implement stuff so that code can declare that they only depend on the CFG of
  a function, not on anything else.  This speeds up GCCAS a lot.

llvm-svn: 3155
2002-07-30 16:27:02 +00:00
Chris Lattner e87f65ec1b Implement constant propogation of logical (and, or, xor) expressions.
Fixes testcase: test/Regression/Transforms/ConstProp/logicaltest.ll

llvm-svn: 3154
2002-07-30 16:24:28 +00:00
Chris Lattner 73503173ed * Eliminate the Provided set. All Passes now finally just automatically
provide themselves.

llvm-svn: 3125
2002-07-29 21:03:38 +00:00
Chris Lattner f356bcc824 * Eliminate the Provided set. All Passes now finally just automatically
provide themselves.

llvm-svn: 3124
2002-07-29 21:03:33 +00:00
Chris Lattner 071577d66d * PassInfo is allowed to be missing now (ie, not all passes need be registered)
* getPassName uses PassInfo if it's available
* PassInfo is now cached in the pass so that it is only looked up once, maximum

llvm-svn: 3123
2002-07-29 21:02:31 +00:00
Chris Lattner 2675007573 * Standardize how analysis results/passes as printed with the print() virtual
methods
* Eliminate AnalysisID:  Now it is just a typedef for const PassInfo*
* Simplify how AnalysisID's are initialized
* Eliminate Analysis/Writer.cpp/.h: incorporate printing functionality into
  the analyses themselves.

llvm-svn: 3116
2002-07-27 01:12:17 +00:00
Chris Lattner c8b7092e54 * Add support for different "PassType's"
* Add new RegisterOpt/RegisterAnalysis templates for registering passes that
  are to show up in opt or analyze
* Register Analyses now
* Change optimizations to use RegisterOpt instead of RegisterPass
* Add support for different "PassType's"
* Add new RegisterOpt/RegisterAnalysis templates for registering passes that
  are to show up in opt or analyze
* Register Analyses now
* Change optimizations to use RegisterOpt instead of RegisterPass
* Remove getPassName implementations from various subclasses

llvm-svn: 3113
2002-07-26 21:12:46 +00:00
Chris Lattner a2c0985980 * Add support for different "PassType's"
* Add new RegisterOpt/RegisterAnalysis templates for registering passes that
  are to show up in opt or analyze
* Register Analyses now
* Change optimizations to use RegisterOpt instead of RegisterPass
* Add support for different "PassType's"
* Add new RegisterOpt/RegisterAnalysis templates for registering passes that
  are to show up in opt or analyze
* Register Analyses now
* Change optimizations to use RegisterOpt instead of RegisterPass
* Remove getPassName implementations from various subclasses

llvm-svn: 3112
2002-07-26 21:12:44 +00:00
Chris Lattner 64eea749de *** empty log message ***
llvm-svn: 3105
2002-07-26 18:40:14 +00:00
Chris Lattner d9a36a63bb Fix typeo that caused bug:
test/Regression/Assembler/2002-07-25-ReturnPtrFunction.llx

llvm-svn: 3103
2002-07-25 20:58:51 +00:00
Chris Lattner 5fa3a4429f Fix bug: test/Regression/Assembler/2002-07-25-QuoteInString.llx
llvm-svn: 3099
2002-07-25 20:32:37 +00:00
Chris Lattner c678c17211 Avoid warning: control reaches end of non-void function
llvm-svn: 3081
2002-07-25 15:39:05 +00:00
Chris Lattner a82ee2df46 *** empty log message ***
llvm-svn: 3065
2002-07-24 22:08:53 +00:00
Chris Lattner 00fb26cf4d Register Verifier pass
llvm-svn: 3019
2002-07-23 18:08:17 +00:00
Chris Lattner 37d3c95b0f Add support for pass registration
llvm-svn: 3018
2002-07-23 18:08:00 +00:00
Chris Lattner 7f8845a523 Register Writer passes
llvm-svn: 3017
2002-07-23 18:07:49 +00:00
Chris Lattner a896b08b5c *** empty log message ***
llvm-svn: 3002
2002-07-23 17:52:38 +00:00
Chris Lattner f5cad15a67 *** empty log message ***
llvm-svn: 2985
2002-07-22 02:10:13 +00:00
Chris Lattner d9f4ac6680 * ConstExpr::getelementptr now takes a vector of Constants not Values
* Assert things instead of printing an error and returning null.

llvm-svn: 2949
2002-07-18 00:14:50 +00:00
Chris Lattner 7af3ee9840 Add a hack to check for a subset of true dominance properties
llvm-svn: 2947
2002-07-18 00:13:42 +00:00
Anand Shukla 991873fc5c added std:: to cerr and endl
llvm-svn: 2914
2002-07-16 00:02:17 +00:00
Vikram S. Adve 4c48533b57 Ensure ConstExpr constants are unique using a
map of <opcode, operands> to ConstExpr.

llvm-svn: 2909
2002-07-15 18:19:33 +00:00
Vikram S. Adve b952b5485f Added support to write out ConstantExpr nodes.
Also, avoid asserting out when writing out an invalid tree
since the assembly writer is used when debugging.

llvm-svn: 2902
2002-07-14 23:14:45 +00:00
Vikram S. Adve 4e537b27d8 Added subclass ConstantExpr to represent expressions consructed from
constants using operators such as cast, getelementptr, add, shl, etc.
Note that a ConstantExpr can be of any type, so classof() in most
other subclasses (that discriminate by type) have to check that it
is also not a ConstantExpr.

llvm-svn: 2901
2002-07-14 23:13:17 +00:00