Commit Graph

3673 Commits

Author SHA1 Message Date
Chris Lattner 4329e078ac add a 'LLVMConstIntOfArbitraryPrecision' api to the C api,
patch by Greg Pfeil!

llvm-svn: 119989
2010-11-23 02:47:22 +00:00
Benjamin Kramer ddd1b7b801 Simplify code. No change in functionality.
llvm-svn: 119908
2010-11-20 18:43:35 +00:00
Owen Anderson 2e831897d6 Fix an order-of-deallocation issue where the AttrListImpl could be deallocated before the global
LLVMContext, causing memory errors.  Patch by Peter Collingbourne.

llvm-svn: 119721
2010-11-18 18:59:13 +00:00
Duncan Sands ec7a6ecb92 Now that hasConstantValue has been made simpler, it may return the
phi node itself if it occurs in an unreachable basic block.  Protect
against this.  Hopefully this will fix some more buildbots.

llvm-svn: 119493
2010-11-17 10:23:23 +00:00
Chris Lattner b0e36085c4 now that AsmPrinter::EmitInlineAsm is factored right, we can eliminate the
cookie argument to the SourceMgr diagnostic stuff.  This cleanly separates
LLVMContext's inlineasm handler from the sourcemgr error handling 
definition, increasing type safety and cleaning things up.

llvm-svn: 119486
2010-11-17 08:13:01 +00:00
Chris Lattner baf00151ba fix PR8613 - Copy constructor of SwitchInst does not call SwitchInst::init
llvm-svn: 119463
2010-11-17 05:41:46 +00:00
Duncan Sands 7412f6e53d Fix a layering violation: hasConstantValue, which is part of the PHINode
class, uses DominatorTree which is an analysis.  This change moves all of
the tricky hasConstantValue logic to SimplifyInstruction, and replaces it
with a very simple literal implementation.  I already taught users of
hasConstantValue that need tricky stuff to use SimplifyInstruction instead.
I didn't update InlineFunction because the IR looks like it might be in a
funky state at the point it calls hasConstantValue, which makes calling
SimplifyInstruction dangerous since it can in theory do a lot of tricky
reasoning.  This may be a pessimization, for example in the case where
all phi node operands are either undef or a fixed constant.

llvm-svn: 119459
2010-11-17 04:30:22 +00:00
Duncan Sands b99f39b9f6 If dom tree information is available, make it possible to pass
it to get better phi node simplification.

llvm-svn: 119055
2010-11-14 18:36:10 +00:00
Nick Lewycky add50b0a5d Doxygenify
llvm-svn: 118846
2010-11-11 21:51:44 +00:00
Dan Gohman a826a88755 Factor out Instruction::isSafeToSpeculativelyExecute's code for
testing for dereferenceable pointers into a helper function,
isDereferenceablePointer.  Teach it how to reason about GEPs
with simple non-zero indices.

Also eliminate ArgumentPromtion's IsAlwaysValidPointer,
which didn't check for weak externals or out of range gep
indices.

llvm-svn: 118840
2010-11-11 21:23:25 +00:00
Dan Gohman f51d06bbee Include ImmutablePass passes in -debug-pass=Arguments.
llvm-svn: 118790
2010-11-11 16:32:17 +00:00
Owen Anderson 91bfeb10b5 Last try to get this reference counting right, I swear.
llvm-svn: 118589
2010-11-09 17:47:10 +00:00
Owen Anderson 2d335436f3 Really fix the leak in the attributes list. Thanks to Benjamin Kramer for pointing out how I was being stupid.
llvm-svn: 118588
2010-11-09 17:46:38 +00:00
Owen Anderson 25e84078e3 Fix leak in my recent fix for PR8442.
llvm-svn: 118490
2010-11-09 05:17:47 +00:00
Owen Anderson 9b14a25126 Fix PR8441, a race condition in the static attributes list. While the reference counting was itself threadsafe,
the implicit removal of each object from the global list was not.  Make this operation atomic.

llvm-svn: 118461
2010-11-09 00:27:03 +00:00
Dale Johannesen e81d45d2cc Apply patch for use-after-free in InlineAsm constant handling,
PR 8522 / 8616046.  Test reduction, analysis and patch by Tim Deegan!
(However, review by someone who understands the classes here better
is welcome.  John Krum will return!)

llvm-svn: 118030
2010-11-02 17:59:27 +00:00
John Thompson e8360b7182 Inline asm multiple alternative constraints development phase 2 - improved basic logic, added initial platform support.
llvm-svn: 117667
2010-10-29 17:29:13 +00:00
Charles Davis 22fe18625d Add a new 'hotpatch' attribute. This attribute will insert a two-byte no-op
instruction at the beginning of each function that has the attribute, allowing
the function to be easily hooked and/or patched.

llvm-svn: 117264
2010-10-25 15:37:09 +00:00
Benjamin Kramer 9192e7ab12 Make some symbols static, move classes into anonymous namespaces.
llvm-svn: 117111
2010-10-22 17:35:07 +00:00
Owen Anderson 80b8817c9d Attempt to fix valgrind complaining about (benign) leaks in pass registration by having PassRegistry
delete PassInfo objects that were created using new-style initialization.

llvm-svn: 116959
2010-10-20 22:22:30 +00:00
Tobias Grosser 23c8341c3d Add RegionPass support.
A RegionPass is executed like a LoopPass but on the regions detected by the
RegionInfo pass instead of the loops detected by the LoopInfo pass.

llvm-svn: 116905
2010-10-20 01:54:44 +00:00
Owen Anderson 6c18d1aac0 Get rid of static constructors for pass registration. Instead, every pass exposes an initializeMyPassFunction(), which
must be called in the pass's constructor.  This function uses static dependency declarations to recursively initialize
the pass's dependencies.

Clients that only create passes through the createFooPass() APIs will require no changes.  Clients that want to use the
CommandLine options for passes will need to manually call the appropriate initialization functions in PassInitialization.h
before parsing commandline arguments.

I have tested this with all standard configurations of clang and llvm-gcc on Darwin.  It is possible that there are problems
with the static dependencies that will only be visible with non-standard options.  If you encounter any crash in pass
registration/creation, please send the testcase to me directly.

llvm-svn: 116820
2010-10-19 17:21:58 +00:00
Benjamin Kramer 74beb4278a std::string cleanup.
llvm-svn: 116686
2010-10-17 07:39:34 +00:00
Oscar Fuentes 889c1e7d80 Build with RTTI and exceptions disabled. Only in GCC for now.
llvm-svn: 116682
2010-10-17 02:26:16 +00:00
Owen Anderson 8ac477ffb5 Begin adding static dependence information to passes, which will allow us to
perform initialization without static constructors AND without explicit initialization
by the client.  For the moment, passes are required to initialize both their
(potential) dependencies and any passes they preserve.  I hope to be able to relax
the latter requirement in the future.

llvm-svn: 116334
2010-10-12 19:48:12 +00:00
Dan Gohman cb85497ddf Delete a redundant check.
llvm-svn: 116280
2010-10-12 00:19:24 +00:00
Dan Gohman 060d5ba248 More SmallVectorImpls.
llvm-svn: 116279
2010-10-12 00:15:27 +00:00
Dan Gohman c450d2caa5 Shrink a SmallVector with a known maximum size.
llvm-svn: 116278
2010-10-12 00:13:43 +00:00
Dan Gohman c8da21b04c Constify.
llvm-svn: 116277
2010-10-12 00:12:29 +00:00
Dan Gohman 7224bcef8f Use SmallVectorImpl in a bunch of places.
llvm-svn: 116276
2010-10-12 00:11:18 +00:00
Dan Gohman 844dd0ad00 Fix the pass manager's search order for immutable passes, and make it
stop searching when it has found a match.

llvm-svn: 116262
2010-10-11 23:19:01 +00:00
Owen Anderson df7a4f2515 Now with fewer extraneous semicolons!
llvm-svn: 115996
2010-10-07 22:25:06 +00:00
Owen Anderson 44621e4306 Add initialization routines for VMCore.
llvm-svn: 115963
2010-10-07 19:51:21 +00:00
Owen Anderson 4698c5d7f7 Next step on the getting-rid-of-static-ctors train: begin adding per-library
initialization functions that initialize the set of passes implemented in
that library.  Add C bindings for these functions as well.

llvm-svn: 115927
2010-10-07 17:55:47 +00:00
Owen Anderson c25b39702f Another step towards getting rid of static ctors for pass registration: have INITIALIZE_PASS AND INITIALIZE_AG_PASS
expand to an initializeMyPass() function (in additional to the extant static ctors).  Eventually, these will be called
from a big InitializeAllPasses() function, and the PassInfo's they create (which would be leaked if this code were used
at the moment) will be handed off to a PassRegistry for ownership.

llvm-svn: 115703
2010-10-05 22:58:16 +00:00
Bill Wendling 402e54822b The pshufw instruction came about in MMX2 when SSE was introduced. Don't place
it in with the SSSE3 instructions.

Steward! Could you place this chair by the aft sun deck? I'm trying to get away
from the Astors. They are such boors!

llvm-svn: 115552
2010-10-04 20:24:01 +00:00
Bill Wendling 3c70417883 Cleanup. Get rid of extraneous variable.
llvm-svn: 115453
2010-10-03 00:46:06 +00:00
Dale Johannesen dec83c3fd9 Attempt to outwit overly smart compiler.
llvm-svn: 115251
2010-10-01 00:21:24 +00:00
Dale Johannesen dd224d2333 Massive rewrite of MMX:
The x86_mmx type is used for MMX intrinsics, parameters and
return values where these use MMX registers, and is also
supported in load, store, and bitcast.

Only the above operations generate MMX instructions, and optimizations
do not operate on or produce MMX intrinsics. 

MMX-sized vectors <2 x i32> etc. are lowered to XMM or split into
smaller pieces.  Optimizations may occur on these forms and the
result casted back to x86_mmx, provided the result feeds into a
previous existing x86_mmx operation.

The point of all this is prevent optimizations from introducing
MMX operations, which is unsafe due to the EMMS problem.

llvm-svn: 115243
2010-09-30 23:57:10 +00:00
Talin 2e59f142cc Allow llvm.gcroot to work with non-pointer allocas.
llvm-svn: 115198
2010-09-30 20:23:47 +00:00
Dan Gohman 25a7bc9081 When an MDNode changes to become identical to another MDNode,
delete the MDNode that changed, rather than the other MDNode.
This is less work, because it doesn't require the changed node
to be re-inserted into the uniquing map and it doesn't require
the is-function-local flag to be recomputed. Also, it avoids
trouble when the existing node is part of a complicated
data structure.

llvm-svn: 114996
2010-09-28 22:07:19 +00:00
Dan Gohman 1a450c61f3 Scope a varible inside an if statement, to make it clear that
it's not used afterwards.

llvm-svn: 114986
2010-09-28 21:02:55 +00:00
Dan Gohman 9a1b8598ef Make this code 65-bit clean.
llvm-svn: 114828
2010-09-27 15:15:44 +00:00
Che-Liang Chiou 299479020a Add ret instruction to PTX backend
llvm-svn: 114788
2010-09-25 07:46:17 +00:00
Owen Anderson c2fcd6106c Allow the PassRegistry mutex to be lazily initialized, and clean up the global namespace at the same time.
llvm-svn: 114131
2010-09-16 23:44:50 +00:00
Owen Anderson d878827107 Do not expose the locking for the PassRegistry in the header. Be careful to
synchronize any method that might lazily initialize the pImpl.

llvm-svn: 114130
2010-09-16 23:32:35 +00:00
Owen Anderson 8dc0b04106 Fix a threaded LLVM bug due the need for operator= on reference counted AttrListImpl's. It might
be possible to implement this very carefully to allow a lock-free implementation while still
avoiding illegal interleavings, but I haven't been able to figure one out.

llvm-svn: 114046
2010-09-16 00:27:35 +00:00
Owen Anderson d4754971b4 Since PassRegistry is currently a shared global object, it needs locking. While it might intuitively seem
that all the setup of this class currently happens at static initialization time, this misses the fact
that some later events can cause mutation of the PassRegistrationListeners list, and thus cause race issues.

llvm-svn: 114036
2010-09-15 23:03:33 +00:00
Dale Johannesen 51fd822634 Add x86MMX a few more places.
llvm-svn: 113914
2010-09-15 00:52:23 +00:00
Dan Gohman 41f14cf3e9 Remove the experimental AliasAnalysis::getDependency interface, which
isn't a good level of abstraction for memdep. Instead, generalize
AliasAnalysis::alias and related interfaces with a new Location
class for describing a memory location. For now, this is the same
Pointer and Size as before, plus an additional field for a TBAA tag.

Also, introduce a fixed MD_tbaa metadata tag kind.

llvm-svn: 113858
2010-09-14 21:25:10 +00:00