oota-llvm.git
20 years agoOnly spit out warning for functions that take pointers, not for sin and the like
Chris Lattner [Mon, 16 Feb 2004 22:57:19 +0000 (22:57 +0000)]
Only spit out warning for functions that take pointers, not for sin and the like
Add more special case handling for stdio functions.  I feel dirty, how about you?

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11506 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoMove the folding of gep null, 0, 0, 0 to a place where it can be shared and
Chris Lattner [Mon, 16 Feb 2004 20:46:13 +0000 (20:46 +0000)]
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

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11505 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoPreSelection isn't optional and so there's no way to turn it off.
Misha Brukman [Mon, 16 Feb 2004 19:45:26 +0000 (19:45 +0000)]
PreSelection isn't optional and so there's no way to turn it off.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11504 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agomemset and bcopy and now unified by the llvm.memset intrinsic
Chris Lattner [Mon, 16 Feb 2004 18:37:40 +0000 (18:37 +0000)]
memset and bcopy and now unified by the llvm.memset intrinsic

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11503 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd some ADD instructions that take memory operands for Alkis
Chris Lattner [Mon, 16 Feb 2004 18:19:31 +0000 (18:19 +0000)]
Add some ADD instructions that take memory operands for Alkis

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11502 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdjusted the default pathname to the SPEC95 benchmarks. The new directory
John Criswell [Mon, 16 Feb 2004 14:55:18 +0000 (14:55 +0000)]
Adjusted the default pathname to the SPEC95 benchmarks.  The new directory
has source code corrections and some input files pre-filtered for use
with the LLVM test suite.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11498 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd LeakDetection to MachineInstr.
Alkis Evlogimenos [Mon, 16 Feb 2004 07:17:43 +0000 (07:17 +0000)]
Add LeakDetection to MachineInstr.

Move out of line member functions of MachineBasicBlock to
MachineBasicBlock.cpp.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11497 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoImplement test/Regression/Transforms/SimplifyCFG/UncondBranchToReturn.ll,
Chris Lattner [Mon, 16 Feb 2004 06:35:48 +0000 (06:35 +0000)]
Implement test/Regression/Transforms/SimplifyCFG/UncondBranchToReturn.ll,
see the testcase for the reasoning.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11496 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoNew testcase, details in the comments
Chris Lattner [Mon, 16 Feb 2004 06:35:19 +0000 (06:35 +0000)]
New testcase, details in the comments

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11495 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoFold PHI nodes of constants which are only used by a single cast. This implements
Chris Lattner [Mon, 16 Feb 2004 05:07:08 +0000 (05:07 +0000)]
Fold PHI nodes of constants which are only used by a single cast.  This implements
phi.ll:test4

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11494 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoA new testcase for a situation that occurs in 181.mcf
Chris Lattner [Mon, 16 Feb 2004 05:06:36 +0000 (05:06 +0000)]
A new testcase for a situation that occurs in 181.mcf

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11493 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoTeach LLVM to unravel the "swap idiom". This implements:
Chris Lattner [Mon, 16 Feb 2004 03:54:20 +0000 (03:54 +0000)]
Teach LLVM to unravel the "swap idiom".  This implements:
Regression/Transforms/InstCombine/xor.ll:test20

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11492 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd a test for the "swap idiom", which LLVM should be able to unravel.
Chris Lattner [Mon, 16 Feb 2004 03:53:44 +0000 (03:53 +0000)]
Add a test for the "swap idiom", which LLVM should be able to unravel.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11491 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoImplement Transforms/InstCombine/xor.ll:test19
Chris Lattner [Mon, 16 Feb 2004 01:20:27 +0000 (01:20 +0000)]
Implement Transforms/InstCombine/xor.ll:test19

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11490 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoTest for xor chains
Chris Lattner [Mon, 16 Feb 2004 01:19:52 +0000 (01:19 +0000)]
Test for xor chains

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11489 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoFix a bug in the recent rewrite of the leakdetector that caused all of the
Chris Lattner [Sun, 15 Feb 2004 23:33:48 +0000 (23:33 +0000)]
Fix a bug in the recent rewrite of the leakdetector that caused all of the
nightly tests to be really messed up.  The problem was that the new leakdetector
was depending on undefined behavior: the order of destruction of static objects.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11488 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoDocument bug 240
Chris Lattner [Sun, 15 Feb 2004 23:09:07 +0000 (23:09 +0000)]
Document bug 240

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11487 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoMoving CBE tests to test/Regression/CodeGen/CBackend
Chris Lattner [Sun, 15 Feb 2004 22:56:30 +0000 (22:56 +0000)]
Moving CBE tests to test/Regression/CodeGen/CBackend

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11486 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoWhen the user runs 'llc foo.bc -march=c', write the output to "foo.cbe.c", not
Chris Lattner [Sun, 15 Feb 2004 22:54:19 +0000 (22:54 +0000)]
When the user runs 'llc foo.bc -march=c', write the output to "foo.cbe.c", not
to "foo.s".

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11485 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoNow that the lowerinvoke pass inserts calls to llvm.setjmp/llvm.longjmp, some
Chris Lattner [Sun, 15 Feb 2004 22:51:47 +0000 (22:51 +0000)]
Now that the lowerinvoke pass inserts calls to llvm.setjmp/llvm.longjmp, some
hacks can be banished.  Also, this gives us the opportunity to emit special code
for the setjmp/longjmps which alows the elimination of one GCC warning for every
setjmp/longjmp site (which is often THOUSANDS in C++ programs).  Yaay!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11484 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoBy default, llvm.setjmp/llvm.longjmp intrinsics get lowered to their libc counterparts
Chris Lattner [Sun, 15 Feb 2004 22:24:51 +0000 (22:24 +0000)]
By default, llvm.setjmp/llvm.longjmp intrinsics get lowered to their libc counterparts

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11483 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoInstead of producing calls to setjmp/longjmp, produce uses of the
Chris Lattner [Sun, 15 Feb 2004 22:24:27 +0000 (22:24 +0000)]
Instead of producing calls to setjmp/longjmp, produce uses of the
llvm.setjmp/llvm.longjmp intrinsics.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11482 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoRefactor code. Now the intrinsic lowering pass tries to recycle preexisting
Chris Lattner [Sun, 15 Feb 2004 22:16:39 +0000 (22:16 +0000)]
Refactor code.  Now the intrinsic lowering pass tries to recycle preexisting
prototypes, even if they don't precisely match what it would prefer to use.
This fixes: CBackend/2004-02-15-PreexistingExternals.llx compiling it into:

  ltmp_0_30 = memcpy(l14_C, 4u, 17);
  ltmp_1_30 = memcpy(((int *)l27_A), ((unsigned )(long)l27_B), ((int )123u));

instead of:

  ltmp_0_30 = memcpy(l14_C, 4u, 17);
  ltmp_1_27 = l43_memcpy(l27_A, l27_B, 123u);

Which does the wrong thing as you could imagine.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11481 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoNew testcase
Chris Lattner [Sun, 15 Feb 2004 22:14:46 +0000 (22:14 +0000)]
New testcase

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11480 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoThis pass should not require phi elimination or live variable
Alkis Evlogimenos [Sun, 15 Feb 2004 21:50:32 +0000 (21:50 +0000)]
This pass should not require phi elimination or live variable
analysis. It should only preserve them and update LiveVariables if it
already ran.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11479 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoFinegrainify namespacification
Chris Lattner [Sun, 15 Feb 2004 21:38:28 +0000 (21:38 +0000)]
Finegrainify namespacification
Remove one of the operands of a two operand instruction

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11478 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoMake dense maps keyed on physical registers smallerusing
Alkis Evlogimenos [Sun, 15 Feb 2004 21:37:17 +0000 (21:37 +0000)]
Make dense maps keyed on physical registers smallerusing
MRegisterInfo::getNumRegs() instead of
MRegisterInfo::FirstVirtualRegister.

Also use MRegisterInfo::is{Physical,Virtual}Register where
appropriate.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11477 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoEliminate the use of spill (reserved) registers.
Alkis Evlogimenos [Sun, 15 Feb 2004 10:24:21 +0000 (10:24 +0000)]
Eliminate the use of spill (reserved) registers.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11476 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoQOI bug fixed
Chris Lattner [Sun, 15 Feb 2004 06:00:04 +0000 (06:00 +0000)]
QOI bug fixed

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11475 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdjustments to support the new ConstantAggregateZero class
Chris Lattner [Sun, 15 Feb 2004 05:55:15 +0000 (05:55 +0000)]
Adjustments to support the new ConstantAggregateZero class

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11474 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd support for the new ConstantAggregateZero class
Chris Lattner [Sun, 15 Feb 2004 05:54:27 +0000 (05:54 +0000)]
Add support for the new ConstantAggregateZero class

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11473 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoMake the JIT zero out globals with memset instead of an element at a time. This
Chris Lattner [Sun, 15 Feb 2004 05:54:06 +0000 (05:54 +0000)]
Make the JIT zero out globals with memset instead of an element at a time.  This
should speed it up a bit on a lot of programs

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11472 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoNo need to scan zero initializers. This should make DSA a bit faster.
Chris Lattner [Sun, 15 Feb 2004 05:53:42 +0000 (05:53 +0000)]
No need to scan zero initializers.  This should make DSA a bit faster.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11471 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd a new ConstantAggregateZero class, to fix PR239. This makes zero
Chris Lattner [Sun, 15 Feb 2004 05:53:04 +0000 (05:53 +0000)]
Add a new ConstantAggregateZero class, to fix PR239.  This makes zero
initializers for constant structs and arrays take constant space, instead of
space proportinal to the number of elements.  This reduces the memory usage of
the LLVM compiler by hundreds of megabytes when compiling some nasty SPEC95
benchmarks.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11470 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agofinegrainify namespacification
Chris Lattner [Sun, 15 Feb 2004 05:52:36 +0000 (05:52 +0000)]
finegrainify namespacification

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11469 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoBug fixed
Chris Lattner [Sun, 15 Feb 2004 05:47:37 +0000 (05:47 +0000)]
Bug fixed

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11468 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoConstantArray::get and ConstantStruct::get now just return pointers to
Chris Lattner [Sun, 15 Feb 2004 04:14:47 +0000 (04:14 +0000)]
ConstantArray::get and ConstantStruct::get now just return pointers to
'Constant', instead of specific subclass pointers.  In the future, these will
return an instance of ConstantAggregateZero if all of the inputs are zeros.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11467 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoRemove dependence on return type of ConstantStruct::get
Chris Lattner [Sun, 15 Feb 2004 04:07:32 +0000 (04:07 +0000)]
Remove dependence on return type of ConstantStruct::get

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11466 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoRemove dependence on the return type of ConstantArray::get
Chris Lattner [Sun, 15 Feb 2004 04:05:58 +0000 (04:05 +0000)]
Remove dependence on the return type of ConstantArray::get

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11463 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoKeep a cache of non-abstract null arrays and structs. This speeds up llvm-dis
Chris Lattner [Sun, 15 Feb 2004 02:46:46 +0000 (02:46 +0000)]
Keep a cache of non-abstract null arrays and structs.  This speeds up llvm-dis
from 16.57 -> 13.46s on 129.compress.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11462 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoTestcase for PR205: [llvmgcc] C front-end does not emit 'zeroinitializer' when possible
Chris Lattner [Sun, 15 Feb 2004 02:19:08 +0000 (02:19 +0000)]
Testcase for PR205: [llvmgcc] C front-end does not emit 'zeroinitializer' when possible

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11461 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoTest for actual support, not just for lack of crashage
Chris Lattner [Sun, 15 Feb 2004 01:21:39 +0000 (01:21 +0000)]
Test for actual support, not just for lack of crashage

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11460 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoImplement llvm.(frame|return)address(0) correctly. They are used by the LLVM JIT...
Chris Lattner [Sun, 15 Feb 2004 01:04:03 +0000 (01:04 +0000)]
Implement llvm.(frame|return)address(0) correctly.  They are used by the LLVM JIT, among other
applications

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11459 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoFix problem in DEBUG code. I could be pointing past the end and
Alkis Evlogimenos [Sun, 15 Feb 2004 00:46:41 +0000 (00:46 +0000)]
Fix problem in DEBUG code. I could be pointing past the end and
dereferencing it causes an assertion error.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11458 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoNew code quality fix
Chris Lattner [Sun, 15 Feb 2004 00:23:15 +0000 (00:23 +0000)]
New code quality fix

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11457 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoNew testcase for PR237: [x86] wierd stack/frame pointer manipulation
Chris Lattner [Sun, 15 Feb 2004 00:21:04 +0000 (00:21 +0000)]
New testcase for PR237: [x86] wierd stack/frame pointer manipulation

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11456 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoFix the 'have a framepointer' case, so that the frame pointer always points
Chris Lattner [Sun, 15 Feb 2004 00:15:37 +0000 (00:15 +0000)]
Fix the 'have a framepointer' case, so that the frame pointer always points
to the old saved EBP.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11455 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAllow for fixed objects to reside in the local area, and if they don't to not
Chris Lattner [Sun, 15 Feb 2004 00:14:20 +0000 (00:14 +0000)]
Allow for fixed objects to reside in the local area, and if they don't to not
clobber them by allocating other objects in the same space!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11454 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd back machine code deleter pass until we get a MachineCode pass
Alkis Evlogimenos [Sun, 15 Feb 2004 00:03:15 +0000 (00:03 +0000)]
Add back machine code deleter pass until we get a MachineCode pass
that will be responsible for the creation of MachineFunctions and will
be required by all MachineFunctionPass passes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11453 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoModularize implementation of LeakDetector into a typed template
Alkis Evlogimenos [Sat, 14 Feb 2004 23:33:39 +0000 (23:33 +0000)]
Modularize implementation of LeakDetector into a typed template
implementation class. This makes the code simpler and allows for more
types to be added easily. It also implements caching for generic
objects (it was only available for llvm objects).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11452 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd support for the 'pop' instruction
Chris Lattner [Sat, 14 Feb 2004 21:06:02 +0000 (21:06 +0000)]
Add support for the 'pop' instruction

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11451 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoThere is no reason to align the stack pointer if there are no callees of this
Chris Lattner [Sat, 14 Feb 2004 20:11:07 +0000 (20:11 +0000)]
There is no reason to align the stack pointer if there are no callees of this
function!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11450 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoThere is no reason to align the stack pointer if there are no callees of this function!
Chris Lattner [Sat, 14 Feb 2004 20:10:59 +0000 (20:10 +0000)]
There is no reason to align the stack pointer if there are no callees of this function!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11449 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoThe prologue/epilogue related method calls have no reason to return a value,
Chris Lattner [Sat, 14 Feb 2004 19:49:54 +0000 (19:49 +0000)]
The prologue/epilogue related method calls have no reason to return a value,
make them return void.

This allows us to avoid some costly MBB.size() calls

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11448 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoThe prologue/epilogue related method calls have no reason to return a value,
Chris Lattner [Sat, 14 Feb 2004 19:49:05 +0000 (19:49 +0000)]
The prologue/epilogue related method calls have no reason to return a value,
make them return void.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11447 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agofix typeo
Chris Lattner [Sat, 14 Feb 2004 19:27:26 +0000 (19:27 +0000)]
fix typeo

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11446 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agofinegrainify namespacification, fix 80col prob
Chris Lattner [Sat, 14 Feb 2004 06:00:36 +0000 (06:00 +0000)]
finegrainify namespacification, fix 80col prob

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11445 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoA target that doesn't support these intrinsics will still meet spec (the
Chris Lattner [Sat, 14 Feb 2004 04:52:06 +0000 (04:52 +0000)]
A target that doesn't support these intrinsics will still meet spec (the
intrinsic will always produce zero), but it will behave unexpectedly, so
warn like GCC does.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11444 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoUrg, right. These need an input value...
Chris Lattner [Sat, 14 Feb 2004 04:47:23 +0000 (04:47 +0000)]
Urg, right.  These need an input value...

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11443 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoCodegen llvm.memset into rep stos[bwd]. Simplify code for llvm.memcpy
Chris Lattner [Sat, 14 Feb 2004 04:46:05 +0000 (04:46 +0000)]
Codegen llvm.memset into rep stos[bwd].  Simplify code for llvm.memcpy

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11442 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoadd 'rep stos[bwd]' instructions
Chris Lattner [Sat, 14 Feb 2004 04:45:37 +0000 (04:45 +0000)]
add 'rep stos[bwd]' instructions

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11441 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agonew testcase for llvm.memset generation
Chris Lattner [Sat, 14 Feb 2004 04:26:15 +0000 (04:26 +0000)]
new testcase for llvm.memset generation

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11440 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoThese should be mentioned
Chris Lattner [Sat, 14 Feb 2004 04:12:41 +0000 (04:12 +0000)]
These should be mentioned

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11439 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoDocument new intrinsics
Chris Lattner [Sat, 14 Feb 2004 04:08:35 +0000 (04:08 +0000)]
Document new intrinsics

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11438 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agobuiltin_return/frame_address now work
Chris Lattner [Sat, 14 Feb 2004 04:08:29 +0000 (04:08 +0000)]
builtin_return/frame_address now work

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11437 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoTestcase for builtin frame/return address
Chris Lattner [Sat, 14 Feb 2004 04:06:48 +0000 (04:06 +0000)]
Testcase for builtin frame/return address

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11436 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoTest that the CFE compiles memset and bzero to llvm.memset
Chris Lattner [Sat, 14 Feb 2004 03:46:37 +0000 (03:46 +0000)]
Test that the CFE compiles memset and bzero to llvm.memset

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11435 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd a testcase for the frameaddress and returnaddress intrinsics.
Chris Lattner [Sat, 14 Feb 2004 02:55:47 +0000 (02:55 +0000)]
Add a testcase for the frameaddress and returnaddress intrinsics.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11434 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd support for the returnaddress and frameaddress intrinsics
Chris Lattner [Sat, 14 Feb 2004 02:55:36 +0000 (02:55 +0000)]
Add support for the returnaddress and frameaddress intrinsics

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11433 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoTest that the code generator supports these intrinsics
Chris Lattner [Sat, 14 Feb 2004 02:51:40 +0000 (02:51 +0000)]
Test that the code generator supports these intrinsics

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11432 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd llvm.memset/frameaddress/returnaddress intrinsics.
Chris Lattner [Sat, 14 Feb 2004 02:47:17 +0000 (02:47 +0000)]
Add llvm.memset/frameaddress/returnaddress intrinsics.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11431 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoUse newly added next() and prior() utility functions.
Alkis Evlogimenos [Sat, 14 Feb 2004 01:18:34 +0000 (01:18 +0000)]
Use newly added next() and prior() utility functions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11430 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd next() and prior() iterator utility functions. Unlike std::advance
Alkis Evlogimenos [Sat, 14 Feb 2004 01:17:28 +0000 (01:17 +0000)]
Add next() and prior() iterator utility functions. Unlike std::advance
they do not modify the passed iterator but return a copy.

next(myIt) returns copy of myIt incremented once
next(myIt, n) returns copy of myIt incremented n times
prior(myIt) returns copy of myIt decremented once
prior(myIt, n) returns copy of myIt decremented n times

While at it remove obsolete implementation of mapped_iterator.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11429 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoUpdates for the C backend's movement
Chris Lattner [Sat, 14 Feb 2004 01:07:17 +0000 (01:07 +0000)]
Updates for the C backend's movement

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11428 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoUse std::numeric_limits<float>::infinity() instead of
Alkis Evlogimenos [Sat, 14 Feb 2004 00:44:07 +0000 (00:44 +0000)]
Use std::numeric_limits<float>::infinity() instead of
std::numeric_limits<float>::max() for weighting preallocated
intervals.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11427 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoUse intrinsic lowering like the rest of the backends. Get rid of crufty hacky
Chris Lattner [Sat, 14 Feb 2004 00:31:10 +0000 (00:31 +0000)]
Use intrinsic lowering like the rest of the backends.  Get rid of crufty hacky
code that worked around problems in the mangler

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11426 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd method
Chris Lattner [Sat, 14 Feb 2004 00:30:31 +0000 (00:30 +0000)]
Add method

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11425 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoFix the logic in the name mangler. If there are two symbols named 'X', and one
Chris Lattner [Sat, 14 Feb 2004 00:30:23 +0000 (00:30 +0000)]
Fix the logic in the name mangler.  If there are two symbols named 'X', and one
is external, make sure to mangle the *internal* one, not external one

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11424 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoBreak a line that's over 80cols into two.
Misha Brukman [Fri, 13 Feb 2004 23:46:47 +0000 (23:46 +0000)]
Break a line that's over 80cols into two.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11423 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoThere is no need to emit a shift if the size is constant, which is common
Chris Lattner [Fri, 13 Feb 2004 23:36:47 +0000 (23:36 +0000)]
There is no need to emit a shift if the size is constant, which is common

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11420 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoMake sure to provide a prototype for the cbackend
Chris Lattner [Fri, 13 Feb 2004 23:36:03 +0000 (23:36 +0000)]
Make sure to provide a prototype for the cbackend

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11419 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoUpdate tests
Chris Lattner [Fri, 13 Feb 2004 23:33:17 +0000 (23:33 +0000)]
Update tests

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11418 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoMoved directory, update makefile
Chris Lattner [Fri, 13 Feb 2004 23:31:12 +0000 (23:31 +0000)]
Moved directory, update makefile

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11417 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoCBackend is no longer here
Chris Lattner [Fri, 13 Feb 2004 23:29:37 +0000 (23:29 +0000)]
CBackend is no longer here

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11416 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoCBackend now lives here
Chris Lattner [Fri, 13 Feb 2004 23:29:20 +0000 (23:29 +0000)]
CBackend now lives here

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11415 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoDo not advertise our -c option anymore
Chris Lattner [Fri, 13 Feb 2004 23:24:46 +0000 (23:24 +0000)]
Do not advertise our -c option anymore

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11413 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoMercilessly rip the cbackend out of llvm-dis. Leave a helpful error message
Chris Lattner [Fri, 13 Feb 2004 23:22:40 +0000 (23:22 +0000)]
Mercilessly rip the cbackend out of llvm-dis.  Leave a helpful error message
for those who have not heard the news.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11412 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoThe cbackend has never had anything to do with llvm assembly writing
Chris Lattner [Fri, 13 Feb 2004 23:19:51 +0000 (23:19 +0000)]
The cbackend has never had anything to do with llvm assembly writing

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11411 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd support for -march=c
Chris Lattner [Fri, 13 Feb 2004 23:19:09 +0000 (23:19 +0000)]
Add support for -march=c

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11410 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoConvert the C backend into a target, for use with LLC. This allows us to use
Chris Lattner [Fri, 13 Feb 2004 23:18:48 +0000 (23:18 +0000)]
Convert the C backend into a target, for use with LLC.  This allows us to use
the lowerallocations pass to eliminate malloc/free warnings and hackish code

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11409 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoChange how we create the cwriter, and add a buttload of libraries that it now
Chris Lattner [Fri, 13 Feb 2004 23:01:14 +0000 (23:01 +0000)]
Change how we create the cwriter, and add a buttload of libraries that it now
needs.  This will be fixed shortly

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11407 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoChange access to the cwriter
Chris Lattner [Fri, 13 Feb 2004 23:00:45 +0000 (23:00 +0000)]
Change access to the cwriter

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11406 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoMake the cwriter use the lowerinvoke pass so that it can either use "disabled excepti...
Chris Lattner [Fri, 13 Feb 2004 23:00:29 +0000 (23:00 +0000)]
Make the cwriter use the lowerinvoke pass so that it can either use "disabled exceptions" or
"expensive exceptions" controlled by an option.  Also refactor and eliminate a bunch of cruft.
This is a temporary solution and causes millions of warnings to pour out of programs that use
exceptions, but it should fix the problem with sparc and the 'write' declaration (PR190).
Subsequent changes will make this stink much less

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11405 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoNew testcase. The CFE should not generate illegal LLVM intrinsics, even if
Chris Lattner [Fri, 13 Feb 2004 22:04:41 +0000 (22:04 +0000)]
New testcase.  The CFE should not generate illegal LLVM intrinsics, even if
the input program is horribly broken (like 126.gcc).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11401 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdded check for target machine endian-ness and put the result into
John Criswell [Fri, 13 Feb 2004 21:57:29 +0000 (21:57 +0000)]
Added check for target machine endian-ness and put the result into
Makefile.config (ENDIAN variable is set to big or little).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11398 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd support for a bunch more functions
Chris Lattner [Fri, 13 Feb 2004 21:21:48 +0000 (21:21 +0000)]
Add support for a bunch more functions

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11395 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoRemove getAllocatedRegNum(). Use getReg() instead.
Alkis Evlogimenos [Fri, 13 Feb 2004 21:01:20 +0000 (21:01 +0000)]
Remove getAllocatedRegNum(). Use getReg() instead.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11393 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoMark MachineBasicBlock::operator[] deprecated.
Alkis Evlogimenos [Fri, 13 Feb 2004 20:05:56 +0000 (20:05 +0000)]
Mark MachineBasicBlock::operator[] deprecated.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11392 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoDefine DEPRECATED so that it can be used in function and variable
Alkis Evlogimenos [Fri, 13 Feb 2004 20:05:44 +0000 (20:05 +0000)]
Define DEPRECATED so that it can be used in function and variable
declarations.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11391 91177308-0d34-0410-b5e6-96231b3b80d8

20 years agoAdd support for fopen/fclose. Specifically with fopen, we were marking all of the
Chris Lattner [Fri, 13 Feb 2004 20:05:32 +0000 (20:05 +0000)]
Add support for fopen/fclose.  Specifically with fopen, we were marking all of the
operands as incomplete, though fopen is known to only read them.  This just adds
fclose for symmetry, though it doesn't gain anything.  This makes the dsgraphs for
181.mcf much more precise.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11390 91177308-0d34-0410-b5e6-96231b3b80d8