Commit Graph

7520 Commits

Author SHA1 Message Date
Dan Gohman
206ad10c14 Propogate the Depth argument when calling
TLI.computeMaskedBitsForTargetNode from ComputeMaskedBits, since
the former may call back into the latter. This fixes a major
compile time problem on a testcase that happnened to hit this
in a particularly bad way, PR4643.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@78023 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-04 00:24:42 +00:00
Ted Kremenek
f6e75c3435 Update CMake files.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@78020 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 23:44:01 +00:00
Chris Lattner
290c2f56ce eliminate CurrentSection, rename CurrentSection_ -> CurrentSection, make it private,
eliminate IsInTextSection.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@78017 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 23:20:21 +00:00
Chris Lattner
c6fdced3db Kill off SwitchToDataSection and SwitchToTextSection, woo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@78015 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 23:10:34 +00:00
David Greene
b525766479 Re-apply LiveInterval index dumping patch, with fixes suggested by Bill
and others.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@78003 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 21:55:09 +00:00
Jakob Stoklund Olesen
980daea857 Fix Bug 4657: register scavenger asserts with subreg lowering
When LowerSubregsInstructionPass::LowerInsert eliminates an INSERT_SUBREG
instriction because it is an identity copy, make sure that the same registers
are alive before and after the elimination.

When the super-register is marked <undef> this requires inserting an
IMPLICIT_DEF instruction to make sure the super register is live.

Fix a related bug where a kill flag on the inserted sub-register was not transferred properly.

Finally, clear the undef flag in MachineInstr::addRegisterKilled. Undef implies dead and kill implies live, so they cant both be valid.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77989 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 20:08:18 +00:00
Chris Lattner
38c398808c make getObjFileLowering() return a non-const reference.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77984 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 19:12:26 +00:00
Bob Wilson
6e7a85c075 Revert 77974. It breaks 3 of the ARM tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77982 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 19:06:29 +00:00
Chris Lattner
d54a6b8c58 make SwitchToSection accept null sections for now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77976 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 18:04:42 +00:00
Sanjiv Gupta
2caf1b212e Allow targets to custom handle softening of results or operands before trying the standard stuff.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77974 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 17:35:21 +00:00
Benjamin Kramer
d5fe92efbc llvm_report_error already prints "LLVM ERROR:". So stop reporting errors like "LLVM ERROR: llvm: error:" or "LLVM ERROR: ERROR:".
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77971 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 13:33:33 +00:00
Evan Cheng
eed0ff147c Fix a coaelescer bug. If a copy val# is extended to eliminate a non-trivially coalesced copy, and the copy kills its source register. Trim the source register's live range to the last use if possible. This fixes up kill marker to make the scavenger happy.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77967 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 08:41:59 +00:00
Daniel Dunbar
275872e799 Fix some comments referring to std::cerr.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77931 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-03 01:02:24 +00:00
Jakob Stoklund Olesen
b36eb9df20 Fix issue in regscavenger when scavenging a callee-saved register that has not been spilled.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77912 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 20:29:41 +00:00
Jakob Stoklund Olesen
ece4818516 Never add a kill flag to a constrained physical register in a two-addr instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77906 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 19:13:03 +00:00
Jakob Stoklund Olesen
0e6a4efdd4 Scavenger asserts.
Allow imp-def and imp-use of anything in the scavenger asserts, just like the machine code verifier.
Allow redefinition of a sub-register of a live register.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77904 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 18:28:41 +00:00
Dan Gohman
4ea480499c Avoid forming a SELECT_CC in a type that the target doesn't
support. This isn't immediately interesting, because Legalize
ends up lowering SELECT_CC if the target doesn't support it,
but this simplifies the process.

Also, if the SELECT_CC would be expanded in Legalize, it
can potentially end up with two copies of the condition
expression. By leaving it as SELECT+SETCC, the SELECT can be
expanded into two SELECTs that use a single SETCC.

The two comparisons are usually CSE'd, but depending on
when various expressions get legalized, the comparison
expression could involve calls to library functions, such
that the comparison expression may not be able to be CSE'd.
This will be needed by a future patch.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77896 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 16:19:38 +00:00
Jakob Stoklund Olesen
bcdbcb9f95 Handle <undef> flag in machine code verifier.
Use of an <undef> register is treated like an <imp-use>. It is not an error to use a dead <undef> register.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77890 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 07:38:21 +00:00
Chris Lattner
18a4c16726 move dwarf debug info section selection stuff from TAI to
TLOF, unifying all the dwarf targets at the same time.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77889 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 07:24:22 +00:00
Chris Lattner
35039ac241 convert EHFrameSection to be managed by TLOF instead of TAI.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77888 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 06:52:36 +00:00
Chris Lattner
9ba8c6872d remove the dead ELFTargetAsmInfo.h/cpp file. TargetAsmInfo
defaults to being ELF.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77866 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 04:33:09 +00:00
Chris Lattner
ad88bc4076 Fix some fixme's in #if 0'd code by making it dependent on the structural
behavior of the LSDA section instead of on some random target hook that
needs to be kept in synch with other points of truth.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77855 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 03:59:56 +00:00
Chris Lattner
d5bbb07ec8 move getDwarfExceptionSection from TAI to TLOF and rename it to
getLSDASection() to be more specific.  This makes it pretty obvious
that the ELF LSDA section is being specified wrong in PIC mode.  We're
probably getting a lot of startup-time relocations to a readonly page,
which is expensive and bad.

Someone who cares about ELF C++ should investigate this.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77847 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 01:34:32 +00:00
Chris Lattner
80ec2792b2 convert ctors/dtors section to be in TLOF instead of
TAI.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77842 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-02 00:34:36 +00:00
Chris Lattner
2798119ab4 (re)introduce new simpler apis for creation sectionkinds.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77834 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 23:57:16 +00:00
Chris Lattner
83d77faf6e Remove "JumpTableDataSection" from TAI, instead, have AsmPrinter
compute it based on what it knows.  As part of this, rename getSectionForMergeableConstant
to getSectionForConstant because it works for non-mergable constants also.

The only functionality change from this is that Xcore will start dropping
its jump tables into readonly section instead of data section in -static mode.
This should be fine as the linker resolves the relocations.  If this is a
problem, let me know and we'll come up with another solution.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77833 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 23:46:12 +00:00
Chris Lattner
2f03f65205 loweringinfo is always non-null.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77821 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 22:48:40 +00:00
Chris Lattner
c19ee610f6 fix a problem Eli noticed where we would compile the attached ptrtoint
to:

.quad X

even on a 32-bit system, where X is not 64-bits.  There isn't much that
we can do here, so we just print:

.quad	((X) & 4294967295)

instead.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77818 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 22:25:12 +00:00
Chris Lattner
968ff11967 Change SectionKind to be a property that is true of a *section*, it
should have no state that is specific to particular globals in the
section.  In this case, it means the removal of the "isWeak" and
"ExplicitSection" bits.  MCSection uses the new form of SectionKind.

To handle isWeak, I introduced a new SectionInfo class, which is
SectionKind + isWeak, and it is used by the part of the code generator
that does classification of a specific global.

The ExplicitSection disappears.  It is moved onto MCSection as a new
"IsDirective" bit.  Since the Name of a section is either a section
or directive, it makes sense to keep this bit in MCSection.  Ultimately
the creator of MCSection should canonicalize (e.g.) .text to whatever
the actual section is.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77803 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 21:11:14 +00:00
Dan Gohman
1944da1c5a Print the target flags as an int instead of a char, as they aren't
actually characters.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77794 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 19:13:38 +00:00
Dan Gohman
f94083333c Give MachineFunctionAnalysis a destructor so it can verify that
that it released its allocated memory.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77775 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 04:19:43 +00:00
Dan Gohman
8a0336a44b Delete a redundant variable.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77774 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 04:18:29 +00:00
Dan Gohman
f7d6cd46e9 Minor code simplifications.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77769 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 03:51:09 +00:00
Dan Gohman
c32304e56e SelectionDAGISel no longer needs to check hasAvailableExternallyLinkage,
as it is now a MachineFunctionPass, and MachineFunctionPass now handles
this.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77760 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 00:42:23 +00:00
Dan Gohman
4f3cfba575 Make UnreachableMachineBlockElim preserve MachineDominatorTree and
MachineLoopInfo.

llc now runs MachineLoopInfo and MachineDominatorTree only twice, instead
of three times.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77759 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 00:34:30 +00:00
Dan Gohman
845012e6d3 Use setPreservesAll and setPreservesCFG in CodeGen passes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77754 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 23:37:33 +00:00
Dan Gohman
a3477fe06d SelectionDAGISel does not "preserve all", since it makes lots of changes
to the MachineFunction.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77753 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 23:36:22 +00:00
Dan Gohman
2d147c4401 Use a range insert instead of an explicit loop.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77752 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 23:36:06 +00:00
Bob Wilson
8d91955d2b Allow target intrinsics that return multiple values, i.e., struct types,
in SelectionDAGLowering::visitTargetIntrinsic.

This removes a bit of special-case code for vector types.  After staring
at it for a while, I managed to convince myself that it is not necessary.
The only case where TLI.getValueType() differs from MVT::getMVT is for iPTR,
so this code could potentially make a difference for a vector of pointers.
But, it looks like that is not supported.  Calling TLI.getValueType() on
a vector of pointers leads to the following sequence of calls:

TargetLowering::getValueType
MVT::getMVT
MVT::getVectorVT(iPTR, num elements)
MVT::getExtendedVectorVT
MVT::getTypeForMVT for iPTR
assertion fails "Type is not extended!"

So, unless I'm really missing something, this bit of code is irrelevant to
the current version of LLVM, which is consistent with the fact that I don't
see this code in other similar places.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77747 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 22:41:21 +00:00
Chris Lattner
46b754c76b add some comments on how this is *supposed* to work. We don't
need the PreferredEHDataFormat hook, but I have yet-more refactoring to
do before I can zap it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77742 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 22:18:14 +00:00
Chris Lattner
81c9a06937 rearrange a conditional. Even if this weren't #if 0'd out, this would
have no functionality change.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77741 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 22:03:47 +00:00
David Greene
bef8768bd0 Simplify operand padding by keying off tabs in the asm stream. If
padding is disabled, tabs get replaced by spaces except in the case of
the first operand, where the tab is output to line up the operands after
the mnemonics.

Add some better comments and eliminate redundant code.

Fix some testcases to not assume tabs.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77740 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 21:57:10 +00:00
Chris Lattner
7a01e95915 PreferredEHDataFormat is always call with data and global, but this whole
thing is #if0'd out anyway.  Just simplify the code by reducing the interface.
Not deleting this is essential for Bill's continuing happiness.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77736 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 21:39:55 +00:00
Chris Lattner
26630c1180 move emitUsedDirectiveFor to TargetLoweringObjectFile and rename it to
indicate that it is a predicate, not an emitter.  This eliminates TAI
dependencies on Mangler and GlobalValue.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77726 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 20:52:39 +00:00
Owen Anderson
a7235ea724 Move a few more APIs back to 2.5 forms. The only remaining ones left to change back are
metadata related, which I'm waiting on to avoid conflicting with Devang.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77721 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 20:28:14 +00:00
Ted Kremenek
f83146fd11 Update CMake files.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77709 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 18:50:22 +00:00
Chris Lattner
a87dea4f8c switch off of 'Section' onto MCSection. We're not properly using
MCSection subclasses yet, but this is a step in the right direction.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77708 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 18:48:30 +00:00
Dan Gohman
f266f89634 MachineFunction no longer needs Annotation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77704 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 18:35:51 +00:00
Dan Gohman
ad2afc2a42 Reapply r77654 with a fix: MachineFunctionPass's getAnalysisUsage
shouldn't do AU.setPreservesCFG(), because even though CodeGen passes
don't modify the LLVM IR CFG, they may modify the MachineFunction CFG,
and passes like MachineLoop are registered with isCFGOnly set to true.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77691 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 18:16:33 +00:00
Chris Lattner
f26e03bc7e refactor section construction in TLOF to be through an explicit
initialize method, which can be called when an MCContext is available.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77687 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-31 17:42:42 +00:00