llvm-svn: 7210

llvm-svn: 7209

llvm-svn: 7208

x86 or Sparc, LLC will automatically default to that platform, no guessing
required. On another platform, it will default to `noarch' and will have to
guess which architecture to compile to.

llvm-svn: 7207

llvm-svn: 7206

Avoid a fall-through in the (stubby) treatment of the longjmp intrinsic
call which causes llc & lli to core-dump.

Add a sort-of treatment of cast double to ulong. I am not really sure
what a user should expect to see upon casting a negative FP value to
unsigned long long. But with what is given here, I was able to write
a program that could cast -123.456 to ulong and back and get -123.0,
which seems like a step in the right direction.  GCC seems to give you
0. I don't know if I'd consider that useful.

These cases were coming up in GNU coreutils-5.0.

llvm-svn: 7205

to pass to dlsym() -- Linux/x86 wants 0 while Sparc/Solaris wants RTLD_SELF,
which is not zero. Thanks to Chris for the suggestion.

llvm-svn: 7204

llvm-svn: 7203

llvm-svn: 7202

llvm-svn: 7201

try to load or store through a bool*.

llvm-svn: 7195

llvm-svn: 7194

an external constant.  Since external constants don't have initializers, we
were failing on an assert() call in llvm/GlobalVariable.h.

llvm-svn: 7193

llvm-svn: 7192

llvm-svn: 7191

    to clone the subgraph reachable from a set of root nodes, into the
    current graph, merging the global nodes into those in the current graph.
(2) Added DSGraph::updateFromGlobalGraph() to rematerialize nodes from the
    globals graph into the current graph in both BU and TD passes.
(3) Added hash_set<const GlobalValue*> InlinedGlobals: a set of globals to
    track which globals have been inlined into the current graph from
    callers or callees.  In the TD pass, such globals are up-to-date and
    do not need to be rematerialized from the GlobalsGraph.
(4) Added StripIncompleteBit/KeepIncompleteBit to remove incomplete bit
    when cloning nodes into the globals graph.

llvm-svn: 7190

after all callees are inlined into the current graph.

NOTE: There's also a major bug fix for the BU pass in DataStructure.cpp,
which ensures that resolvable indirect calls are not moved out to the
globals graph, so that they are eventually inlined (if possible).

llvm-svn: 7189

    after all callers are inlined into the current graph.
(2) Optimize the way a graph is inlined into its callees in the TD phase:
    (a) Use DSGraph::cloneReachableSubgraph to clone only a subgraph at
        each call site, for faster inlining.
    (b) Clone separately for the same callee at different call sites,
        since only the reachable subgraph is being cloned, not the entire
        caller graph.

llvm-svn: 7188

and (2) faster inlining by cloning only reachable nodes.  In particular:
(1) Added DSGraph::cloneReachableSubgraph and DSGraph::cloneReachableNodes
    to clone the subgraph reachable from a set of root nodes, into the
    current graph, merging the global nodes into thos in the current graph.
    The TD pass now uses this for faster inlining, and so does the
    next function.
(2) Added DSGraph::updateFromGlobalGraph() to rematerialize nodes from the
    globals graph into the current graph in both BU and TD passes.
(3) `I' flags are removed from all nodes in the globals graph, because they
    are difficult to maintain correctly and are not needed anyway.
(4) Aux. function calls are only removed to the globals graph if they
    will never be resovled.  (This is what fixed gap.)  The immediate
    reason is that if we took these out of a function (and moved them to
    the globals graph) we would need to rematerialize these nodes into the
    function graph for every function in the BU pass.  The longer term
    problem is that we would need to find a way to remove them from the
    globals graph iff they have been resolved on all paths through the
    call graph.

llvm-svn: 7187

isUnresolvableFunc() (I thought I needed this externally.
I don't, but it's still nicer this way.)

llvm-svn: 7186

now works in instructions which require a 2-bit or 3-bit INTcc code.

Incidentally, that means that the representation of INTcc registers is now the
same in both integer and FP instructions. Thus, code became much simpler and
cleaner.

llvm-svn: 7185

no change in functionality.

llvm-svn: 7184

up builds a little bit since it will prevent GNU make from matching these
phony targets against implicit rules.

llvm-svn: 7183

llvm-svn: 7182

llvm-svn: 7181

Also, placed DEBUG() guards around debug information so that the generated file
is much smaller and hence should be faster to preprocess/compile.

llvm-svn: 7180

masking and shifting operands directly into their place in the instruction,
instead of the old-fashioned way of ORing in each bit separately.

llvm-svn: 7179

allow, i.e. make a sequence of instructions to enable an indirect call using
jump-and-link and 2 temporary registers (which we save and ultimately restore).

Warning: if the delay slot of a function call is used to do meaningful work and
not just a NOP, this behavior is incorrect. However, the Sparc backend does not
yet utilize the delay slots effectively, so it is not necessary to make an
overly complicated algorithm for something that's not used.

llvm-svn: 7178

usage of the special file handle RTLD_SELF on Sparc/Solaris vs. 0 on Linux/x86.

llvm-svn: 7177

the program's executing image, not 0 as it is on Linux/x86 and possibly other
systems.

llvm-svn: 7176

* FP double registers are now coded correctly
* Removed function which converted registers based on register types, it was
  broken (because regTypes are broken)

llvm-svn: 7175

llvm-svn: 7174

llvm-svn: 7173

llvm-svn: 7172

Specifically, this updates libtool to version 1.5 and adds the following:
	- Added the -only-static option that we added in our previous libtool.
	- Modified the autoconf macros so that libtool uses the -G option when
	  linking on Solaris.  This allows libraries with global variables with
	  constructors to automatically run those constructors when the
	  library is dlopened().

llvm-svn: 7171

Also, make Pointer type depend on architecture.

llvm-svn: 7170

remembered in valuesStoredInFunction, but never traced at function return,
and that's too late to be finding the error anyway).
Stores trace both the value and the address being stored to,
but after some experience I think only values should be traced.
The pointer hash table just fills up far too quickly if every
store address were traced.

llvm-svn: 7169

out the entire llvm disassembly for the function at global constant-output
time, which caused the assembler to barf in 164.gzip. This fixes that
particular problem (though 164.gzip has other problems with X86 llc.)

llvm-svn: 7168

Fhourstones, McCat-vor, and many others...)

Printer.cpp: Print implicit uses for AddRegFrm instructions.  Break gas
bug workarounds up into separate stanzas of code for each bug.  Add new
workarounds for fild and fistp.

X86InstrInfo.def: Add O_ST0 implicit uses for more FP instrs where they
obviously apply. Also add PrintImplUses flags for FP instrs where they
are necessary for gas to understand the output.

llvm-svn: 7165

llvm-svn: 7161