Instead of emitting a JIT stub that looks like this:
internal void %l1_main_entry_2E_ce_wrapper(int) {
header:
%resolver = call sbyte* %getPointerToNamedFunction( sbyte* getelementptr ([20 x sbyte]* %l1_main_entry_2E_ce_name, int 0, int 0) ) ; <sbyte*> [#uses=1]
%resolverCast = cast sbyte* %resolver to void (int)* ; <void (int)*> [#uses=1]
call void %resolverCast( int %0 )
ret void
}
Emit one that looks like this:
internal void %l1_main_entry_2E_ce_wrapper(int) {
Entry:
%fpcache = load void (int)** %l1_main_entry_2E_ce.fpcache ; <void (int)*> [#uses=2]
%isNull = seteq void (int)* %fpcache, null ; <bool> [#uses=1]
br bool %isNull, label %lookupfp, label %usecache
usecache: ; preds = %lookupfp, %Entry
%fp = phi void (int)* [ %resolverCast, %lookupfp ], [ %fpcache, %Entry ] ; <void (int)*> [#uses=1]
call void %fp( int %0 )
ret void
lookupfp: ; preds = %Entry
%resolver = call sbyte* %getPointerToNamedFunction( sbyte* getelementptr ([20 x sbyte]* %l1_main_entry_2E_ce_name, int 0, int 0) ) ; <sbyte*> [#uses=1]
%resolverCast = cast sbyte* %resolver to void (int)* ; <void (int)*> [#uses=2]
store void (int)* %resolverCast, void (int)** %l1_main_entry_2E_ce.fpcache
br label %usecache
}
This makes the JIT debugger *MUCH* faster on large programs, as
getPointerToNamedFunction takes time linear with the size of the program, and
before we would call it every time a function in the text module was called from
the safe module (ouch!).
llvm-svn: 22387
allows objdump to know which function we are emitting to:
00000000 <foo>: <----
0: b8 01 00 00 00 mov $0x1,%eax
5: 03 44 24 04 add 0x4(%esp,1),%eax
9: c3 ret
... and allows .o files to be useful for linking :)
llvm-svn: 22378
Add a *VERY INITIAL* machine code emitter class. This is enough to take
this C function:
int foo(int X) { return X +1; }
and make objdump produce the following:
$ objdump -d t-llvm.o
t-llvm.o: file format elf32-i386
Disassembly of section .text:
00000000 <.text>:
0: b8 01 00 00 00 mov $0x1,%eax
5: 03 44 24 04 add 0x4(%esp,1),%eax
9: c3 ret
Anything using branches or refering to the constant pool or requiring
relocations will not work yet.
llvm-svn: 22375
This is the last MVTSDNode.
This allows us to eliminate a bunch of special case code for handling
MVTSDNodes.
Also, remove some uses of dyn_cast that should really be cast (which is
cheaper in a release build).
llvm-svn: 22368
MVTSDNode class. This class is used to provide an operand to operators
that require an extra type. We start by converting FP_ROUND_INREG and
SIGN_EXTEND_INREG over to using it.
llvm-svn: 22364
1. Use isValid() to check validity of the resulting path name in the
eraseSuffix even though we can't think of a case where eraseSuffix could
possibly cause an invalid path name.
2. Rewrite isValid() to not use the deprecated realpath function any more.
It now just uses isascii to make sure all the characters are legit.
llvm-svn: 22359
This chagne just renames some sys::Path methods to ensure they are not
misused. The Path documentation now divides methods into two dimensions:
Path/Disk and accessor/mutator. Path accessors and mutators only operate
on the Path object itself without making any disk accesses. Disk accessors
and mutators will also access or modify the file system. Because of the
potentially destructive nature of disk mutators, it was decided that all
such methods should end in the work "Disk" to ensure the user recognizes
that the change will occur on the file system. This patch makes that
change. The method name changes are:
makeReadable -> makeReadableOnDisk
makeWriteable -> makeWriteableOnDisk
makeExecutable -> makeExecutableOnDisk
setStatusInfo -> setStatusInfoOnDisk
createDirectory -> createDirectoryOnDisk
createFile -> createFileOnDisk
createTemporaryFile -> createTemporaryFileOnDisk
destroy -> eraseFromDisk
rename -> renamePathOnDisk
These changes pass the Linux Deja Gnu tests.
llvm-svn: 22354
Get rid of the difference between file paths and directory paths. The Path
class now simply stores a path that can refer to either a file or a
directory. This required various changes in the implementation and interface
of the class with the corresponding impact to its users. Doxygen comments were
also updated to reflect these changes. Interface changes are:
appendDirectory -> appendComponent
appendFile -> appendComponent
elideDirectory -> eraseComponent
elideFile -> eraseComponent
elideSuffix -> eraseSuffix
renameFile -> rename
setDirectory -> set
setFile -> set
Changes pass Dejagnu and llvm-test/SingleSource tests.
llvm-svn: 22349
Because the instcombine has to scan the entire function when it starts up
to begin with, we might as well do it in DFO so we can nuke unreachable code.
This fixes: Transforms/InstCombine/2005-07-07-DeadPHILoop.ll
llvm-svn: 22348
module to the ELF file. Test it by adding support for emitting common
symbols. This allows us to compile this:
%X = weak global int 0
%Y = weak global int 0
%Z = weak global int 0
to an elf file that 'readelf's this:
Symbol table '.symtab' contains 4 entries:
Num: Value Size Type Bind Vis Ndx Name
0: 00000000 0 NOTYPE LOCAL DEFAULT UND
1: 00000004 4 OBJECT GLOBAL DEFAULT COM X
2: 00000004 4 OBJECT GLOBAL DEFAULT COM Y
3: 00000004 4 OBJECT GLOBAL DEFAULT COM Z
llvm-svn: 22343
XMM registers. There are many known deficiencies and fixmes, which will be
addressed ASAP. The major benefit of this work is that it will allow the
LLVM register allocator to allocate FP registers across basic blocks.
The x86 backend will still default to x87 style FP. To enable this work,
you must pass -enable-sse-scalar-fp and either -sse2 or -sse3 to llc.
An example before and after would be for:
double foo(double *P) { double Sum = 0; int i; for (i = 0; i < 1000; ++i)
Sum += P[i]; return Sum; }
The inner loop looks like the following:
x87:
.LBB_foo_1: # no_exit
fldl (%esp)
faddl (%eax,%ecx,8)
fstpl (%esp)
incl %ecx
cmpl $1000, %ecx
#FP_REG_KILL
jne .LBB_foo_1 # no_exit
SSE2:
addsd (%eax,%ecx,8), %xmm0
incl %ecx
cmpl $1000, %ecx
#FP_REG_KILL
jne .LBB_foo_1 # no_exit
llvm-svn: 22340
1. Pass Value*'s into lowering methods so that the proper pointers can be
added to load/stores from the valist
2. Intrinsics that return void should only return a token chain, not a token
chain/retval pair.
3. Rename LowerVAArgNext -> LowerVAArg, because VANext is long gone.
4. Now that we have Value*'s available in the lowering methods, pass them
into any load/stores from the valist that are emitted
llvm-svn: 22339
1. Pass Value*'s into lowering methods so that the proper pointers can be
added to load/stores from the valist
2. Intrinsics that return void should only return a token chain, not a token
chain/retval pair.
3. Rename LowerVAArgNext -> LowerVAArg, because VANext is long gone.
llvm-svn: 22338
Chris Lattner