Commit Graph

10681 Commits

Author SHA1 Message Date
Chris Lattner
fdf6a3c8cc Add prototype
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11837 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 21:34:51 +00:00
Chris Lattner
079236d1c9 My faith in programmers has been found to be totally misplaced. One would
assume that if they don't intend to write to a global variable, that they
would mark it as constant.  However, there are people that don't understand
that the compiler can do nice things for them if they give it the information
it needs.

This pass looks for blatently obvious globals that are only ever read from.
Though it uses a trivially simple "alias analysis" of sorts, it is still able
to do amazing things to important benchmarks.  253.perlbmk, for example,
contains several ***GIANT*** function pointer tables that are not marked
constant and should be.  Marking them constant allows the optimizer to turn
a whole bunch of indirect calls into direct calls.  Note that only a link-time
optimizer can do this transformation, but perlbmk does have several strings
and other minor globals that can be marked constant by this pass when run
from GCCAS.

176.gcc has a ton of strings and large tables that are marked constant, both
at compile time (38 of them) and at link time (48 more).  Other benchmarks
give similar results, though it seems like big ones have disproportionally
more than small ones.

This pass is extremely quick and does good things.  I'm going to enable it
in gccas & gccld.  Not bad for 50 SLOC.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11836 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 21:34:36 +00:00
Misha Brukman
757df02826 SparcV8 regs are really 32-bit, not 64! Thanks, Chris.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11835 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 21:03:02 +00:00
Misha Brukman
e07c2aa67c Clean up the tablegen descriptions for SparcV8.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11834 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 21:02:21 +00:00
Misha Brukman
5914bf6ef5 Fix the SparcV8 register definitions that were imported from PPC template.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11833 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 21:00:05 +00:00
Misha Brukman
3dff822988 SparcV8 has different types of instructions, but F1 is only used for CALL.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11832 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 20:52:20 +00:00
Brian Gaeke
f44f72715d Note that this test is currently expected to fail.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11831 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 20:34:02 +00:00
Chris Lattner
95c34f2efd Add an assertion
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11830 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 19:37:44 +00:00
Chris Lattner
87124425d0 Fix failures in 099.go due to the cfgsimplify pass creating switch instructions
where there did not used to be any before


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11829 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 19:30:19 +00:00
Brian Gaeke
e785e531f4 SparcV8 skeleton
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11828 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 19:28:19 +00:00
Brian Gaeke
150666fd82 Great renaming part II: Sparc --> SparcV9 (also includes command-line options and Makefiles)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11827 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 19:08:12 +00:00
Brian Gaeke
e3d6807ab5 Great renaming: Sparc --> SparcV9
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11826 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 18:44:15 +00:00
Chris Lattner
adc1efe81c Add a bunch more functions used by perlbmk
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11824 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 17:43:20 +00:00
John Criswell
cac4b8ea04 Updated to use llc to generate CBE code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11823 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 17:15:02 +00:00
Chris Lattner
a39719ff24 Substantial improvements and cleanups for the release notes. We were missing
a bunch of stuff!  :)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11822 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 16:36:51 +00:00
Chris Lattner
940ff563f7 Fix incorrect debug code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11821 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 15:15:04 +00:00
Chris Lattner
5f2c7b1975 Teach the instruction selector how to transform 'array' GEP computations into X86
scaled indexes.  This allows us to compile GEP's like this:

int* %test([10 x { int, { int } }]* %X, int %Idx) {
        %Idx = cast int %Idx to long
        %X = getelementptr [10 x { int, { int } }]* %X, long 0, long %Idx, ubyte 1, ubyte 0
        ret int* %X
}

Into a single address computation:

test:
        mov %EAX, DWORD PTR [%ESP + 4]
        mov %ECX, DWORD PTR [%ESP + 8]
        lea %EAX, DWORD PTR [%EAX + 8*%ECX + 4]
        ret

Before it generated:
test:
        mov %EAX, DWORD PTR [%ESP + 4]
        mov %ECX, DWORD PTR [%ESP + 8]
        shl %ECX, 3
        add %EAX, %ECX
        lea %EAX, DWORD PTR [%EAX + 4]
        ret

This is useful for things like int/float/double arrays, as the indexing can be folded into
the loads&stores, reducing register pressure and decreasing the pressure on the decode unit.
With these changes, I expect our performance on 256.bzip2 and gzip to improve a lot.  On
bzip2 for example, we go from this:

10665 asm-printer           - Number of machine instrs printed
   40 ra-local              - Number of loads/stores folded into instructions
 1708 ra-local              - Number of loads added
 1532 ra-local              - Number of stores added
 1354 twoaddressinstruction - Number of instructions added
 1354 twoaddressinstruction - Number of two-address instructions
 2794 x86-peephole          - Number of peephole optimization performed

to this:
9873 asm-printer           - Number of machine instrs printed
  41 ra-local              - Number of loads/stores folded into instructions
1710 ra-local              - Number of loads added
1521 ra-local              - Number of stores added
 789 twoaddressinstruction - Number of instructions added
 789 twoaddressinstruction - Number of two-address instructions
2142 x86-peephole          - Number of peephole optimization performed

... and these types of instructions are often in tight loops.

Linear scan is also helped, but not as much.  It goes from:

8787 asm-printer           - Number of machine instrs printed
2389 liveintervals         - Number of identity moves eliminated after coalescing
2288 liveintervals         - Number of interval joins performed
3522 liveintervals         - Number of intervals after coalescing
5810 liveintervals         - Number of original intervals
 700 spiller               - Number of loads added
 487 spiller               - Number of stores added
 303 spiller               - Number of register spills
1354 twoaddressinstruction - Number of instructions added
1354 twoaddressinstruction - Number of two-address instructions
 363 x86-peephole          - Number of peephole optimization performed

to:

7982 asm-printer           - Number of machine instrs printed
1759 liveintervals         - Number of identity moves eliminated after coalescing
1658 liveintervals         - Number of interval joins performed
3282 liveintervals         - Number of intervals after coalescing
4940 liveintervals         - Number of original intervals
 635 spiller               - Number of loads added
 452 spiller               - Number of stores added
 288 spiller               - Number of register spills
 789 twoaddressinstruction - Number of instructions added
 789 twoaddressinstruction - Number of two-address instructions
 258 x86-peephole          - Number of peephole optimization performed

Though I'm not complaining about the drop in the number of intervals.  :)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11820 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 07:00:55 +00:00
Chris Lattner
b6bac51351 * Make the previous patch more efficient by not allocating a temporary MachineInstr
to do analysis.

*** FOLD getelementptr instructions into loads and stores when possible,
    making use of some of the crazy X86 addressing modes.

For example, the following C++ program fragment:

struct complex {
    double re, im;
    complex(double r, double i) : re(r), im(i) {}
};
inline complex operator+(const complex& a, const complex& b) {
    return complex(a.re+b.re, a.im+b.im);
}
complex addone(const complex& arg) {
    return arg + complex(1,0);
}

Used to be compiled to:
_Z6addoneRK7complex:
        mov %EAX, DWORD PTR [%ESP + 4]
        mov %ECX, DWORD PTR [%ESP + 8]
***     mov %EDX, %ECX
        fld QWORD PTR [%EDX]
        fld1
        faddp %ST(1)
***     add %ECX, 8
        fld QWORD PTR [%ECX]
        fldz
        faddp %ST(1)
***     mov %ECX, %EAX
        fxch %ST(1)
        fstp QWORD PTR [%ECX]
***     add %EAX, 8
        fstp QWORD PTR [%EAX]
        ret

Now it is compiled to:
_Z6addoneRK7complex:
        mov %EAX, DWORD PTR [%ESP + 4]
        mov %ECX, DWORD PTR [%ESP + 8]
        fld QWORD PTR [%ECX]
        fld1
        faddp %ST(1)
        fld QWORD PTR [%ECX + 8]
        fldz
        faddp %ST(1)
        fxch %ST(1)
        fstp QWORD PTR [%EAX]
        fstp QWORD PTR [%EAX + 8]
        ret

Other programs should see similar improvements, across the board.  Note that
in addition to reducing instruction count, this also reduces register pressure
a lot, always a good thing on X86.  :)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11819 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 06:13:04 +00:00
Chris Lattner
2e68037187 Add a helper to create an addressing mode given all of the pieces.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11818 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 06:01:07 +00:00
Chris Lattner
985fe3df6f add an inefficient way of folding structure and constant array indexes together
into a single LEA instruction.  This should improve the code generated for
things like X->A.B.C[12].D.

The bigger benefit is still coming though.  Note that this uses an LEA instruction
instead of an add, giving the register allocator more freedom.  We should probably
never generate ADDri32's.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11817 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 03:45:50 +00:00
Chris Lattner
5a83096d6a Implement special case for storing an immediate into memory so that we don't need
an intermediate register.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11816 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 02:56:58 +00:00
Brian Gaeke
bb92f6fbf2 Cygwin defines log2 as a macro. Undef it here IFF it has already been defined,
so that we always get the inline function instead. Remember, kids, like it says
in the GCC manual, "An Inline Function is As Fast As a Macro."


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11815 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-25 01:53:45 +00:00
Brian Gaeke
2051fe3df5 small portability fix.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11814 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 22:58:31 +00:00
Chris Lattner
39bb2dc557 Add support for 'rename'
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11813 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 22:17:00 +00:00
Chris Lattner
2a11653fa9 Make the verifier a little more explicit about this problem.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11811 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 22:06:07 +00:00
Chris Lattner
d561209a47 Add support for remove, fwrite, and fread
Also fix problem where we didn't check to see if a node pointer was null.
Though fclose(null) doesn't make a lot of sense, 300.twolf does it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11810 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 22:02:48 +00:00
John Criswell
878be7dbe9 Added the VTune tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11809 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 21:43:38 +00:00
Brian Gaeke
748fba141f FunctionLiveVarInfo.h moved: include/llvm/CodeGen -> lib/Target/Sparc/LiveVar
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11804 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 19:46:00 +00:00
Chris Lattner
2bcab1f900 Fix some unexpected fallout from the config.h changes. Because the CBE no
longer was getting this #include, it always fell back on the less precise
floating point initializer values, causing some testsuite failures.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11803 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 18:34:10 +00:00
Chris Lattner
d38f208cd9 Fix a faulty optimization on FP values
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11801 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 18:10:14 +00:00
John Criswell
6e4ca61438 Fixed minor typos.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11800 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 16:13:56 +00:00
Chris Lattner
8e509dd5e8 If a block is made dead, make sure to promptly remove it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11799 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 16:09:21 +00:00
Alkis Evlogimenos
0d6c5b6489 Move machine code rewriter and spiller outside the register
allocator.

The implementation is completely rewritten and now employs several
optimizations not exercised before. For example for 164.gzip we have
997 loads and 699 stores vs the 1221 loads and 880 stores we have
before.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11798 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 08:58:30 +00:00
Chris Lattner
d52c261cf8 Implement SimplifyCFG/switch_switch_fold.ll
This case occurs many times in various benchmarks, especially when combined
with the previous patch.  This allows it to get stuff like:
  if (X == 4 || X == 3)
    if (X == 5 || X == 8)

and

switch (X) {
case 4: case 5: case 6:
  if (X == 4 || X == 5)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11797 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 07:23:58 +00:00
Chris Lattner
2189a76024 New testcase. Switch instructions that go to switch instructions should be
merged.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11796 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 07:21:09 +00:00
Alkis Evlogimenos
ec8b8bb9ab Add predicates for checking if a virtual register has a physical
register mapping or a stack slot mapping.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11795 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 06:30:36 +00:00
Chris Lattner
b4a482a098 Add some helpful methods for dealing with switch instructions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11794 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 06:26:00 +00:00
Chris Lattner
e14ea0804a Rearrange code a bit
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11793 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 05:54:22 +00:00
Chris Lattner
0d56008f53 Implement: test/Regression/Transforms/SimplifyCFG/switch_create.ll
This turns code like this:
  if (X == 4 | X == 7)
and
  if (X != 4 & X != 7)
into switch instructions.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11792 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 05:38:11 +00:00
Chris Lattner
d1eaebef49 The simplifycfg pass should be able to turn stuff like:
if (X == 4 || X == 7)
and
  if (X != 4 && X != 7)

into switch instructions.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11791 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 05:34:44 +00:00
Chris Lattner
c88c17b0f6 Wow, the description of the 'switch' instruction was out of date.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11790 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 04:54:45 +00:00
Chris Lattner
d7d6af9bef we no longer include boost
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11789 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 04:02:20 +00:00
Chris Lattner
bf81213693 Hrm, my find must have been faulty. It didn't remove these as well.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11788 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 03:54:22 +00:00
Chris Lattner
e35b06c6c4 Boost is now unneeded, thanks to the fix for PR253, contributed by Reid Spencer!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11787 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 03:53:00 +00:00
Chris Lattner
1e1a78a899 Now that's a new feature!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11786 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 03:50:24 +00:00
Chris Lattner
7b3d73b7ec Use the new LLVM is_class template instead of the boost one, allowing us to
remove our dependency on boost!  Thanks to Reid Spencer for making this possible!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11785 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 03:50:05 +00:00
Chris Lattner
5c4d53a4d1 Check in a new type_traits header which provides the mysterious is_class
template.  Thanks go out to Reid Spencer for skillfully extracting this
from boost!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11784 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 03:49:29 +00:00
Chris Lattner
c50879dfe6 Noone cares about similarity to boost
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11783 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-24 03:47:25 +00:00
Alkis Evlogimenos
e8124b9ddb Make enum private as it is an implementation detail.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11782 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-23 23:49:40 +00:00
Alkis Evlogimenos
797428719f Remove '4Virt' from member function names as it is obvious.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@11781 91177308-0d34-0410-b5e6-96231b3b80d8
2004-02-23 23:47:10 +00:00