Commit Graph

30466 Commits

Author SHA1 Message Date
Chris Lattner
6e50f8cfe9 switch to smallvector
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34633 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 07:59:53 +00:00
Chris Lattner
ab143beb93 another missing header :( :( :(
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34632 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 07:51:00 +00:00
Chris Lattner
14dd4c9735 initial hack at splitting the x86-64 calling convention info out from the
mechanics that process it.  I'm still not happy with this, but it's a step
in the right direction.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34631 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 07:50:02 +00:00
Reid Spencer
ebf4ebd691 Make isNegative() a const function since it doesn't modify the APInt.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34630 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 07:45:40 +00:00
Reid Spencer
47fbe9e00c 1. Remove redundant calls to clearUsedBits().
2. Fix countTrailingZeros to use a faster algorithm.
3. Simplify sext() slightly by using isNegative().
4. Implement ashr using word-at-a-time logic instead of bit-at-a-time
5. Rename locals named isNegative so they don't clash with method name.
6. Fix fromString to compute negated value correctly.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34629 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 07:44:38 +00:00
Chris Lattner
cc7b07a0ff the truncate must always be done, it's only the assert that is conditional.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34628 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 05:21:05 +00:00
Chris Lattner
00f23ec5c2 reapply my previous patch with a bugfix.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34627 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 05:02:39 +00:00
Chris Lattner
e2c2b76c8a revert my previous change, something strange is happening.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34626 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 04:43:19 +00:00
Chris Lattner
54e3efde46 add an accessor.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34625 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 04:01:25 +00:00
Chris Lattner
114e5dd464 emit an enum value for the # of target registers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34624 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 03:34:38 +00:00
Chris Lattner
df5cdb5288 in X86-64 CCC, i8/i16 arguments are already properly zext/sext'd on input.
Capture this so that downstream zext/sext's are optimized out.  This
compiles:
  int test(short X) { return (int)X; }

to:

_test:
        movl %edi, %eax
        ret

instead of:

_test:
        movswl %di, %eax
        ret


GCC produces this bizarre code:

_test:
        movw    %di, -12(%rsp)
        movswl  -12(%rsp),%eax
        ret


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34623 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 03:18:56 +00:00
Chris Lattner
6c0880c624 new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34622 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 03:16:20 +00:00
Chris Lattner
2255887574 Fold (sext (truncate x)) more aggressively, by avoiding creation of a
sextinreg if not needed.   This is useful in two cases: before legalize,
it avoids creating a sextinreg that will be trivially removed.  After legalize
if the target doesn't support sextinreg, the trunc/sext would not have been
removed before.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34621 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 03:13:59 +00:00
Chris Lattner
ddf53e45b2 track signedness of formal argument, though we have a fixme here.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34620 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 02:56:58 +00:00
Reid Spencer
dcffd5c3d9 Add an isNegative method to determine if the APInt's value is negative.
This is much less expensive than a test against zero.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34619 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 01:20:59 +00:00
Reid Spencer
ba81c2b871 Rewrite lshr to not do bit by bit copy but to copy and shift whole words.
This makes it much more efficient.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34618 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-26 01:19:48 +00:00
Reid Spencer
f30b1885ae Fix sext operation. Shifting by zero would leave an incorrect mask.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34617 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 23:54:00 +00:00
Reid Spencer
9eec241347 1. Fix the flip() method to correctly flip all words of the APInt.
2. Implement the trunc, sext, and zext operations.
3. Improve fromString to accept negative values as input.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34616 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 23:44:53 +00:00
Chris Lattner
c30654cc96 Fix an X86-64 abi bug. We now compile:
void foo(short);
void bar(unsigned short A) {
  foo(A);
}

into:

_bar:
        subq $8, %rsp
        movswl %di, %edi
        call _foo
        addq $8, %rsp
        ret

instead of:

_bar:
        subq $8, %rsp
        call _foo
        addq $8, %rsp
        ret

Testcase here: test/CodeGen/X86/x86-64-shortint.ll


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34615 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 23:10:46 +00:00
Chris Lattner
df6b57ad48 new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34614 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 23:08:29 +00:00
Chris Lattner
7e22977c0f fix CodeGen/X86/2007-02-25-FastCCStack.ll, a regression from my patch last
night:  fastcc returns should only go in XMM0 if we have SSE2 or above.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34613 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 22:23:46 +00:00
Chris Lattner
f82435d4d7 new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34612 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 22:23:15 +00:00
Chris Lattner
876e5942bc new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34611 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 22:02:01 +00:00
Jim Laskey
183f47fb1a Fix for PR1224.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34610 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 21:43:59 +00:00
Jim Laskey
98770d4ed2 Test for PR1224.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34609 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 21:43:21 +00:00
Chris Lattner
1cf5574636 Rework GlobalValue::removeDeadConstantUsers to always remove dead constant
exprs hanging off a global, even if the global is not otherwise dead.  This
requires some tricky iterator gymnastics.

This implements Transforms/GlobalOpt/constantexpr-dangle.ll by deleting a
constantexpr that made it appear that the address of the function was taken.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34608 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 21:06:13 +00:00
Chris Lattner
2aef09a06b new testcase. @foo should be marked fastcc by globalopt
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34607 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 21:04:39 +00:00
Chris Lattner
3f3098c8ca disable some noisy debug output
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34606 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 20:42:59 +00:00
Chris Lattner
4ddf7a4ca6 no really, this is the right patch
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34605 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 20:01:40 +00:00
Chris Lattner
640c0ac01d always promote float varargs to double.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34604 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 19:59:18 +00:00
Reid Spencer
5d0d05c9b8 1. Provide more detail in file comment.
2. Move comments for methods to .h file, delete them in .cpp file.
3. All places that were doing manual clear of high order bits now call the
   clearUnusedBits() method in order to not depend on undefined behavior
   of the >> operator when the number of bits shifted equals the word size.
4. Reduced # of loc by using the new result of clearUnusedBits() method.
5. Simplified logic (decreased indentation) in a few places.
6. Added code comments to larger functions that needed them.
7. Added FIXME notes about weak implementations of things (e.g. bit-by-bit
   shift right is sub-optimal).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34603 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 19:32:03 +00:00
Reid Spencer
c68c2243dc Fix clearUnusedBits to not depend on "undefined behavior" of >> operator
when the bit size is equal to the word size. This happens to work out okay
on x86, but might not on other platforms. The change just detects when
there are no bits to clear (because BitWidth is a multiple of the word size)
and returns early.

Also, move some comments from .cpp file into header.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34602 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 19:26:01 +00:00
Chris Lattner
1fa3d9ef93 one important bugfix: PPC32 didn't have both elf and macho support for
external symbols and global addresses.  Add the missing ones.

one important workaround: PPCISD::CALL is matched by both PPCcall_ELF
and PPCcall_Macho, disable the _ELF patterns for now.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34601 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 19:20:53 +00:00
Chris Lattner
c25904f23a add -enable-eh
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34600 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 18:50:48 +00:00
Chris Lattner
251db1890c optimize duplicate ValueMap lookups
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34599 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 18:40:32 +00:00
Dale Johannesen
ce74de46dd cosmetic changes from review of last patch. obvious
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34598 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 18:31:31 +00:00
Evan Cheng
de4e942faa A couple of more places where a register liveness has been extended and its last kill should be updated accordingly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34597 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:51:27 +00:00
Evan Cheng
0badfea274 Add an assertion.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34596 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:47:31 +00:00
Evan Cheng
48ef398ebd Fix a couple of bugs related IsDead back propagation during coalescing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34595 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:46:31 +00:00
Evan Cheng
16191f0333 If the liveinterval of the source instruction has been extended, remove the IsKill marker.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34594 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:41:59 +00:00
Evan Cheng
bc025fbb64 Only add liveinterval to livein set if it isn't assigned a stack slot.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34593 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:39:02 +00:00
Chris Lattner
5e265b29ea fastcc functions that return double values now return them in xmm0 on x86-32.
This implements CodeGen/X86/fp-stack-ret.ll:test[23]


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34592 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:31:16 +00:00
Chris Lattner
113296ba51 verify that double is returned in XMM0 if the function is fastcc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34591 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:30:03 +00:00
Chris Lattner
4dfccf4b48 allow vectors to be passed to stdcall/fastcall functions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34590 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:14:25 +00:00
Chris Lattner
2a9bdd74f2 move LowerRET into the 'Return Value Calling Convention Implementation'
section of the file.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34589 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:12:39 +00:00
Chris Lattner
339b439d47 make all Lower*CallTo implementations use LowerCallResult to handle their
result value stuff.  This eliminates a bunch of duplicated code and now
GetRetValueLocs is the sole place that decides where a value is returned.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34588 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:10:05 +00:00
Chris Lattner
09c75a4372 pass the calling convention into Lower*CallTo, instead of using ad-hoc flags.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34587 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 09:06:15 +00:00
Chris Lattner
3085e15117 factor a bunch of code out of LowerCCCCallTo into a new LowerCallResult
function.  This function now uses GetRetValueLocs to determine *where*
the result values are located and concerns itself with *how* to pull the
values out.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34586 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 08:59:22 +00:00
Chris Lattner
2b02a4409f move some code around, pass in calling conv, even though it is unused
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34585 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 08:29:00 +00:00
Chris Lattner
753d9cb50e fold trivial token factor nodes. This allows us to compile
test/CodeGen/X86/fp-stack-ret.ll into:

        movl 4(%esp), %eax
        fldl (%eax)
        ret

instead of:

        subl $12, %esp
        movl 16(%esp), %eax
        movsd (%eax), %xmm0
        movsd %xmm0, (%esp)
        fldl (%esp)
        addl $12, %esp
        ret

by eliminating a token factor that blocked a check.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@34584 91177308-0d34-0410-b5e6-96231b3b80d8
2007-02-25 08:24:27 +00:00