Commit Graph

2035 Commits

Author SHA1 Message Date
Dale Johannesen
e0e6fac5e9 Redo previous patch so optimization only done for i1.
Simpler and safer.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44663 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 17:53:31 +00:00
Chris Lattner
d3fd6d2b25 third time around: instead of disabling this completely,
only disable it if we don't know it will be obviously profitable.
Still fixme, but less so. :)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44658 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 07:47:55 +00:00
Chris Lattner
e00cbad78d Actually, disable this code for now. More analysis and improvements to
the X86 backend are needed before this should be enabled by default.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44657 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 07:44:31 +00:00
Chris Lattner
e70da20f9a implement a readme entry, compiling the code into:
_foo:
	movl	$12, %eax
	andl	4(%esp), %eax
	movl	_array(%eax), %eax
	ret

instead of:

_foo:
	movl	4(%esp), %eax
	shrl	$2, %eax
	andl	$3, %eax
	movl	_array(,%eax,4), %eax
	ret

As it turns out, this triggers all the time, in a wide variety of
situations, for example, I see diffs like this in various programs:

-       movl    8(%eax), %eax
-       shll    $2, %eax
-       andl    $1020, %eax
-       movl    (%esi,%eax), %eax
+       movzbl  8(%eax), %eax
+       movl    (%esi,%eax,4), %eax


-       shll    $2, %edx
-       andl    $1020, %edx
-       movl    (%edi,%edx), %edx
+       andl    $255, %edx
+       movl    (%edi,%edx,4), %edx

Unfortunately, I also see stuff like this, which can be fixed in the
X86 backend:

-       andl    $85, %ebx
-       addl    _bit_count(,%ebx,4), %ebp
+       shll    $2, %ebx
+       andl    $340, %ebx
+       addl    _bit_count(%ebx), %ebp



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44656 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 07:33:36 +00:00
Chris Lattner
72b2df1f77 implement the rest of the functionality from SelectionDAGLegalize::ScalarizeVectorOp
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44654 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 05:53:43 +00:00
Dale Johannesen
59f1e97ef3 Fix PR1842.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44649 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 01:43:46 +00:00
Chris Lattner
60fb5ecccc more scalarization
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44608 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-05 07:45:02 +00:00
Chris Lattner
d36db029c8 scalarize vector binops
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44607 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-05 07:36:58 +00:00
Chris Lattner
eb1819d1e5 Implement framework for scalarizing node results. This is sufficient
to codegen this:

define float @test_extract_elt(<1 x float> * %P) {
	%p = load <1 x float>* %P
	%R = extractelement <1 x float> %p, i32 0
	ret float %R
}




git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44570 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-04 07:48:46 +00:00
Chris Lattner
13c207b5c8 start providing framework for scalarizing vectors.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44569 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-04 07:29:51 +00:00
Duncan Sands
a3355ffb3d Rather than having special rules like "intrinsics cannot
throw exceptions", just mark intrinsics with the nounwind
attribute.  Likewise, mark intrinsics as readnone/readonly
and get rid of special aliasing logic (which didn't use
anything more than this anyway).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44544 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-03 20:06:50 +00:00
Duncan Sands
afa3b6da11 Add some convenience methods for querying attributes, and
use them.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44403 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-28 17:07:01 +00:00
Nate Begeman
d73ab8884f Support returning non-power-of-2 vectors to unblock some work
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44371 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-27 19:28:48 +00:00
Duncan Sands
dc024674ff Fix PR1146: parameter attributes are longer part of
the function type, instead they belong to functions
and function calls.  This is an updated and slightly
corrected version of Reid Spencer's original patch.
The only known problem is that auto-upgrading of
bitcode files doesn't seem to work properly (see
test/Bitcode/AutoUpgradeIntrinsics.ll).  Hopefully
a bitcode guru (who might that be? :) ) will fix it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44359 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-27 13:23:08 +00:00
Chris Lattner
b3d9cdb677 err, no really.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44352 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-27 06:14:32 +00:00
Chris Lattner
5ebfaa2c9d don't depend on ADL.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44351 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-27 06:14:12 +00:00
Dan Gohman
77003040b2 Don't lower srem/urem X%C to X-X/C*C unless the division is actually
optimized. This avoids creating illegal divisions when the combiner is
running after legalize; this fixes PR1815. Also, it produces better
code in the included testcase by avoiding the subtract and multiply
when the division isn't optimized.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44341 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-26 23:46:11 +00:00
Chris Lattner
c58d558a79 Implement expand support for MERGE_VALUEs that only produces one result.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44304 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-24 19:12:15 +00:00
Chris Lattner
87726ac0f9 Implement support for custom legalization in DAGTypeLegalizer::ExpandOperand.
Improve a comment.
Unbreak Duncan's carefully written path compression where I didn't realize
what was happening!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44301 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-24 18:11:42 +00:00
Chris Lattner
27a6c7380f Several changes:
1) Change the interface to TargetLowering::ExpandOperationResult to 
   take and return entire NODES that need a result expanded, not just
   the value.  This allows us to handle things like READCYCLECOUNTER,
   which returns two values.
2) Implement (extremely limited) support in LegalizeDAG::ExpandOp for MERGE_VALUES.
3) Reimplement custom lowering in LegalizeDAGTypes in terms of the new
   ExpandOperationResult.  This makes the result simpler and fully 
   general.
4) Implement (fully general) expand support for MERGE_VALUES in LegalizeDAGTypes.
5) Implement ExpandOperationResult support for ARM f64->i64 bitconvert and ARM
   i64 shifts, allowing them to work with LegalizeDAGTypes.
6) Implement ExpandOperationResult support for X86 READCYCLECOUNTER and FP_TO_SINT,
   allowing them to work with LegalizeDAGTypes.

LegalizeDAGTypes now passes several more X86 codegen tests when enabled and when
type legalization in LegalizeDAG is ifdef'd out.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44300 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-24 07:07:01 +00:00
Duncan Sands
baccd5872e Fix a bug in which node A is replaced by node B, but later
node A gets back into the DAG again because it was hiding in
one of the node maps: make sure that node replacement happens
in those maps too.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44263 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-21 16:43:19 +00:00
Chris Lattner
e400af83b3 ExpandUnalignedLoad doesn't handle vectors right at all apparently.
Fix a couple of problems:
1. Don't assume the VT-1 is a VT that is half the size.
2. Treat vectors of FP in the vector path, not the FP path.

This has a couple of remaining problems before it will work with
the code in PR1811: the code below this change assumes that it can
use extload/shift/or to construct the result, which isn't right for
vectors.

This also doesn't handle vectors of 1 or vectors that aren't pow-2.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44243 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-19 21:38:03 +00:00
Chris Lattner
6c9c6800b8 Implement vector expand support for shuffle_vector. This fixes PR1811.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44242 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-19 21:16:54 +00:00
Chris Lattner
daf9bc8eee Implement splitting of UNDEF nodes. This is the first step towards fixing PR1811
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44239 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-19 20:21:32 +00:00
Dan Gohman
089617d9e3 Add support in SplitVectorOp for remainder operators.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44233 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-19 15:15:03 +00:00
Nate Begeman
b348d18caf Add support for vectors to int <-> float casts.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44204 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-17 03:58:34 +00:00
Anton Korobeynikov
917c2a6ff2 Implement necessary bits for flt_rounds gcc builtin.
Codegen bits and llvm-gcc support will follow.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44182 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-15 23:25:33 +00:00
Nate Begeman
5db1afb462 Basic non-power-of-2 vector support
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44181 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-15 21:15:26 +00:00
Duncan Sands
560a737e07 This assertion was bogus.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44167 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-15 09:54:37 +00:00
Bill Wendling
0f8d9c04d9 Unify CALLSEQ_{START,END}. They take 4 parameters: the chain, two stack
adjustment fields, and an optional flag. If there is a "dynamic_stackalloc" in
the code, make sure that it's bracketed by CALLSEQ_START and CALLSEQ_END. If
not, then there is the potential for the stack to be changed while the stack's
being used by another instruction (like a call).

This can only result in tears...


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44037 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-13 00:44:25 +00:00
Duncan Sands
fd617d0143 Move MinAlign to MathExtras.h.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43944 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-09 13:41:39 +00:00
Duncan Sands
c6fa170b4d Fix some load/store logic that would be wrong for
apints on big-endian machines if the bitwidth is
not a multiple of 8.  Introduce a new helper,
MVT::getStoreSizeInBits, and use it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43934 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-09 08:57:19 +00:00
Evan Cheng
cc41586b9d Much improved pic jumptable codegen:
Then:
        call    "L1$pb"
"L1$pb":
        popl    %eax
		...
LBB1_1: # entry
        imull   $4, %ecx, %ecx
        leal    LJTI1_0-"L1$pb"(%eax), %edx
        addl    LJTI1_0-"L1$pb"(%ecx,%eax), %edx
        jmpl    *%edx

        .align  2
        .set L1_0_set_3,LBB1_3-LJTI1_0
        .set L1_0_set_2,LBB1_2-LJTI1_0
        .set L1_0_set_5,LBB1_5-LJTI1_0
        .set L1_0_set_4,LBB1_4-LJTI1_0
LJTI1_0:
        .long    L1_0_set_3
        .long    L1_0_set_2

Now:
        call    "L1$pb"
"L1$pb":
        popl    %eax
		...
LBB1_1: # entry
        addl    LJTI1_0-"L1$pb"(%eax,%ecx,4), %eax
        jmpl    *%eax

		.align  2
		.set L1_0_set_3,LBB1_3-"L1$pb"
		.set L1_0_set_2,LBB1_2-"L1$pb"
		.set L1_0_set_5,LBB1_5-"L1$pb"
		.set L1_0_set_4,LBB1_4-"L1$pb"
LJTI1_0:
        .long    L1_0_set_3
        .long    L1_0_set_2


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43924 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-09 01:32:10 +00:00
Evan Cheng
2dfdefd282 Didn't mean to check these in.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43923 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-09 01:28:33 +00:00
Evan Cheng
7da8f399bf Bug fix. Passive nodes are not in SUnitMap.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43922 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-09 01:27:11 +00:00
Evan Cheng
4471194d69 If both parts of smul_lohi, etc. are used, don't simplify. If only one part is used, try simplify it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43888 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-08 09:25:29 +00:00
Dan Gohman
0d97426b7b Remainder operations must be either integer or floating-point.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43781 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-06 22:11:54 +00:00
Evan Cheng
1fd15ba961 Add pseudo dependency to force two-address instruction to be scheduled after
other uses. There was a overly restricted check that prevented some obvious
cases.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43762 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-06 08:44:59 +00:00
Dan Gohman
80176319f3 Add support for vector remainder operations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43744 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-05 23:35:22 +00:00
Rafael Espindola
f1ba1cad38 Move the LowerMEMCPY and LowerMEMCPYCall to a common place.
Thanks for the suggestions Bill :-)



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43742 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-05 23:12:20 +00:00
Dale Johannesen
eb57ea7ea2 Make labels work in asm blocks; allow labels as
parameters.  Rename ValueRefList to ParamList
in AsmParser, since its only use is for parameters.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43734 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-05 21:20:28 +00:00
Dan Gohman
111c4f897e Add std:: to sort calls.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43652 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-02 22:24:01 +00:00
Dan Gohman
b61f2f061f Change illegal uses of ++ to uses of STLExtra.h's next function.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43651 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-02 22:22:02 +00:00
Duncan Sands
e414cdacb4 Fix a thinko.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43639 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-02 15:18:06 +00:00
Duncan Sands
514ab348fd Executive summary: getTypeSize -> getTypeStoreSize / getABITypeSize.
The meaning of getTypeSize was not clear - clarifying it is important
now that we have x86 long double and arbitrary precision integers.
The issue with long double is that it requires 80 bits, and this is
not a multiple of its alignment.  This gives a primitive type for
which getTypeSize differed from getABITypeSize.  For arbitrary precision
integers it is even worse: there is the minimum number of bits needed to
hold the type (eg: 36 for an i36), the maximum number of bits that will
be overwriten when storing the type (40 bits for i36) and the ABI size
(i.e. the storage size rounded up to a multiple of the alignment; 64 bits
for i36).

This patch removes getTypeSize (not really - it is still there but
deprecated to allow for a gradual transition).  Instead there is:

(1) getTypeSizeInBits - a number of bits that suffices to hold all
values of the type.  For a primitive type, this is the minimum number
of bits.  For an i36 this is 36 bits.  For x86 long double it is 80.
This corresponds to gcc's TYPE_PRECISION.

(2) getTypeStoreSizeInBits - the maximum number of bits that is
written when storing the type (or read when reading it).  For an
i36 this is 40 bits, for an x86 long double it is 80 bits.  This
is the size alias analysis is interested in (getTypeStoreSize
returns the number of bytes).  There doesn't seem to be anything
corresponding to this in gcc.

(3) getABITypeSizeInBits - this is getTypeStoreSizeInBits rounded
up to a multiple of the alignment.  For an i36 this is 64, for an
x86 long double this is 96 or 128 depending on the OS.  This is the
spacing between consecutive elements when you form an array out of
this type (getABITypeSize returns the number of bytes).  This is
TYPE_SIZE in gcc.

Since successive elements in a SequentialType (arrays, pointers
and vectors) need to be aligned, the spacing between them will be
given by getABITypeSize.  This means that the size of an array
is the length times the getABITypeSize.  It also means that GEP
computations need to use getABITypeSize when computing offsets.
Furthermore, if an alloca allocates several elements at once then
these too need to be aligned, so the size of the alloca has to be
the number of elements multiplied by getABITypeSize.  Logically
speaking this doesn't have to be the case when allocating just
one element, but it is simpler to also use getABITypeSize in this
case.  So alloca's and mallocs should use getABITypeSize.  Finally,
since gcc's only notion of size is that given by getABITypeSize, if
you want to output assembler etc the same as gcc then getABITypeSize
is the size you want.

Since a store will overwrite no more than getTypeStoreSize bytes,
and a read will read no more than that many bytes, this is the
notion of size appropriate for alias analysis calculations.

In this patch I have corrected all type size uses except some of
those in ScalarReplAggregates, lib/Codegen, lib/Target (the hard
cases).  I will get around to auditing these too at some point,
but I could do with some help.

Finally, I made one change which I think wise but others might
consider pointless and suboptimal: in an unpacked struct the
amount of space allocated for a field is now given by the ABI
size rather than getTypeStoreSize.  I did this because every
other place that reserves memory for a type (eg: alloca) now
uses getABITypeSize, and I didn't want to make an exception
for unpacked structs, i.e. I did it to make things more uniform.
This only effects structs containing long doubles and arbitrary
precision integers.  If someone wants to pack these types more
tightly they can always use a packed struct.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43620 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-01 20:53:16 +00:00
Duncan Sands
7169a2f9e8 Promotion of sdiv/srem/udiv/urem.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43551 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-31 08:57:43 +00:00
Dale Johannesen
25f1d08619 Make i64=expand_vector_elt(v2i64) work in 32-bit mode.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43535 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-31 00:32:36 +00:00
Evan Cheng
de1631bc85 Typo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43511 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-30 20:11:21 +00:00
Duncan Sands
01405f9284 Add support for expanding trunc stores. Consider
storing an i170 on a 32 bit machine.  This is first
promoted to a trunc-i170 store of an i256.  On a
little-endian machine this expands to a store of
an i128 and a trunc-i42 store of an i128.  The
trunc-i42 store is further expanded to a trunc-i42
store of an i64, then to a store of an i32 and a
trunc-i10 store of an i32.  At this point the operand
type is legal (i32) and expansion stops (legalization
of the trunc-i10 needs to be handled in LegalizeDAG.cpp).
On big-endian machines the high bits are stored first,
and some bit-fiddling is needed in order to generate
aligned stores.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43499 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-30 12:50:39 +00:00
Duncan Sands
ba3b1d10fd If a call to getTruncStore is for a normal store,
offload to getStore rather than trying to handle
both cases at once (the assertions for example
assume the store really is truncating).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43498 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-30 12:40:58 +00:00