RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-28 14:32:51 +01:00
Code Issues Projects Releases Wiki Activity
37,310 Commits 4 Branches 5 Tags 704 MiB
02f86c6890
Commit Graph

4702 Commits

Author SHA1 Message Date
Evan Cheng
02f86c6890 Refactor ExpandConstantFP so it can optimize load from constpool of types larger than f64 into extload from smaller types. 
llvm-svn: 47883
 2008-03-04 08:05:30 +00:00
Bill Wendling
c1b2e17a2c Did I say 'e = getNumOperands()'? I meant --e, of course. 
llvm-svn: 47875
 2008-03-04 00:48:15 +00:00
Evan Cheng
b5b16810ac Rename isOperand() to isOperandOf() (and other similar methods). It always confuses me. 
llvm-svn: 47872
 2008-03-04 00:41:45 +00:00
Bill Wendling
aa5ca3b98d Miscellaneous clean-ups based on Evan's feedback: 
- Cleaned up how the prologue-epilogue inserter loops over the instructions.
- Instead of restarting the processing of an instruction if we remove an
  implicit kill, just update the end iterator and make sure that the iterator
  isn't incremented.

llvm-svn: 47870
 2008-03-03 23:57:28 +00:00
Dan Gohman
ef264d521d Misc. APInt-ification in the DAGCombiner. 
llvm-svn: 47869
 2008-03-03 23:51:38 +00:00
Dan Gohman
5792bdb74e More APInt-ification. 
llvm-svn: 47868
 2008-03-03 23:35:36 +00:00
Dan Gohman
48e3b966fe Yet more APInt-ification. 
llvm-svn: 47867
 2008-03-03 22:37:52 +00:00
Dan Gohman
f372d4d982 More APInt-ification. 
llvm-svn: 47866
 2008-03-03 22:22:56 +00:00
Dan Gohman
d8d1ad659a More APInt-ification. 
llvm-svn: 47864
 2008-03-03 22:20:46 +00:00
Bill Wendling
0f7bd2c0e5 Go through the machine instruction's operands to make sure that we're not 
marking both a super- and sub-register as "killed". This removes implicit uses
that are marked as "killed".

llvm-svn: 47862
 2008-03-03 22:14:33 +00:00
Bill Wendling
b3761b1a39 Make the register scavenger update the bookkeeping values for sub/super 
registers.

llvm-svn: 47861
 2008-03-03 22:12:25 +00:00
Bill Wendling
f6b0fc44fe Multiple instructions can be inserted when eliminating frame indexes. We need 
the register scavenger to process all of those new instructions instead of just
the last one inserted.

llvm-svn: 47860
 2008-03-03 22:11:16 +00:00
Andrew Lenharth
b91c664226 all but CAS working on x86 
llvm-svn: 47798
 2008-03-01 21:52:34 +00:00
Dale Johannesen
cfd005f2ea Add MVT::is128BitVector and is64BitVector. Shrink 
unaligned load/store code using them.  Per review
of unaligned load/store vector patch.

llvm-svn: 47782
 2008-03-01 03:40:57 +00:00
Evan Cheng
1ce9ac19d3 Refactor / clean up code; remove td list scheduler special tie breaker (no real benefit). 
llvm-svn: 47779
 2008-03-01 00:39:47 +00:00
Evan Cheng
5beb036eab Don't fill eh frames even though these are text sections. 
llvm-svn: 47765
 2008-02-29 19:36:59 +00:00
Bill Wendling
d0f85a5ca3 If we reload a virtual register that's already been assigned, we want to mark 
that instruction as its "last use". This fixes PR1925.

llvm-svn: 47758
 2008-02-29 18:52:01 +00:00
Evan Cheng
3fbaf52776 Fix PR2112: don't run loop aligner if target doesn't have a TargetLowering object. 
llvm-svn: 47755
 2008-02-29 17:52:15 +00:00
Evan Cheng
f6d76f2e52 No need for coalescer to update kills. Only copies are coalesced and those instructions will be deleted. Doh. 
llvm-svn: 47749
 2008-02-29 02:50:03 +00:00
Evan Cheng
65d1b38875 Remove redundant #include. 
llvm-svn: 47748
 2008-02-29 02:49:15 +00:00
Dan Gohman
2f3d378bf6 More APInt-ification. 
llvm-svn: 47746
 2008-02-29 01:47:35 +00:00
Dan Gohman
35d58dcc00 Use the new convertFromAPInt instead of convertFromZeroExtendedInteger, 
which allows more of the surrounding arithmetic to be done with APInt
instead of uint64_t.

llvm-svn: 47745
 2008-02-29 01:44:25 +00:00
Dan Gohman
43f51f5cb2 Use the new APInt-enabled form of getConstant instead of converting 
an APInt into a uint64_t to call getConstant.

llvm-svn: 47742
 2008-02-29 01:41:59 +00:00
Evan Cheng
ea1f2d1110 Added option -align-loops=<true/false> to disable loop aligner pass. 
llvm-svn: 47736
 2008-02-28 23:29:57 +00:00
Dale Johannesen
310201ba32 Interface of getByValTypeAlignment differed between 
generic & x86 versions; change generic to follow x86
and improve comments.  Add PPC version (not right
for non-Darwin.)

llvm-svn: 47734
 2008-02-28 22:31:51 +00:00
Dale Johannesen
8e6309464b Fix an assertion message. 
llvm-svn: 47722
 2008-02-28 18:36:51 +00:00
Evan Cheng
c390fa5e5e Keep track how many commutes are performed by the scheduler. 
llvm-svn: 47710
 2008-02-28 07:40:24 +00:00
Chris Lattner
a592f5c85b implement expand for ISD::DECLARE by just deleting it. 
llvm-svn: 47708
 2008-02-28 05:53:40 +00:00
Evan Cheng
f8b1257d2e Add a quick and dirty "loop aligner pass". x86 uses it to align its loops to 16-byte boundaries. 
llvm-svn: 47703
 2008-02-28 00:43:03 +00:00
Dale Johannesen
8039d40b76 Handle load/store of misaligned vectors that are the 
same size as an int type by doing a bitconvert of
load/store of the int type (same algorithm as floating point).
This makes them work for ppc Altivec.  There was some
code that purported to handle loads of (some) vectors
by splitting them into two smaller vectors, but getExtLoad
rejects subvector loads, so this could never have worked;
the patch removes it.

llvm-svn: 47696
 2008-02-27 22:36:00 +00:00
Evan Cheng
da92e34fe3 Fix a bug in dead spill slot elimination. 
llvm-svn: 47687
 2008-02-27 19:57:11 +00:00
Dan Gohman
16ba74da61 Remove the `else', at Evan's insistence. 
llvm-svn: 47686
 2008-02-27 19:44:57 +00:00
Duncan Sands
0139087442 Add a FIXME about the VECTOR_SHUFFLE evil hack. 
llvm-svn: 47676
 2008-02-27 17:39:13 +00:00
Duncan Sands
77ff6715ad LegalizeTypes support for EXTRACT_VECTOR_ELT. The 
approach taken is different to that in LegalizeDAG
when it is a question of expanding or promoting the
result type: for example, if extracting an i64 from
a <2 x i64>, when i64 needs expanding, it bitcasts
the vector to <4 x i32>, extracts the appropriate
two i32's, and uses those for the Lo and Hi parts.
Likewise, when extracting an i16 from a <4 x i16>,
and i16 needs promoting, it bitcasts the vector to
<2 x i32>, extracts the appropriate i32, twiddles
the bits if necessary, and uses that as the promoted
value.  This puts more pressure on bitcast legalization,
and I've added the appropriate cases.  They needed to
be added anyway since users can generate such bitcasts
too if they want to.  Also, when considering various
cases (Legal, Promote, Expand, Scalarize, Split) it is
a pain that expand can correspond to Expand, Scalarize
or Split, so I've changed the LegalizeTypes enum so it
lists those different cases - now Expand only means
splitting a scalar in two.
The code produced is the same as by LegalizeDAG for
all relevant testcases, except for
2007-10-31-extractelement-i64.ll, where the code seems
to have improved (see below; can an expert please tell
me if it is better or not).
Before < vs after >.

<       subl    $92, %esp
<       movaps  %xmm0, 64(%esp)
<       movaps  %xmm0, (%esp)
<       movl    4(%esp), %eax
<       movl    %eax, 28(%esp)
<       movl    (%esp), %eax
<       movl    %eax, 24(%esp)
<       movq    24(%esp), %mm0
<       movq    %mm0, 56(%esp)
---
>       subl    $44, %esp
>       movaps  %xmm0, 16(%esp)
>       pshufd  $1, %xmm0, %xmm1
>       movd    %xmm1, 4(%esp)
>       movd    %xmm0, (%esp)
>       movq    (%esp), %mm0
>       movq    %mm0, 8(%esp)

<       subl    $92, %esp
<       movaps  %xmm0, 64(%esp)
<       movaps  %xmm0, (%esp)
<       movl    12(%esp), %eax
<       movl    %eax, 28(%esp)
<       movl    8(%esp), %eax
<       movl    %eax, 24(%esp)
<       movq    24(%esp), %mm0
<       movq    %mm0, 56(%esp)
---
>       subl    $44, %esp
>       movaps  %xmm0, 16(%esp)
>       pshufd  $3, %xmm0, %xmm1
>       movd    %xmm1, 4(%esp)
>       movhlps %xmm0, %xmm0
>       movd    %xmm0, (%esp)
>       movq    (%esp), %mm0
>       movq    %mm0, 8(%esp)

<       subl    $92, %esp
<       movaps  %xmm0, 64(%esp)
---
>       subl    $44, %esp

<       movl    16(%esp), %eax
<       movl    %eax, 48(%esp)
<       movl    20(%esp), %eax
<       movl    %eax, 52(%esp)
<       movaps  %xmm0, (%esp)
<       movl    4(%esp), %eax
<       movl    %eax, 60(%esp)
<       movl    (%esp), %eax
<       movl    %eax, 56(%esp)
---
>       pshufd  $1, %xmm0, %xmm1
>       movd    %xmm1, 4(%esp)
>       movd    %xmm0, (%esp)
>       movd    %xmm1, 12(%esp)
>       movd    %xmm0, 8(%esp)

<       subl    $92, %esp
<       movaps  %xmm0, 64(%esp)
---
>       subl    $44, %esp

<       movl    24(%esp), %eax
<       movl    %eax, 48(%esp)
<       movl    28(%esp), %eax
<       movl    %eax, 52(%esp)
<       movaps  %xmm0, (%esp)
<       movl    12(%esp), %eax
<       movl    %eax, 60(%esp)
<       movl    8(%esp), %eax
<       movl    %eax, 56(%esp)
---
>       pshufd  $3, %xmm0, %xmm1
>       movd    %xmm1, 4(%esp)
>       movhlps %xmm0, %xmm0
>       movd    %xmm0, (%esp)
>       movd    %xmm1, 12(%esp)
>       movd    %xmm0, 8(%esp)

llvm-svn: 47672
 2008-02-27 13:34:40 +00:00
Duncan Sands
4c1f3c0c5b LegalizeTypes support for legalizing the mask 
operand of a VECTOR_SHUFFLE.  The mask is a
vector of constant integers.  The code in
LegalizeDAG doesn't bother to legalize the
mask, since it's basically just storage for
a bunch of constants, however LegalizeTypes
is more picky.  The problem is that there may
not exist any legal vector-of-integers type
with a legal element type, so it is impossible
to create a legal mask!  Unless of course you
cheat by creating a BUILD_VECTOR where the
operands have a different type to the element
type of the vector being built...  This is
pretty ugly but works - all relevant tests in
the testsuite pass, and produce the same
assembler with and without LegalizeTypes.

llvm-svn: 47670
 2008-02-27 13:03:44 +00:00
Duncan Sands
cadfe810f3 LegalizeTypes support for INSERT_VECTOR_ELT. 
llvm-svn: 47669
 2008-02-27 10:18:23 +00:00
Evan Cheng
295ae42ede Don't track max alignment during stack object allocations since they can be deleted later. Let PEI compute it. 
llvm-svn: 47668
 2008-02-27 10:04:56 +00:00
Duncan Sands
f8ac836240 Support for legalizing MEMBARRIER. 
llvm-svn: 47667
 2008-02-27 08:53:44 +00:00
Bill Wendling
2cae66e28b Final de-tabification. 
llvm-svn: 47663
 2008-02-27 06:33:05 +00:00
Evan Cheng
7553230e3a Spiller now remove unused spill slots. 
llvm-svn: 47657
 2008-02-27 03:04:06 +00:00
Dan Gohman
2042802d32 Teach Legalize how to expand an EXTRACT_ELEMENT. 
llvm-svn: 47656
 2008-02-27 01:52:30 +00:00
Dan Gohman
938e74654b Convert the last remaining users of the non-APInt form of 
ComputeMaskedBits to use the APInt form, and remove the
non-APInt form.

llvm-svn: 47654
 2008-02-27 01:23:58 +00:00
Dan Gohman
689d8cac04 Convert SimplifyDemandedMask and ShrinkDemandedConstant to use APInt. 
Change several cases in SimplifyDemandedMask that don't ever do any
simplifying to reuse the logic in ComputeMaskedBits instead of
duplicating it.

llvm-svn: 47648
 2008-02-27 00:25:32 +00:00
Chris Lattner
6318b4aee9 Use a smallvector for inactiveCounts and initialize it lazily 
instead of init'ing it maximally to zeros on entry.  getFreePhysReg
is pretty hot and only a few elements are typically used.  This speeds
up linscan by 5% on 176.gcc.

llvm-svn: 47631
 2008-02-26 22:08:41 +00:00
Bill Wendling
8fb166bf6c Rename PrintableName to Name. 
llvm-svn: 47629
 2008-02-26 21:47:57 +00:00
Bill Wendling
50f5c4be14 Change "Name" to "AsmName" in the target register info. Gee, a refactoring tool 
would have been a Godsend here!

llvm-svn: 47625
 2008-02-26 21:11:01 +00:00
Evan Cheng
701b6a1dc3 Enable -coalescer-commute-instrs by default. 
llvm-svn: 47623
 2008-02-26 20:40:22 +00:00
Dan Gohman
8a8f3fe7e0 Avoid aborting on invalid shift counts. 
llvm-svn: 47612
 2008-02-26 18:50:50 +00:00
Chris Lattner
1a461075ef Fix PR2096, a regression introduced with my patch last night. This 
also fixes cfrac, flops, and 175.vpr

llvm-svn: 47605
 2008-02-26 17:09:59 +00:00
Duncan Sands
c63bc1577a Fix a nasty bug in LegalizeTypes (spotted in 
CodeGen/PowerPC/illegal-element-type.ll): suppose
a node X is processed, and processing maps it to
a node Y.  Then X continues to exist in the DAG,
but with no users.  While processing some other
node, a new node may be created that happens to
be equal to X, and thus X will be reused rather
than a truly new node.  This can cause X to
"magically reappear", and since it is in the
Processed state in will not be reprocessed, so
at the end of type legalization the illegal node
X can still be present.  The solution is to replace
X with Y whenever X gets resurrected like this.

llvm-svn: 47601
 2008-02-26 11:21:42 +00:00