6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2024-11-01 00:11:00 +00:00
Code Issues Projects Releases Wiki Activity
23,815 Commits 39 Branches 0 Tags 277 MiB
a1c3538537
Commit Graph

5645 Commits

Author SHA1 Message Date
Evan Cheng
ed4ca7f6c3 Missed X86::isUNPCKHMask 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27222 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 08:27:15 +00:00
Evan Cheng
9bbfd4f68c movlps and movlpd should be modeled as two address code. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27221 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 07:01:28 +00:00
Evan Cheng
6a6d354d4c Update 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27220 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 06:55:45 +00:00
Evan Cheng
be296ac5ca Typo 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27219 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 06:53:49 +00:00
Evan Cheng
2064a2b47e * Prefer using operation of matching types. e.g unpcklpd rather than movlhps. 
* Bug fixes.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27218 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 06:50:32 +00:00
Nate Begeman
816cee2216 Fix a couple typos 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27216 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 04:18:18 +00:00
Nate Begeman
98e70cc124 Add a few more altivec intrinsics 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27215 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 04:15:58 +00:00
Evan Cheng
4c4a2e2a0d Added a couple of entries about movhps and movlhps. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27212 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 02:49:12 +00:00
Evan Cheng
000e4dd54b All unpack cases are now being handled. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27211 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 02:44:05 +00:00
Evan Cheng
4fcb922c70 - Clean up / consoladate various shuffle masks. 
- Some misc. bug fixes.
- Use MOVHPDrm to load from m64 to upper half of a XMM register.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27210 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 02:43:26 +00:00
Chris Lattner
ecc219b8d4 implement a bunch more intrinsics. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27209 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 02:29:37 +00:00
Chris Lattner
7f20b13518 Use normal lvx for scalar_to_vector instead of lve*x. They do the exact 
same thing and we have a dag node for the former.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27205 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 01:43:22 +00:00
Chris Lattner
48b61a729d Tblgen doesn't like multiple SDNode<> definitions that map to the sameenum value. Split them into separate enums. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27201 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 00:40:33 +00:00
Evan Cheng
0038e59803 Model unpack lower and interleave as vector_shuffle so we can lower the 
intrinsics as such.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27200 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-28 00:39:58 +00:00
Jim Laskey
414e682bac Translate llvm target registers to dwarf register numbers properly. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27180 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-27 20:18:45 +00:00
Chris Lattner
749b758b2e unbreak the build 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27174 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-27 16:52:45 +00:00
Chris Lattner
3ee9ffb0e5 Add a bunch of notes from my journey thus far. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27170 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-27 07:41:00 +00:00
Chris Lattner
b86bd2cee2 Split out altivec notes into their own README 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27168 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-27 07:04:16 +00:00
Evan Cheng
a0b3afbe14 Use pcmpeq to generate vector of all ones. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27167 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-27 07:00:16 +00:00
Evan Cheng
999f3b538f Changed isBuildVectorAllOnesInteger to isBuildVectorAllOnes. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27166 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-27 06:59:32 +00:00
Chris Lattner
fb143ce459 Fix the JIT encoding of VSEL 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27160 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-27 03:34:17 +00:00
Chris Lattner
eeaf72af39 Fix the JIT encoding of VSPLTI* 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27159 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-27 03:28:57 +00:00
Nate Begeman
f15485a8d0 SelectionDAGISel can now natively handle Switch instructions, in the same 
manner that the LowerSwitch LLVM to LLVM pass does: emitting a binary
search tree of basic blocks.  The new approach has several advantages:
it is faster, it generates significantly smaller code in many cases, and
it paves the way for implementing dense switch tables as a jump table by
handling switches directly in the instruction selector.

This functionality is currently only enabled on x86, but should be safe for
every target.  In anticipation of making it the default, the cfg is now
properly updated in the x86, ppc, and sparc select lowering code.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27156 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-27 01:32:24 +00:00
Chris Lattner
bd6be6f52d add vsel 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27153 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-26 22:38:43 +00:00
Nate Begeman
c02e5a8668 Readme note 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27152 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-26 19:19:27 +00:00
Chris Lattner
6d92caddc4 Codegen vector predicate compares. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27151 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-26 10:06:40 +00:00
Evan Cheng
ffea91e522 Remove X86:isZeroVector, use ISD::isBuildVectorAllZeros instead; some fixes / cleanups 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27150 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-26 09:53:12 +00:00
Evan Cheng
5b6a01b59c Remove PPC:isZeroVector, use ISD::isBuildVectorAllZeros instead 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27149 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-26 09:52:32 +00:00
Evan Cheng
e506f9ae35 Add immAllZerosV helper 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27148 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-26 09:51:39 +00:00
Chris Lattner
b8a45c2798 Add all of the altivec comparison instructions. Add patterns for the 
non-predicate altivec compare intrinsics.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27143 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-26 04:57:17 +00:00
Chris Lattner
5d72907e00 Add and 8/16-bit adds, add all integer subtracts, add saturating subtract 
intrinsics.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27142 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-26 02:39:02 +00:00
Chris Lattner
e7d959c069 implement the vsldoi intrinsic. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27139 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-26 00:41:48 +00:00
Chris Lattner
af9136bc0c fix the pattern for vandc, it's NOT vnand 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27136 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 23:10:40 +00:00
Chris Lattner
6509ae859a add patterns for VANDC/VNOR, implementing 
CodeGen/PowerPC/eqv-andc-orc-nor.ll:VNOR/VANDC


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27135 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 23:05:29 +00:00
Chris Lattner
c985d828bc add a vnot helper node for matching 'not' on vectors 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27132 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 23:00:08 +00:00
Chris Lattner
2430a5f0c7 Add some logical operations 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27127 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 22:16:05 +00:00
Evan Cheng
6e16ee5634 Added missing (any_extend (load ...)) patterns. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27120 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 09:45:48 +00:00
Evan Cheng
c60bd97b94 Build arbitrary vector with more than 2 distinct scalar elements with a 
series of unpack and interleave ops.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27119 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 09:37:23 +00:00
Chris Lattner
984f38bf4f implement a bunch of intrinsics 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27118 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 08:01:02 +00:00
Chris Lattner
b22a04d881 Move all Altivec stuff out into a new PPCInstrAltivec.td file. 
Add a bunch of patterns for different datatypes, e.g. bit_convert, undef and
zero vector support.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27117 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 07:51:43 +00:00
Chris Lattner
8d052bc711 Add some basic patterns for other datatypes 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27116 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 07:39:07 +00:00
Chris Lattner
150ffa7842 add all supported formats to the vector register file 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27115 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 07:36:56 +00:00
Chris Lattner
5a2025465b Add support for __builtin_altivec_vnmsubfp /vmaddfp 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27112 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 07:05:55 +00:00
Chris Lattner
420736dc85 #include Intrinsics.h into all dag isels 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27109 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 06:47:10 +00:00
Chris Lattner
9c61dcf1aa Codegen things like: 
<int -1, int -1, int -1, int -1>
and
 <int 65537, int 65537, int 65537, int 65537>

Using things like:
  vspltisb v0, -1
and:
  vspltish v0, 1

instead of using constant pool loads.

This implements CodeGen/PowerPC/vec_splat.ll:splat_imm_i{32|16}.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27106 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 06:12:06 +00:00
Evan Cheng
ecac9cb959 Added SSE cachebility ops 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27103 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 06:03:26 +00:00
Evan Cheng
cc4f047dca Instruction encoding bug 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27102 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 06:00:03 +00:00
Chris Lattner
ea93f63964 Add new intrinsic node definitions for tblgen use 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27100 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 02:29:35 +00:00
Evan Cheng
7b1d34bc6c Added 128-bit packed integer subtraction. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27096 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 01:33:37 +00:00
Evan Cheng
3246e06f84 Added CVTTPS2PI. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27095 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 01:31:59 +00:00
Evan Cheng
7dda4052f5 Added CVTSS2SI. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27094 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-25 01:00:18 +00:00
Evan Cheng
bc4832bc64 Support for scalar to vector with zero extension. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27091 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 23:15:12 +00:00
Jim Laskey
ff70fe61ed D'oh - should be even numbered. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27088 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 22:48:02 +00:00
Evan Cheng
c653d48022 Added LDMXCSR 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27087 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 22:28:37 +00:00
Chris Lattner
ac53eadc29 plug the intrinsics into the patterns for movmsk* 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27083 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 21:49:18 +00:00
Jim Laskey
47622e3721 Add dwarf register numbering to register data. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27081 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 21:15:58 +00:00
Jim Laskey
8da17b242f Add support for dwarf register numbering. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27080 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 21:13:21 +00:00
Chris Lattner
057f09bc0b add another note 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27077 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 20:04:27 +00:00
Chris Lattner
e6cd96d467 add a note 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27076 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 19:59:17 +00:00
Chris Lattner
da10f19d5c Shuffle some includes around 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27073 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 18:52:35 +00:00
Chris Lattner
da91bdcfb0 expose intrinsic info to the targets. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27070 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 18:44:11 +00:00
Chris Lattner
8edd11f33d Fix a bad JIT encoding of VPERM. Why is VPERM D,A,B,C but vfmadd is D,A,C,B ?? 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27069 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 18:24:43 +00:00
Chris Lattner
54e869e18c Like the comment says, prefer to use the implicit add done by [r+r] addressing 
modes than emitting an explicit add and using a base of r0.  This implements
Regression/CodeGen/PowerPC/mem-rr-addr-mode.ll


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27068 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 17:58:06 +00:00
Jim Laskey
98a6979b25 Clean up some commentary. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27064 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 10:00:56 +00:00
Chris Lattner
7fbcef7102 Disable the i32->float G5 optimization. It is unsafe, as documented in the 
comment.

This fixes 177.mesa, and McCat/09-vor with the td scheduler.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27060 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 07:53:47 +00:00
Chris Lattner
64b3a08bc6 add support for using vxor to build zero vectors. This implements 
Regression/CodeGen/PowerPC/vec_zero.ll


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27059 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 07:48:08 +00:00
Evan Cheng
386031a06f Handle BUILD_VECTOR with all zero elements. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27056 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 07:29:27 +00:00
Chris Lattner
9d5da1d96c Gabor points out that we can't spell. :) 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27049 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 07:12:19 +00:00
Evan Cheng
5217a5b58c All v2f64 shuffle cases can be handled. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27044 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 06:40:32 +00:00
Evan Cheng
2c0dbd01d2 More efficient v2f64 shuffle using movlhps, movhlps, unpckhpd, and unpcklpd. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27040 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 02:58:06 +00:00
Evan Cheng
b20aaceb52 A new entry 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27039 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 02:57:03 +00:00
Reid Spencer
cf22612fad Ignore the burg output files. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27033 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 02:21:35 +00:00
Evan Cheng
14aed5e66b Handle more shuffle cases with SHUFP* instructions. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27024 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-24 01:18:28 +00:00
Chris Lattner
716aefcd91 add a note 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27000 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 21:28:44 +00:00
Evan Cheng
8fc23cd0e9 Typo 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26997 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 20:26:04 +00:00
Chris Lattner
dc6af72781 Add PPC vector bit-convert support 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26995 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 19:54:27 +00:00
Jim Laskey
f1d78e8335 Add support to locate local variables in frames (early version.) 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26994 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 18:12:57 +00:00
Jim Laskey
99db0442f0 Change interface to DwarfWriter. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26991 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 18:09:44 +00:00
Jim Laskey
580418e082 Modify how CBE handles #lines. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26990 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 18:08:29 +00:00
Chris Lattner
29b4dd0c9c Fix the encodings of these new instructions, hopefully fixing the JIT 
failures from last night


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26981 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 16:13:50 +00:00
Evan Cheng
24dc1f5975 Following icc's lead: use movdqa to load / store 128-bit integer vectors 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26980 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 07:44:07 +00:00
Chris Lattner
bc641b9d8b Eliminate IntrinsicLowering from TargetMachine. 
Make the CBE and V9 backends create their own, since they're the only ones that use it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26974 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 05:43:16 +00:00
Chris Lattner
ef98691ca3 remove always-null IntrinsicLowering argument. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26971 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 05:28:02 +00:00
Evan Cheng
3b047f7bfa Add v4i32 <-> v4f32 bitconvert patterns. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26969 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 02:36:37 +00:00
Evan Cheng
a971f6f967 Add 128-bit integer vector load and add (for testing). 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26967 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 01:57:24 +00:00
Nate Begeman
ce9448218a Add support for 8 bit immediates with 16/32 bit cmp instructions 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26966 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-23 01:29:48 +00:00
Evan Cheng
ca6e8eafd2 Added a ValueType operand to isShuffleMaskLegal(). For now, x86 will not do 
64-bit vector shuffle.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26964 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 22:07:06 +00:00
Evan Cheng
0cea6d2b9c SHUFP* are two address code. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26959 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 20:08:18 +00:00
Evan Cheng
a88973f826 Some clean up. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26957 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 19:22:18 +00:00
Evan Cheng
1bffadd7fb - Supposely movlhps is faster / better than unpcklpd. 
- Don't forget pshufd is only available with sse2.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26956 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 19:16:21 +00:00
Evan Cheng
0188ecba85 - Implement X86ISelLowering::isShuffleMaskLegal(). We currently only support 
splat and PSHUFD cases.
- Clean up shuffle / splat matching code.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26954 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 18:59:22 +00:00
Evan Cheng
63d3300da1 - VECTOR_SHUFFLE of v4i32 / v4f32 with undef second vector always matches 
PSHUFD. We can make permutes entries which point to the undef pointing
  anything we want.
- Change some names to appease Chris.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26951 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 08:01:21 +00:00
Chris Lattner
c04b423f14 add a note 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26950 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 07:33:46 +00:00
Evan Cheng
2da953f77a Fix PSHUF* and SHUF* jit code emission problems 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26949 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 07:10:28 +00:00
Chris Lattner
9d86a9dff2 This has been implemented. Tweak it into another note 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26944 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 05:33:23 +00:00
Chris Lattner
ecfe55e65b When possible, custom lower 32-bit SINT_TO_FP to this: 
_foo2:
        extsw r2, r3
        std r2, -8(r1)
        lfd f0, -8(r1)
        fcfid f0, f0
        frsp f1, f0
        blr

instead of this:

_foo2:
        lis r2, ha16(LCPI2_0)
        lis r4, 17200
        xoris r3, r3, 32768
        stw r3, -4(r1)
        stw r4, -8(r1)
        lfs f0, lo16(LCPI2_0)(r2)
        lfd f1, -8(r1)
        fsub f0, f1, f0
        frsp f1, f0
        blr

This speeds up Misc/pi from 2.44s->2.09s with LLC and from 3.01->2.18s
with llcbeta (16.7% and 38.1% respectively).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26943 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 05:30:33 +00:00
Chris Lattner
e5ba580ab0 Add support for "ri" addressing modes where the immediate is a 14-bit field 
which is shifted left two bits before use.  Instructions like STD use this
addressing mode.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26942 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 05:26:03 +00:00
Chris Lattner
6df1154644 fix a warning 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26941 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 04:18:34 +00:00
Evan Cheng
b9df0ca67b Some splat and shuffle support. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26940 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 02:53:00 +00:00
Evan Cheng
a9f2a717e9 Add a couple more pseudo instructions. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26939 91177308-0d34-0410-b5e6-96231b3b80d8
 2006-03-22 02:52:03 +00:00