nicolas/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
110,634 Commits 41 Branches 151 Tags
e33c95d39b4a315be12c79f710cb1550650cb457
Commit Graph

18593 Commits

Author SHA1 Message Date
Owen Anderson
e33c95d39b Correct immediate range for shifter operands. Patch by James Molloy, with additional encoding fixes added by me. 
llvm-svn: 137322
 2011-08-11 18:41:59 +00:00
Owen Anderson
ed25385227 Improve error checking in the new ARM disassembler. Patch by James Molloy. 
llvm-svn: 137320
 2011-08-11 18:24:51 +00:00
Jim Grosbach
27ad83d8a9 ARM push of a single register encodes as pre-indexed STR. 
Per the ARM ARM, a 'push' of a single register encodes as an STR,
not an STM.

llvm-svn: 137318
 2011-08-11 18:07:11 +00:00
Jim Grosbach
8ba76c6d5c ARM pop of a single register encodes as post-indexed LDR. 
Per the ARM ARM, a 'pop' of a single register encodes as an LDR,
not an LDM.

llvm-svn: 137316
 2011-08-11 17:35:48 +00:00
Nadav Rotem
efdd183f52 Add a comment, per Bruno's CR. 
llvm-svn: 137313
 2011-08-11 17:05:47 +00:00
Nadav Rotem
1542d5a00a [AVX] If the data which is going to be saved is already in two XMM registers 
(for example, after integer operation), do not pack the registers into a YMM
before saving. Its better to save as two XMM registers.

Before:
                vinsertf128         $1, %xmm3, %ymm0, %ymm3
                vinsertf128         $0, %xmm1, %ymm3, %ymm1
                vmovaps              %ymm1, 416(%rsp)

After:
                vmovaps              %xmm3, 416+16(%rsp)
                vmovaps              %xmm1, 416(%rsp)

llvm-svn: 137308
 2011-08-11 16:41:21 +00:00
Bruno Cardoso Lopes
dbd1352c80 Cleanup: Remove Int_ CVTSS2SI* forms 
llvm-svn: 137297
 2011-08-11 02:52:36 +00:00
Bruno Cardoso Lopes
a2d8bb97b9 Splats for v8i32/v8f32 can be handled by VPERMILPSY. This was causing 
infinite recursive calls in legalize. Fix PR10562

llvm-svn: 137296
 2011-08-11 02:49:44 +00:00
Bruno Cardoso Lopes
572c9aaf53 Use the splat index to generate the desired shuffle. Otherwise we 
could only get undefs and the vector shuffle becomes an undef,
generating wrong code.

llvm-svn: 137295
 2011-08-11 02:49:41 +00:00
Eli Friedman
3ae39f8ad1 Fix X86TargetLowering::LowerExternalSymbol so that it actually works in non-trivial cases. This hasn't been an issue before because the function isn't normally called (but apparently is used to generate a tail-call to sin() on ELF x86-32 with PIC and SSE2). 
Fixes PR9693.

llvm-svn: 137292
 2011-08-11 01:48:05 +00:00
Jim Grosbach
d5d6359785 ARM LDRT assembly parsing and encoding. 
llvm-svn: 137282
 2011-08-10 23:43:54 +00:00
Jim Grosbach
d3f7bcd43c Tidy up. 80 columns. 
llvm-svn: 137277
 2011-08-10 23:23:47 +00:00
Jim Grosbach
cd4dd255c0 ARM LDRH(immediate) assembly parsing and encoding support. 
llvm-svn: 137260
 2011-08-10 22:42:16 +00:00
Jim Grosbach
1d9d5e93d1 ARM LDRD(register) assembly parsing and encoding. 
Add support for literal encoding of #-0 along the way.

llvm-svn: 137254
 2011-08-10 21:56:18 +00:00
Jim Grosbach
f7164b2cfd Fix typo. Not quite sure how that slipped in there. 
llvm-svn: 137245
 2011-08-10 20:49:18 +00:00
Jim Grosbach
5b96b80644 ARM LDRD(immediate) assembly parsing and encoding support. 
llvm-svn: 137244
 2011-08-10 20:29:19 +00:00
Nadav Rotem
410a11fe82 When performing a truncating store, it is sometimes possible to rearrange the 
data in-register prior to saving to memory.  When we reorder the data in memory
we prevent the need to save multiple scalars to memory, making a single regular
store.

llvm-svn: 137238
 2011-08-10 19:30:14 +00:00
Owen Anderson
c86a5bd219 Add initial support for decoding NEON instructions in Thumb2 mode. 
llvm-svn: 137236
 2011-08-10 19:01:10 +00:00
Bruno Cardoso Lopes
3ff111c12d The following X86 pattern is incorrect: 
def : Pat<(X86Movss VR128:$src1,
                   (bc_v4i32 (v2i64 (load addr:$src2)))),
          (MOVLPSrm VR128:$src1, addr:$src2)>;
This matches a MOVSS dag with a MOVLPS instruction. However, MOVSS will replace only the low 32 bits of the register, while the MOVLPS instruction will replace the low 64 bits. A testcase is added and illustrates the bug and also modified the one that was already present. Patch by Tanya Lattner.

llvm-svn: 137227
 2011-08-10 17:45:17 +00:00
Owen Anderson
1531e5cd2b Tabs --> spaces. 
llvm-svn: 137225
 2011-08-10 17:38:05 +00:00
Owen Anderson
5d69f63bbb Cleanups based on Nick Lewycky's feedback. 
llvm-svn: 137224
 2011-08-10 17:36:48 +00:00
Owen Anderson
732f82c463 Rewrite some ARM InstrInfo functions to be most accepting of arbitrary register subclasses. Hopefully this fixes some buildbots. 
llvm-svn: 137223
 2011-08-10 17:21:20 +00:00
Rafael Espindola
36a3abc671 Add support for the R and Q constraints. 
llvm-svn: 137217
 2011-08-10 16:26:42 +00:00
Bruno Cardoso Lopes
278ffd7d8e Fix a bug in vpermilps mask checking. Fix PR10560 
llvm-svn: 137194
 2011-08-10 01:54:17 +00:00
Owen Anderson
8059f0cf8d Push GPRnopc through a large number of instruction definitions to tighten operand decoding. 
llvm-svn: 137189
 2011-08-10 00:03:03 +00:00
Jakob Stoklund Olesen
6a14dc01ff Promote VMOVS to VMOVD when possible. 
On Cortex-A8, we use the NEON v2f32 instructions for f32 arithmetic. For
better latency, we also send D-register copies down the NEON pipeline by
translating them to vorr instructions.

This patch promotes even S-register copies to D-register copies when
possible so they can also go down the NEON pipeline.  Example:

        vldr.32 s0, LCPI0_0
    loop:
        vorr    d1, d0, d0
    loop2:
        ...
        vadd.f32        d1, d1, d16

The vorr instruction looked like this after regalloc:

    %S2<def> = COPY %S0, %D1<imp-def>

Copies involving odd S-registers, and copies that don't define the full
D-register are left alone.

llvm-svn: 137182
 2011-08-09 23:41:44 +00:00
Owen Anderson
92b942b1b5 Tighten operand checking of register-shifted-register operands. 
llvm-svn: 137180
 2011-08-09 23:33:27 +00:00
Bruno Cardoso Lopes
72323966c8 Add 256-bit support for v8i32, v4i64 and v4f64 ISD::SELECT. Fix PR10556 
llvm-svn: 137179
 2011-08-09 23:27:13 +00:00
Owen Anderson
e008931bf6 Tighten operand checking on memory barrier instructions. 
llvm-svn: 137176
 2011-08-09 23:25:42 +00:00
Owen Anderson
3d2e0e9db6 Tighten operand checking on CPS instructions. 
llvm-svn: 137172
 2011-08-09 23:05:39 +00:00
Owen Anderson
042619f97d Create a new register class for the set of all GPRs except the PC. Use it to tighten our decoding of BFI. 
llvm-svn: 137168
 2011-08-09 22:48:45 +00:00
Bruno Cardoso Lopes
fc481959d2 Add v16i16 and v32i8 store patterns 
llvm-svn: 137166
 2011-08-09 22:39:53 +00:00
Bruno Cardoso Lopes
6963062a99 Use fp unpack instructions to unpack int types. Until we have AVX2, this 
is the best we can do for these patterns. This fix PR10554.

llvm-svn: 137161
 2011-08-09 22:18:37 +00:00
Eli Friedman
4ef2426b87 Fix a couple ridiculous copy-paste errors. rdar://9914773 . 
llvm-svn: 137160
 2011-08-09 22:17:39 +00:00
Benjamin Kramer
406dc1755f ARM Disassembler: sign extend branch immediates. 
Not sure about BLXi, but this is what the old disassembler did.

llvm-svn: 137156
 2011-08-09 22:02:50 +00:00
Owen Anderson
d151b09921 Silence an false-positive warning. 
llvm-svn: 137154
 2011-08-09 21:38:14 +00:00
Owen Anderson
d770f6c110 Don't generate the old-style disassembler in CMake builds either. 
llvm-svn: 137153
 2011-08-09 21:36:11 +00:00
Benjamin Kramer
de2c381331 The new ARM disassembler disassembles "bx lr" as a special BX_ret instruction so target specific analysis isn't needed anymore. 
llvm-svn: 137151
 2011-08-09 21:34:19 +00:00
Owen Anderson
982aa05017 Don't continue generating the old-style decoder file. 
llvm-svn: 137150
 2011-08-09 21:30:29 +00:00
Jim Grosbach
5e80abbb5d ARM fix typo in pre-indexed store lowering. 
rdar://9915869

llvm-svn: 137148
 2011-08-09 21:22:41 +00:00
Owen Anderson
c7afd84322 Attempt to fix CMake build. 
llvm-svn: 137147
 2011-08-09 21:09:59 +00:00
Owen Anderson
7a2401dbf0 Tighten Thumb1 branch predicate decoding. 
llvm-svn: 137146
 2011-08-09 21:07:45 +00:00
Owen Anderson
e0152a73c2 Replace the existing ARM disassembler with a new one based on the FixedLenDecoderEmitter. 
This new disassembler can correctly decode all the testcases that the old one did, though
some "expected failure" testcases are XFAIL'd for now because it is not (yet) as strict in
operand checking as the old one was.

llvm-svn: 137144
 2011-08-09 20:55:18 +00:00
Bill Wendling
d7f41b7f66 Revert r137134. It breaks some code as Eli pointed out. 
llvm-svn: 137135
 2011-08-09 18:56:35 +00:00
Bill Wendling
84ec8f65d1 Print out the variable declaration only if it is a declaration. Otherwise, a 
'static' variable will be emitted twice.
PR10081

llvm-svn: 137134
 2011-08-09 18:31:50 +00:00
Bruno Cardoso Lopes
bed48dc8ff Reapply a more appropriate solution than in r137114. AVX supports 
v4f64 = sitofp v4i32. This fix PR10559.
Also add support for v4i32 = fptosi v4f64.

llvm-svn: 137128
 2011-08-09 17:39:13 +00:00
Bruno Cardoso Lopes
24dd1d4a27 Revert r137114 
llvm-svn: 137127
 2011-08-09 17:39:01 +00:00
Justin Holewinski
db05c2b963 PTX: Add initial support for device function calls 
- Calls are supported on SM 2.0+ for function with no return values

llvm-svn: 137125
 2011-08-09 17:36:31 +00:00
Renato Golin
faff512536 Emitting ARM build attributes and values as ULEB, rather than char. 
llvm-svn: 137115
 2011-08-09 09:50:10 +00:00
Bruno Cardoso Lopes
ad3453cf2d Handle sitofp between v4f64 <- v4i32. Fix PR10559 
llvm-svn: 137114
 2011-08-09 05:48:01 +00:00