1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-30 23:42:52 +01:00
Commit Graph

8200 Commits

Author SHA1 Message Date
Daniel Sanders
9a3de1f604 [mips][msa] Added support for matching comparisons from normal IR (i.e. not intrinsics)
MIPS SelectionDAG changes:
* Added VCEQ, VCL[ET]_[SU] nodes to represent vector comparisons that produce a bitmask.

llvm-svn: 191286
2013-09-24 10:46:19 +00:00
Daniel Sanders
362149b5a7 [mips][msa] Added support for matching slli, srai, and srli from normal IR (i.e. not intrinsics)
llvm-svn: 191285
2013-09-24 10:28:18 +00:00
NAKAMURA Takumi
3b910496bf llvm/test/CodeGen/AArch64/neon-scalar-reduce-pairwise.ll: Use -mtriple here, or aach64-pecoff might be misassumed on win32 hosts.
llvm-svn: 191275
2013-09-24 04:14:29 +00:00
Jiangning Liu
5867567c41 Initial support for Neon scalar instructions.
Patch by Ana Pazos.

1.Added support for v1ix and v1fx types.
2.Added Scalar Pairwise Reduce instructions.
3.Added initial implementation of Scalar Arithmetic instructions.

llvm-svn: 191263
2013-09-24 02:47:27 +00:00
Michael Gottesman
a2ef7dd057 [stackprotector] Forgot to add in PR number to test case.
llvm-svn: 191261
2013-09-24 02:10:55 +00:00
Michael Gottesman
2ec63d27a9 [stackprotector] Allow for copies from vreg -> vreg to be in a terminator sequence.
Sometimes a copy from a vreg -> vreg sneaks into the middle of a terminator
sequence. It is safe to slice this into the stack protector success bb.

This fixes PR16979.

llvm-svn: 191260
2013-09-24 01:50:26 +00:00
Bill Wendling
339b0f39aa Selecting the address from a very long chain of GEPs can blow the stack.
The recursive nature of the address selection code can cause the stack to
explode if there is a long chain of GEPs. Convert the recursive bit into a
iterative method to avoid this.

<rdar://problem/12445434>

llvm-svn: 191252
2013-09-24 00:13:08 +00:00
Reed Kotler
ed09a36fb5 Make nomips16 mask not repeat if it ends with a '.'.
This mask is purely for debugging and testing.

llvm-svn: 191231
2013-09-23 22:36:11 +00:00
Ben Langmuir
706a7ccbeb Add sha intrinsic tests
These should have been included with r190864, but I forgot to use svn add.

llvm-svn: 191208
2013-09-23 16:57:52 +00:00
Daniel Sanders
ced4e4005c [mips][msa] Added support for matching addvi, and subvi from normal IR (i.e. not intrinsics)
llvm-svn: 191203
2013-09-23 14:29:55 +00:00
Daniel Sanders
34cb8f3e4d [mips][msa] Added support for matching insert and copy from normal IR (i.e. not intrinsics)
Changes to MIPS SelectionDAG:
* Added nodes VEXTRACT_[SZ]EXT_ELT to represent extract and extend in a single
  operation and implemented the DAG combines necessary to fold sign/zero
  extends into the extract.

llvm-svn: 191199
2013-09-23 14:03:12 +00:00
Daniel Sanders
d1df1263eb [mips][msa] Added support for matching pcnt from normal IR (i.e. not intrinsics)
llvm-svn: 191198
2013-09-23 13:40:21 +00:00
Daniel Sanders
7d945d142d [mips][msa] Added support for matching nor from normal IR (i.e. not intrinsics)
llvm-svn: 191195
2013-09-23 13:22:24 +00:00
Daniel Sanders
91c78d1d33 [mips][msa] Added support for matching and, or, and xor from normal IR (i.e. not intrinsics)
llvm-svn: 191194
2013-09-23 12:57:42 +00:00
Daniel Sanders
d3c403c386 [mips][msa] Implemented build_vector using ldi, fill, and custom SelectionDAG nodes (VSPLAT and VSPLATD)
Note: There's a later patch on my branch that re-implements this to select
build_vector without the custom SelectionDAG nodes. The future patch avoids
the constant-folding problems stemming from the custom node (i.e. it doesn't
need to re-implement all the DAG combines related to BUILD_VECTOR).

Changes to MIPS specific SelectionDAG nodes:
* Added VSPLAT
    This is a special case of BUILD_VECTOR that covers the case the
    BUILD_VECTOR is a splat operation.
* Added VSPLATD
    This is a special case of VSPLAT that handles the cases when v2i64 is legal

llvm-svn: 191191
2013-09-23 12:02:46 +00:00
Tim Northover
c9a7e47164 ISelDAG: spot chain cycles involving MachineNodes
Previously, the DAGISel function WalkChainUsers was spotting that it
had entered already-selected territory by whether a node was a
MachineNode (amongst other things). Since it's fairly common practice
to insert MachineNodes during ISelLowering, this was not the correct
check.

Looking around, it seems that other nodes get their NodeId set to -1
upon selection, so this makes sure the same thing happens to all
MachineNodes and uses that characteristic to determine whether we
should stop looking for a loop during selection.

This should fix PR15840.

llvm-svn: 191165
2013-09-22 08:21:56 +00:00
Venkatraman Govindaraju
ae9ddc5768 [Sparc] Add support for TLS in sparc.
llvm-svn: 191164
2013-09-22 06:48:52 +00:00
Venkatraman Govindaraju
df68ba133b [SPARC] Make functions with GLOBAL_OFFSET_TABLE access as non-leaf functions.
llvm-svn: 191160
2013-09-22 01:40:24 +00:00
Venkatraman Govindaraju
e3ed207140 [Sparc] Emit .register directive to declare the use of global registers %g2, %g4, %g6 and %g7.
llvm-svn: 191158
2013-09-22 00:42:30 +00:00
Venkatraman Govindaraju
54744c0b41 [Sparc] Fix lowering FABS on fp128 (long double) on pre-v9 targets.
llvm-svn: 191154
2013-09-21 23:51:08 +00:00
Juergen Ributzka
b55735e2d8 Revert "SelectionDAG: Teach the legalizer to split SETCC if VSELECT needs splitting too."
This reverts commit r191130.

llvm-svn: 191138
2013-09-21 15:09:46 +00:00
Juergen Ributzka
32cca125e1 [X86] Emulate AVX 256bit MIN/MAX support by splitting the vector.
In AVX 256bit vectors are valid vectors and therefore the Type Legalizer doesn't
split the VSELECT and SETCC nodes. AVX only supports MIN/MAX on 128bit vectors
and this fix enables vector splitting for this special case in the X86 DAG
Combiner.

This fix is related to PR16695, PR17002, and <rdar://problem/14594431>.

llvm-svn: 191131
2013-09-21 04:55:22 +00:00
Juergen Ributzka
67e5289ff2 SelectionDAG: Teach the legalizer to split SETCC if VSELECT needs splitting too.
The Type Legalizer recognizes that VSELECT needs to be split, because the type
is to wide for the given target. The same does not always apply to SETCC,
because less space is required to encode the result of a comparison. As a result
VSELECT is split and SETCC is unrolled into scalar comparisons.

This commit fixes the issue by checking for VSELECT-SETCC patterns in the DAG
Combiner. If a matching pattern is found, then the result mask of SETCC is
promoted to the expected vector mask for the given target. This mask has usually
te same size as the VSELECT return type (except for Intel KNL). Now the type
legalizer will split both VSELECT and SETCC.

This allows the following X86 DAG Combine code to sucessfully detect the MIN/MAX
pattern. This fixes PR16695, PR17002, and <rdar://problem/14594431>.

llvm-svn: 191130
2013-09-21 04:55:18 +00:00
NAKAMURA Takumi
81025ea333 Initialize BSSSection explicitly in InitMachOMCObjectFileInfo() to appease msvc.
This can revert r191087.

llvm-svn: 191128
2013-09-21 02:34:45 +00:00
Reed Kotler
c8c27affb8 Set .reorder for the stub so that gas takes care of delay slot processing.
llvm-svn: 191125
2013-09-21 01:37:52 +00:00
NAKAMURA Takumi
69102d22a5 llvm/test: Mark 3 tests as XFAIL:msvc.
llvm-svn: 191087
2013-09-20 12:57:34 +00:00
Kai Nacke
64a3fccd60 PR16726: extend rol/ror matching
C-like languages promote types like unsigned short to unsigned int before
performing an arithmetic operation. Currently the rotate matcher in the
DAGCombiner does not consider this situation.

This commit extends the DAGCombiner in the way that the pattern

(or (shl ([az]ext x), (*ext y)), (srl ([az]ext x), (*ext (sub 32, y))))

is folded into

([az]ext (rotl x, y))

The matching is restricted to aext and zext because in this cases the upper
bits are either undefined or known. Test case is included.

This fixes PR16726.

llvm-svn: 191049
2013-09-19 23:00:28 +00:00
Kai Nacke
21c0476931 Revert PR16726: extend rol/ror matching
There is a buildbot failure. Need to investigate this.

llvm-svn: 191048
2013-09-19 22:53:36 +00:00
Kai Nacke
fe02753846 PR16726: extend rol/ror matching
C-like languages promote types like unsigned short to unsigned int before
performing an arithmetic operation. Currently the rotate matcher in the
DAGCombiner does not consider this situation.

This commit extends the DAGCombiner in the way that the pattern

(or (shl ([az]ext x), (*ext y)), (srl ([az]ext x), (*ext (sub 32, y))))

is folded into

([az]ext (rotl x, y))

The matching is restricted to aext and zext because in this cases the upper
bits are either undefined or known. Test case is included.

This fixes PR16726.

llvm-svn: 191045
2013-09-19 22:36:39 +00:00
Bill Wendling
01837a249d Add testcase to make sure we don't generate too many jumps for a une compare.
<rdar://problem/7859988>

llvm-svn: 191040
2013-09-19 21:58:20 +00:00
Benjamin Kramer
f939b2b330 DAGCombiner: Don't fold vector muls with constants that look like a splat of a power of 2 but differ in bit width.
PR17283.

llvm-svn: 191000
2013-09-19 13:28:20 +00:00
Justin Holewinski
940e5b09fe [NVPTX] Make constant vector test case endian-independent
llvm-svn: 190998
2013-09-19 13:14:44 +00:00
Justin Holewinski
69db0d8365 [NVPTX] Support constant vector globals
llvm-svn: 190997
2013-09-19 12:51:46 +00:00
Amara Emerson
7ad0409c56 [ARMv8] Add support for the v8 cryptography extensions.
llvm-svn: 190996
2013-09-19 11:59:01 +00:00
Tim Northover
89d57eb12b X86: FrameIndex addressing modes do have a base register.
When selecting the DAG (add (WrapperRIP ...), (FrameIndex ...)), X86 code had
spotted the FrameIndex possibility and was working out whether it could fold
the WrapperRIP into this.

The test for forming a %rip version is notionally whether we already have a
base or index register (%rip precludes both), but we were forgetting to account
for the register that would be inserted later to access the frame.

rdar://problem/15024520

llvm-svn: 190995
2013-09-19 11:33:53 +00:00
Reed Kotler
03a3269d15 Fix two issues regarding Got pointer (GP) setup.
1) make sure that the first two instructions of the sequence cannot
separate from each other. The linker requires that they be sequential.
If they get separated, it can still work but it will not work in all
cases because the first of the instructions mostly involves the hi part
of the pc relative offset and that part changes slowly. You would have
to be at the right boundary for this to matter.
2) make sure that this sequence begins  on a longword boundary. 
There appears to be a bug in binutils which makes some of these calculations
get messed up if the instruction sequence does not begin on a longword
boundary. This is being investigated with the appropriate binutils folks.

llvm-svn: 190966
2013-09-18 22:46:09 +00:00
Preston Gurd
efaca57b70 Attempt to fix llvm-ppc64-linux2 buildbot failure by adding
-march=x86 to SLM test.

llvm-svn: 190958
2013-09-18 21:39:33 +00:00
Preston Gurd
1800994293 Verify that llvm can generate the prefetchw instruction when the CPU is
Atom Silvermont.

Patch by Sriram Murali.

llvm-svn: 190957
2013-09-18 21:08:09 +00:00
Richard Sandiford
76d1801e90 [SystemZ] Add unsigned compare-and-branch instructions
For some reason I never got around to adding these at the same time as
the signed versions.  No idea why.

I'm not sure whether this SystemZII::BranchC* stuff is useful, or whether
it should just be replaced with an "is normal" flag.  I'll leave that
for later though.

There are some boundary conditions that can be tweaked, such as preferring
unsigned comparisons for equality with [128, 256), and "<= 255" over "< 256",
but again I'll leave those for a separate patch.

llvm-svn: 190930
2013-09-18 09:56:40 +00:00
Craig Topper
5d022196de Lift alignment restrictions for load/store folding on VINSERTF128/VEXTRACTF128. Fixes PR17268.
llvm-svn: 190916
2013-09-18 03:55:53 +00:00
Reid Kleckner
130539949d COFF: Ensure that objects produced by LLVM link with /safeseh
Summary:
We indicate that the object files are safe by emitting a @feat.00
absolute address symbol.  The address is presumably interpreted as a
bitfield of features that the compiler would like to enable.  Bit 0 is
documented in the PE COFF spec to opt in to "registered SEH", which is
what /safeseh enables.

LLVM's object files are safe by default because LLVM doesn't know how to
produce SEH handlers.

Reviewers: Bigcheese

CC: llvm-commits

Differential Revision: http://llvm-reviews.chandlerc.com/D1691

llvm-svn: 190898
2013-09-17 23:18:05 +00:00
Bill Schmidt
f26512486a [PowerPC] Fix problems with large code model (PR17169).
Large code model on PPC64 requires creating and referencing TOC entries when
using the addis/ld form of addressing.  This was not being done in all cases.
The changes in this patch to PPCAsmPrinter::EmitInstruction() fix this.  Two
test cases are also modified to reflect this requirement.

Fast-isel was not creating correct code for loading floating-point constants
using large code model.  This also requires the addis/ld form of addressing.
Previously we were using the addis/lfd shortcut which is only applicable to
medium code model.  One test case is modified to reflect this requirement.

llvm-svn: 190882
2013-09-17 20:03:25 +00:00
Kevin Qin
3be5824550 Implement 3 AArch64 neon instructions : umov smov ins.
llvm-svn: 190839
2013-09-17 02:21:02 +00:00
Quentin Colombet
475709bdc2 [SelectionDAG] Teach the vector scalarizer about TRUNCATE.
When a truncate node defines a legal vector type but uses an illegal
vector type, the legalization process was splitting the vector until
<1 x vector> type, but then it was failing to scalarize the node because
it did not know how to handle TRUNCATE.

<rdar://problem/14989896>

llvm-svn: 190830
2013-09-17 00:26:56 +00:00
Preston Gurd
6efe5100eb Add Atom Silvermont (slm) tests
- check that -mcpu=slm uses the call register indirect optimization
- check that -mcpu=slm runs the scheduler 
- check that -mcpu=slm supports the movbe instruction

llvm-svn: 190814
2013-09-16 22:22:07 +00:00
Richard Sandiford
ea2b1a8b94 [SystemZ] Improve extload handling
The port originally had special patterns for extload, mapping them to the
same instructions as sextload.  It seemed neater to have patterns that
match "an extension that is allowed to be signed" and "an extension that
is allowed to be unsigned".

This was originally meant to be a clean-up, but it does improve the handling
of promoted integers a little, as shown by args-06.ll.

llvm-svn: 190777
2013-09-16 09:03:10 +00:00
Peter Collingbourne
cf3b1a2910 Implement function prefix data as an IR feature.
Previous discussion:
http://lists.cs.uiuc.edu/pipermail/llvmdev/2013-July/063909.html

Differential Revision: http://llvm-reviews.chandlerc.com/D1191

llvm-svn: 190773
2013-09-16 01:08:15 +00:00
Hal Finkel
5bb449bca0 PPC: Don't restrict lvsl generation to after type legalization
This is a re-commit of r190764, with an extra check to make sure that we're not
performing the transformation on illegal types (a small test case has been
added for this as well).

Original commit message:

The PPC backend uses a target-specific DAG combine to turn unaligned Altivec
loads into a permutation-based sequence when possible. Unfortunately, the
target-specific DAG combine is not always called on all loads of interest
(sometimes the routines in DAGCombine call CombineTo such that the new node and
users are not added to the worklist); allowing the combine to trigger early
(before type legalization) mitigates this problem. Because the autovectorizers
only create legal vector types, I don't expect a lot of cases where this
optimization is enabled by type legalization in practice.

llvm-svn: 190771
2013-09-15 22:09:58 +00:00
Hal Finkel
c45bfe85cc Revert r190764: PPC: Don't restrict lvsl generation to after type legalization
This is causing test-suite failures.

Original commit message:

The PPC backend uses a target-specific DAG combine to turn unaligned Altivec
loads into a permutation-based sequence when possible. Unfortunately, the
target-specific DAG combine is not always called on all loads of interest
(sometimes the routines in DAGCombine call CombineTo such that the new node and
users are not added to the worklist); allowing the combine to trigger early
(before type legalization) mitigates this problem. Because the autovectorizers
only create legal vector types, I don't expect a lot of cases where this
optimization is enabled by type legalization in practice.

llvm-svn: 190765
2013-09-15 15:41:11 +00:00
Hal Finkel
ae7feec56e PPC: Don't restrict lvsl generation to after type legalization
The PPC backend uses a target-specific DAG combine to turn unaligned Altivec
loads into a permutation-based sequence when possible. Unfortunately, the
target-specific DAG combine is not always called on all loads of interest
(sometimes the routines in DAGCombine call CombineTo such that the new node and
users are not added to the worklist); allowing the combine to trigger early
(before type legalization) mitigates this problem. Because the autovectorizers
only create legal vector types, I don't expect a lot of cases where this
optimization is enabled by type legalization in practice.

llvm-svn: 190764
2013-09-15 15:20:54 +00:00