Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 177277

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 177276

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 177275

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 177274

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 177273

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 177272

Unfortunately the previous fix for inserting waits for unordered
defines wasn't sufficient, cause it's possible that even ordered
defines are only partially used (or not used at all).

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 177271

MinGW is almost completely compatible to MSVC, with the exception of the _tls_array global not being available.

Patch by David Nadlinger!

llvm-svn: 177257

llvm-svn: 177243

llvm-svn: 177242

llvm-svn: 177234

This change cleans up two issues with Altivec register spilling:

  1. The spilling code was inefficient (using two instructions, and add and a
     load, when just one would do)

  2. The code assumed that r0 would always be available (true for now, but this
     will change)

The new code handles VR spilling just like GPR spills but forced into r+r mode.
As a result, when any VR spills are present, we must now always allocate the
register-scavenger spill slot.

llvm-svn: 177231

As a follow-up to r158719, remove PPCRegisterInfo::avoidWriteAfterWrite.
Jakob pointed out in response to r158719 that this callback is currently unused
and so this has no effect (and the speedups that I thought that I had observed
as a result of implementing this function must have been noise).

llvm-svn: 177228

Previously we weren't skipping the VVVV encoded register. Based on patch by Michael Liao.

llvm-svn: 177221

Since almost all X86 instructions can fold loads, use a multiclass to
define register/memory pairs of SchedWrites.

An X86FoldableSchedWrite represents the register version of an
instruction. It holds a reference to the SchedWrite to use when the
instruction folds a load.

This will be used inside multiclasses that define rr and rm instruction
versions together.

llvm-svn: 177210

I was too pessimistic in r177105. Vector selects that fit into a legal register
type lower just fine. I was mislead by the code fragment that I was using. The
stores/loads that I saw in those cases came from lowering the conditional off
an address.

Changing the code fragment to:

%T0_3 = type <8 x i18>
%T1_3 = type <8 x i1>

define void @func_blend3(%T0_3* %loadaddr, %T0_3* %loadaddr2,
                         %T1_3* %blend, %T0_3* %storeaddr) {
  %v0 = load %T0_3* %loadaddr
  %v1 = load %T0_3* %loadaddr2
==> FROM:
  ;%c = load %T1_3* %blend
==> TO:
  %c = icmp slt %T0_3 %v0, %v1
==> USE:
  %r = select %T1_3 %c, %T0_3 %v0, %T0_3 %v1

  store %T0_3 %r, %T0_3* %storeaddr
  ret void
}

revealed this mistake.

radar://13403975

llvm-svn: 177170

Adding an A15 specific optimization pass for interactions between S/D/Q registers. The pass handles all the required transformations pre-regalloc.

llvm-svn: 177169

Fixes PR15520.

llvm-svn: 177167

Unaligned access is supported on PPC for non-vector types, and is generally
more efficient than manually expanding the loads and stores.

A few of the existing test cases were using expanded unaligned loads and stores
to test other features (like load/store with update), and for these test cases,
unaligned access remains disabled.

llvm-svn: 177160

A vector fptrunc and fpext simply gets split into scalar instructions.

radar://13192358

llvm-svn: 177159

In preparation for the addition of other SIMD ISA extensions (such as QPX) we
need to make sure that all Altivec patterns are properly predicated on having
Altivec support.

No functionality change intended (one test case needed to be updated b/c it
assumed that Altivec intrinsics would be supported without enabling Altivec
support).

llvm-svn: 177152

For spills into a large stack frame, the FI-elimination code uses the register
scavenger to obtain a free GPR for use with an r+r-addressed load or store.
When there are no available GPRs, the scavenger gets one by using its spill
slot. Previously, we were not always allocating that spill slot and the RS
would assert when the spill slot was needed.

I don't currently have a small test that triggered the assert, but I've
created a small regression test that verifies that the spill slot is now
added when the stack frame is sufficiently large.

llvm-svn: 177140

llvm-svn: 177135

llvm-svn: 177130

(these were added in r177089)

llvm-svn: 177129

llvm-svn: 177128

The new InstrSchedModel is easier to use than the instruction
itineraries. It will be used to model instruction latency and throughput
in modern Intel microarchitectures like Sandy Bridge.

InstrSchedModel should be able to coexist with instruction itinerary
classes, but for cleanliness we should switch the Atom processor model
to the new InstrSchedModel as well.

llvm-svn: 177122

See the Mips16ISetLowering.cpp patch to see a use of this.
For now now the extra code in Mips16ISetLowering.cpp is a nop but is
used for test purposes. Mips32 registers are setup and then removed and
then the Mips16 registers are setup. 

Normally you need to add register classes and then call
computeRegisterProperties.

llvm-svn: 177120

the win64 calling convention.
rdar://13423768

llvm-svn: 177113

This is a generic function (derived from PEI); moving it into
MachineFrameInfo eliminates a current redundancy between the ARM and AArch64
backends, and will allow it to be used by the PowerPC target code.

No functionality change intended.

llvm-svn: 177111

Add the current PEI register scavenger as a parameter to the
processFunctionBeforeFrameFinalized callback.

This change is necessary in order to allow the PowerPC target code to
set the register scavenger frame index after the save-area offset
adjustments performed by processFunctionBeforeFrameFinalized. Only
after these adjustments have been made is it possible to estimate
the size of the stack frame.

llvm-svn: 177108

Make requiresFrameIndexScavenging return true, and create virtual registers in
the spilling code instead of using the register scavenger directly. This makes
the target-level code simpler, and importantly, delays the scavenging until
after callee-saved register processing (which will be important for later
changes).

Also cleans up trackLivenessAfterRegAlloc (makes it inline in the header with
the other related functions). This makes it clear that it always returns true.

No functionality change intended.

llvm-svn: 177107

We used to add a spill slot for the register scavenger whenever the function
has a frame pointer. This is unnecessarily conservative: We may need the spill
slot for dynamic stack allocations, and functions with dynamic stack
allocations always have a FP, but we might also have a FP for other reasons
(such as the user explicitly disabling frame-pointer elimination), and we don't
necessarily need a spill slot for those functions.

The structsinregs test needed adjustment because it disables FP elimination.

llvm-svn: 177106

By terrible I mean we store/load from the stack.

This matters on PAQp8 in _Z5trainPsS_ii (which is inlined into Mixer::update)
where we decide to vectorize a loop with a VF of 8 resulting in a 25%
degradation on a cortex-a8.

LV: Found an estimated cost of 2 for VF 8 For instruction:   icmp slt i32
LV: Found an estimated cost of 2 for VF 8 For instruction:   select i1, i32, i32

The bug that tracks the CodeGen part is PR14868.

radar://13403975

llvm-svn: 177105

No functionality changes.

llvm-svn: 177104

We are warning the user about the alignment, so we should not assert.

llvm-svn: 177103

llvm-svn: 177096

No intended functionality changes.

llvm-svn: 177095

I don't think that it is otherwise clear how the overlapping offsets
are processed into distinct spill slots. Comment that this is done
in processFunctionBeforeFrameFinalized.

llvm-svn: 177094

llvm-svn: 177092