38 Commits

Author SHA1 Message Date
James Molloy
3a106a2813 [AArch64] Add an FP load balancing pass for Cortex-A57
For best-case performance on Cortex-A57, we should try to use a balanced mix of odd and even D-registers when performing a critical sequence of independent, non-quadword FP/ASIMD floating-point multiply or multiply-accumulate operations.

This pass attempts to detect situations where the register allocation may adversely affect this load balancing and to change the registers used so as to better utilize the CPU.

Ideally we'd just take each multiply or multiply-accumulate in turn and allocate it alternating even or odd registers. However, multiply-accumulates are most efficiently performed in the same functional unit as their accumulation operand. Therefore this pass tries to find maximal sequences ("Chains") of multiply-accumulates linked via their accumulation operand, and assign them all the same "color" (oddness/evenness).

This optimization affects S-register and D-register floating point multiplies and FMADD/FMAs, as well as vector (floating point only) muls and FMADD/FMA. Q register instructions (and 128-bit vector instructions) are not affected.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@215199 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-08 12:33:21 +00:00
Gerolf Hoflehner
e4fa341dde MachineCombiner Pass for selecting faster instruction sequence on AArch64
Re-commit of r214832,r21469 with a work-around that
avoids the previous problem with gcc build compilers

The work-around is to use SmallVector instead of ArrayRef
of basic blocks in preservesResourceLen()/MachineCombiner.cpp



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@215151 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-07 21:40:58 +00:00
James Molloy
e0243fb42d [AArch64] Add a testcase for r214957.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@214965 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-06 13:31:32 +00:00
Kevin Qin
6739735812 Revert "r214832 - MachineCombiner Pass for selecting faster instruction"
It broke compiling of most Benchmark and internal test, as clang got
clashed by segmentation fault or assertion.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@214845 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-05 05:43:47 +00:00
Gerolf Hoflehner
c2328d552c MachineCombiner Pass for selecting faster instruction
sequence on AArch64

Re-commit of r214669 without changes to test cases
LLVM::CodeGen/AArch64/arm64-neon-mul-div.ll and
LLVM:: CodeGen/AArch64/dp-3source.ll
This resolves the reported compfails of the original commit.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@214832 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-05 01:16:13 +00:00
Kevin Qin
534100b31e Revert "r214669 - MachineCombiner Pass for selecting faster instruction"
This commit broke "make check" for several hours, so get it reverted.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@214697 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-04 05:10:33 +00:00
Gerolf Hoflehner
48e1bd7287 MachineCombiner Pass for selecting faster instruction
sequence -  AArch64 target support

 This patch turns off madd/msub generation in the DAGCombiner and generates
 them in the MachineCombiner instead. It replaces the original code sequence
 with the combined sequence when it is beneficial to do so.

 When there is no machine model support it always generates the madd/msub
 instruction. This is true also when the objective is to optimize for code
 size: when the combined sequence is shorter is always chosen and does not
 get evaluated.

 When there is a machine model the combined instruction sequence
 is evaluated for critical path and resource length using machine
 trace metrics and the original code sequence is replaced when it is
 determined to be faster.

 rdar://16319955



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@214669 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-03 22:03:40 +00:00
Benjamin Kramer
ce63ab327a Run sort_includes.py on the AArch64 backend.
No functionality change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@213938 91177308-0d34-0410-b5e6-96231b3b80d8
2014-07-25 11:42:14 +00:00
Tim Northover
8b6257629a AArch64: remove "arm64_be" support in favour of "aarch64_be".
There really is no arm64_be: it was a useful fiction to test big-endian support
while both backends existed in parallel, but now the only platform that uses
the name (iOS) doesn't have a big-endian variant, let alone one called
"arm64_be".

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@213748 91177308-0d34-0410-b5e6-96231b3b80d8
2014-07-23 12:58:11 +00:00
Duncan P. N. Exon Smith
9b4509a759 AArch64: Re-enable AArch64AddressTypePromotion
This reverts commits r212189 and r212190.

While this pass was accidentally disabled (until r212073), r205437
slipped in a use of `auto` that should have been `auto&`.

This fixes PR20188.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@212201 91177308-0d34-0410-b5e6-96231b3b80d8
2014-07-02 18:17:40 +00:00
Duncan P. N. Exon Smith
2a56dd87c8 AArch64: Temporarily disable AArch64AddressTypePromotion
Temporarily disable AArch64AddressTypePromotion, which was effectively
re-enabled in r212073 and r212075, while I look into PR20188.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@212189 91177308-0d34-0410-b5e6-96231b3b80d8
2014-07-02 17:03:16 +00:00
Eric Christopher
fe80185273 Move AArch64TargetLowering to AArch64Subtarget.
This currently necessitates a TargetMachine for the TargetLowering
constructor and TLOF.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210605 91177308-0d34-0410-b5e6-96231b3b80d8
2014-06-10 23:26:45 +00:00
Eric Christopher
159e408563 Move AArch64InstrInfo to AArch64Subtarget.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210599 91177308-0d34-0410-b5e6-96231b3b80d8
2014-06-10 22:57:25 +00:00
Eric Christopher
30ad4cb09b Move AArch64SelectionDAGInfo down to the subtarget.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210557 91177308-0d34-0410-b5e6-96231b3b80d8
2014-06-10 18:21:53 +00:00
Eric Christopher
b8dca33a05 Have AArch64SelectionDAGInfo take a DataLayout parameter rather
than a TargetMachine.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210554 91177308-0d34-0410-b5e6-96231b3b80d8
2014-06-10 18:06:28 +00:00
Eric Christopher
fc0f6e8cc5 Move DataLayout onto the AArch64 subtarget.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210552 91177308-0d34-0410-b5e6-96231b3b80d8
2014-06-10 18:06:23 +00:00
Eric Christopher
c4f9370391 Move AArch64FrameLowering into the subtarget.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210549 91177308-0d34-0410-b5e6-96231b3b80d8
2014-06-10 17:44:12 +00:00
Eric Christopher
8128a7965f Remove the uses of AArch64TargetMachine and AArch64Subtarget from
AArch64FrameLowering.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210548 91177308-0d34-0410-b5e6-96231b3b80d8
2014-06-10 17:33:39 +00:00
Tim Northover
d0dbe02fd2 ARM & AArch64: make use of common cmpxchg idioms after expansion
The C and C++ semantics for compare_exchange require it to return a bool
indicating success. This gets mapped to LLVM IR which follows each cmpxchg with
an icmp of the value loaded against the desired value.

When lowered to ldxr/stxr loops, this extra comparison is redundant: its
results are implicit in the control-flow of the function.

This commit makes two changes: it replaces that icmp with appropriate PHI
nodes, and then makes sure earlyCSE is called after expansion to actually make
use of the opportunities revealed.

I've also added -{arm,aarch64}-enable-atomic-tidy options, so that
existing fragile tests aren't perturbed too much by the change. Many
of them either rely on undef/unreachable too pervasively to be
restored to something well-defined (particularly while making sure
they test the same obscure assert from many years ago), or depend on a
particular CFG shape, which is disrupted by SimplifyCFG.

rdar://problem/16227836

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@209883 91177308-0d34-0410-b5e6-96231b3b80d8
2014-05-30 10:09:59 +00:00
Tim Northover
29f94c7201 AArch64/ARM64: move ARM64 into AArch64's place
This commit starts with a "git mv ARM64 AArch64" and continues out
from there, renaming the C++ classes, intrinsics, and other
target-local objects for consistency.

"ARM64" test directories are also moved, and tests that began their
life in ARM64 use an arm64 triple, those from AArch64 use an aarch64
triple. Both should be equivalent though.

This finishes the AArch64 merge, and everyone should feel free to
continue committing as normal now.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@209577 91177308-0d34-0410-b5e6-96231b3b80d8
2014-05-24 12:50:23 +00:00
Tim Northover
9105f66d6f AArch64/ARM64: remove AArch64 from tree prior to renaming ARM64.
I'm doing this in two phases for a better "git blame" record. This
commit removes the previous AArch64 backend and redirects all
functionality to ARM64. It also deduplicates test-lines and removes
orphaned AArch64 tests.

The next step will be "git mv ARM64 AArch64" and rewire most of the
tests.

Hopefully LLVM is still functional, though it would be even better if
no-one ever had to care because the rename happens straight
afterwards.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@209576 91177308-0d34-0410-b5e6-96231b3b80d8
2014-05-24 12:42:26 +00:00
Craig Topper
0fd57f4b56 [C++11] Add 'override' keywords and remove 'virtual'. Additionally add 'final' and leave 'virtual' on some methods that are marked virtual without overriding anything and have no obvious overrides themselves. AArch64 edition
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@207510 91177308-0d34-0410-b5e6-96231b3b80d8
2014-04-29 07:58:34 +00:00
Jiangning Liu
0240286c23 [AArch64] Enable global merge pass.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@206861 91177308-0d34-0410-b5e6-96231b3b80d8
2014-04-22 03:33:26 +00:00
Christian Pirker
d7e12561a5 Add AArch64 big endian Target (aarch64_be)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@202024 91177308-0d34-0410-b5e6-96231b3b80d8
2014-02-24 11:34:50 +00:00
Chad Rosier
fda2dfbca7 [AArch64] Add support for TargetTransformInfo Analysis.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201793 91177308-0d34-0410-b5e6-96231b3b80d8
2014-02-20 16:00:08 +00:00
Christian Pirker
eb5fe7002a Test commit - remove the new line to lib/Target/AArch64/AArch64TargetMachine.cpp.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201698 91177308-0d34-0410-b5e6-96231b3b80d8
2014-02-19 16:58:28 +00:00
Christian Pirker
f47b32ed76 Test commit - added a new line to lib/Target/AArch64/AArch64TargetMachine.cpp.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201692 91177308-0d34-0410-b5e6-96231b3b80d8
2014-02-19 16:07:32 +00:00
Chandler Carruth
974a445bd9 Re-sort all of the includes with ./utils/sort_includes.py so that
subsequent changes are easier to review. About to fix some layering
issues, and wanted to separate out the necessary churn.

Also comment and sink the include of "Windows.h" in three .inc files to
match the usage in Memory.inc.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198685 91177308-0d34-0410-b5e6-96231b3b80d8
2014-01-07 11:48:04 +00:00
Rafael Espindola
8e0f67dcec Make the llvm mangler depend only on DataLayout.
Before this patch any program that wanted to know the final symbol name of a
GlobalValue had to link with Target.

This patch implements a compromise solution where the mangler uses DataLayout.
This way, any tool that already links with Target (llc, clang) gets the exact
behavior as before and new IR files can be mangled without linking with Target.

With this patch the mangler is constructed with just a DataLayout and DataLayout
is extended to include the information the Mangler needs.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198438 91177308-0d34-0410-b5e6-96231b3b80d8
2014-01-03 19:21:54 +00:00
Rafael Espindola
33cc3f81c1 Remove the 's' DataLayout specification
During the years there have been some attempts at figuring out how to
align byval arguments. A look at the commit log suggests that they
were

* Use the ABI alignment.
* When that was not sufficient for x86-64, I added the 's' specification to
  DataLayout.
* When that was not sufficient Evan added the virtual getByValTypeAlignment.
* When even that was not sufficient, we just got the FE to add the alignment
  to the byval.

This patch is just a simple cleanup that removes my first attempt at fixing the
problem. I also added an AArch64 implementation of getByValTypeAlignment to
make sure this patch is a nop. I also left the 's' parsing for backward
compatibility.

I will send a short email to llvmdev about the change for anyone maintaining
an out of tree target.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198287 91177308-0d34-0410-b5e6-96231b3b80d8
2014-01-01 22:29:43 +00:00
Rafael Espindola
ea53c6ae30 The preferred alignment defaults to the abi alignment. Omit if it is the same.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197400 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-16 18:01:51 +00:00
Rafael Espindola
7ef5520e8c On DataLayout, omit the default of p:64:64:64.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197397 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-16 17:15:29 +00:00
Rafael Espindola
1cf81e6244 Simplify the datalayout string of ARM and AArch64.
No functionality change.

Reviewed by Tim Northover.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197172 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-12 17:43:37 +00:00
Rafael Espindola
4a971705bc Remove the MachineMove class.
It was just a less powerful and more confusing version of
MCCFIInstruction. A side effect is that, since MCCFIInstruction uses
dwarf register numbers, calls to getDwarfRegNum are pushed out, which
should allow further simplifications.

I left the MachineModuleInfo::addFrameMove interface unchanged since
this patch was already fairly big.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181680 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-13 01:16:13 +00:00
Tim Northover
85d2760c8e AArch64: add branch fixup pass.
This is essentially a stripped-down version of the ConstandIslands pass (which
always had these two functions), providing just the features necessary for
correctness.

In particular there needs to be a way to resolve the situation where a
conditional branch's destination block ends up out of range.

This issue crops up when self-hosting for AArch64.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175269 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-15 14:32:20 +00:00
Tim Northover
1e8839302b AArch64: remove ConstantIsland pass & put literals in separate section.
This implements the review suggestion to simplify the AArch64 backend. If we
later discover that we *really* need the extra complexity of the
ConstantIslands pass for performance reasons it can be resurrected.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175258 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-15 09:33:43 +00:00
Tim Northover
5bd6cb2dab AArch64: add block comments where missing
Only comments affected. No code change at all.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175169 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-14 16:17:01 +00:00
Tim Northover
72062f5744 Add AArch64 as an experimental target.
This patch adds support for AArch64 (ARM's 64-bit architecture) to
LLVM in the "experimental" category. Currently, it won't be built
unless requested explicitly.

This initial commit should have support for:
    + Assembly of all scalar (i.e. non-NEON, non-Crypto) instructions
      (except the late addition CRC instructions).
    + CodeGen features required for C++03 and C99.
    + Compilation for the "small" memory model: code+static data <
      4GB.
    + Absolute and position-independent code.
    + GNU-style (i.e. "__thread") TLS.
    + Debugging information.

The principal omission, currently, is performance tuning.

This patch excludes the NEON support also reviewed due to an outbreak of
batshit insanity in our legal department. That will be committed soon bringing
the changes to precisely what has been approved.

Further reviews would be gratefully received.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174054 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-31 12:12:40 +00:00