1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-20 03:23:01 +02:00
Commit Graph

290 Commits

Author SHA1 Message Date
Robin Morisset
b2dd60f27d Rename AtomicExpandLoadLinked into AtomicExpand
AtomicExpandLoadLinked is currently rather ARM-specific. This patch is the first of
a group that aim at making it more target-independent. See
http://lists.cs.uiuc.edu/pipermail/llvmdev/2014-August/075873.html
for details

The command line option is "atomic-expand"

llvm-svn: 216231
2014-08-21 21:50:01 +00:00
Jonathan Roelofs
31a7253462 Add a thread-model knob for lowering atomics on baremetal & single threaded systems
http://reviews.llvm.org/D4984

llvm-svn: 216182
2014-08-21 14:35:47 +00:00
Eric Christopher
378bc328f0 Temporarily Revert "Nuke the old JIT." as it's not quite ready to
be deleted. This will be reapplied as soon as possible and before
the 3.6 branch date at any rate.

Approved by Jim Grosbach, Lang Hames, Rafael Espindola.

This reverts commits r215111, 215115, 215116, 215117, 215136.

llvm-svn: 215154
2014-08-07 22:02:54 +00:00
Rafael Espindola
e9ebbe5559 Nuke the old JIT.
I am sure we will be finding bits and pieces of dead code for years to
come, but this is a good start.

Thanks to Lang Hames for making MCJIT a good replacement!

llvm-svn: 215111
2014-08-07 14:21:18 +00:00
Eric Christopher
da2b8b72dd Move all of the ARM subtarget features down onto the subtarget
rather than the target machine.

llvm-svn: 211799
2014-06-26 19:30:02 +00:00
Eric Christopher
3dac9feafc Add a new subtarget hook for whether or not we'd like to enable
the atomic load linked expander pass to run for a particular
subtarget. This requires a check of the subtarget and so save
the TargetMachine rather than only TargetLoweringInfo and update
all callers.

llvm-svn: 211314
2014-06-19 21:03:04 +00:00
Eric Christopher
5321af5276 Move ARMJITInfo off of the TargetMachine and down onto the subtarget.
This required untangling a mess of headers that included around.

This a recommit of r210953 with a fix for the removed accessor
for JITInfo.

llvm-svn: 211233
2014-06-18 22:48:09 +00:00
James Molloy
26c8f2b1cd Refactor the disabling of Thumb-1 LDM/STM generation
Originally I switched the LD/ST optimizer off in TargetMachine as it was previously, but Eric has suggested he'd prefer that it be short-circuited in the pass itself.

No functionality change.

llvm-svn: 211037
2014-06-16 16:42:53 +00:00
Eric Christopher
ac850efcf4 Temporarily revert r210953 in an attempt to bring the ARM buildbots
back.

llvm-svn: 210996
2014-06-15 19:55:14 +00:00
Eric Christopher
f650ca8a5b Remove InstrItineraryData off of the TargetMachine - it's already
on the subtarget and just forward the accessor.

llvm-svn: 210955
2014-06-13 23:11:13 +00:00
Eric Christopher
395ff9e8de Move ARMJITInfo off of the TargetMachine and down onto the subtarget.
This required untangling a mess of headers that included around.

llvm-svn: 210953
2014-06-13 23:04:46 +00:00
Eric Christopher
825e0c609b Move ARMSelectionDAGInfo from the TargetMachine to the subtarget.
llvm-svn: 210862
2014-06-13 00:20:39 +00:00
Eric Christopher
19f9acc9ab Move to a private function to initialize subtarget dependencies
so we can use initializer lists for the ARMSubtarget and then
use this to initialize a moved DataLayout on the subtarget from
the TargetMachine.

llvm-svn: 210861
2014-06-13 00:20:35 +00:00
Eric Christopher
234efcaac4 Have ARMSelectionDAGInfo take a DataLayout as it's argument as the
DAG has access to the subtarget and TargetSelectionDAGInfo only
needs a DataLayout.

llvm-svn: 210859
2014-06-12 23:39:49 +00:00
James Molloy
caa8dab3ef Disable the load/store optimization pass for Thumb-1.
Moritz's changes have improved codegen a lot, but further testing showed significant correctness problems. Disable by default until these have been worked out.

Patch by Moritz Roth!

llvm-svn: 210789
2014-06-12 15:18:33 +00:00
Tim Northover
3bb84c9bcc 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

llvm-svn: 209883
2014-05-30 10:09:59 +00:00
James Molloy
f2173e4edc Enable the Load/Store optimization pass for Thumb1 but make it return immediately for now.
Patch by Moritz Roth!

llvm-svn: 208991
2014-05-16 14:11:38 +00:00
Tim Northover
fa11ed01b6 Atomics: promote ARM's IR-based atomics pass to CodeGen.
Still only 32-bit ARM using it at this stage, but the promotion allows
direct testing via opt and is a reasonably self-contained patch on the
way to switching ARM64.

At this point, other targets should be able to make use of it without
too much difficulty if they want. (See ARM64 commit coming soon for an
example).

llvm-svn: 206485
2014-04-17 18:22:47 +00:00
Tim Northover
fa95942ab3 ARM: expand atomic ldrex/strex loops in IR
The previous situation where ATOMIC_LOAD_WHATEVER nodes were expanded
at MachineInstr emission time had grown to be extremely large and
involved, to account for the subtly different code needed for the
various flavours (8/16/32/64 bit, cmpxchg/add/minmax).

Moving this transformation into the IR clears up the code
substantially, and makes future optimisations much easier:

1. an atomicrmw followed by using the *new* value can be more
   efficient. As an IR pass, simple CSE could handle this
   efficiently.
2. Making use of cmpxchg success/failure orderings only has to be done
   in one (simpler) place.
3. The common "cmpxchg; did we store?" idiom can be exposed to
   optimisation.

I intend to gradually improve this situation within the ARM backend
and make sure there are no hidden issues before moving the code out
into CodeGen to be shared with (at least ARM64/AArch64, though I think
PPC & Mips could benefit too).

llvm-svn: 205525
2014-04-03 11:44:58 +00:00
Renato Golin
479ae7528b Remove duplicated DMB instructions
ARM specific optimiztion, finding places in ARM machine code where 2 dmbs
follow one another, and eliminating one of them.

Patch by Reinoud Elhorst.

llvm-svn: 205409
2014-04-02 09:03:43 +00:00
Christian Pirker
fb4268517c ARM: rename ARMle/ARMbe with ARMLE/ARMBE, and Thumble/Thumbbe with ThumbLE/ThumbBE
llvm-svn: 205317
2014-04-01 15:19:30 +00:00
Saleem Abdulrasool
b33e13b89d ARM: fix typo
llvm-svn: 205233
2014-03-31 18:09:10 +00:00
Christian Pirker
487402828a Add ARM big endian Target (armeb, thumbeb)
Reviewed at http://llvm-reviews.chandlerc.com/D3095

llvm-svn: 205007
2014-03-28 14:35:30 +00:00
Craig Topper
982736c41f Prune includes in ARM target.
llvm-svn: 204548
2014-03-22 23:51:00 +00:00
Craig Topper
e7c9ce2777 [C++11] Add 'override' keyword to virtual methods that override their base class.
llvm-svn: 203433
2014-03-10 02:09:33 +00:00
Tim Northover
83bbdcb246 GlobalMerge: move "-global-merge" option to the pass itself.
It's rather odd to have the flag enabling and disabling this pass only affect a
single target.

llvm-svn: 201559
2014-02-18 11:17:29 +00:00
Rafael Espindola
eae6386a1e 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.

llvm-svn: 198438
2014-01-03 19:21:54 +00:00
Rafael Espindola
859cb122ba Synchronize the NaCl DataLayout strings with the ones in clang.
Patch by Derek Schuff.

llvm-svn: 197640
2013-12-19 00:44:37 +00:00
Tim Northover
c9f20609f9 ARM: update comment to match reality
llvm-svn: 197570
2013-12-18 14:18:36 +00:00
Tim Northover
fe4e45a5b0 ARM: set default float ABI based on triple.
Clang sets the float-abi target option manually, but no longer
annotates each function with its ABI. This can lead to confusing
mistmatch between "clang -emit-llvm | llc" and normal clang
invocations.

Besides which, gnueabihf actually *is* hard-float. Defaulting to soft
was just perverse.

llvm-svn: 197554
2013-12-18 09:27:33 +00:00
Rafael Espindola
2b8c8dce53 On APCS, only try to align aggregates to 32 bits instead of 64.
This matches clang's behavior and since it is only a preference, it is not
an ABI issue.

llvm-svn: 197526
2013-12-17 21:36:54 +00:00
Rafael Espindola
0e6563fe28 Handle i64 first for clarity. No functionality change.
llvm-svn: 197524
2013-12-17 21:28:36 +00:00
Rafael Espindola
525165cec2 One last cleanup of LLVM's DataLayout strings.
Produce them in the same order on every target. The order is that of
getStringRepresentation: e|E-i*-f*-v*-a*-s*-n*-S*.

llvm-svn: 197411
2013-12-16 19:31:14 +00:00
Rafael Espindola
559bceac20 The preferred alignment defaults to the abi alignment. Omit if it is the same.
llvm-svn: 197400
2013-12-16 18:01:51 +00:00
Rafael Espindola
b5a53246ce Simplify the datalayout string of ARM and AArch64.
No functionality change.

Reviewed by Tim Northover.

llvm-svn: 197172
2013-12-12 17:43:37 +00:00
Rafael Espindola
3a54357741 Add comments documenting the ARM datalayout string.
llvm-svn: 196850
2013-12-10 00:37:37 +00:00
Rafael Espindola
b9446f75e7 Simplify further.
Thanks to Jim Grosbach for noticing it.

llvm-svn: 196846
2013-12-10 00:15:35 +00:00
Rafael Espindola
eb7e18aa45 Refactor the construction of the DataLayout string on ARM.
llvm-svn: 196843
2013-12-09 23:56:41 +00:00
Weiming Zhao
1594297410 Enable generating legacy IT block for AArch32
By default, the behavior of IT block generation will be determinated
dynamically base on the arch (armv8 vs armv7). This patch adds backend
options: -arm-restrict-it and -arm-no-restrict-it.  The former one
restricts the generation of IT blocks (the same behavior as thumbv8) for
both arches. The later one allows the generation of legacy IT block (the
same behavior as ARMv7 Thumb2) for both arches.

Clang will support -mrestrict-it and -mno-restrict-it, which is
compatible with GCC.

llvm-svn: 194592
2013-11-13 18:29:49 +00:00
Joey Gouly
03af45ccfe [ARMv8] Prevent generation of deprecated IT blocks on ARMv8 in Thumb mode.
IT blocks can only be one instruction lonf, and can only contain a subset of
the 16 instructions.

Patch by Artyom Skrobov!

llvm-svn: 190309
2013-09-09 14:21:49 +00:00
Silviu Baranga
5aac9ffdd0 Allow generation of vmla.f32 instructions when targeting Cortex-A15. The patch also adds the VFP4 feature to Cortex-A15 and fixes the DontUseFusedMAC predicate so that we can still generate vmla.f32 instructions on non-darwin targets with VFP4.
llvm-svn: 187349
2013-07-29 09:25:50 +00:00
Bill Wendling
4d82ecded8 Access the TargetLoweringInfo from the TargetMachine object instead of caching it. The TLI may change between functions. No functionality change.
llvm-svn: 184352
2013-06-19 21:07:11 +00:00
Bill Wendling
1919cdf3c7 Access the TargetLoweringInfo from the TargetMachine object instead of caching it. The TLI may change between functions. No functionality change.
llvm-svn: 184349
2013-06-19 20:51:24 +00:00
Rafael Espindola
237980d752 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.

llvm-svn: 181680
2013-05-13 01:16:13 +00:00
Silviu Baranga
bd61af84a7 Enabling the generation of dependency breakers for partial updates on Cortex-A15. Also fixing a small bug in getting the update clearence for VLD1LNd32.
llvm-svn: 178134
2013-03-27 12:38:44 +00:00
Renato Golin
0854fd9bef Avoid NEON SP-FP unless unsafe-math or Darwin
NEON is not IEEE 754 compliant, so we should avoid lowering single-precision
floating point operations with NEON unless unsafe-math is turned on. The
equivalent VFP instructions are IEEE 754 compliant, but in some cores they're
much slower, so some archs/OSs might still request it to be on by default,
such as Swift and Darwin.

llvm-svn: 177651
2013-03-21 18:47:47 +00:00
Silviu Baranga
ff316abe9d 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
2013-03-15 18:28:25 +00:00
Jim Grosbach
3cf2dcd376 ARM: Fix a few copy-paste errors.
s/X86/ARM/

llvm-svn: 171789
2013-01-07 21:12:13 +00:00
Chandler Carruth
30bd563e01 Switch TargetTransformInfo from an immutable analysis pass that requires
a TargetMachine to construct (and thus isn't always available), to an
analysis group that supports layered implementations much like
AliasAnalysis does. This is a pretty massive change, with a few parts
that I was unable to easily separate (sorry), so I'll walk through it.

The first step of this conversion was to make TargetTransformInfo an
analysis group, and to sink the nonce implementations in
ScalarTargetTransformInfo and VectorTargetTranformInfo into
a NoTargetTransformInfo pass. This allows other passes to add a hard
requirement on TTI, and assume they will always get at least on
implementation.

The TargetTransformInfo analysis group leverages the delegation chaining
trick that AliasAnalysis uses, where the base class for the analysis
group delegates to the previous analysis *pass*, allowing all but tho
NoFoo analysis passes to only implement the parts of the interfaces they
support. It also introduces a new trick where each pass in the group
retains a pointer to the top-most pass that has been initialized. This
allows passes to implement one API in terms of another API and benefit
when some other pass above them in the stack has more precise results
for the second API.

The second step of this conversion is to create a pass that implements
the TargetTransformInfo analysis using the target-independent
abstractions in the code generator. This replaces the
ScalarTargetTransformImpl and VectorTargetTransformImpl classes in
lib/Target with a single pass in lib/CodeGen called
BasicTargetTransformInfo. This class actually provides most of the TTI
functionality, basing it upon the TargetLowering abstraction and other
information in the target independent code generator.

The third step of the conversion adds support to all TargetMachines to
register custom analysis passes. This allows building those passes with
access to TargetLowering or other target-specific classes, and it also
allows each target to customize the set of analysis passes desired in
the pass manager. The baseline LLVMTargetMachine implements this
interface to add the BasicTTI pass to the pass manager, and all of the
tools that want to support target-aware TTI passes call this routine on
whatever target machine they end up with to add the appropriate passes.

The fourth step of the conversion created target-specific TTI analysis
passes for the X86 and ARM backends. These passes contain the custom
logic that was previously in their extensions of the
ScalarTargetTransformInfo and VectorTargetTransformInfo interfaces.
I separated them into their own file, as now all of the interface bits
are private and they just expose a function to create the pass itself.
Then I extended these target machines to set up a custom set of analysis
passes, first adding BasicTTI as a fallback, and then adding their
customized TTI implementations.

The fourth step required logic that was shared between the target
independent layer and the specific targets to move to a different
interface, as they no longer derive from each other. As a consequence,
a helper functions were added to TargetLowering representing the common
logic needed both in the target implementation and the codegen
implementation of the TTI pass. While technically this is the only
change that could have been committed separately, it would have been
a nightmare to extract.

The final step of the conversion was just to delete all the old
boilerplate. This got rid of the ScalarTargetTransformInfo and
VectorTargetTransformInfo classes, all of the support in all of the
targets for producing instances of them, and all of the support in the
tools for manually constructing a pass based around them.

Now that TTI is a relatively normal analysis group, two things become
straightforward. First, we can sink it into lib/Analysis which is a more
natural layer for it to live. Second, clients of this interface can
depend on it *always* being available which will simplify their code and
behavior. These (and other) simplifications will follow in subsequent
commits, this one is clearly big enough.

Finally, I'm very aware that much of the comments and documentation
needs to be updated. As soon as I had this working, and plausibly well
commented, I wanted to get it committed and in front of the build bots.
I'll be doing a few passes over documentation later if it sticks.

Commits to update DragonEgg and Clang will be made presently.

llvm-svn: 171681
2013-01-07 01:37:14 +00:00
Chandler Carruth
a490793037 Use the new script to sort the includes of every file under lib.
Sooooo many of these had incorrect or strange main module includes.
I have manually inspected all of these, and fixed the main module
include to be the nearest plausible thing I could find. If you own or
care about any of these source files, I encourage you to take some time
and check that these edits were sensible. I can't have broken anything
(I strictly added headers, and reordered them, never removed), but they
may not be the headers you'd really like to identify as containing the
API being implemented.

Many forward declarations and missing includes were added to a header
files to allow them to parse cleanly when included first. The main
module rule does in fact have its merits. =]

llvm-svn: 169131
2012-12-03 16:50:05 +00:00
Rafael Espindola
b6c97c170b Remove TargetELFWriterInfo.
All the credit goes to Jan Voung for noticing it was dead!

llvm-svn: 166902
2012-10-28 21:34:43 +00:00
Nadav Rotem
4a918454d8 Implement a basic VectorTargetTransformInfo interface to be used by the loop and bb vectorizers for modeling the cost of instructions.
llvm-svn: 166593
2012-10-24 17:22:41 +00:00
Nadav Rotem
3cce3abf28 Reapply the TargerTransformInfo changes, minus the changes to LSR and Lowerinvoke.
llvm-svn: 166248
2012-10-18 23:22:48 +00:00
Bob Wilson
b6adb70bdd Temporarily revert the TargetTransform changes.
The TargetTransform changes are breaking LTO bootstraps of clang.  I am
working with Nadav to figure out the problem, but I am reverting it for now
to get our buildbots working.

This reverts svn commits: 165665 165669 165670 165786 165787 165997
and I have also reverted clang svn 165741

llvm-svn: 166168
2012-10-18 05:43:52 +00:00
Nadav Rotem
b82a3821f7 Add a new interface to allow IR-level passes to access codegen-specific information.
llvm-svn: 165665
2012-10-10 22:04:55 +00:00
Micah Villmow
bb1a25cd67 Move TargetData to DataLayout.
llvm-svn: 165402
2012-10-08 16:38:25 +00:00
Jush Lu
ff46f6b0c6 [arm-fast-isel] Add support for ELF PIC.
This is a preliminary step towards ELF support; currently ARMFastISel hasn't
been used for ELF object files yet.

llvm-svn: 164759
2012-09-27 05:21:41 +00:00
Silviu Baranga
11ff2a551d This patch introduces A15 as a target in LLVM.
llvm-svn: 163803
2012-09-13 15:05:10 +00:00
Bob Wilson
7d344104a7 Consistently use AnalysisID types in TargetPassConfig.
This makes it possible to just use a zero value to represent "no pass", so
the phony NoPassID global variable is no longer needed.

llvm-svn: 159568
2012-07-02 19:48:37 +00:00
Bob Wilson
0a1ef38836 Add all codegen passes to the PassManager via TargetPassConfig.
This is a preliminary step toward having TargetPassConfig be able to
start and stop the compilation at specified passes for unit testing
and debugging.  No functionality change.

llvm-svn: 159567
2012-07-02 19:48:31 +00:00
Bill Wendling
003b1bf46c Change the PassManager from a reference to a pointer.
The TargetPassManager's default constructor wants to initialize the PassManager
to 'null'. But it's illegal to bind a null reference to a null l-value. Make the
ivar a pointer instead.
PR12468

llvm-svn: 155902
2012-05-01 08:27:43 +00:00
Jim Grosbach
249356cbf3 Tidy up. 80 columns.
llvm-svn: 154226
2012-04-06 23:43:50 +00:00
Jakob Stoklund Olesen
ebee7e5cff Preserve implicit defs in ARMLoadStoreOptimizer.
When a number of sub-register VLRDS instructions are combined into a
VLDM, preserve any super-register implicit defs. This is required to
keep the register scavenger and machine code verifier happy.

Enable machine code verification after ARMLoadStoreOptimizer.
ARM/2012-01-26-CopyPropKills.ll was failing because of this.

llvm-svn: 153610
2012-03-28 22:50:56 +00:00
Andrew Trick
b9d2e9e81d Codegen pass definition cleanup. No functionality.
Moving toward a uniform style of pass definition to allow easier target configuration.
Globally declare Pass ID.
Globally declare pass initializer.
Use INITIALIZE_PASS consistently.
Add a call to the initializer from CodeGen.cpp.
Remove redundant "createPass" functions and "getPassName" methods.

While cleaning up declarations, cleaned up comments (sorry for large diff).

llvm-svn: 150100
2012-02-08 21:23:13 +00:00
Andrew Trick
fb574d1c19 TargetPassConfig: confine the MC configuration to TargetMachine.
Passes prior to instructon selection are now split into separate configurable stages.
Header dependencies are simplified.
The bulk of this diff is simply removal of the silly DisableVerify flags.

Sorry for the target header churn. Attempting to stabilize them.

llvm-svn: 149754
2012-02-04 02:56:59 +00:00
Andrew Trick
da0c52e742 Added TargetPassConfig. The first little step toward configuring codegen passes.
Allows command line overrides to be centralized in LLVMTargetMachine.cpp.
LLVMTargetMachine can intercept common passes and give precedence to command line overrides.
Allows adding "internal" target configuration options without touching TargetOptions.
Encapsulates the PassManager.
Provides a good point to initialize all CodeGen passes so that Pass ID's can be used in APIs.
Allows modifying the target configuration hooks without rebuilding the world.

llvm-svn: 149672
2012-02-03 05:12:41 +00:00
David Blaikie
576aba04f1 Unweaken vtables as per http://llvm.org/docs/CodingStandards.html#ll_virtual_anch
llvm-svn: 146960
2011-12-20 02:50:00 +00:00
Evan Cheng
68ba5536f3 - Add MachineInstrBundle.h and MachineInstrBundle.cpp. This includes a function
to finalize MI bundles (i.e. add BUNDLE instruction and computing register def
  and use lists of the BUNDLE instruction) and a pass to unpack bundles.
- Teach more of MachineBasic and MachineInstr methods to be bundle aware.
- Switch Thumb2 IT block to MI bundles and delete the hazard recognizer hack to
  prevent IT blocks from being broken apart.

llvm-svn: 146542
2011-12-14 02:11:42 +00:00
Nick Lewycky
7d0d3c2d58 Move global variables in TargetMachine into new TargetOptions class. As an API
change, now you need a TargetOptions object to create a TargetMachine. Clang
patch to follow.

One small functionality change in PTX. PTX had commented out the machine
verifier parts in their copy of printAndVerify. That now calls the version in
LLVMTargetMachine. Users of PTX who need verification disabled should rely on
not passing the command-line flag to enable it.

llvm-svn: 145714
2011-12-02 22:16:29 +00:00
Evan Cheng
2b239cbcf6 Sink codegen optimization level into MCCodeGenInfo along side relocation model
and code model. This eliminates the need to pass OptLevel flag all over the
place and makes it possible for any codegen pass to use this information.

llvm-svn: 144788
2011-11-16 08:38:26 +00:00
Devang Patel
8f9c569a13 svn mv Target/ARM/ARMGlobalMerge.cpp Transforms/Scalar/GlobalMerge.cpp
There is no reason to have simple IR level pass in lib/Target.

llvm-svn: 142200
2011-10-17 17:17:43 +00:00
Lang Hames
be4997db2f Add a natural stack alignment field to TargetData, and prevent InstCombine from
promoting allocas to preferred alignments that exceed the natural
alignment. This avoids some potentially expensive dynamic stack realignments.

The natural stack alignment is set in target data strings via the "S<size>"
option. Size is in bits and must be a multiple of 8. The natural stack alignment
defaults to "unspecified" (represented by a zero value), and the "unspecified"
value does not prevent any alignment promotions. Target maintainers that care
about avoiding promotions should explicitly add the "S<size>" option to their
target data strings.

llvm-svn: 141599
2011-10-10 23:42:08 +00:00
Jakob Stoklund Olesen
cfd5280df5 Use ExecutionDepsFix instead of NEONMoveFix.
This enables NEON domain tracking across basic blocks, but should
otherwise do the same thing.

llvm-svn: 140772
2011-09-29 02:48:41 +00:00
Bill Wendling
cc7e50a465 Move to ISelLowering.
llvm-svn: 140754
2011-09-29 01:13:55 +00:00
Bill Wendling
37088b0660 Ahem...actually *add* the ARMSjLjLowering pass to the pass manager.
llvm-svn: 140718
2011-09-28 20:29:28 +00:00
Bill Wendling
c63b31e1ec This is the start of the new SjLj EH preparation pass, which will replace the
current IR-level pass.

The old SjLj EH pass has some problems, especially with the new EH model. Most
significantly, it violates some of the new restrictions the new model has. For
instance, the 'dispatch' table wants to jump to the landing pad, but we cannot
allow that because only an invoke's unwind edge can jump to a landing pad. This
requires us to mangle the code something awful. In addition, we need to keep the
now dead landingpad instructions around instead of CSE'ing them because the
DWARF emitter uses that information (they are dead because no control flow edge
will execute them - the control flow edge from an invoke's unwind is superceded
by the edge coming from the dispatch).

Basically, this pass belongs not at the IR level where SSA is king, but at the
code-gen level, where we have more flexibility.

llvm-svn: 140646
2011-09-27 22:14:12 +00:00
Evan Cheng
7de8defb9c Hide -global-merge option.
llvm-svn: 138540
2011-08-25 01:22:49 +00:00
Evan Cheng
6c89acc6d1 Add a command line option to disable global merge pass.
llvm-svn: 138536
2011-08-25 01:00:36 +00:00
Evan Cheng
9a2607f3b9 Remove a out-of-place comment.
llvm-svn: 138534
2011-08-25 00:54:42 +00:00
Evan Cheng
420bf5446c Move TargetRegistry and TargetSelect from Target to Support where they belong.
These are strictly utilities for registering targets and components.

llvm-svn: 138450
2011-08-24 18:08:43 +00:00
Evan Cheng
41aa34c417 Sink ARM mc routines into MCTargetDesc.
llvm-svn: 135825
2011-07-23 00:00:19 +00:00
Evan Cheng
55d7fcc5f7 - Move CodeModel from a TargetMachine global option to MCCodeGenInfo.
- Introduce JITDefault code model. This tells targets to set different default
  code model for JIT. This eliminates the ugly hack in TargetMachine where
  code model is changed after construction.

llvm-svn: 135580
2011-07-20 07:51:56 +00:00
Evan Cheng
bfc0cac54d Introduce MCCodeGenInfo, which keeps information that can affect codegen
(including compilation, assembly). Move relocation model Reloc::Model from
TargetMachine to MCCodeGenInfo so it's accessible even without TargetMachine.

llvm-svn: 135468
2011-07-19 06:37:02 +00:00
Evan Cheng
9e8f90a020 Rename createAsmInfo to createMCAsmInfo and move registration code to MCTargetDesc to prepare for next round of changes.
llvm-svn: 135219
2011-07-14 23:50:31 +00:00
Evan Cheng
50f2d8d304 Eliminate asm parser's dependency on TargetMachine:
- Each target asm parser now creates its own MCSubtatgetInfo (if needed).
- Changed AssemblerPredicate to take subtarget features which tablegen uses
  to generate asm matcher subtarget feature queries. e.g.
  "ModeThumb,FeatureThumb2" is translated to
  "(Bits & ModeThumb) != 0 && (Bits & FeatureThumb2) != 0".

llvm-svn: 134678
2011-07-08 01:53:10 +00:00
Evan Cheng
9581f645b3 Factor ARM triple parsing out of ARMSubtarget. Another step towards making ARM subtarget info available to MC.
llvm-svn: 134569
2011-07-07 00:08:19 +00:00
Evan Cheng
034261674b Fix the ridiculous SubtargetFeatures API where it implicitly expects CPU name to
be the first encoded as the first feature. It then uses the CPU name to look up
features / scheduling itineray even though clients know full well the CPU name
being used to query these properties.

The fix is to just have the clients explictly pass the CPU name!

llvm-svn: 134127
2011-06-30 01:53:36 +00:00
Evan Cheng
76b1239b24 Remove TargetOptions.h dependency from ARMSubtarget.
llvm-svn: 133738
2011-06-23 18:15:17 +00:00
Daniel Dunbar
1c0a0fde5c ADT/Triple: Move a variety of clients to using isOSDarwin() and isOSWindows()
predicates.

llvm-svn: 129816
2011-04-19 21:14:45 +00:00
Bob Wilson
3daeb462cb This patch combines several changes from Evan Cheng for rdar://8659675.
Making use of VFP / NEON floating point multiply-accumulate / subtraction is
difficult on current ARM implementations for a few reasons.
1. Even though a single vmla has latency that is one cycle shorter than a pair
   of vmul + vadd, a RAW hazard during the first (4? on Cortex-a8) can cause
   additional pipeline stall. So it's frequently better to single codegen
   vmul + vadd.
2. A vmla folowed by a vmul, vmadd, or vsub causes the second fp instruction to
   stall for 4 cycles. We need to schedule them apart.
3. A vmla followed vmla is a special case. Obvious issuing back to back RAW
   vmla + vmla is very bad. But this isn't ideal either:
     vmul
     vadd
     vmla
   Instead, we want to expand the second vmla:
     vmla
     vmul
     vadd
   Even with the 4 cycle vmul stall, the second sequence is still 2 cycles
   faster.

Up to now, isel simply avoid codegen'ing fp vmla / vmls. This works well enough
but it isn't the optimial solution. This patch attempts to make it possible to
use vmla / vmls in cases where it is profitable.

A. Add missing isel predicates which cause vmla to be codegen'ed.
B. Make sure the fmul in (fadd (fmul)) has a single use. We don't want to
   compute a fmul and a fmla.
C. Add additional isel checks for vmla, avoid cases where vmla is feeding into
   fp instructions (except for the #3 exceptional case).
D. Add ARM hazard recognizer to model the vmla / vmls hazards.
E. Add a special pre-regalloc case to expand vmla / vmls when it's likely the
   vmla / vmls will trigger one of the special hazards.

Enable these fp vmlx codegen changes for Cortex-A9.

llvm-svn: 129775
2011-04-19 18:11:57 +00:00
Jim Grosbach
0510dc2765 Tidy up.
llvm-svn: 129034
2011-04-06 22:35:47 +00:00
NAKAMURA Takumi
00228d0c2c Triple::MinGW64 is deprecated and removed. We can use Triple::MinGW32 generally.
No one uses *-mingw64. mingw-w64 is represented as {i686|x86_64}-w64-mingw32. In llvm side, i686 and x64 can be treated as similar way.

llvm-svn: 125747
2011-02-17 12:24:17 +00:00
Rafael Espindola
547873da60 Add support for the --noexecstack option.
llvm-svn: 124077
2011-01-23 17:55:27 +00:00
Anton Korobeynikov
cf5967630b Rename TargetFrameInfo into TargetFrameLowering. Also, put couple of FIXMEs and fixes here and there.
llvm-svn: 123170
2011-01-10 12:39:04 +00:00
Evan Cheng
fc78767730 Making use of VFP / NEON floating point multiply-accumulate / subtraction is
difficult on current ARM implementations for a few reasons.
1. Even though a single vmla has latency that is one cycle shorter than a pair
   of vmul + vadd, a RAW hazard during the first (4? on Cortex-a8) can cause
   additional pipeline stall. So it's frequently better to single codegen
   vmul + vadd.
2. A vmla folowed by a vmul, vmadd, or vsub causes the second fp instruction to
   stall for 4 cycles. We need to schedule them apart.
3. A vmla followed vmla is a special case. Obvious issuing back to back RAW
   vmla + vmla is very bad. But this isn't ideal either:
     vmul
     vadd
     vmla
   Instead, we want to expand the second vmla:
     vmla
     vmul
     vadd
   Even with the 4 cycle vmul stall, the second sequence is still 2 cycles
   faster.

Up to now, isel simply avoid codegen'ing fp vmla / vmls. This works well enough
but it isn't the optimial solution. This patch attempts to make it possible to
use vmla / vmls in cases where it is profitable.

A. Add missing isel predicates which cause vmla to be codegen'ed.
B. Make sure the fmul in (fadd (fmul)) has a single use. We don't want to
   compute a fmul and a fmla.
C. Add additional isel checks for vmla, avoid cases where vmla is feeding into
   fp instructions (except for the #3 exceptional case).
D. Add ARM hazard recognizer to model the vmla / vmls hazards.
E. Add a special pre-regalloc case to expand vmla / vmls when it's likely the
   vmla / vmls will trigger one of the special hazards.

Work in progress, only A+B are enabled.

llvm-svn: 120960
2010-12-05 22:04:16 +00:00
Chris Lattner
4e53e6e198 tidy up
llvm-svn: 119462
2010-11-17 05:41:32 +00:00
Anton Korobeynikov
76c52dcf44 First step of huge frame-related refactoring: move emit{Prologue,Epilogue} out of TargetRegisterInfo to TargetFrameInfo, which is definitely much better suitable place
llvm-svn: 119097
2010-11-15 00:06:54 +00:00
Eric Christopher
474a93ff76 Revert the accidental commit I made reverting the previous commit.
llvm-svn: 118835
2010-11-11 20:50:14 +00:00
Eric Christopher
e7f27cf66a Revert this temporarily.
llvm-svn: 118827
2010-11-11 19:47:02 +00:00
Rafael Espindola
49d508fd19 Jim Asked us to move DataLayout on ARM back to the most specialized classes. Do
so and also change X86 for consistency.

Investigating if this can be improved a bit.

llvm-svn: 115469
2010-10-03 18:59:45 +00:00