Add some generic SchedWrites and assign resources for Swift and Cortex A9.
Reapply of r183257. (Removed empty InstRW for division on swift)
llvm-svn: 183319
An instruction with less than 3 inputs is trivially a fast immediate shift.
Reapply of 183256, should not have caused the tablegen segfault on linux either.
llvm-svn: 183314
The ARM backend did not expect LDRBi12 to hold a constant pool operand.
Allow for LLVM to deal with the instruction similar to how it deals with
LDRi12.
This fixes PR16215.
llvm-svn: 183238
NOTE: If this broke your out-of-tree backend, in *RegisterInfo.td, change
the instances of SubRegIndex that have a comps template arg to use the
ComposedSubRegIndex class instead.
In TableGen land, this adds Size and Offset attributes to SubRegIndex,
and the ComposedSubRegIndex class, for which the Size and Offset are
computed by TableGen. This also adds an accessor in MCRegisterInfo, and
Size/Offsets for the X86 and ARM subreg indices.
llvm-svn: 183020
These instructions are deprecated oddities, but we still need to be able to
disassemble (and reassemble) them if and when they're encountered.
Patch by Amaury de la Vieuville.
llvm-svn: 183011
The disassembly of VEXT instructions was too lax in the bits checked. This
fixes the case where the instruction affects Q-registers but a misaligned lane
was specified (should be UNDEFINED).
Patch by Amaury de la Vieuville
llvm-svn: 183003
Fixes PR16146: gdb.base__call-ar-st.exp fails after
pre-RA-sched=source fixes.
Patch by Xiaoyi Guo!
This also fixes an unsupported dbg.value test case. Codegen was
previously incorrect but the test was passing by luck.
llvm-svn: 182885
FastISel was only enabled for iOS ARM and Thumb2, this patch enables it
for ARM (not Thumb2) on Linux and NaCl.
Thumb2 support needs a bit more work, mainly around register class
restrictions.
The patch punts to SelectionDAG when doing TLS relocation on non-Darwin
targets. I will fix this and other FastISel-to-SelectionDAG failures in
a separate patch.
The patch also forces FastISel to retain frame pointers: iOS always
keeps them for backtracking (so emitted code won't change because of
this), but Linux was getting much worse code that was incorrect when
using big frames (such as test-suite's lencod). I'll also fix this in a
later patch, it will probably require a peephole so that FastISel
doesn't rematerialize frame pointers back-to-back.
The test changes are straightforward, similar to:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20130513/174279.html
They also add a vararg test that got dropped in that change.
I ran all of test-suite on A15 hardware with --optimize-option=-O0 and
all the tests pass.
llvm-svn: 182877
Tidy up three places where the register class for ARM and Thumb wasn't
restrictive enough:
- No PC dest for reg-reg add/orr/sub.
- No PC dest for shifts.
- No PC or SP for Thumb2 reg-imm add.
I encountered this while combining FastISel with
-verify-machineinstrs. These instructions defined registers whose
classes weren't restrictive enough, and the uses failed
verification. They're also undefined in the ISA, or would produce code
that FastISel wouldn't want. This doesn't fix the register class
narrowing issue (where uses should restrict definitions), and isn't
thorough, but it's a small step in the right direction.
llvm-svn: 182863
There was exactly one caller using this API right, the others were relying on
specific behavior of the default implementation. Since it's too hard to use it
right just remove it and standardize on the default behavior.
Defines away PR16132.
llvm-svn: 182636
This patch builds on some existing code to do CFG reconstruction from
a disassembled binary:
- MCModule represents the binary, and has a list of MCAtoms.
- MCAtom represents either disassembled instructions (MCTextAtom), or
contiguous data (MCDataAtom), and covers a specific range of addresses.
- MCBasicBlock and MCFunction form the reconstructed CFG. An MCBB is
backed by an MCTextAtom, and has the usual successors/predecessors.
- MCObjectDisassembler creates a module from an ObjectFile using a
disassembler. It first builds an atom for each section. It can also
construct the CFG, and this splits the text atoms into basic blocks.
MCModule and MCAtom were only sketched out; MCFunction and MCBB were
implemented under the experimental "-cfg" llvm-objdump -macho option.
This cleans them up for further use; llvm-objdump -d -cfg now generates
graphviz files for each function found in the binary.
In the future, MCObjectDisassembler may be the right place to do
"intelligent" disassembly: for example, handling constant islands is just
a matter of splitting the atom, using information that may be available
in the ObjectFile. Also, better initial atom formation than just using
sections is possible using symbols (and things like Mach-O's
function_starts load command).
This brings two minor regressions in llvm-objdump -macho -cfg:
- The printing of a relocation's referenced symbol.
- An annotation on loop BBs, i.e., which are their own successor.
Relocation printing is replaced by the MCSymbolizer; the basic CFG
annotation will be superseded by more related functionality.
llvm-svn: 182628
This is a basic first step towards symbolization of disassembled
instructions. This used to be done using externally provided (C API)
callbacks. This patch introduces:
- the MCSymbolizer class, that mimics the same functions that were used
in the X86 and ARM disassemblers to symbolize immediate operands and
to annotate loads based off PC (for things like c string literals).
- the MCExternalSymbolizer class, which implements the old C API.
- the MCRelocationInfo class, which provides a way for targets to
translate relocations (either object::RelocationRef, or disassembler
C API VariantKinds) to MCExprs.
- the MCObjectSymbolizer class, which does symbolization using what it
finds in an object::ObjectFile. This makes simple symbolization (with
no fancy relocation stuff) work for all object formats!
- x86-64 Mach-O and ELF MCRelocationInfos.
- A basic ARM Mach-O MCRelocationInfo, that provides just enough to
support the C API VariantKinds.
Most of what works in otool (the only user of the old symbolization API
that I know of) for x86-64 symbolic disassembly (-tvV) works, namely:
- symbol references: call _foo; jmp 15 <_foo+50>
- relocations: call _foo-_bar; call _foo-4
- __cf?string: leaq 193(%rip), %rax ## literal pool for "hello"
Stub support is the main missing part (because libObject doesn't know,
among other things, about mach-o indirect symbols).
As for the MCSymbolizer API, instead of relying on the disassemblers
to call the tryAdding* methods, maybe this could be done automagically
using InstrInfo? For instance, even though PC-relative LEAs are used
to get the address of string literals in a typical Mach-O file, a MOV
would be used in an ELF file. And right now, the explicit symbolization
only recognizes PC-relative LEAs. InstrInfo should have already have
most of what is needed to know what to symbolize, so this can
definitely be improved.
I'd also like to remove object::RelocationRef::getValueString (it seems
only used by relocation printing in objdump), as simply printing the
created MCExpr is definitely enough (and cleaner than string concats).
llvm-svn: 182625
This implements the @llvm.readcyclecounter intrinsic as the specific
MRC instruction specified in the ARM manuals for CPUs with the Power
Management extensions.
Older CPUs had slightly different methods which may also have to be
implemented eventually, but this should cover all v7 cases.
rdar://problem/13939186
llvm-svn: 182603
Performance monitors, including a basic cycle counter, are an official
extension in the ARMv7 specification. This adds support for enabling and
disabling them, orthogonally from CPU selection.
rdar://problem/13939186
llvm-svn: 182602
Introduction:
In case when stack alignment is 8 and GPRs parameter part size is not N*8:
we add padding to GPRs part, so part's last byte must be recovered at
address K*8-1.
We need to do it, since remained (stack) part of parameter starts from
address K*8, and we need to "attach" "GPRs head" without gaps to it:
Stack:
|---- 8 bytes block ----| |---- 8 bytes block ----| |---- 8 bytes...
[ [padding] [GPRs head] ] [ ------ Tail passed via stack ------ ...
FIX:
Note, once we added padding we need to correct *all* Arg offsets that are going
after padded one. That's why we need this fix: Arg offsets were never corrected
before this patch. See new test-cases included in patch.
We also don't need to insert padding for byval parameters that are stored in GPRs
only. We need pad only last byval parameter and only in case it outsides GPRs
and stack alignment = 8.
Though, stack area, allocated for recovered byval params, must satisfy
"Size mod 8 = 0" restriction.
This patch reduces stack usage for some cases:
We can reduce ArgRegsSaveArea since inner N*4 bytes sized byval params my be
"packed" with alignment 4 in some cases.
llvm-svn: 182237
This patch matches GCC behavior: the code used to only allow unaligned
load/store on ARM for v6+ Darwin, it will now allow unaligned load/store
for v6+ Darwin as well as for v7+ on Linux and NaCl.
The distinction is made because v6 doesn't guarantee support (but LLVM
assumes that Apple controls hardware+kernel and therefore have
conformant v6 CPUs), whereas v7 does provide this guarantee (and
Linux/NaCl behave sanely).
The patch keeps the -arm-strict-align command line option, and adds
-arm-no-strict-align. They behave similarly to GCC's -mstrict-align and
-mnostrict-align.
I originally encountered this discrepancy in FastIsel tests which expect
unaligned load/store generation. Overall this should slightly improve
performance in most cases because of reduced I$ pressure.
llvm-svn: 182175
This patch matches GCC behavior: the code used to only allow unaligned
load/store on ARM for v6+ Darwin, it will now allow unaligned load/store for
v6+ Darwin as well as for v7+ on other targets.
The distinction is made because v6 doesn't guarantee support (but LLVM assumes
that Apple controls hardware+kernel and therefore have conformant v6 CPUs),
whereas v7 does provide this guarantee (and Linux behaves sanely).
Overall this should slightly improve performance in most cases because of
reduced I$ pressure.
Patch by JF Bastien
llvm-svn: 181897
The transformation happening here is that we want to turn a
"mul(ext(X), ext(X))" into a "vmull(X, X)", stripping off the extension. We have
to make sure that X still has a valid vector type - possibly recreate an
extension to a smaller type. In case of a extload of a memory type smaller than
64 bit we used create a ext(load()). The problem with doing this - instead of
recreating an extload - is that an illegal type is exposed.
This patch fixes this by creating extloads instead of ext(load()) sequences.
Fixes PR15970.
radar://13871383
llvm-svn: 181842
return values are bitcasts.
The chain had previously been being clobbered with the entry node to
the dag, which sometimes caused other code in the function to be
erroneously deleted when tailcall optimization kicked in.
<rdar://problem/13827621>
llvm-svn: 181696
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
This commit implements the AsmParser for fnstart, fnend,
cantunwind, personality, handlerdata, pad, setfp, save, and
vsave directives.
This commit fixes some minor issue in the ARMELFStreamer:
* The switch back to corresponding section after the .fnend
directive.
* Emit the unwind opcode while processing .fnend directive
if there is no .handlerdata directive.
* Emit the unwind opcode to .ARM.extab while processing
.handlerdata even if .personality directive does not exist.
llvm-svn: 181603
indirect branch at the end of the BB. Otherwise if-converter, branch folding
pass may incorrectly update its successor info if it consider BB as fallthrough
to the next BB.
rdar://13782395
llvm-svn: 181161
Now even the small structures could be passed within byval (small enough
to be stored in GPRs).
In regression tests next function prototypes are checked:
PR15293:
%artz = type { i32 }
define void @foo(%artz* byval %s)
define void @foo2(%artz* byval %s, i32 %p, %artz* byval %s2)
foo: "s" stored in R0
foo2: "s" stored in R0, "s2" stored in R2.
Next AAPCS rules are checked:
5.5 Parameters Passing, C.4 and C.5,
"ParamSize" is parameter size in 32bit words:
-- NSAA != 0, NCRN < R4 and NCRN+ParamSize > R4.
Parameter should be sent to the stack; NCRN := R4.
-- NSAA != 0, and NCRN < R4, NCRN+ParamSize < R4.
Parameter stored in GPRs; NCRN += ParamSize.
llvm-svn: 181148
Build attribute sections can now be read if they exist via ELFObjectFile, and
the llvm-readobj tool has been extended with an option to dump this information
if requested. Regression tests are also included which exercise these features.
Also update the docs with a fixed ARM ABI link and a new link to the Addenda
which provides the build attributes specification.
llvm-svn: 181009
1. VarArgStyleRegisters: functionality that emits "store" instructions for byval regs moved out into separated method "StoreByValRegs". Before this patch VarArgStyleRegisters had confused use-cases. It was used for both variadic functions and for regular functions with byval parameters. In last case it created new stack-frame and registered it as VarArg frame, that is wrong.
This patch replaces VarArgsStyleRegisters usage for byval parameters with StoreByValRegs method.
2. In ARMMachineFunctionInfo, "get/setVarArgsRegSaveSize" was renamed to "get/setArgRegsSaveSize". By the same reason. Sometimes it was used for variadic functions, and sometimes for byval parameters in regular functions. Actually, this property means the size of registers, that keeps arguments, and thats why it was renamed.
3. In ARMISelLowering.cpp, ARMTargetLowering class, in methods computeRegArea and StoreByValRegs, VARegXXXXXX was renamed to ArgRegsXXXXXX still by the same reasons.
llvm-svn: 180774
"hint" space for Thumb actually overlaps the encoding space of the CPS
instruction. In actuality, hints can be defined as CPS instructions where imod
and M bits are all nil.
Handle decoding of permitted nop-compatible hints (i.e. nop, yield, wfi, wfe,
sev) in DecodeT2CPSInstruction.
This commit adds a proper diagnostic message for Imm0_4 and updates all tests.
Patch by Mihail Popa <Mihail.Popa@arm.com>.
llvm-svn: 180617
-- C.4 and C.5 statements, when NSAA is not equal to SP.
-- C.1.cp statement for VA functions. Note: There are no VFP CPRCs in a
variadic procedure.
Before this patch "NSAA != 0" means "don't use GPRs anymore ". But there are
some exceptions in AAPCS.
1. For non VA function: allocate all VFP regs for CPRC. When all VFPs are allocated
CPRCs would be sent to stack, while non CPRCs may be still allocated in GRPs.
2. Check that for VA functions all params uses GPRs and then stack.
No exceptions, no CPRCs here.
llvm-svn: 180011
Rather than just splitting the input type and hoping for the best, apply
a bit more cleverness. Just splitting the types until the source is
legal often leads to an illegal result time, which is then widened and a
scalarization step is introduced which leads to truly horrible code
generation. With the loop vectorizer, these sorts of operations are much
more common, and so it's worth extra effort to do them well.
Add a legalization hook for the operands of a TRUNCATE node, which will
be encountered after the result type has been legalized, but if the
operand type is still illegal. If simple splitting of both types
ends up with the result type of each half still being legal, just
do that (v16i16 -> v16i8 on ARM, for example). If, however, that would
result in an illegal result type (v8i32 -> v8i8 on ARM, for example),
we can get more clever with power-two vectors. Specifically,
split the input type, but also widen the result element size, then
concatenate the halves and truncate again. For example on ARM,
To perform a "%res = v8i8 trunc v8i32 %in" we transform to:
%inlo = v4i32 extract_subvector %in, 0
%inhi = v4i32 extract_subvector %in, 4
%lo16 = v4i16 trunc v4i32 %inlo
%hi16 = v4i16 trunc v4i32 %inhi
%in16 = v8i16 concat_vectors v4i16 %lo16, v4i16 %hi16
%res = v8i8 trunc v8i16 %in16
This allows instruction selection to generate three VMOVN instructions
instead of a sequences of moves, stores and loads.
Update the ARMTargetTransformInfo to take this improved legalization
into account.
Consider the simplified IR:
define <16 x i8> @test1(<16 x i32>* %ap) {
%a = load <16 x i32>* %ap
%tmp = trunc <16 x i32> %a to <16 x i8>
ret <16 x i8> %tmp
}
define <8 x i8> @test2(<8 x i32>* %ap) {
%a = load <8 x i32>* %ap
%tmp = trunc <8 x i32> %a to <8 x i8>
ret <8 x i8> %tmp
}
Previously, we would generate the truly hideous:
.syntax unified
.section __TEXT,__text,regular,pure_instructions
.globl _test1
.align 2
_test1: @ @test1
@ BB#0:
push {r7}
mov r7, sp
sub sp, sp, #20
bic sp, sp, #7
add r1, r0, #48
add r2, r0, #32
vld1.64 {d24, d25}, [r0:128]
vld1.64 {d16, d17}, [r1:128]
vld1.64 {d18, d19}, [r2:128]
add r1, r0, #16
vmovn.i32 d22, q8
vld1.64 {d16, d17}, [r1:128]
vmovn.i32 d20, q9
vmovn.i32 d18, q12
vmov.u16 r0, d22[3]
strb r0, [sp, #15]
vmov.u16 r0, d22[2]
strb r0, [sp, #14]
vmov.u16 r0, d22[1]
strb r0, [sp, #13]
vmov.u16 r0, d22[0]
vmovn.i32 d16, q8
strb r0, [sp, #12]
vmov.u16 r0, d20[3]
strb r0, [sp, #11]
vmov.u16 r0, d20[2]
strb r0, [sp, #10]
vmov.u16 r0, d20[1]
strb r0, [sp, #9]
vmov.u16 r0, d20[0]
strb r0, [sp, #8]
vmov.u16 r0, d18[3]
strb r0, [sp, #3]
vmov.u16 r0, d18[2]
strb r0, [sp, #2]
vmov.u16 r0, d18[1]
strb r0, [sp, #1]
vmov.u16 r0, d18[0]
strb r0, [sp]
vmov.u16 r0, d16[3]
strb r0, [sp, #7]
vmov.u16 r0, d16[2]
strb r0, [sp, #6]
vmov.u16 r0, d16[1]
strb r0, [sp, #5]
vmov.u16 r0, d16[0]
strb r0, [sp, #4]
vldmia sp, {d16, d17}
vmov r0, r1, d16
vmov r2, r3, d17
mov sp, r7
pop {r7}
bx lr
.globl _test2
.align 2
_test2: @ @test2
@ BB#0:
push {r7}
mov r7, sp
sub sp, sp, #12
bic sp, sp, #7
vld1.64 {d16, d17}, [r0:128]
add r0, r0, #16
vld1.64 {d20, d21}, [r0:128]
vmovn.i32 d18, q8
vmov.u16 r0, d18[3]
vmovn.i32 d16, q10
strb r0, [sp, #3]
vmov.u16 r0, d18[2]
strb r0, [sp, #2]
vmov.u16 r0, d18[1]
strb r0, [sp, #1]
vmov.u16 r0, d18[0]
strb r0, [sp]
vmov.u16 r0, d16[3]
strb r0, [sp, #7]
vmov.u16 r0, d16[2]
strb r0, [sp, #6]
vmov.u16 r0, d16[1]
strb r0, [sp, #5]
vmov.u16 r0, d16[0]
strb r0, [sp, #4]
ldm sp, {r0, r1}
mov sp, r7
pop {r7}
bx lr
Now, however, we generate the much more straightforward:
.syntax unified
.section __TEXT,__text,regular,pure_instructions
.globl _test1
.align 2
_test1: @ @test1
@ BB#0:
add r1, r0, #48
add r2, r0, #32
vld1.64 {d20, d21}, [r0:128]
vld1.64 {d16, d17}, [r1:128]
add r1, r0, #16
vld1.64 {d18, d19}, [r2:128]
vld1.64 {d22, d23}, [r1:128]
vmovn.i32 d17, q8
vmovn.i32 d16, q9
vmovn.i32 d18, q10
vmovn.i32 d19, q11
vmovn.i16 d17, q8
vmovn.i16 d16, q9
vmov r0, r1, d16
vmov r2, r3, d17
bx lr
.globl _test2
.align 2
_test2: @ @test2
@ BB#0:
vld1.64 {d16, d17}, [r0:128]
add r0, r0, #16
vld1.64 {d18, d19}, [r0:128]
vmovn.i32 d16, q8
vmovn.i32 d17, q9
vmovn.i16 d16, q8
vmov r0, r1, d16
bx lr
llvm-svn: 179989
Previously, when spilling 64-bit paired registers, an LDMIA with both
a FrameIndex and an offset was produced. This kind of instruction
shouldn't exist, and the extra operand was being confused with the
predicate, causing aborts later on.
This removes the invalid 0-offset from the instruction being
produced.
llvm-svn: 179956
I think it's almost impossible to fold atomic fences profitably under
LLVM/C++11 semantics. As a result, this is now unused and just
cluttering up the target interface.
llvm-svn: 179940
variant/dialect. Addresses a FIXME in the emitMnemonicAliases function.
Use and test case to come shortly.
rdar://13688439 and part of PR13340.
llvm-svn: 179804
The reference manual defines only 5 permitted values for the immediate field of the "hint" instruction:
1. nop (imm == 0)
2. yield (imm == 1)
3. wfe (imm == 2)
4. wfi (imm == 3)
5. sev (imm == 4)
Therefore, restrict the permitted values for the "hint" instruction to 0 through 4.
Patch by Mihail Popa <Mihail.Popa@arm.com>
llvm-svn: 179707
According to the ARM reference manual, constant offsets are mandatory for pre-indexed addressing modes.
The MC disassembler was not obeying this when the offset is 0.
It was producing instructions like: str r0, [r1]!.
Correct syntax is: str r0, [r1, #0]!.
This change modifies the dumping of operands so that the offset is always printed, regardless of its value, when pre-indexed addressing mode is used.
Patch by Mihail Popa <Mihail.Popa@arm.com>
llvm-svn: 179398
These instructions aren't universally available, but depend on a specific
extension to the normal ARM architecture (rather than, say, v6/v7/...) so a new
feature is appropriate.
This also enables the feature by default on A-class cores which usually have
these extensions, to avoid breaking existing code and act as a sensible
default.
llvm-svn: 179171
The Thumb2SizeReduction pass avoids false CPSR dependencies, except it
still aggressively creates tMOVi8 instructions because they are so
common.
Avoid creating false CPSR dependencies even for tMOVi8 instructions when
the the CPSR flags are known to have high latency. This allows integer
computation to overlap floating point computations.
Also process blocks in a reverse post-order and propagate high-latency
flags to successors.
<rdar://problem/13468102>
llvm-svn: 178773
Reapply r177968:
After commit 178074 we can now have undefined scheduler variants.
Move the CortexA9 resources into the CortexA9 SchedModel namespace. Define
resource mappings under the CortexA9 SchedModel. Define resources and mappings
for the SwiftModel.
Incooperate Andrew's feedback.
llvm-svn: 178460
This reverts commit r177968. It is causing failures in a local build bot.
"fatal error: error in backend: Expected a variant SchedClass"
Original commit message:
Move the CortexA9 resources into the CortexA9 SchedModel namespace. Define
resource mappings under the CortexA9 SchedModel. Define resources and mappings
for the SwiftModel.
llvm-svn: 178028
If PC or SP is the destination, the disassembler erroneously failed with the
invalid encoding, despite the manual saying that both are fine.
This patch addresses failure to decode encoding T4 of LDR (A8.8.62) which is a
postindexed load, where the offset 0xc is applied to SP after the load occurs.
llvm-svn: 178017
Move the CortexA9 resources into the CortexA9 SchedModel namespace. Define
resource mappings under the CortexA9 SchedModel. Define resources and mappings
for the SwiftModel.
llvm-svn: 177968
This is very much work in progress. Please send me a note if you start to depend
on the added abstract read/write resources. They are subject to change until
further notice.
The old itinerary is still the default.
llvm-svn: 177967
sure the base register and would-be writeback register don't conflict for
stores. This was already being done for loads.
Unfortunately, it is rather difficult to create a test case for this issue. It
was exposed in 450.soplex at LTO and requires unlucky register allocation.
<rdar://13394908>
llvm-svn: 177874
This patch lets the register scavenger make use of multiple spill slots in
order to guarantee that it will be able to provide multiple registers
simultaneously.
To support this, the RS's API has changed slightly: setScavengingFrameIndex /
getScavengingFrameIndex have been replaced by addScavengingFrameIndex /
isScavengingFrameIndex / getScavengingFrameIndices.
In forthcoming commits, the PowerPC backend will use this capability in order
to implement the spilling of condition registers, and some special-purpose
registers, without relying on r0 being reserved. In some cases, spilling these
registers requires two GPRs: one for addressing and one to hold the value being
transferred.
llvm-svn: 177774
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
The ARM backend currently has poor codegen for long sext/zext
operations, such as v8i8 -> v8i32. This patch addresses this
by performing a custom expansion in ARMISelLowering. It also
adds/changes the cost of such lowering in ARMTTI.
This partially addresses PR14867.
Patch by Pete Couperus
llvm-svn: 177380
The default logic marks them as too expensive.
For example, before this patch we estimated:
cost of 16 for instruction: %r = uitofp <4 x i16> %v0 to <4 x float>
While this translates to:
vmovl.u16 q8, d16
vcvt.f32.u32 q8, q8
All other costs are left to the values assigned by the fallback logic. Theses
costs are mostly reasonable in the sense that they get progressively more
expensive as the instruction sequences emitted get longer.
radar://13445992
llvm-svn: 177334
Fix cost of some "cheap" cast instructions. Before this patch we used to
estimate for example:
cost of 16 for instruction: %r = fptoui <4 x float> %v0 to <4 x i16>
While we would emit:
vcvt.s32.f32 q8, q8
vmovn.i32 d16, q8
vuzp.8 d16, d17
All other costs are left to the values assigned by the fallback logic. Theses
costs are mostly reasonable in the sense that they get progressively more
expensive as the instruction sequences emitted get longer.
radar://13434072
llvm-svn: 177333
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
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
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
Increase the cost of v8/v16-i8 to v8/v16-i32 casts and truncates as the backend
currently lowers those using stack accesses.
This was responsible for a significant degradation on
MultiSource/Benchmarks/Trimaran/enc-pc1/enc-pc1
where we vectorize one loop to a vector factor of 16. After this patch we select
a vector factor of 4 which will generate reasonable code.
unsigned char cle[32];
void test(short c) {
unsigned short compte;
for (compte = 0; compte <= 31; compte++) {
cle[compte] = cle[compte] ^ c;
}
}
radar://13220512
llvm-svn: 176898
The VDUP instruction source register doesn't allow a non-constant lane
index, so make sure we don't construct a ARM::VDUPLANE node asking it to
do so.
rdar://13328063
http://llvm.org/bugs/show_bug.cgi?id=13963
llvm-svn: 176413
dispatch code. As far as I can tell the thumb2 code is behaving as expected.
I was able to compile and run the associated test case for both arm and thumb1.
rdar://13066352
llvm-svn: 176363
This fixes an issue where trying to assemlbe valid ADR instructions would cause
LLVM to hit a failed assertion.
Patch by Keith Walker.
llvm-svn: 176189
to TargetFrameLowering, where it belongs. Incidentally, this allows us
to delete some duplicated (and slightly different!) code in TRI.
There are potentially other layering problems that can be cleaned up
as a result, or in a similar manner.
The refactoring was OK'd by Anton Korobeynikov on llvmdev.
Note: this touches the target interfaces, so out-of-tree targets may
be affected.
llvm-svn: 175788
It is possible that frame pointer is not found in the
callee saved info, thus FramePtrSpillFI may be incorrect
if we don't check the result of hasFP(MF).
Besides, if we enable the stack coloring algorithm, there
will be an assertion to ensure the slot is live. But in
the test case, %var1 is not live in the prologue of the
function, and we will get the assertion failure.
Note: There is similar code in ARMFrameLowering.cpp.
llvm-svn: 175616
In my previous commit:
"Merge a f32 bitcast of a v2i32 extractelt
A vectorized sitfp on doubles will get scalarized to a sequence of an
extract_element of <2 x i32>, a bitcast to f32 and a sitofp.
Due to the the extract_element, and the bitcast we will uneccessarily generate
moves between scalar and vector registers."
I added a pattern containing a copy_to_regclass. The copy_to_regclass is
actually not needed.
radar://13191881
llvm-svn: 175555
When creating an allocation hint for a register pair, make sure the hint
for the physical register reference is still in the allocation order.
rdar://13240556
llvm-svn: 175541
A vectorized sitfp on doubles will get scalarized to a sequence of an
extract_element of <2 x i32>, a bitcast to f32 and a sitofp.
Due to the the extract_element, and the bitcast we will uneccessarily generate
moves between scalar and vector registers.
The patch fixes this by using a COPY_TO_REGCLASS and a EXTRACT_SUBREG to extract
the element from the vector instead.
radar://13191881
llvm-svn: 175520
assembler should also accept a two arg form, as the docuemntation specifies that
the first (destination) register is optional.
This patch uses TwoOperandAliasConstraint to add the two argument form.
It also fixes an 80-column formatting problem in:
test/MC/ARM/neon-bitwise-encoding
<rdar://problem/12909419> Clang rejects ARM NEON assembly instructions
llvm-svn: 175221
The parser will now accept instructions with alignment specifiers written like
vld1.8 {d16}, [r0:64]
, while also still accepting the incorrect syntax
vld1.8 {d16}, [r0, :64]
llvm-svn: 175164
Lower reverse shuffles to a vrev64 and a vext instruction instead of the default
legalization of storing and loading to the stack. This is important because we
generate reverse shuffles in the loop vectorizer when we reverse store to an
array.
uint8_t Arr[N];
for (i = 0; i < N; ++i)
Arr[N - i - 1] = ...
radar://13171760
llvm-svn: 174929
function is successfully handled by fast-isel. That's because function
arguments are *always* handled by SDISel. Introduce FastLowerArguments to
allow each target to provide hook to handle formal argument lowering.
As a proof-of-concept, add ARMFastIsel::FastLowerArguments to handle
functions with 4 or fewer scalar integer (i8, i16, or i32) arguments. It
completely eliminates the need for SDISel for trivial functions.
rdar://13163905
llvm-svn: 174855
Adds a function to target transform info to query for the cost of address
computation. The cost model analysis pass now also queries this interface.
The code in LoopVectorize adds the cost of address computation as part of the
memory instruction cost calculation. Only there, we know whether the instruction
will be scalarized or not.
Increase the penality for inserting in to D registers on swift. This becomes
necessary because we now always assume that address computation has a cost and
three is a closer value to the architecture.
radar://13097204
llvm-svn: 174713
Use the validateTargetOperandClass() hook to match literal '#0' operands in
InstAlias definitions. Previously this required per-instruction C++ munging of the
operand list, but not is handled as a natural part of the matcher. Much better.
No additional tests are required, as the pre-existing tests for these instructions
exercise the new behaviour as being functionally equivalent to the old.
llvm-svn: 174488
Swift has a renaming dependency if we load into D subregisters. We don't have a
way of distinguishing between insertelement operations of values from loads and
other values. Therefore, we are pessimistic for now (The performance problem
showed up in example 14 of gcc-loops).
radar://13096933
llvm-svn: 174300
infrastructure on MCStreamer to test for whether there is an
MCELFStreamer object available.
This is just a cleanup on the AsmPrinter side of things, moving ad-hoc
tests of random APIs to a direct type query. But the AsmParser
completely broken. There were no tests, it just blindly cast its
streamer to an MCELFStreamer and started manipulating it.
I don't have a test case -- this actually failed on LLVM's own
regression test suite. Unfortunately the failure only appears when the
stars, compilers, and runtime align to misbehave when we read a pointer
to a formatted_raw_ostream as-if it were an MCAssembler. =/
UBSan would catch this immediately.
Many thanks to Matt for doing about 80% of the debugging work here in
GDB, Jim for helping to explain how exactly to fix this, and others for
putting up with the hair pulling that ensued during debugging it.
llvm-svn: 174118
isa<> and dyn_cast<>. In several places, code is already hacking around
the absence of this, and there seem to be several interfaces that might
be lifted and/or devirtualized using this.
This change was based on a discussion with Jim Grosbach about how best
to handle testing for specific MCStreamer subclasses. He said that this
was the correct end state, and everything else was too hacky so
I decided to just make it so.
No functionality should be changed here, this is just threading the kind
through all the constructors and setting up the classof overloads.
llvm-svn: 174113
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.
llvm-svn: 174054
and update ELF header e_flags.
Currently gathering information such as symbol,
section and data is done by collecting it in an
MCAssembler object. From MCAssembler and MCAsmLayout
objects ELFObjectWriter::WriteObject() forms and
streams out the ELF object file.
This patch just adds a few members to the MCAssember
class to store and access the e_flag settings. It
allows for runtime additions to the e_flag by
assembler directives. The standalone assembler can
get to MCAssembler from getParser().getStreamer().getAssembler().
This patch is the generic infrastructure and will be
followed by patches for ARM and Mips for their target
specific use.
Contributer: Jack Carter
llvm-svn: 173882
Changing ARMBaseTargetMachine to return ARMTargetLowering intead of
the generic one (similar to x86 code).
Tests showing which instructions were added to cast when necessary
or cost zero when not. Downcast to 16 bits are not lowered in NEON,
so costs are not there yet.
llvm-svn: 173849
The ARM and Thumb variants of LDREXD and STREXD have different constraints and
take different operands. Previously the code expanding atomic operations didn't
take this into account and asserted in Thumb mode.
llvm-svn: 173780
conditions are met:
1. They share the same operand and are in the same BB.
2. Both outputs are used.
3. The target has a native instruction that maps to ISD::FSINCOS node or
the target provides a sincos library call.
Implemented the generic optimization in sdisel and enabled it for
Mac OSX. Also added an additional optimization for x86_64 Mac OSX by
using an alternative entry point __sincos_stret which returns the two
results in xmm0 / xmm1.
rdar://13087969
PR13204
llvm-svn: 173755
This was an experimental option, but needs to be defined
per-target. e.g. PPC A2 needs to aggressively hide latency.
I converted some in-order scheduling tests to A2. Hal is working on
more test cases.
llvm-svn: 171946
This is necessary not only for representing empty ranges, but for handling
multibyte characters in the input. (If the end pointer in a range refers to
a multibyte character, should it point to the beginning or the end of the
character in a char array?) Some of the code in the asm parsers was already
assuming this anyway.
llvm-svn: 171765
Absent a Contributor's License Agreement (CLA) with an LLVM legal entity and as
reviewed and agreed with Chris Lattner, add a patent license covering future
contributions from ARM until there is a CLA. This is to make explicit ARM's
grant of patent rights to recipients of LLVM containing ARM-contributed
material.
llvm-svn: 171721
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
into their new header subdirectory: include/llvm/IR. This matches the
directory structure of lib, and begins to correct a long standing point
of file layout clutter in LLVM.
There are still more header files to move here, but I wanted to handle
them in separate commits to make tracking what files make sense at each
layer easier.
The only really questionable files here are the target intrinsic
tablegen files. But that's a battle I'd rather not fight today.
I've updated both CMake and Makefile build systems (I think, and my
tests think, but I may have missed something).
I've also re-sorted the includes throughout the project. I'll be
committing updates to Clang, DragonEgg, and Polly momentarily.
llvm-svn: 171366
Arnold Schwaighofer