This introduces a custom lowering for ISD::SETCCE (introduced in r253572)
that allows us to emit a short code sequence for 64-bit compares.
Before:
push {r7, lr}
cmp r0, r2
mov.w r0, #0
mov.w r12, #0
it hs
movhs r0, #1
cmp r1, r3
it ge
movge.w r12, #1
it eq
moveq r12, r0
cmp.w r12, #0
bne .LBB1_2
@ BB#1: @ %bb1
bl f
pop {r7, pc}
.LBB1_2: @ %bb2
bl g
pop {r7, pc}
After:
push {r7, lr}
subs r0, r0, r2
sbcs.w r0, r1, r3
bge .LBB1_2
@ BB#1: @ %bb1
bl f
pop {r7, pc}
.LBB1_2: @ %bb2
bl g
pop {r7, pc}
Saves around 80KB in Chromium's libchrome.so.
Some notes on this patch:
- I don't much like the ARMISD::BRCOND and ARMISD::CMOV combines I
introduced (nothing else needs them). However, they are necessary in
order to avoid poor codegen, and they seem similar to existing combines
in other backends (e.g. X86 combines (brcond (cmp (setcc Compare))) to
(brcond Compare)).
- No support for Thumb-1. This is in principle possible, but we'd need
to implement ARMISD::SUBE for Thumb-1.
Differential Revision: http://reviews.llvm.org/D15256
llvm-svn: 263962
- Rename getATOMIC to getSYNC, as llvm will soon be able to emit both
'__sync' libcalls and '__atomic' libcalls, and this function is for
the '__sync' ones.
- getInsertFencesForAtomic() has been replaced with
shouldInsertFencesForAtomic(Instruction), so that the decision can be
made per-instruction. This functionality will be used soon.
- emitLeadingFence/emitTrailingFence are no longer called if
shouldInsertFencesForAtomic returns false, and thus don't need to
check the condition themselves.
llvm-svn: 263665
Add support for TLS access for Windows on ARM. This generates a similar access
to MSVC for ARM.
The changes to the tablegen data is needed to support loading an external symbol
global that is not for a call. The adjustments to the DAG to DAG transforms are
needed to preserve the 32-bit move.
llvm-svn: 259676
Darwin TLS accesses most closely resemble ELF's general-dynamic situation,
since they have to be able to handle all possible situations. The descriptors
and so on are obviously slightly different though.
llvm-svn: 257039
Building on r253865 the crash is not limited to signed overflows.
Disable custom handling of unsigned 32-bit and 64-bit integer divide.
Add test cases for both 32-bit and 64-bit unsigned integer overflow.
llvm-svn: 254158
Disable custom handling of signed 32-bit and 64-bit integer divide.
Add test cases for both 32-bit and 64-bit integer overflow crashes.
llvm-svn: 253865
ARM V6T2 has instructions for efficient count-leading/trailing-zeros, so this should be
considered a cheap operation (and therefore fair game for speculation) for any ARM V6T2
implementation.
The net result of allowing this speculation for the regression tests in this patch is
that we get this code:
ctlz:
clz r0, r0
bx lr
cttz:
rbit r0, r0
clz r0, r0
bx lr
Instead of:
ctlz:
cmp r0, #0
moveq r0, #32
clzne r0, r0
bx lr
cttz:
cmp r0, #0
moveq r0, #32
rbitne r0, r0
clzne r0, r0
bx lr
This will help solve a general speculation/despeculation problem noted in PR24818:
https://llvm.org/bugs/show_bug.cgi?id=24818
Differential Revision: http://reviews.llvm.org/D14469
llvm-svn: 252639
Added fixes for stage2 failures: CMOV is not commutable; commuting the operands results in the condition being flipped! d'oh!
Original commit message:
If we have a CMOV, OR and AND combination such as:
if (x & CN)
y |= CM;
And:
* CN is a single bit;
* All bits covered by CM are known zero in y;
Then we can convert this to a sequence of BFI instructions. This will always be a win if CM is a single bit, will always be no worse than the TST & OR sequence if CM is two bits, and for thumb will be no worse if CM is three bits (due to the extra IT instruction).
llvm-svn: 252606
Summary:
The CLR's personality routine passes these in rdx/edx, not rax/eax.
Make getExceptionPointerRegister a virtual method parameterized by
personality function to allow making this distinction.
Similarly make getExceptionSelectorRegister a virtual method parameterized
by personality function, for symmetry.
Reviewers: pgavlin, majnemer, rnk
Subscribers: jyknight, dsanders, llvm-commits
Differential Revision: http://reviews.llvm.org/D14344
llvm-svn: 252383
If we have a CMOV, OR and AND combination such as:
if (x & CN)
y |= CM;
And:
* CN is a single bit;
* All bits covered by CM are known zero in y;
Then we can convert this to a sequence of BFI instructions. This will always be a win if CM is a single bit, will always be no worse than the TST & OR sequence if CM is two bits, and for thumb will be no worse if CM is three bits (due to the extra IT instruction).
llvm-svn: 252057
In PIC mode we were previously computing global variable addresses (or GOT
entry addresses) by adding the PC, the PC-relative GOT displacement and
the GOT-relative symbol/GOT entry displacement. Because the latter two
displacements are fixed, we ended up performing one more addition than
necessary.
This change causes us to compute addresses using a single PC-relative
displacement, resulting in a shorter code sequence. This reduces code size
by about 4% in a recent build of Chromium for Android.
As a result of this change we no longer need to compute the GOT base address
in the ARM backend, which allows us to remove the Global Base Reg pass and
SDAG lowering for the GOT.
We also now no longer use the GOT when addressing a symbol which is known
to be defined in the same linkage unit. Specifically, the symbol must have
either hidden visibility or a strong definition in the current module in
order to not use the the GOT.
This is a change from the previous behaviour where we would use the GOT to
address externally visible symbols defined in the same module. I think the
only cases where this could matter are cases involving symbol interposition,
but we don't really support that well anyway.
Differential Revision: http://reviews.llvm.org/D13650
llvm-svn: 251322
We were previously codegen'ing memcpy as regular load/store operations and
hoping that the register allocator would allocate registers in ascending order
so that we could apply an LDM/STM combine after register allocation. According
to the commit that first introduced this code (r37179), we planned to teach the
register allocator to allocate the registers in ascending order. This never got
implemented, and up to now we've been stuck with very poor codegen.
A much simpler approach for achieving better codegen is to create MEMCPY pseudo
instructions, attach scratch virtual registers to them and then, post register
allocation, expand the MEMCPYs into LDM/STM pairs using the scratch registers.
The register allocator will have picked arbitrary registers which we sort when
expanding the MEMCPY. This approach also avoids the need to repeatedly calculate
offsets which ultimately ought to be eliminated pre-RA in order to decrease
register pressure.
Fixes PR9199 and PR23768.
[This is based on Peter Collingbourne's r238473 which was reverted.]
Differential Revision: http://reviews.llvm.org/D13239
Change-Id: I727543c2e94136e0f80b8e22d5642d7b9ee5b458
Author: Peter Collingbourne <peter@pcc.me.uk>
llvm-svn: 249322
We now emit the compiler generated divide by zero check that was needed for the
MSVC routines. We construct a psuedo-instruction for the DBZ check as the
operation requires splitting up the BB. For the 64-bit operations, we need to
custom expand the node as we need to insert the DBZ check and then emit the
libcall to the appropriate name. Because this is target specific, it seemed
better to reproduce the expansion operation from the target-agnostic type
legalization rather than sink this there to avoid the duplication. The division
library calls now match MSVC semantically.
llvm-svn: 248561
ARM counterpart to r248291:
In the comparison failure block of a cmpxchg expansion, the initial
ldrex/ldxr will not be followed by a matching strex/stxr.
On ARM/AArch64, this unnecessarily ties up the execution monitor,
which might have a negative performance impact on some uarchs.
Instead, release the monitor in the failure block.
The clrex instruction was designed for this: use it.
Also see ARMARM v8-A B2.10.2:
"Exclusive access instructions and Shareable memory locations".
Differential Revision: http://reviews.llvm.org/D13033
llvm-svn: 248294
We used to have this magic "hasLoadLinkedStoreConditional()" callback,
which really meant two things:
- expand cmpxchg (to ll/sc).
- expand atomic loads using ll/sc (rather than cmpxchg).
Remove it, and, instead, introduce explicit callbacks:
- bool shouldExpandAtomicCmpXchgInIR(inst)
- AtomicExpansionKind shouldExpandAtomicLoadInIR(inst)
Differential Revision: http://reviews.llvm.org/D12557
llvm-svn: 247429
llvm.eh.sjlj.setjmp was used as part of the SjLj exception handling
style but is also used in clang to implement __builtin_setjmp. The ARM
backend needs to output additional dispatch tables for the SjLj
exception handling style, these tables however can't be emitted if
llvm.eh.sjlj.setjmp is simply used for __builtin_setjmp and no actual
landing pad blocks exist.
To solve this issue a new llvm.eh.sjlj.setup_dispatch intrinsic is
introduced which is used instead of llvm.eh.sjlj.setjmp in the SjLj
exception handling lowering, so we can differentiate between the case
where we actually need to setup a dispatch table and the case where we
just need the __builtin_setjmp semantic.
Differential Revision: http://reviews.llvm.org/D9313
llvm-svn: 242481
This patch allows the read_register and write_register intrinsics to
read/write the RBP/EBP registers on X86 iff the targeted register is
the frame pointer for the containing function.
Differential Revision: http://reviews.llvm.org/D10977
llvm-svn: 241827
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.
Reviewers: echristo
Subscribers: jholewinski, llvm-commits, rafael, yaron.keren
Differential Revision: http://reviews.llvm.org/D11040
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241778
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.
Reviewers: echristo
Subscribers: jholewinski, ted, yaron.keren, rafael, llvm-commits
Differential Revision: http://reviews.llvm.org/D11028
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241775
There is some functional change here because it changes target code from
atoi(3) to StringRef::getAsInteger which has error checking. For valid
constraints there should be no difference.
llvm-svn: 241411
The patch is generated using this command:
tools/clang/tools/extra/clang-tidy/tool/run-clang-tidy.py -fix \
-checks=-*,llvm-namespace-comment -header-filter='llvm/.*|clang/.*' \
llvm/lib/
Thanks to Eugene Kosov for the original patch!
llvm-svn: 240137
Summary:
But still handle them the same way since I don't know how they differ on
this target.
Of these, /U[qytnms]/ do not have backend tests but are accepted by clang.
No functional change intended.
Reviewers: t.p.northover
Reviewed By: t.p.northover
Subscribers: t.p.northover, aemerson, llvm-commits
Differential Revision: http://reviews.llvm.org/D8203
llvm-svn: 238921
This is important because of different addressing modes
depending on the address space for GPU targets.
This only adds the argument, and does not update
any of the uses to provide the correct address space.
llvm-svn: 238723
We were previously codegen'ing these as regular load/store operations and
hoping that the register allocator would allocate registers in ascending order
so that we could apply an LDM/STM combine after register allocation. According
to the commit that first introduced this code (r37179), we planned to teach
the register allocator to allocate the registers in ascending order. This
never got implemented, and up to now we've been stuck with very poor codegen.
A much simpler approach for achiveing better codegen is to create LDM/STM
instructions with identical sets of virtual registers, let the register
allocator pick arbitrary registers and order register lists when printing an
MCInst. This approach also avoids the need to repeatedly calculate offsets
which ultimately ought to be eliminated pre-RA in order to decrease register
pressure.
This is implemented by lowering the memcpy intrinsic to a series of SD-only
MCOPY pseudo-instructions which performs a memory copy using a given number
of registers. During SD->MI lowering, we lower MCOPY to LDM/STM. This is a
little unusual, but it avoids the need to encode register lists in the SD,
and we can take advantage of SD use lists to decide whether to use the _UPD
variant of the instructions.
Fixes PR9199.
Differential Revision: http://reviews.llvm.org/D9508
llvm-svn: 238473
to use the information in the module rather than TargetOptions.
We've had and clang has used the use-soft-float attribute for some
time now so have the backends set a subtarget feature based on
a particular function now that subtargets are created based on
functions and function attributes.
For the one middle end soft float check go ahead and create
an overloadable TargetLowering::useSoftFloat function that
just checks the TargetSubtargetInfo in all cases.
Also remove the command line option that hard codes whether or
not soft-float is set by using the attribute for all of the
target specific test cases - for the generic just go ahead and
add the attribute in the one case that showed up.
llvm-svn: 237079
Anton tried this 5 years ago but it was reverted due to extra VMOVs
being emitted. This can be easily fixed with a liberal application
of patterns - matching loads/stores and extractelts.
llvm-svn: 232958
Memcpy, and other memory intrinsics, typically tries to use LDM/STM if
the source and target addresses are 4-byte aligned. In CodeGenPrepare
look for calls to memory intrinsics and, if the object is on the
stack, 4-byte align it if it's large enough that we expect that memcpy
would want to use LDM/STM to copy it.
Differential Revision: http://reviews.llvm.org/D7908
llvm-svn: 232627
Summary:
This is instead of doing this in target independent code and is the last
non-functional change before targets begin to distinguish between
different memory constraints when selecting code for the ISD::INLINEASM
node.
Next, each target will individually move away from the idea that all
memory constraints behave like 'm'.
Subscribers: jholewinski, llvm-commits
Differential Revision: http://reviews.llvm.org/D8173
llvm-svn: 232373
The main issue being fixed here is that APCS targets handling a "byval align N"
parameter with N > 4 were miscounting what objects were where on the stack,
leading to FrameLowering setting the frame pointer incorrectly and clobbering
the stack.
But byval handling had grown over many years, and had multiple layers of cruft
trying to compensate for each other and calculate padding correctly. This only
really needs to be done once, in the HandleByVal function. Elsewhere should
just do what it's told by that call.
I also stripped out unnecessary APCS/AAPCS distinctions (now that Clang emits
byvals with the correct C ABI alignment), which simplified HandleByVal.
rdar://20095672
llvm-svn: 231959
This commit enables forming vector extloads for ARM.
It only does so for legal types, and when we can't fold the extension
in a wide/long form of the user instruction.
Enabling it for larger types isn't as good an idea on ARM as it is on
X86, because:
- we pretend that extloads are legal, but end up generating vld+vmov
- we have instructions like vld {dN, dM}, which can't be generated
when we "manually expand" extloads to vld+vmov.
For legal types, the combine doesn't fire that often: in the
integration tests only in a big endian testcase, where it removes a
pointless AND.
Related to rdar://19723053
Differential Revision: http://reviews.llvm.org/D7423
llvm-svn: 231396
Summary:
In PNaCl, most atomic instructions have their own @llvm.nacl.atomic.* function, each one, with a few exceptions, represents a consistent behaviour across all NaCl-supported targets. Unfortunately, the atomic RMW operations nand, [u]min, and [u]max aren't directly represented by any such @llvm.nacl.atomic.* function. This patch refines shouldExpandAtomicRMWInIR in TargetLowering so that a future `Le32TargetLowering` class can selectively inform the caller how the target desires the atomic RMW instruction to be expanded (ie via load-linked/store-conditional for ARM/AArch64, via cmpxchg for X86/others?, or not at all for Mips) if at all.
This does not represent a behavioural change and as such no tests were added.
Patch by: Richard Diamond.
Reviewers: jfb
Reviewed By: jfb
Subscribers: jfb, aemerson, t.p.northover, llvm-commits
Differential Revision: http://reviews.llvm.org/D7713
llvm-svn: 231250
a lookup, pass that in rather than use a naked call to getSubtargetImpl.
This involved passing down and around either a TargetMachine or
TargetRegisterInfo. Update all callers/definitions around the targets
and SelectionDAG.
llvm-svn: 230699
This required plumbing a TargetRegisterInfo through computeRegisterProperties
and into findRepresentativeClass which uses it for register class
iteration. This required passing a subtarget into a few target specific
initializations of TargetLowering.
llvm-svn: 230583
It was previously using the subtarget to get values for the global
offset without actually checking each function as it was generating
code. Go ahead and solidify the current behavior and make the
existing FIXMEs more prominent.
As a note the ARM backend previously had a thumb1 and non-thumb1
set of defaults. Only the former was tested so I've changed the
behavior to only use that for now.
llvm-svn: 230245
This patch adds an optimization in CodeGenPrepare to move an extractelement
right before a store when the target can combine them.
The optimization may promote any scalar operations to vector operations in the
way to make that possible.
** Context **
Some targets use different register files for both vector and scalar operations.
This means that transitioning from one domain to another may incur copy from one
register file to another. These copies are not coalescable and may be expensive.
For example, according to the scheduling model, on cortex-A8 a vector to GPR
move is 20 cycles.
** Motivating Example **
Let us consider an example:
define void @foo(<2 x i32>* %addr1, i32* %dest) {
%in1 = load <2 x i32>* %addr1, align 8
%extract = extractelement <2 x i32> %in1, i32 1
%out = or i32 %extract, 1
store i32 %out, i32* %dest, align 4
ret void
}
As it is, this IR generates the following assembly on armv7:
vldr d16, [r0] @vector load
vmov.32 r0, d16[1] @ cross-register-file copy: 20 cycles
orr r0, r0, #1 @ scalar bitwise or
str r0, [r1] @ scalar store
bx lr
Whereas we could generate much faster code:
vldr d16, [r0] @ vector load
vorr.i32 d16, #0x1 @ vector bitwise or
vst1.32 {d16[1]}, [r1:32] @ vector extract + store
bx lr
Half of the computation made in the vector is useless, but this allows to get
rid of the expensive cross-register-file copy.
** Proposed Solution **
To avoid this cross-register-copy penalty, we promote the scalar operations to
vector operations. The penalty will be removed if we manage to promote the whole
chain of computation in the vector domain.
Currently, we do that only when the chain of computation ends by a store and the
target is able to combine an extract with a store.
Stores are the most likely candidates, because other instructions produce values
that would need to be promoted and so, extracted as some point[1]. Moreover,
this is customary that targets feature stores that perform a vector extract (see
AArch64 and X86 for instance).
The proposed implementation relies on the TargetTransformInfo to decide whether
or not it is beneficial to promote a chain of computation in the vector domain.
Unfortunately, this interface is rather inaccurate for this level of details and
although this optimization may be beneficial for X86 and AArch64, the inaccuracy
will lead to the optimization being too aggressive.
Basically in TargetTransformInfo, everything that is legal has a cost of 1,
whereas, even if a vector type is legal, usually a vector operation is slightly
more expensive than its scalar counterpart. That will lead to too many
promotions that may not be counter balanced by the saving of the
cross-register-file copy. For instance, on AArch64 this penalty is just 4
cycles.
For now, the optimization is just enabled for ARM prior than v8, since those
processors have a larger penalty on cross-register-file copies, and the scope is
limited to basic blocks. Because of these two factors, we limit the effects of
the inaccuracy. Indeed, I did not want to build up a fancy cost model with block
frequency and everything on top of that.
[1] We can imagine targets that can combine an extractelement with other
instructions than just stores. If we want to go into that direction, the current
interfaces must be augmented and, moreover, I think this becomes a global isel
problem.
Differential Revision: http://reviews.llvm.org/D5921
<rdar://problem/14170854>
llvm-svn: 220978
Manman Ren