The .b8 operations in PTX are far more limiting than I first thought. The mov operation isn't even supported, so there's no way of converting a .pred value into a .b8 without going via .b16, which is
not sensible. An improved implementation needs to use the fact that loads and stores automatically extend and truncate to implement support for EXTLOAD and TRUNCSTORE in order to correctly support
boolean values.
llvm-svn: 133873
Move the target-specific RecordRelocation logic out of the generic MC
MachObjectWriter and into the target-specific object writers. This allows
nuking quite a bit of target knowledge from the supposedly target-independent
bits in lib/MC.
llvm-svn: 133844
The fixup value comes in as the whole 32-bit value, so for the lo16 fixup,
the upper bits need to be masked off. Previously we assumed the masking had
already been done and asserted.
rdar://9635991
llvm-svn: 133818
The i8 type is required for boolean values, but can only use ld, st and mov instructions. The i1 type continues to be used for predicates.
llvm-svn: 133814
instructions can be used to match combinations of multiply/divide and VCVT
(between floating-point and integer, Advanced SIMD). Basically the VCVT
immediate operand that specifies the number of fraction bits corresponds to a
floating-point multiply or divide by the corresponding power of 2.
For example, VCVT (floating-point to fixed-point, Advanced SIMD) can replace a
combination of VMUL and VCVT (floating-point to integer) as follows:
Example (assume d17 = <float 8.000000e+00, float 8.000000e+00>):
vmul.f32 d16, d17, d16
vcvt.s32.f32 d16, d16
becomes:
vcvt.s32.f32 d16, d16, #3
Similarly, VCVT (fixed-point to floating-point, Advanced SIMD) can replace a
combinations of VCVT (integer to floating-point) and VDIV as follows:
Example (assume d17 = <float 8.000000e+00, float 8.000000e+00>):
vcvt.f32.s32 d16, d16
vdiv.f32 d16, d17, d16
becomes:
vcvt.f32.s32 d16, d16, #3
llvm-svn: 133813
.file and .loc directives.
Ideally, we would utilize the existing support in AsmPrinter for this, but
I cannot find a way to get .file and .loc directives to print without the
rest of the associated DWARF sections, which ptxas cannot handle.
llvm-svn: 133812
enables SelectionDAG::getLoad at MipsISelLowering.cpp:1914 to return a
pre-existing node instead of redundantly create a new node every time it is
called.
llvm-svn: 133811
target machine from those that are only needed by codegen. The goal is to
sink the essential target description into MC layer so we can start building
MC based tools without needing to link in the entire codegen.
First step is to refactor TargetRegisterInfo. This patch added a base class
MCRegisterInfo which TargetRegisterInfo is derived from. Changed TableGen to
separate register description from the rest of the stuff.
llvm-svn: 133782
parameters if SM >= 2.0
- Update test cases to be more robust against register allocation changes
- Bump up the number of registers to 128 per type
- Include Python script to re-generate register file with any number of
registers
llvm-svn: 133736
It has only one user. This eliminates the last include of
config.h from the public headers -- ideally, config.h
shouldn't even be installed by `make install` anymore.
llvm-svn: 133713
"Reinstate r133435 and r133449 (reverted in r133499) now that the clang
self-hosted build failure has been fixed (r133512)."
Due to some additional warnings.
llvm-svn: 133700
register allocation if it has a indirectbr or if we can duplicate it to
every predecessor.
This fixes the SingleSource/Benchmarks/Shootout-C++/matrix.cpp regression but
keeps the previous improvements to sunspider.
llvm-svn: 133682
If the linker supports it, this will hold the CIE and FDE information in a
compact format. The implementation of the compact unwinding emission is coming
soon.
llvm-svn: 133658
representing a constant reference to ValType. Normally this is just
"const ValType &", but when ValType is a std::vector we want to use
ArrayRef as the reference type.
llvm-svn: 133611
be one with only one unconditional branch and no phis. Duplicating the phis in this case
is possible, but requeres liveness analysis or breaking edges.
llvm-svn: 133607
1. (((x) & 0xFF00) >> 8) | (((x) & 0x00FF) << 8)
=> (bswap x) >> 16
2. ((x&0xff)<<8)|((x&0xff00)>>8)|((x&0xff000000)>>8)|((x&0x00ff0000)<<8))
=> (rotl (bswap x) 16)
This allows us to eliminate most of the def : Pat patterns for ARM rev16
revsh instructions. It catches many more cases for ARM and x86.
rdar://9609108
llvm-svn: 133503
ops.
This is a rewrite of the IV simplification algorithm used by
-disable-iv-rewrite. To avoid perturbing the default mode, I
temporarily split the driver and created SimplifyIVUsersNoRewrite. The
idea is to avoid doing opcode/pattern matching inside
IndVarSimplify. SCEV already does it. We want to optimize with the
full generality of SCEV, but optimize def-use chains top down on-demand rather
than rewriting the entire expression bottom-up. This was easy to do
for operations that SCEV can prove are identity function. So we're now
eliminating bitmasks and zero extends this way.
A result of this rewrite is that indvars -disable-iv-rewrite no longer
requires IVUsers.
llvm-svn: 133502
The current implementation generates stack loads/stores, which are
really just mov instructions from/to "special" registers. This may
not be the most efficient implementation, compared to an approach where
the stack registers are directly folded into instructions, but this is
easier to implement and I have yet to see a case where ptxas is unable
to see through this kind of register usage and know what is really
going on.
llvm-svn: 133443
Change PHINodes to store simple pointers to their incoming basic blocks,
instead of full-blown Uses.
Note that this loses an optimization in SplitCriticalEdge(), because we
can no longer walk the use list of a BasicBlock to find phi nodes. See
the comment I removed starting "However, the foreach loop is slow for
blocks with lots of predecessors".
Extend replaceAllUsesWith() on a BasicBlock to also update any phi
nodes in the block's successors. This mimics what would have happened
when PHINodes were proper Users of their incoming blocks. (Note that
this only works if OldBB->replaceAllUsesWith(NewBB) is called when
OldBB still has a terminator instruction, so it still has some
successors.)
llvm-svn: 133435
Change various bits of code to make better use of the existing PHINode
API, to insulate them from forthcoming changes in how PHINodes store
their operands.
llvm-svn: 133434
I don't think the AugmentedUse struct buys us much, either in
correctness or in ease of use. Ditch it, and simplify Use::getUser() and
User::allocHungoffUses().
llvm-svn: 133433
* Don't introduce a duplicated bb in the CFG
* When making a branch unconditional, clear the PredCond array so that it
is really unconditional.
llvm-svn: 133432
dragonegg buildbots back to life. Original commit message:
Teach early dup how to duplicate basic blocks with one successor and only phi instructions
into more complex blocks.
llvm-svn: 133430
all over the place in different styles and variants. Standardize on two
preferred entrypoints: one that takes a StructType and ArrayRef, and one that
takes StructType and varargs.
In cases where there isn't a struct type convenient, we now add a
ConstantStruct::getAnon method (whose name will make more sense after a few
more patches land).
It would be "really really nice" if the ConstantStruct::get and
ConstantVector::get methods didn't make temporary std::vectors.
llvm-svn: 133412
top level type without a specified number. This asmprinter has never
generated this, as you can tell by no tests being updated. It also isn't
documented.
llvm-svn: 133368
In cases such as the attached test, where the case value for a switch
destination is used in a phi node that follows the destination, it
might be better to replace that value with the condition value of the
switch, so that more blocks can be folded away with
TryToSimplifyUncondBranchFromEmptyBlock because there are less
conflicts in the phi node.
llvm-svn: 133344
type's bitwidth matches the (allocated) size of the alloca. This severely
pessimizes vector scalar replacement when the only vector type being used is
something like <3 x float> on x86 or ARM whose allocated size matches a
<4 x float>.
I hope to fix some of the flawed assumptions about allocated size throughout
scalar replacement and reenable this in most cases.
llvm-svn: 133338
for pre-2.9 bitcode files. We keep x86 unaligned loads, movnt, crc32, and the
target indep prefetch change.
As usual, updating the testsuite is a PITA.
llvm-svn: 133337
The LSDA is a bit difficult for the non-initiated to read. Even with comments,
it's not always clear what's going on. This wraps the ASM streamer in a class
that retains the LSDA and then emits a human-readable description of what's
going on in it.
So instead of having to make sense of:
Lexception1:
.byte 255
.byte 155
.byte 168
.space 1
.byte 3
.byte 26
Lset0 = Ltmp7-Leh_func_begin1
.long Lset0
Lset1 = Ltmp812-Ltmp7
.long Lset1
Lset2 = Ltmp913-Leh_func_begin1
.long Lset2
.byte 3
Lset3 = Ltmp812-Leh_func_begin1
.long Lset3
Lset4 = Leh_func_end1-Ltmp812
.long Lset4
.long 0
.byte 0
.byte 1
.byte 0
.byte 2
.byte 125
.long __ZTIi@GOTPCREL+4
.long __ZTIPKc@GOTPCREL+4
you can read this instead:
## Exception Handling Table: Lexception1
## @LPStart Encoding: omit
## @TType Encoding: indirect pcrel sdata4
## @TType Base: 40 bytes
## @CallSite Encoding: udata4
## @Action Table Size: 26 bytes
## Action 1:
## A throw between Ltmp7 and Ltmp812 jumps to Ltmp913 on an exception.
## For type(s): __ZTIi@GOTPCREL+4 __ZTIPKc@GOTPCREL+4
## Action 2:
## A throw between Ltmp812 and Leh_func_end1 does not have a landing pad.
llvm-svn: 133286
* We should change the generated code because of a debug use.
* Avoid creating debug uses of undef, as they become a kill.
Test to follow.
llvm-svn: 133255
* rounding modes for fp add, mul, sub now use .rn
* float -> int rounding correctly uses .rzi not .rni
* 32bit fdiv for sm13 uses div.rn (instead of div.approx)
* 32bit fdiv for sm10 now uses div (instead of div.approx)
Approx is not IEEE 754 compatible (and should be optionally set by a flag to the backend instead). The .rn rounding modifier is the PTX default anyway, but it's better to be explicit.
All these modifiers should be available by using __fmul_rz functions for example, but support will need to be added for this in the backend.
Patch by Dan Bailey
llvm-svn: 133253
The reserved R14-R15 are always saved in the prolog, and using CSRs
starting from R13 allows them to be saved in one instruction.
Thanks to Anton for explaining this.
llvm-svn: 133233
the old malloc/free instructions, and for 'sext' and 'zext' as function
attributes (they are spelled signext/zeroext now), and support for result
value attributes being specified after a function.
Additionally, diagnose invalid attributes on functions with an error message
instead of an abort in the verifier.
llvm-svn: 133229
Also switch the return type to ArrayRef<unsigned> which works out nicely
for ARM's implementation of this function because of the clever ArrayRef
constructors.
The name change indicates that the returned allocation order may contain
reserved registers as has been the case for a while.
llvm-svn: 133216
In Thumb mode we cannot handle GPR virtual registers, even though some
instructions can. When isel is lowering a CopyFromReg, it should limit
itself to subclasses of getRegClassFor(VT).
<rdar://problem/9624323>
llvm-svn: 133210
I think PBQP could use RegisterClassInfo, but it didn't fit neatly with
the external interfaces that PBQP uses, so I'll leave that to Lang.
llvm-svn: 133186
BranchProbabilityInfo (expect setEdgeWeight which is not available here).
Branch Weights are kept in MachineBasicBlocks. To turn off this analysis
set -use-mbpi=false.
llvm-svn: 133184
This is intended to support using REG_SEQUENCE SDNode's with type MVT::untyped, and is part of the long road to eliminating some of the hacks we currently use to support register pairs and other strange constraints, particularly on ARM NEON.
llvm-svn: 133178
This virtual function will replace allocation_order_begin/end as the one
to override when implementing custom allocation orders. It is simpler to
have one function return an ArrayRef than having two virtual functions
computing different ends of the same array.
Use getRawAllocationOrder() in place of allocation_order_begin() where
it makes sense, but leave some clients that look like they really want
the filtered allocation orders from RegisterClassInfo.
llvm-svn: 133170
accumulator forwarding. Specifically (from SVN log entry):
Distribute (A + B) * C to (A * C) + (B * C) to make use of NEON multiplier
accumulator forwarding:
vadd d3, d0, d1
vmul d3, d3, d2
=>
vmul d3, d0, d2
vmla d3, d1, d2
Make sure it catches cases where operand 1 is add/fadd/sub/fsub, which was
intended in the original revision.
llvm-svn: 133127
The limit in this patch is probably too high, but it is enough to stop DSE from going completely insane on a testcase I have (which has a single block with around 50,000 non-aliasing stores in it).
rdar://9471075
llvm-svn: 133111
This simplifies many of the target description files since it is common
for register classes to be related or contain sequences of numbered
registers.
I have verified that this doesn't change the files generated by TableGen
for ARM and X86. It alters the allocation order of MBlaze GPR and Mips
FGR32 registers, but I believe the change is benign.
llvm-svn: 133105
optimizations when emitting calls to the function; instead those calls may
use faster relocations which require the function to be immediately resolved
upon loading the dynamic object featuring the call. This is useful when it
is known that the function will be called frequently and pervasively and
therefore there is no merit in delaying binding of the function.
Currently only implemented for x86-64, where it turns into a call through
the global offset table.
Patch by Dan Gohman, who assures me that he's going to add LangRef documentation
for this once it's committed.
llvm-svn: 133080
Re-apply 133010, with fixes for inline assembler.
Original commit message:
"When an assembler local symbol is used but not defined in a module, a
Darwin assembler wants to issue a diagnostic to that effect."
Added fix to only perform the check when finalizing, as otherwise we're not
done and undefined symbols may simply not have been encountered yet.
Passes "make check" and a self-host check on Darwin.
llvm-svn: 133071
Note that this actually changes code generation, and someone who
understands this target better should check the changes.
- R12Q is now allocatable. I think it was omitted from the allocation
order by mistake since it isn't reserved. It as apparently used as a
GOT pointer sometimes, and it should probably be reserved if that is
the case.
- The GR64 registers are allocated in a different order now. The
register allocator will automatically put the CSRs last. There were
other changes to the order that may have been significant.
The test fix is because r0 and r1 swapped places in the allocation order.
llvm-svn: 133067
GetDemandBits (which must operate on the vector element type).
Fix the a usage of getZeroExtendInReg which must also be done on scalar types.
llvm-svn: 133052
At the time I wrote this code (circa 2007), TargetRegisterInfo was using a std::set to perform these queries. Switching to the static hashtables was an obvious improvement, but in reality there's no reason to do anything other than scan.
With this change, total LLC time on a whole-program 403.gcc is reduced by approximately 1.5%, almost all of which comes from a 15% reduction in LiveVariables time. It also reduces the binary size of LLC by 86KB, thanks to eliminating a bunch of very large static tables.
llvm-svn: 133051
Unfortunately we can't follow what the rest of the language does (wrapping it
in double-quotes) because that would cause an ambiguity with metadata strings,
so instead we escape any unusual characters with \xx escaping.
llvm-svn: 133050
the bits being cleared by the AND are not demanded by the BFI.
The previous BFI dag combine rule was actually incorrect (or used to be
correct until BFI representation changed).
rdar://9609030
llvm-svn: 133034
toString() now takes an optional bool argument that,
depending on the radix, adds the appropriate prefix
to the integer's string representation that makes it into a
meaningful C literal, e.g.:
hexademical: '-f' becomes '-0xf'
octal: '77' becomes '077'
binary: '110' becomes '0b110'
Patch by nobled@dreamwidth.org!
llvm-svn: 133032
converted to add x,x if x is a undef. add undef, undef does not guarantee
that the resulting low order bit is zero.
Fixes <rdar://problem/9453156> and <rdar://problem/9487392>.
llvm-svn: 133022
When an assembler local symbol is used but not defined in a module, a
Darwin assembler wants to issue a diagnostic to that effect.
rdar://9559714
llvm-svn: 133010
Dan noted that this would work on the case shown on the commit message. I think
the case that was failing was a bb ending with a redundant conditional jump:
...
jne foo
foo:
...
I was unable to find any such case in the tests or in a debug build of clang,
so I will revert this part of the patch and watch the bots.
llvm-svn: 133004