Massive check in. This changes the "-fast" flag to "-O#" in llc. If you want to
use the old behavior, the flag is -O0. This change allows for finer-grained
control over which optimizations are run at different -O levels.
Most of this work was pretty mechanical. The majority of the fixes came from
verifying that a "fast" variable wasn't used anymore. The JIT still uses a
"Fast" flag. I'll change the JIT with a follow-up patch.
llvm-svn: 70343
use the old behavior, the flag is -O0. This change allows for finer-grained
control over which optimizations are run at different -O levels.
Most of this work was pretty mechanical. The majority of the fixes came from
verifying that a "fast" variable wasn't used anymore. The JIT still uses a
"Fast" flag. I'm not 100% sure if it's necessary to change it there...
llvm-svn: 70270
- Add patterns for h-register extract, which avoids a shift and mask,
and in some cases a temporary register.
- Add address-mode matching for turning (X>>(8-n))&(255<<n), where
n is a valid address-mode scale value, into an h-register extract
and a scaled-offset address.
- Replace X86's MOV32to32_ and related instructions with the new
target-independent COPY_TO_SUBREG instruction.
On x86-64 there are complicated constraints on h registers, and
CodeGen doesn't currently provide a high-level way to express all of them,
so they are handled with a bunch of special code. This code currently only
supports extracts where the result is used by a zero-extend or a store,
though these are fairly common.
These transformations are not always beneficial; since there are only
4 h registers, they sometimes require extra move instructions, and
this sometimes increases register pressure because it can force out
values that would otherwise be in one of those registers. However,
this appears to be relatively uncommon.
llvm-svn: 68962
ISD::SIGN_EXTEND_INREG. Tablegen-generated code can handle
these cases, and the scheduling issues observed earlier
appear to be resolved now.
llvm-svn: 68959
builds.
--- Reverse-merging (from foreign repository) r68552 into '.':
U test/CodeGen/X86/tls8.ll
U test/CodeGen/X86/tls10.ll
U test/CodeGen/X86/tls2.ll
U test/CodeGen/X86/tls6.ll
U lib/Target/X86/X86Instr64bit.td
U lib/Target/X86/X86InstrSSE.td
U lib/Target/X86/X86InstrInfo.td
U lib/Target/X86/X86RegisterInfo.cpp
U lib/Target/X86/X86ISelLowering.cpp
U lib/Target/X86/X86CodeEmitter.cpp
U lib/Target/X86/X86FastISel.cpp
U lib/Target/X86/X86InstrInfo.h
U lib/Target/X86/X86ISelDAGToDAG.cpp
U lib/Target/X86/AsmPrinter/X86ATTAsmPrinter.cpp
U lib/Target/X86/AsmPrinter/X86IntelAsmPrinter.cpp
U lib/Target/X86/AsmPrinter/X86ATTAsmPrinter.h
U lib/Target/X86/AsmPrinter/X86IntelAsmPrinter.h
U lib/Target/X86/X86ISelLowering.h
U lib/Target/X86/X86InstrInfo.cpp
U lib/Target/X86/X86InstrBuilder.h
U lib/Target/X86/X86RegisterInfo.td
llvm-svn: 68560
This introduces a small regression on the generated code
quality in the case we are just computing addresses, not
loading values.
Will work on it and on X86-64 support.
llvm-svn: 68552
operand is a signed 32-bit immediate. Unlike with the 8-bit
signed immediate case, it isn't actually smaller to fold a
32-bit signed immediate instead of a load. In fact, it's
larger in the case of 32-bit unsigned immediates, because
they can be materialized with movl instead of movq.
llvm-svn: 67001
operands can't both be fully folded at the same time. For example,
in the included testcase, a global variable is being added with
an add of two values. The global variable wants RIP-relative
addressing, so it can't share the address with another base
register, but it's still possible to fold the initial add.
llvm-svn: 66865
and into the ScheduleDAGInstrs class, so that they don't get
destructed and re-constructed for each block. This fixes a
compile-time hot spot in the post-pass scheduler.
To help facilitate this, tidy and do some minor reorganization
in the scheduler constructor functions.
llvm-svn: 62275
special-purpose hook to a new pass. Also, add check to see if any
x87 virtual registers are used, to avoid doing any work in the
common case that no x87 code is needed.
llvm-svn: 59190
to be sign-extended when it is promoted to 64 bits for intermediate
offset calculations. The offset calculations are done as uint64_t so that
overflow conditions are well defined.
This fixes a problem which is currently hidden by the x86 AsmPrinter but
which was exposed by r58917 (which is temporarily reverted). See PR3027
for details.
llvm-svn: 59044
priority function. Instead, just iterate over the AllNodes list, which is
already in topological order. This eliminates a fair amount of bookkeeping,
and speeds up the isel phase by about 15% on many testcases.
The impact on most targets is that AddToISelQueue calls can be simply removed.
In the x86 target, there are two additional notable changes.
The rule-bending AND+SHIFT optimization in MatchAddress that creates new
pre-isel nodes during isel is now a little more verbose, but more robust.
Instead of either creating an invalid DAG or creating an invalid topological
sort, as it has historically done, it can now just insert the new nodes into
the node list at a position where they will be consistent with the topological
ordering.
Also, the address-matching code has logic that checked to see if a node was
"already selected". However, when a node is selected, it has all its uses
taken away via ReplaceAllUsesWith or equivalent, so it won't recieve any
further visits from MatchAddress. This code is now removed.
llvm-svn: 58748
flag. Then in a debugger developers can set breakpoints at these calls
to see waht is about to be selected and what the resulting subgraph
looks like. This really helps when debugging instruction selection.
llvm-svn: 58278
and add a TargetLowering hook for it to use to determine when this
is legal (i.e. not in PIC mode, etc.)
This allows instruction selection to emit folded constant offsets
in more cases, such as the included testcase, eliminating the need
for explicit arithmetic instructions.
This eliminates the need for the C++ code in X86ISelDAGToDAG.cpp
that attempted to achieve the same effect, but wasn't as effective.
Also, fix handling of offsets in GlobalAddressSDNodes in several
places, including changing GlobalAddressSDNode's offset from
int to int64_t.
The Mips, Alpha, Sparc, and CellSPU targets appear to be
unaware of GlobalAddress offsets currently, so set the hook to
false on those targets.
llvm-svn: 57748
create a new DAG node to represent the new shift to keep the
DAG consistent, even though it'll almost always be folded into
the address.
If a user of the resulting address has multiple uses, the
nodes may get revisited by a later MatchAddress call, in which
case DAG inconsistencies do matter.
This fixes PR2849.
llvm-svn: 57465
Instead, just create one, and make sure everything that needs
it can access it. Previously most of the SelectionDAGISel
subclasses all had their own TargetLowering object, which was
redundant with the TargetLowering object in the TargetMachine
subclasses, except on Sparc, where SparcTargetMachine
didn't have a TargetLowering object. Change Sparc to work
more like the other targets here.
llvm-svn: 57016
Completely eliminate the TopOrder std::vector. Instead, sort
the AllNodes list in place. This also eliminates the need to
call AllNodes.size(), a linear-time operation, before
performing the sort.
Also, eliminate the Sources temporary std::vector, since it
essentially duplicates the sorted result as it is being
built.
This also changes the direction of the topological sort
from bottom-up to top-down. The AllNodes list starts out in
roughly top-down order, so this reduces the amount of
reordering needed. Top-down is also more convenient for
Legalize, and ISel needed only minor adjustments.
llvm-svn: 56867
require RIP-relative addressing and use it to fix a bug
in X86FastISel in x86-64 PIC mode, where it was trying to
use base/index registers with RIP-relative addresses. This
fixes a bunch of x86-64 testsuite failures.
llvm-svn: 56676
- Add linkage to SymbolSDNode (default to external).
- Change ISD::ExternalSymbol to ISD::Symbol.
- Change ISD::TargetExternalSymbol to ISD::TargetSymbol
These changes pave the way to allowing SymbolSDNodes with non-external linkage.
llvm-svn: 56249
process up to a higher level. This allows FastISel to leverage
more of SelectionDAGISel's infastructure, such as updating Machine
PHI nodes.
Also, implement transitioning from SDISel back to FastISel in
the middle of a block, so it's now possible to go back and
forth. This allows FastISel to hand individual CallInsts and other
complicated things off to SDISel to handle, while handling the rest
of the block itself.
To help support this, reorganize the SelectionDAG class so that it
is allocated once and reused throughout a function, instead of
being completely reallocated for each block.
llvm-svn: 55219
out of X86ISelDAGToDAG.cpp C++ code and into tablegen code.
Among other things, using tablegen for these things makes them
friendlier to FastISel.
Tablegen can handle the case of i8 subregs on x86-32, but currently
the C++ code for that case uses MVT::Flag in a tricky way, and it
happens to schedule better in some cases. So for now, leave the
C++ code in place to handle the i8 case on x86-32.
llvm-svn: 55078
model, except for external calls; this makes
addressing modes PC-relative. Incomplete.
The assertion at the top of Emitter::runOnMachineFunction
was obviously bogus (always true) so I removed it.
If someone knows what the correct test should be to cover
all the various targets, please fix.
llvm-svn: 54656
replacement of multiple values. This is slightly more efficient
than doing multiple ReplaceAllUsesOfValueWith calls, and theoretically
could be optimized even further. However, an important property of this
new function is that it handles the case where the source value set and
destination value set overlap. This makes it feasible for isel to use
SelectNodeTo in many very common cases, which is advantageous because
SelectNodeTo avoids a temporary node and it doesn't require CSEMap
updates for users of values that don't change position.
Revamp MorphNodeTo, which is what does all the work of SelectNodeTo, to
handle operand lists more efficiently, and to correctly handle a number
of corner cases to which its new wider use exposes it.
This commit also includes a change to the encoding of post-isel opcodes
in SDNodes; now instead of being sandwiched between the target-independent
pre-isel opcodes and the target-dependent pre-isel opcodes, post-isel
opcodes are now represented as negative values. This makes it possible
to test if an opcode is pre-isel or post-isel without having to know
the size of the current target's post-isel instruction set.
These changes speed up llc overall by 3% and reduce memory usage by 10%
on the InstructionCombining.cpp testcase with -fast and -regalloc=local.
llvm-svn: 53728
and better control the abstraction. Rename the type
to MVT. To update out-of-tree patches, the main
thing to do is to rename MVT::ValueType to MVT, and
rewrite expressions like MVT::getSizeInBits(VT) in
the form VT.getSizeInBits(). Use VT.getSimpleVT()
to extract a MVT::SimpleValueType for use in switch
statements (you will get an assert failure if VT is
an extended value type - these shouldn't exist after
type legalization).
This results in a small speedup of codegen and no
new testsuite failures (x86-64 linux).
llvm-svn: 52044
definitions. This adds a new construct, "discard", for indicating
that a named node in the input matching pattern is to be discarded,
instead of corresponding to a node in the output pattern. This
allows tblgen to know where the arguments for the varaible_ops are
supposed to begin.
This fixes "rdar://5791600", whatever that is ;-).
llvm-svn: 51699
idea what this code (findNonImmUse) does, so I'm only guessing
that this is the right thing. It would be really really nice
if this had comments and perhaps switched to SmallPtrSet
(hint hint) :)
This fixes rdar://5886601, a crash on gcc.target/i386/sse4_1-pblendw.c
llvm-svn: 50252
LLVM Value/Use does and MachineRegisterInfo/MachineOperand does.
This allows constant time for all uses list maintenance operations.
The idea was suggested by Chris. Reviewed by Evan and Dan.
Patch is tested and approved by Dan.
On normal use-cases compilation speed is not affected. On very big basic
blocks there are compilation speedups in the range of 15-20% or even better.
llvm-svn: 48822
Change insert/extract subreg instructions to be able to be used in TableGen patterns.
Use the above features to reimplement an x86-64 pseudo instruction as a pattern.
llvm-svn: 48130
result into a MUL late in the X86 codegen process. ISD::MUL is
once again Legal on X86, so this is no longer needed. And, the
hack was suboptimal; see PR1874 for details.
llvm-svn: 47567
Added ISD::DECLARE node type to represent llvm.dbg.declare intrinsic. Now the intrinsic calls are lowered into a SDNode and lives on through out the codegen passes.
For now, since all the debugging information recording is done at isel time, when a ISD::DECLARE node is selected, it has the side effect of also recording the variable. This is a short term solution that should be fixed in time.
llvm-svn: 46659
This case returns the value in ST(0) and then has to convert it to an SSE
register. This causes significant codegen ugliness in some cases. For
example in the trivial fp-stack-direct-ret.ll testcase we used to generate:
_bar:
subl $28, %esp
call L_foo$stub
fstpl 16(%esp)
movsd 16(%esp), %xmm0
movsd %xmm0, 8(%esp)
fldl 8(%esp)
addl $28, %esp
ret
because we move the result of foo() into an XMM register, then have to
move it back for the return of bar.
Instead of hacking ever-more special cases into the call result lowering code
we take a much simpler approach: on x86-32, fp return is modeled as always
returning into an f80 register which is then truncated to f32 or f64 as needed.
Similarly for a result, we model it as an extension to f80 + return.
This exposes the truncate and extensions to the dag combiner, allowing target
independent code to hack on them, eliminating them in this case. This gives
us this code for the example above:
_bar:
subl $12, %esp
call L_foo$stub
addl $12, %esp
ret
The nasty aspect of this is that these conversions are not legal, but we want
the second pass of dag combiner (post-legalize) to be able to hack on them.
To handle this, we lie to legalize and say they are legal, then custom expand
them on entry to the isel pass (PreprocessForFPConvert). This is gross, but
less gross than the code it is replacing :)
This also allows us to generate better code in several other cases. For
example on fp-stack-ret-conv.ll, we now generate:
_test:
subl $12, %esp
call L_foo$stub
fstps 8(%esp)
movl 16(%esp), %eax
cvtss2sd 8(%esp), %xmm0
movsd %xmm0, (%eax)
addl $12, %esp
ret
where before we produced (incidentally, the old bad code is identical to what
gcc produces):
_test:
subl $12, %esp
call L_foo$stub
fstpl (%esp)
cvtsd2ss (%esp), %xmm0
cvtss2sd %xmm0, %xmm0
movl 16(%esp), %eax
movsd %xmm0, (%eax)
addl $12, %esp
ret
Note that we generate slightly worse code on pr1505b.ll due to a scheduling
deficiency that is unrelated to this patch.
llvm-svn: 46307
that "machine" classes are used to represent the current state of
the code being compiled. Given this expanded name, we can start
moving other stuff into it. For now, move the UsedPhysRegs and
LiveIn/LoveOuts vectors from MachineFunction into it.
Update all the clients to match.
This also reduces some needless #includes, such as MachineModuleInfo
from MachineFunction.
llvm-svn: 45467
sometimes emit "zero" and "all one" vectors multiple times,
for example:
_test2:
pcmpeqd %mm0, %mm0
movq %mm0, _M1
pcmpeqd %mm0, %mm0
movq %mm0, _M2
ret
instead of:
_test2:
pcmpeqd %mm0, %mm0
movq %mm0, _M1
movq %mm0, _M2
ret
This patch fixes this by always arranging for zero/one vectors
to be defined as v4i32 or v2i32 (SSE/MMX) instead of letting them be
any random type. This ensures they get trivially CSE'd on the dag.
This fix is also important for LegalizeDAGTypes, as it gets unhappy
when the x86 backend wants BUILD_VECTOR(i64 0) to be legal even when
'i64' isn't legal.
This patch makes the following changes:
1) X86TargetLowering::LowerBUILD_VECTOR now lowers 0/1 vectors into
their canonical types.
2) The now-dead patterns are removed from the SSE/MMX .td files.
3) All the patterns in the .td file that referred to immAllOnesV or
immAllZerosV in the wrong form now use *_bc to match them with a
bitcast wrapped around them.
4) X86DAGToDAGISel::SelectScalarSSELoad is generalized to handle
bitcast'd zero vectors, which simplifies the code actually.
5) getShuffleVectorZeroOrUndef is updated to generate a shuffle that
is legal, instead of generating one that is illegal and expecting
a later legalize pass to clean it up.
6) isZeroShuffle is generalized to handle bitcast of zeros.
7) several other minor tweaks.
This patch is definite goodness, but has the potential to cause random
code quality regressions. Please be on the lookout for these and let
me know if they happen.
llvm-svn: 44310
use ISD::{S,U}DIVREM and ISD::{S,U}MUL_HIO. Move the lowering code
associated with these operators into target-independent in LegalizeDAG.cpp
and TargetLowering.cpp.
llvm-svn: 42762
both results with a single div or idiv instruction. This uses new X86ISD
nodes for DIV and IDIV which are introduced during the legalize phase
so that the SelectionDAG's CSE can automatically eliminate redundant
computations.
llvm-svn: 42308
have situations where an SSE instruction turns into
multiple blocks, with the live range of an x87
register crossing them. To do this correctly make
sure we examine all blocks when inserting
FP_REG_KILL. PR 1697. (This was exposed by my
fix for PR 1681, but the same thing could happen
mixing x87 long double with SSE.)
llvm-svn: 42281
see if the base register is already occupied before assuming it can be
used. This fixes bogus code generation in the accompanying testcase.
llvm-svn: 41049
SSE mode (all but conversions <-> other FP types, I think):
>>Do not mark all-80-bit operations as "Requires[FPStack]"
(which really means "not SSE").
>>Refactor load-and-extend to facilitate this.
>>Update comments.
>>Handle long double in SSE when computing FP_REG_KILL.
llvm-svn: 40906
TargetLowering to SelectionDAG so that they have more convenient
access to the current DAG, in preparation for the ValueType routines
being changed from standalone functions to members of SelectionDAG for
the pre-legalize vector type changes.
llvm-svn: 37704
1) codegen a shift of a register as a shift, not an LEA.
2) teach the RA to convert a shift to an LEA instruction if it wants something
in three-address form.
This gives us asm diffs like:
- leal (,%eax,4), %eax
+ shll $2, %eax
which is faster on some processors and smaller on all of them.
and, more interestingly:
- movl 24(%esi), %eax
- leal (,%eax,4), %edi
+ movl 24(%esi), %edi
+ shll $2, %edi
Without #2, #1 was a significant pessimization in some cases.
This implements CodeGen/X86/shift-codegen.ll
llvm-svn: 35204
X + C to promote LEA formation. We would incorrectly apply it in some cases
(test) and miss it in others.
This fixes CodeGen/X86/2007-02-04-OrAddrMode.ll
llvm-svn: 33884
* PIC-aware internal structures in X86 Codegen have been refactored
* Visibility (default/weak) has been added
* Docs fixes (external weak linkage, visibility, formatting)
llvm-svn: 33136
- New target type "mingw" was introduced
- Same things for both mingw & cygwin are marked as "cygming" (as in
gcc)
- .lcomm is supported here, so allow LLVM to use it
- Correctly use underscored versions of setjmp & _longjmp for both mingw
& cygwin
llvm-svn: 32833
clearing the upper 8-bits instead of issuing two instructions. This also
eliminates the need to target the AH register which can be problematic on
x86-64.
llvm-svn: 31832
SCALAR_TO_VECTOR, even if the hasOneUse() check pass we may end up folding
the load into two instructions. Make sure we check the SCALAR_TO_VECTOR
has only one use as well.
llvm-svn: 31641
a framework for doing it right. This fixes
CodeGen/X86/2006-10-07-ScalarSSEMiscompile.ll.
Once X86DAGToDAGISel::SelectScalarSSELoad is implemented right, this task
will be done.
llvm-svn: 30817
Suppose the TokenFactor can reach the Op:
[Load chain]
^
|
[Load]
^ ^
| |
/ \-
/ |
/ [Op]
/ ^ ^
| .. |
| / |
[TokenFactor] |
^ |
| |
\ /
\ /
[Store]
If we move the Load below the TokenFactor, we would have created a cycle in
the DAG.
llvm-svn: 30040
selection is done. That's rather expensive especially in situations where it
isn't really needed.
Move back to a searching the predecessors, but make use of topological order
to trim the search space.
llvm-svn: 29559
Fold c2 in (x << c1) | c2 where (c2 < c1)
e.g.
int test(int x) {
return (x << 3) + 7;
}
This can be codegen'd as:
leal 7(,%eax,8), %eax
llvm-svn: 28550
If it reads the chain result of the call, then the use, callseq_start,
and call would form a cycle!
- Don't forget handle node replacement!
- There could also be a TokenFactor between the load and the callseq_start.
llvm-svn: 28420
Bill Wendling