because it does not support CMOV of vectors. To implement this efficientlyi, we broadcast the condition bit and use a sequence of NAND-OR
to select between the two operands. This is the same sequence we use for targets that don't have vector BLENDs (like SSE2).
rdar://12201387
llvm-svn: 162926
- Add 'UseSSEx' to force SSE legacy insn not being selected when AVX is
enabled.
As the penalty of inter-mixing SSE and AVX instructions, we need
prevent SSE legacy insn from being generated except explicitly
specified through some intrinsics. For patterns supported by both
SSE and AVX, so far, we force AVX insn will be tried first relying on
AddedComplexity or position in td file. It's error-prone and
introduces bugs accidentally.
'UseSSEx' is disabled when AVX is turned on. For SSE insns inherited
by AVX, we need this predicate to force VEX encoding or SSE legacy
encoding only.
For insns not inherited by AVX, we still use the previous predicates,
i.e. 'HasSSEx'. So far, these insns fall into the following
categories:
* SSE insns with MMX operands
* SSE insns with GPR/MEM operands only (xFENCE, PREFETCH, CLFLUSH,
CRC, and etc.)
* SSE4A insns.
* MMX insns.
* x87 insns added by SSE.
2 test cases are modified:
- test/CodeGen/X86/fast-isel-x86-64.ll
AVX code generation is different from SSE one. 'vcvtsi2sdq' cannot be
selected by fast-isel due to complicated pattern and fast-isel
fallback to materialize it from constant pool.
- test/CodeGen/X86/widen_load-1.ll
AVX code generation is different from SSE one after fixing SSE/AVX
inter-mixing. Exec-domain fixing prefers 'vmovapd' instead of
'vmovaps'.
llvm-svn: 162919
- The root cause is that target constant materialization in X86 fast-isel
creates a PC-rel addressing which may overflow 32-bit range in non-Small code
model if .rodata section is allocated too far away from code segment in
MCJIT, which uses Large code model so far.
- Follow the similar logic to fix non-Small code model in fast-isel by skipping
non-Small code model.
llvm-svn: 162881
We need to reserve space for the mandatory traceback fields,
though leaving them as zero is appropriate for now.
Although the ABI calls for these fields to be filled in fully, no
compiler on Linux currently does this, and GDB does not read these
fields. GDB uses the first word of zeroes during exception handling to
find the end of the function and the size field, allowing it to compute
the beginning of the function. DWARF information is used for everything
else. We need the extra 8 bytes of pad so the size field is found in
the right place.
As a comparison, GCC fills in a few of the fields -- language, number
of saved registers -- but ignores the rest. IBM's proprietary OSes do
make use of the full traceback table facility.
Patch by Bill Schmidt.
llvm-svn: 162854
traceback table on PowerPC64. This helps gdb handle exceptions. The other
mandatory fields are ignored by gdb and harder to implement so just add
there a FIXME.
Patch by Bill Schmidt. PR13641.
llvm-svn: 162778
- Add a target-specific DAG optimization to recognize a pattern PTEST-able.
Such a pattern is a OR'd tree with X86ISD::OR as the root node. When
X86ISD::OR node has only its flag result being used as a boolean value and
all its leaves are extracted from the same vector, it could be folded into an
X86ISD::PTEST node.
llvm-svn: 162735
These extra flags are not required to properly order the atomic
load/store instructions. SelectionDAGBuilder chains atomics as if they
were volatile, and SelectionDAG::getAtomic() sets the isVolatile bit on
the memory operands of all atomic operations.
The volatile bit is enough to order atomic loads and stores during and
after SelectionDAG.
This means we set mayLoad on atomic_load, mayStore on atomic_store, and
mayLoad+mayStore on the remaining atomic read-modify-write operations.
llvm-svn: 162733
Instructions emitted to compute branch offsets now use immediate operands
instead of symbolic labels. This change was needed because there were problems
when R_MIPS_HI16/LO16 relocations were used to make shared objects.
llvm-svn: 162731
In SelectionDAGLegalize::ExpandLegalINT_TO_FP, expand INT_TO_FP nodes without
using any f64 operations if f64 is not a legal type.
Patch by Stefan Kristiansson.
llvm-svn: 162728
Allow load-immediates to be rematerialised in the register coalescer for
PPC. This makes test/CodeGen/PowerPC/big-endian-formal-args.ll fail,
because it relies on a register move getting emitted. The immediate load is
equivalent, so change this test case.
Patch by Tobias von Koch.
llvm-svn: 162727
The 32-bit ABI requires CR bit 6 to be set if the call has fp arguments and
unset if it doesn't. The solution up to now was to insert a MachineNode to
set/unset the CR bit, which produces a CR vreg. This vreg was then copied
into CR bit 6. When the register allocator saw a bunch of these in the same
function, it allocated the set/unset CR bit in some random CR register (1
extra instruction) and then emitted CR moves before every vararg function
call, rather than just setting and unsetting CR bit 6 directly before every
vararg function call. This patch instead inserts a PPCcrset/PPCcrunset
instruction which are then matched by a dedicated instruction pattern.
Patch by Tobias von Koch.
llvm-svn: 162725
The zeroextend IR instruction is lowered to an 'and' node with an immediate
mask operand, which in turn gets legalised to a sequence of ori's & ands.
This can be done more efficiently using the rldicl instruction.
Patch by Tobias von Koch.
llvm-svn: 162724
Previously, instructions without a primary patterns wouldn't get their
properties inferred. Now, we use all single-instruction patterns for
inference, including 'def : Pat<>' instances.
This causes a lot of instruction flags to change.
- Many instructions no longer have the UnmodeledSideEffects flag because
their flags are now inferred from a pattern.
- Instructions with intrinsics will get a mayStore flag if they already
have UnmodeledSideEffects and a mayLoad flag if they already have
mayStore. This is because intrinsics properties are linear.
- Instructions with atomic_load patterns get a mayStore flag because
atomic loads can't be reordered. The correct workaround is to create
pseudo-instructions instead of using normal loads. PR13693.
llvm-svn: 162614
the case of multiple edges from one block to another.
A simple example is a switch statement with multiple values to the same
destination. The definition of an edge is modified from a pair of blocks to
a pair of PredBlock and an index into the successors.
Also set the weight correctly when building SelectionDAG from LLVM IR,
especially when converting a Switch.
IntegersSubsetMapping is updated to calculate the weight for each cluster.
llvm-svn: 162572
IR that hasn't been through SimplifyCFG can look like this:
br i1 %b, label %r, label %r
Make sure we don't create duplicate Machine CFG edges in this case.
Fix the machine code verifier to accept conditional branches with a
single CFG edge.
llvm-svn: 162230
this allows for better code generation.
Added a new DAGCombine transformation to convert FMAX and FMIN to FMANC and
FMINC, which are commutative.
For example:
movaps %xmm0, %xmm1
movsd LC(%rip), %xmm0
minsd %xmm1, %xmm0
becomes:
minsd LC(%rip), %xmm0
llvm-svn: 162187
Add these transformations to the existing add/sub ones:
(and (select cc, -1, c), x) -> (select cc, x, (and, x, c))
(or (select cc, 0, c), x) -> (select cc, x, (or, x, c))
(xor (select cc, 0, c), x) -> (select cc, x, (xor, x, c))
The selects can then be transformed to a single predicated instruction
by peephole.
This transformation will make it possible to eliminate the ISD::CAND,
COR, and CXOR custom DAG nodes.
llvm-svn: 162176
arithmetic instructions. However, when small data types are used, a truncate
node appears between the SETCC node and the arithmetic operation. This patch
adds support for this pattern.
Before:
xorl %esi, %edi
testb %dil, %dil
setne %al
ret
After:
xorb %dil, %sil
setne %al
ret
rdar://12081007
llvm-svn: 162160
PEI can't handle the pseudo-instructions. This can be removed when the
pseudo-instructions are replaced by normal predicated instructions.
Fixes PR13628.
llvm-svn: 162130
I really need to find a way to automate this, but I can't come up with a regex
that has no false positives while handling tricky cases like custom check
prefixes.
llvm-svn: 162097
It is not my plan to duplicate the entire ARM instruction set with
predicated versions. We need a way of representing predicated
instructions in SSA form without requiring a separate opcode.
Then the pseudo-instructions can go away.
llvm-svn: 162061
Without fastcc support, the caller just falls through to CallingConv::C
for fastcc, but callee still uses fastcc, this inconsistency of calling
convention is a problem, and fastcc support can fix it.
llvm-svn: 162013
The ARM select instructions are just predicated moves. If the select is
the only use of an operand, the instruction defining the operand can be
predicated instead, saving one instruction and decreasing register
pressure.
This implementation can turn AND/ORR/EOR instructions into their
corresponding ANDCC/ORRCC/EORCC variants. Ideally, we should be able to
predicate any instruction, but we don't yet support predicated
instructions in SSA form.
llvm-svn: 161994
around. That's not how we do things. Besides, the commit message tells us that
it is covered by the GCC test suite.
------------------------------------------------------------------------
r127497 | zwarich | 2011-03-11 13:51:56 -0800 (Fri, 11 Mar 2011) | 3 lines
Fix the GCC test suite issue exposed by r127477, which was caused by stack
protector insertion not working correctly with unreachable code. Since that
revision was rolled out, this test doesn't actual fail before this fix.
------------------------------------------------------------------------
llvm-svn: 161985
reversed. This leads to wrong codegen for float-to-half conversion
intrinsics which are used to support storage-only fp16 type.
NEON variants of same instructions are fine.
llvm-svn: 161907
- FP_EXTEND only support extending from vectors with matching elements.
This results in the scalarization of extending to v2f64 from v2f32,
which will be legalized to v4f32 not matching with v2f64.
- add X86-specific VFPEXT supproting extending from v4f32 to v2f64.
- add BUILD_VECTOR lowering helper to recover back the original
extending from v4f32 to v2f64.
- test case is enhanced to include different vector width.
llvm-svn: 161894
and allow some optimizations to turn conditional branches into unconditional.
This commit adds a simple control-flow optimization which merges two consecutive
basic blocks which are connected by a single edge. This allows the codegen to
operate on larger basic blocks.
rdar://11973998
llvm-svn: 161852
It is still possible to if-convert if the tail block has extra
predecessors, but the tail phis must be rewritten instead of being
removed.
llvm-svn: 161781
- FCMOV only supports a subset of X86 conditions. Skip boolean
simplification if X86 condition is not valid for FCMOV.
- add a minimal test case for PR13577.
llvm-svn: 161732
FeatureFastUAMem for Nehalem, Westmere and Sandy Bridge.
FeatureFastUAMem is already on if we pass in nehalem or westmere as a command
argument.
rdar: 7252306
llvm-svn: 161717
- if a boolean test (X86ISD::CMP or X86ISD:SUB) checks a boolean value
generated from X86ISD::SETCC, try to simplify the boolean value
generation and checking by reusing the original EFLAGS with proper
condition code
- add hooks to X86 specific SETCC/BRCOND/CMOV, the major 3 places
consuming EFLAGS
part of patches fixing PR12312
llvm-svn: 161687
When replacing Old with New, it can happen that New is already a
successor. Add the old and new edge weights instead of creating a
duplicate edge.
llvm-svn: 161653
This makes it possible to speed up def_iterator by stopping at the first
use. This makes def_empty() and getUniqueVRegDef() much faster when
there are many uses.
In a +Asserts build, LiveVariables is 100x faster in one case because
getVRegDef() has an assertion that would scan to the end of a
def_iterator chain.
Spill weight calculation is significantly faster (300x in one case)
because isTriviallyReMaterializable() calls MRI->isConstantPhysReg(%RIP)
which calls def_empty(%RIP).
llvm-svn: 161634
Use a more conventional doubly linked list where the Prev pointers form
a cycle. This means it is no longer necessary to adjust the Prev
pointers when reallocating the VRegInfo array.
The test changes are required because the register allocation hint is
using the use-list order to break ties.
llvm-svn: 161633
This patch corrects the definition of umlal/smlal instructions and adds support
for matching them to the ARM dag combiner.
Bug 12213
Patch by Yin Ma!
llvm-svn: 161581
We perform the following:
1> Use SUB instead of CMP for i8,i16,i32 and i64 in ISel lowering.
2> Modify MachineCSE to correctly handle implicit defs.
3> Convert SUB back to CMP if possible at peephole.
Removed pattern matching of (a>b) ? (a-b):0 and like, since they are handled
by peephole now.
rdar://11873276
llvm-svn: 161462
Previously, MBP essentially aligned every branch target it could. This
bloats code quite a bit, especially non-looping code which has no real
reason to prefer aligned branch targets so heavily.
As Andy said in review, it's still a bit odd to do this without a real
cost model, but this at least has much more plausible heuristics.
Fixes PR13265.
llvm-svn: 161409
If the result of a common subexpression is used at all uses of the candidate
expression, CSE should not increase the live range of the common subexpression.
rdar://11393714 and rdar://11819721
llvm-svn: 161396
The MFTB instruction itself is being phased out, and its functionality
is provided by MFSPR. According to the ISA docs, using MFSPR works on all known
chips except for the 601 (which did not have a timebase register anyway)
and the POWER3.
Thanks to Adhemerval Zanella for pointing this out!
llvm-svn: 161346
On PPC64, this can be done with a simple TableGen pattern.
To enable this, I've added the (otherwise missing) readcyclecounter
SDNode definition to TargetSelectionDAG.td.
llvm-svn: 161302
This patch is mostly just refactoring a bunch of copy-and-pasted code, but
it also adds a check that the call instructions are readnone or readonly.
That check was already present for sin, cos, sqrt, log2, and exp2 calls, but
it was missing for the rest of the builtins being handled in this code.
llvm-svn: 161282
I noticed that SelectionDAGBuilder::visitCall was missing a check for memcmp
in TargetLibraryInfo, so that it would use custom code for memcmp calls even
with -fno-builtin. I also had to add a new -disable-simplify-libcalls option
to llc so that I could write a test for this.
llvm-svn: 161262
Fast isel doesn't currently have support for translating builtin function
calls to target instructions. For embedded environments where the library
functions are not available, this is a matter of correctness and not
just optimization. Most of this patch is just arranging to make the
TargetLibraryInfo available in fast isel. <rdar://problem/12008746>
llvm-svn: 161232
Add more comments and use early returns to reduce nesting in isLoadFoldable.
Also disable folding for V_SET0 to avoid introducing a const pool entry and
a const pool load.
rdar://10554090 and rdar://11873276
llvm-svn: 161207
Machine CSE and other optimizations can remove instructions so folding
is possible at peephole while not possible at ISel.
This patch is a rework of r160919 and was tested on clang self-host on my local
machine.
rdar://10554090 and rdar://11873276
llvm-svn: 161152
MipsSEFrameLowering.
Implement MipsSEFrameLowering::hasReservedCallFrame. Call frames will not be
reserved if there is a call with a large call frame or there are variable sized
objects on the stack.
llvm-svn: 161090
The frame object which points to the dynamically allocated area will not be
needed after changes are made to cease reserving call frames.
llvm-svn: 161076
arguments to the stack in MipsISelLowering::LowerCall, use stack pointer and
integer offset operands rather than frame object operands.
llvm-svn: 161068
single-precision load and store.
Also avoid selecting LUXC1 and SUXC1 instructions during isel. It is incorrect
to map unaligned floating point load/store nodes to these instructions.
llvm-svn: 161063
One motivating example is to sink an instruction from a basic block which has
two successors: one outside the loop, the other inside the loop. We should try
to sink the instruction outside the loop.
rdar://11980766
llvm-svn: 161062
We are extending live ranges, so kill flags are not accurate. They
aren't needed until they are recomputed after RA anyway.
<rdar://problem/11950722>
llvm-svn: 161023
We branch to the successor with higher edge weight first.
Convert from
je LBB4_8 --> to outer loop
jmp LBB4_14 --> to inner loop
to
jne LBB4_14
jmp LBB4_8
PR12750
rdar: 11393714
llvm-svn: 161018
Machine CSE and other optimizations can remove instructions so folding
is possible at peephole while not possible at ISel.
rdar://10554090 and rdar://11873276
llvm-svn: 160919
It is possible that an instruction can use and update EFLAGS.
When checking the safety, we should check the usage of EFLAGS first before
declaring it is safe to optimize due to the update.
llvm-svn: 160912
These idempotent sub-register indices don't do anything --- They simply
map XMM registers to themselves. They no longer affect register classes
either since the SubRegClasses field has been removed from Target.td.
This patch replaces XMM->XMM EXTRACT_SUBREG and INSERT_SUBREG patterns
with COPY_TO_REGCLASS patterns which simply become COPY instructions.
The number of IMPLICIT_DEF instructions before register allocation is
reduced, and that is the cause of the test case changes.
llvm-svn: 160816
It is redundant; RegisterCoalescer will do the remat if it can't eliminate
the copy. Collected instruction counts before and after this. A few extra
instructions are generated due to spilling but it is normal to see these kinds
of changes with almost any small codegen change, according to Jakob.
This also fixed rdar://11830760 where xor is expected instead of movi0.
llvm-svn: 160749
The long branch pass (fixed in r160601) no longer uses the global base register
to compute addresses of branch destinations, so it is not necessary to reserve
a slot on the stack.
llvm-svn: 160703
struct s {
double x1;
float x2;
};
__attribute__((regparm(3))) struct s f(int a, int b, int c);
void g(void) {
f(41, 42, 43);
}
We need to be able to represent passing the address of s to f (sret) in a
register (inreg). Turns out that all that is needed is to not mark them as
mutually incompatible.
llvm-svn: 160695
are targeting an ELF platform. Only fold gs-relative (and fs-relative) loads
if it is actually sensible to do so for the target platform.
This fixes PR13438.
llvm-svn: 160687
This pass no longer requires that the global pointer value be saved to the
stack or register since it uses bal instruction to compute branch distance.
llvm-svn: 160601
LiveRangeEdit::foldAsLoad() can eliminate a register by folding a load
into its only use. Only do that when the load is safe to move, and it
won't extend any live ranges.
This fixes PR13414.
llvm-svn: 160575
PHIElimination splits critical edges when it predicts it can resolve
interference and eliminate copies. It doesn't split the edge if the
interference wouldn't be resolved anyway because the phi-use register is
live in the critical edge anyway.
Teach PHIElimination to split loop exiting edges with interference, even
if it wouldn't resolve the interference. This removes the necessary
copies from the loop, which is still an improvement from injecting the
copies into the loop.
The test case demonstrates the improvement. Before:
LBB0_1:
cmpb $0, (%rdx)
leaq 1(%rdx), %rdx
movl %esi, %eax
je LBB0_1
After:
LBB0_1:
cmpb $0, (%rdx)
leaq 1(%rdx), %rdx
je LBB0_1
movl %esi, %eax
llvm-svn: 160571
Updated OptimizeCompare in peephole to remove redundant cmp against zero.
We only remove Compare if CF and OF are not used.
rdar://11855129
llvm-svn: 160454
when run on an Intel Atom processor. The failures have arisen due
to changes elsewhere in the trunk over the past 8 weeks or so.
These failures were not detected by the Atom buildbot because the
CPU on the Atom buildbot was not being detected as an Atom CPU.
The fix for this problem is in Host.cpp and X86Subtarget.cpp, but
shall remain commented out until the current set of Atom test failures
are fixed.
Patch by Andy Zhang and Tyler Nowicki!
llvm-svn: 160451
LiveIntervals due to the two-addr pass generating bogus MI code.
The crux of the issue was a loop nesting problem. The intent of the code
which attempts to transform instructions before converting them to
two-addr form is to defer and reprocess any transformed instructions as
the second processing is likely to have more opportunities to coalesce
copies, etc. Unfortunately, there was one section of processing that was
not deferred -- the INSERT_SUBREG rewriting. Due to quirks of how this
rewriting proceeded, not only did it occur early, it removed the bits of
information needed for the deferred processing to correctly generate the
necessary two address form (specifically inserting a copy), but didn't
trigger any immediate assertions and produced what appeared to be
already valid two-address from code. Thus, the assertion only fired much
later in the pipeline.
The fix is to hoist the transformation logic up layer to where it can
more firmly defer all further processing, and to teach the normal
processing to handle an edge case previously handled as part of the
transformation logic. This edge case (already matched tied register
operands) needs to *not* defer any steps.
As has been brought up repeatedly in the process: wow does this code
need refactoring. I *may* squeeze in some time to at least bring sanity
to this loop... but wow... =]
Thanks to Jakob for helpful hints on the way here, and the review.
llvm-svn: 160443
Print the high order register of a double word register operand.
In 32 bit mode, a 64 bit double word integer will be represented
by 2 32 bit registers. This modifier causes the high order register
to be used in the asm expression. It is useful if you are using
doubles in assembler and continue to control register to variable
relationships.
This patch also fixes a related bug in a previous patch:
case 'D': // Second part of a double word register operand
case 'L': // Low order register of a double word register operand
case 'M': // High order register of a double word register operand
I got 'D' and 'M' confused. The second part of a double word operand
will only match 'M' for one of the endianesses. I had 'L' and 'D'
be the opposite twins when 'L' and 'M' are.
llvm-svn: 160429
intrinsics. The second instruction(s) to be handled are the vector versions
of count set bits (ctpop).
The changes here are to clang so that it generates a target independent
vector ctpop when it sees an ARM dependent vector bits set count. The changes
in llvm are to match the target independent vector ctpop and in
VMCore/AutoUpgrade.cpp to update any existing bc files containing ARM
dependent vector pop counts with target-independent ctpops. There are also
changes to an existing test case in llvm for ARM vector count instructions and
to a test for the bitcode upgrade.
<rdar://problem/11892519>
There is deliberately no test for the change to clang, as so far as I know, no
consensus has been reached regarding how to test neon instructions in clang;
q.v. <rdar://problem/8762292>
llvm-svn: 160410
large immediates. Add dag combine logic to recover in case the large
immediates doesn't fit in cmp immediate operand field.
int foo(unsigned long l) {
return (l>> 47) == 1;
}
we produce
%shr.mask = and i64 %l, -140737488355328
%cmp = icmp eq i64 %shr.mask, 140737488355328
%conv = zext i1 %cmp to i32
ret i32 %conv
which codegens to
movq $0xffff800000000000,%rax
andq %rdi,%rax
movq $0x0000800000000000,%rcx
cmpq %rcx,%rax
sete %al
movzbl %al,%eax
ret
TargetLowering::SimplifySetCC would transform
(X & -256) == 256 -> (X >> 8) == 1
if the immediate fails the isLegalICmpImmediate() test. For x86,
that's immediates which are not a signed 32-bit immediate.
Based on a patch by Eli Friedman.
PR10328
rdar://9758774
llvm-svn: 160346
uint32_t hi(uint64_t res)
{
uint_32t hi = res >> 32;
return !hi;
}
llvm IR looks like this:
define i32 @hi(i64 %res) nounwind uwtable ssp {
entry:
%lnot = icmp ult i64 %res, 4294967296
%lnot.ext = zext i1 %lnot to i32
ret i32 %lnot.ext
}
The optimizer has optimize away the right shift and truncate but the resulting
constant is too large to fit in the 32-bit immediate field. The resulting x86
code is worse as a result:
movabsq $4294967296, %rax ## imm = 0x100000000
cmpq %rax, %rdi
sbbl %eax, %eax
andl $1, %eax
This patch teaches the x86 lowering code to handle ult against a large immediate
with trailing zeros. It will issue a right shift and a truncate followed by
a comparison against a shifted immediate.
shrq $32, %rdi
testl %edi, %edi
sete %al
movzbl %al, %eax
It also handles a ugt comparison against a large immediate with trailing bits
set. i.e. X > 0x0ffffffff -> (X >> 32) >= 1
rdar://11866926
llvm-svn: 160312
In the added testcase the constant 55 was behind an AssertZext of type i1, and ComputeDemandedBits
reported that some of the bits were both known to be one and known to be zero.
Together with Michael Kuperstein <michael.m.kuperstein@intel.com>
llvm-svn: 160305
1. FileCheck-ize epilogue.ll and allow another asm instruction to restore %rsp.
2. Remove check in widen_arith-3.ll that was hitting instruction in epilogue instead of
vector add.
llvm-svn: 160274
undef virtual register. The problem is that ProcessImplicitDefs removes the
definition of the register and marks all uses as undef. If we lose the undef
marker then we get a register which has no def, is not marked as undef. The
live interval analysis does not collect information for these virtual
registers and we crash in later passes.
Together with Michael Kuperstein <michael.m.kuperstein@intel.com>
llvm-svn: 160260
It is intended to fix PR11468.
Old prologue and epilogue looked like this:
push %rbp
mov %rsp, %rbp
and $alignment, %rsp
push %r14
push %r15
...
pop %r15
pop %r14
mov %rbp, %rsp
pop %rbp
The problem was to reference the locations of callee-saved registers in exception handling:
locations of callee-saved had to be re-calculated regarding the stack alignment operation. It would
take some effort to implement this in LLVM, as currently MachineLocation can only have the form
"Register + Offset". Funciton prologue and epilogue are now changed to:
push %rbp
mov %rsp, %rbp
push %14
push %15
and $alignment, %rsp
...
lea -$size_of_saved_registers(%rbp), %rsp
pop %r15
pop %r14
pop %rbp
Reviewed by Chad Rosier.
llvm-svn: 160248
Allow the folding of vbroadcastRR to vbroadcastRM, where the memory operand is a spill slot.
PR12782.
Together with Michael Kuperstein <michael.m.kuperstein@intel.com>
llvm-svn: 160230
intrinsics with target-indepdent intrinsics. The first instruction(s) to be
handled are the vector versions of count leading zeros (ctlz).
The changes here are to clang so that it generates a target independent
vector ctlz when it sees an ARM dependent vector ctlz. The changes in llvm
are to match the target independent vector ctlz and in VMCore/AutoUpgrade.cpp
to update any existing bc files containing ARM dependent vector ctlzs with
target-independent ctlzs. There are also changes to an existing test case in
llvm for ARM vector count instructions and a new test for the bitcode upgrade.
<rdar://problem/11831778>
There is deliberately no test for the change to clang, as so far as I know, no
consensus has been reached regarding how to test neon instructions in clang;
q.v. <rdar://problem/8762292>
llvm-svn: 160200
the input vector, it can be bigger (this is helpful for powerpc where <2 x i16>
is a legal vector type but i16 isn't a legal type, IIRC). However this wasn't
being taken into account by ExpandRes_EXTRACT_VECTOR_ELT, causing PR13220.
Lightly tweaked version of a patch by Michael Liao.
llvm-svn: 160116
It is safe if CPSR is killed or re-defined.
When we are done with the basic block, check whether CPSR is live-out.
Do not optimize away cmp if CPSR is live-out.
llvm-svn: 160090
Low order register of a double word register operand. Operands
are defined by the name of the variable they are marked with in
the inline assembler code. This is a way to specify that the
operand just refers to the low order register for that variable.
It is the opposite of modifier 'D' which specifies the high order
register.
Example:
main()
{
long long ll_input = 0x1111222233334444LL;
long long ll_val = 3;
int i_result = 0;
__asm__ __volatile__(
"or %0, %L1, %2"
: "=r" (i_result)
: "r" (ll_input), "r" (ll_val));
}
Which results in:
lui $2, %hi(_gp_disp)
addiu $2, $2, %lo(_gp_disp)
addiu $sp, $sp, -8
addu $2, $2, $25
sw $2, 0($sp)
lui $2, 13107
ori $3, $2, 17476 <-- Low 32 bits of ll_input
lui $2, 4369
ori $4, $2, 8738 <-- High 32 bits of ll_input
addiu $5, $zero, 3 <-- Low 32 bits of ll_val
addiu $2, $zero, 0 <-- High 32 bits of ll_val
#APP
or $3, $4, $5 <-- or i_result, high 32 ll_input, low 32 of ll_val
#NO_APP
addiu $sp, $sp, 8
jr $ra
If not direction is done for the long long for 32 bit variables results
in using the low 32 bits as ll_val shows.
There is an existing bug if 'L' or 'D' is used for the destination register
for 32 bit long longs in that the target value will be updated incorrectly
for the non-specified part unless explicitly set within the inline asm code.
llvm-svn: 160028
X86. Basically, this is a reapplication of r158087 with a few fixes.
Specifically, (1) the stack pointer is restored from the base pointer before
popping callee-saved registers and (2) in obscure cases (see comments in patch)
we must cache the value of the original stack adjustment in the prologue and
apply it in the epilogue.
rdar://11496434
llvm-svn: 160002
multiple scalars and insert them into a vector. Next, we shuffle the elements
into the correct places, as before.
Also fix a small dagcombine bug in SimplifyBinOpWithSameOpcodeHands, when the
migration of bitcasts happened too late in the SelectionDAG process.
llvm-svn: 159991
getCondFromSETOpc, getCondFromCMovOpc, getSETFromCond, getCMovFromCond
No functional change intended.
If we want to update the condition code of CMOV|SET|Jcc, we first analyze the
opcode to get the condition code, then update the condition code, finally
synthesize the new opcode form the new condition code.
llvm-svn: 159955
It is safe if EFLAGS is killed or re-defined.
When we are done with the basic block, check whether EFLAGS is live-out.
Do not optimize away cmp if EFLAGS is live-out.
llvm-svn: 159888
For each Cmp, we check whether there is an earlier Sub which make Cmp
redundant. We handle the case where SUB operates on the same source operands as
Cmp, including the case where the two source operands are swapped.
llvm-svn: 159838
Print the second half of a double word operand.
The include list was cleaned up a bit as well.
Also the test case was modified to test for both
big and little patterns.
llvm-svn: 159787
The CopyToReg nodes that set up the argument registers before a call
must be glued to the call instruction. Otherwise, the scheduler may emit
the physreg copies long before the call, causing long live ranges for
the fixed registers.
Besides disabling good register allocation, that can also expose
problems when EmitInstrWithCustomInserter() splits a basic block during
the live range of a physreg.
llvm-svn: 159721
Implement the TII hooks needed by EarlyIfConversion to create cmov
instructions and estimate their latency.
Early if-conversion is still not enabled by default.
llvm-svn: 159695
inlineasm-cnstrnt-bad-r-1.ll is NOT supposed to fail, so it was removed. This resulted in the removal of a negative test (inlineasm-cnstrnt-bad-r-1.ll)
llvm-svn: 159625
inlineasm-cnstrnt-bad-r-1.ll is NOT supposed to fail, so it was removed. This resulted in the removal of a negative test (inlineasm-cnstrnt-bad-r-1.ll)
llvm-svn: 159610
This is still a work in progress but I believe it is currently good enough
to fix PR13122 "Need unit test driver for codegen IR passes". For example,
you can run llc with -stop-after=loop-reduce to have it dump out the IR after
running LSR. Serializing machine-level IR is not yet supported but we have
some patches in progress for that.
The plan is to serialize the IR to a YAML file, containing separate sections
for the LLVM IR, machine-level IR, and whatever other info is needed. Chad
suggested that we stash the stop-after pass in the YAML file and use that
instead of the start-after option to figure out where to restart the
compilation. I think that's a great idea, but since it's not implemented yet
I put the -start-after option into this patch for testing purposes.
llvm-svn: 159570
another mechanical change accomplished though the power of terrible Perl
scripts.
I have manually switched some "s to 's to make escaping simpler.
While I started this to fix tests that aren't run in all configurations,
the massive number of tests is due to a really frustrating fragility of
our testing infrastructure: things like 'grep -v', 'not grep', and
'expected failures' can mask broken tests all too easily.
Essentially, I'm deeply disturbed that I can change the testsuite so
radically without causing any change in results for most platforms. =/
llvm-svn: 159547
versions of Bash. In addition, I can back out the change to the lit
built-in shell test runner to support this.
This should fix the majority of fallout on Darwin, but I suspect there
will be a few straggling issues.
llvm-svn: 159544
through my perl nets.
With this, the test suite passes even if I force it to run with the
built-in shell test logic, except for a test which REQUIREs shell.
llvm-svn: 159529
This was done through the aid of a terrible Perl creation. I will not
paste any of the horrors here. Suffice to say, it require multiple
staged rounds of replacements, state carried between, and a few
nested-construct-parsing hacks that I'm not proud of. It happens, by
luck, to be able to deal with all the TCL-quoting patterns in evidence
in the LLVM test suite.
If anyone is maintaining large out-of-tree test trees, feel free to poke
me and I'll send you the steps I used to convert things, as well as
answer any painful questions etc. IRC works best for this type of thing
I find.
Once converted, switch the LLVM lit config to use ShTests the same as
Clang. In addition to being able to delete large amounts of Python code
from 'lit', this will also simplify the entire test suite and some of
lit's architecture.
Finally, the test suite runs 33% faster on Linux now. ;]
For my 16-hardware-thread (2x 4-core xeon e5520): 36s -> 24s
llvm-svn: 159525
use FileCheck.
Aside from removing a dependence on TCL-style quoting, this also makes
the tests ... significantly more robust. =] It would be really, *really*
great of the maintainer(s) of the CellSPU backend went through and
systematically rewrite these tests to use FileCheck. There are a lot
more that have nearly this bad of abuses.
Another step along the path to a TclTest-free testsuite.
llvm-svn: 159523
implicit_def, the other instruction can be anything, including instructions
that define multiple values. Be careful about that and don't assume what operand
0 is.
Fixes pr13249.
llvm-svn: 159509
Before this patch in pic 32 bit code we would add the global base register
and not load from that address. This is a really old bug, but before the
introduction of the tls attributes we would never select initial exec for
pic code.
llvm-svn: 159409
Corrected type for index of llvm.x86.avx2.gather.d.pd.256
from 256-bit to 128-bit.
Corrected types for src|dst|mask of llvm.x86.avx2.gather.q.ps.256
from 256-bit to 128-bit.
Support the following intrinsics:
llvm.x86.avx2.gather.d.q, llvm.x86.avx2.gather.q.q
llvm.x86.avx2.gather.d.q.256, llvm.x86.avx2.gather.q.q.256
llvm.x86.avx2.gather.d.d, llvm.x86.avx2.gather.q.d
llvm.x86.avx2.gather.d.d.256, llvm.x86.avx2.gather.q.d.256
llvm-svn: 159402
up to r158925 were handled as processor specific. Making them
generic and putting tests for these modifiers in the CodeGen/Generic
directory caused a number of targets to fail.
This commit addresses that problem by having the targets call
the generic routine for generic modifiers that they don't currently
have explicit code for.
For now only generic print operands 'c' and 'n' are supported.vi
Affected files:
test/CodeGen/Generic/asm-large-immediate.ll
lib/Target/PowerPC/PPCAsmPrinter.cpp
lib/Target/NVPTX/NVPTXAsmPrinter.cpp
lib/Target/ARM/ARMAsmPrinter.cpp
lib/Target/XCore/XCoreAsmPrinter.cpp
lib/Target/X86/X86AsmPrinter.cpp
lib/Target/Hexagon/HexagonAsmPrinter.cpp
lib/Target/CellSPU/SPUAsmPrinter.cpp
lib/Target/Sparc/SparcAsmPrinter.cpp
lib/Target/MBlaze/MBlazeAsmPrinter.cpp
lib/Target/Mips/MipsAsmPrinter.cpp
MSP430 isn't represented because it did not even run with
the long existing 'c' modifier and it was not apparent what
needs to be done to get it inline asm ready.
Contributer: Jack Carter
llvm-svn: 159203
The primary advantage is that loop optimizations will be applied in a
stable order. This helps debugging and unit test creation. It is also
a better overall implementation without pathologically bad performance
on deep functions.
On large functions (llvm-stress --size=200000 | opt -loops)
Before: 0.1263s
After: 0.0225s
On deep functions (after tweaking llvm-stress, thanks Nadav):
Before: 0.2281s
After: 0.0227s
See r158790 for more comments.
The loop tree is now consistently generated in forward order, but loop
passes are applied in reverse order over the program. If we have a
loop optimization that prefers forward order, that can easily be
achieved by adding a different type of LoopPassManager.
llvm-svn: 159183
More condition codes are included when deciding whether to remove cmp after
a sub instruction. Specifically, we extend from GE|LT|GT|LE to
GE|LT|GT|LE|HS|LS|HI|LO|EQ|NE. If we have "sub a, b; cmp b, a; movhs", we
should be able to replace with "sub a, b; movls".
rdar: 11725965
llvm-svn: 159166
Verify that all paths from the entry block to a virtual register read
pass through a def. Enable this check even when MRI->isSSA() is false.
Verify that the live range of a virtual register is live out of all
predecessor blocks, even for PHI-values.
This requires that PHIElimination sometimes inserts IMPLICIT_DEF
instruction in predecessor blocks.
llvm-svn: 159150
Implicitly defined virtual registers can simply have the <undef> bit set
on all uses, and copies can be turned into implicit defs recursively.
Physical registers are a bit trickier. We handle the common case where a
physreg def is used by a nearby instruction in the same basic block. For
more complicated cases, just leave the IMPLICIT_DEF instruction in.
llvm-svn: 159149
The function live-out registers must be live at all function returns,
and %RCX is only used by eh.return. When a function also has a normal
return, only %RAX holds a return value.
This fixes PR13188.
llvm-svn: 159116
This allows the user/front-end to specify a model that is better
than what LLVM would choose by default. For example, a variable
might be declared as
@x = thread_local(initialexec) global i32 42
if it will not be used in a shared library that is dlopen'ed.
If the specified model isn't supported by the target, or if LLVM can
make a better choice, a different model may be used.
llvm-svn: 159077
There are patterns to handle immediates when they fit in the immediate field.
e.g. %sub = add i32 %x, -123
=> sub r0, r0, #123
Add patterns to catch immediates that do not fit but should be materialized
with a single movw instruction rather than movw + movt pair.
e.g. %sub = add i32 %x, -65535
=> movw r1, #65535
sub r0, r0, r1
rdar://11726136
llvm-svn: 159057
The code in X86TargetLowering::LowerEH_RETURN() assumes that a frame
pointer exists, but the frame pointer was forced by the presence of
llvm.eh.unwind.init which isn't guaranteed.
If llvm.eh.unwind.init is actually required in functions calling
eh.return (is it?), we should diagnose that instead of emitting bad
machine code.
This should fix the dragonegg-x86_64-linux-gcc-4.6-test bot.
llvm-svn: 158961
Minor drive by fix to cleanup latency computation. Calling
getOperandLatency with a deliberately incorrect operand index does not
give you the latency you want.
llvm-svn: 158959
boolean flag to an enum: { Fast, Standard, Strict } (default = Standard).
This option controls the creation by optimizations of fused FP ops that store
intermediate results in higher precision than IEEE allows (E.g. FMAs). The
behavior of this option is intended to match the behaviour specified by a
soon-to-be-introduced frontend flag: '-ffuse-fp-ops'.
Fast mode - allows formation of fused FP ops whenever they're profitable.
Standard mode - allow fusion only for 'blessed' FP ops. At present the only
blessed op is the fmuladd intrinsic. In the future more blessed ops may be
added.
Strict mode - allow fusion only if/when it can be proven that the excess
precision won't effect the result.
Note: This option only controls formation of fused ops by the optimizers. Fused
operations that are explicitly requested (e.g. FMA via the llvm.fma.* intrinsic)
will always be honored, regardless of the value of this option.
Internally TargetOptions::AllowExcessFPPrecision has been replaced by
TargetOptions::AllowFPOpFusion.
llvm-svn: 158956
to be generic across architectures. It has the
following description in the gnu sources:
Negate the immediate constant
Several Architectures such as x86 have local implementations
of operand modifier 'n' which go beyond the above description
slightly. This won't affect them.
Affected files:
lib/CodeGen/AsmPrinter/AsmPrinterInlineAsm.cpp
Added 'n' to the switch cases.
test/CodeGen/Generic/asm-large-immediate.ll
Generic compiled test (x86 for me)
test/CodeGen/Mips/asm-large-immediate.ll
Mips compiled version of the generic one
Contributer: Jack Carter
llvm-svn: 158939
to be generic across architectures. It has the
following description in the gnu sources:
Substitute immediate value without immediate syntax
Several Architectures such as x86 have local implementations
of operand modifier 'c' which go beyond the above description
slightly. To make use of the generic modifiers without overriding
local implementation one can make a call to the base class method
for AsmPrinter::PrintAsmOperand() in the locally derived method's
"default" case in the switch statement. That way if it is already
defined locally the generic version will never get called.
This change is needed when test/CodeGen/generic/asm-large-immediate.ll
failed on a native Mips board. The test was assuming a generic
implementation was in place.
Affected files:
lib/Target/Mips/MipsAsmPrinter.cpp:
Changed the default case to call the base method.
lib/CodeGen/AsmPrinter/AsmPrinterInlineAsm.cpp
Added 'c' to the switch cases.
test/CodeGen/Mips/asm-large-immediate.ll
Mips compiled version of the generic one
Contributer: Jack Carter
llvm-svn: 158925
_umodsi3 libcalls if they have the same arguments. This optimization
was apparently broken if one of the node was replaced in place.
rdar://11714607
llvm-svn: 158900
Regunit live ranges are computed on demand, so when mi-sched calls
handleMove, some regunits may not have live ranges yet.
That makes updating them easier: Just skip the non-existing ranges. They
will be computed correctly from the rescheduled machine code when they
are needed.
llvm-svn: 158831
This patch adds DAG combines to form FMAs from pairs of FADD + FMUL or
FSUB + FMUL. The combines are performed when:
(a) Either
AllowExcessFPPrecision option (-enable-excess-fp-precision for llc)
OR
UnsafeFPMath option (-enable-unsafe-fp-math)
are set, and
(b) TargetLoweringInfo::isFMAFasterThanMulAndAdd(VT) is true for the type of
the FADD/FSUB, and
(c) The FMUL only has one user (the FADD/FSUB).
If your target has fast FMA instructions you can make use of these combines by
overriding TargetLoweringInfo::isFMAFasterThanMulAndAdd(VT) to return true for
types supported by your FMA instruction, and adding patterns to match ISD::FMA
to your FMA instructions.
llvm-svn: 158757
The PPC::EXTSW instruction preserves the low 32 bits of its input, just
like some of the x86 instructions. Use it to reduce register pressure
when the low 32 bits have multiple uses.
This requires a small change to PeepholeOptimizer since EXTSW takes a
64-bit input register.
This is related to PR5997.
llvm-svn: 158743
TargetLoweringObjectFileELF. Use this to support it on X86. Unlike ARM,
on X86 it is not easy to find out if .init_array should be used or not, so
the decision is made via TargetOptions and defaults to off.
Add a command line option to llc that enables it.
llvm-svn: 158692
when a compile time constant is known. This occurs when implicitly zero
extending function arguments from 16 bits to 32 bits. The 8 bit case doesn't
need to be handled, as the 8 bit constants are encoded directly, thereby
not needing a separate load instruction to form the constant into a register.
<rdar://problem/11481151>
llvm-svn: 158659
temporarily reverted.
This test is annoyingly overspecified, but I don't know of another way
to thoroughly test the saving and restoring of the registers. While this
will have to be adjusted even with the issue fixed in order to re-apply
r158087, those adjustments should very clearly indicate that it is still
correct (%esp getting restored prior to pops), whereas without it, this
case can easily slip under the radar.
Still, any suggestions for improvements are very welcome.
All credit to Matt Beaumont-Gay for reducing this out of an insane
Address Sanitizer crash to a reasonably small seg-faulting C program
when built with -mstackrealign. I just reduced it to IR, which was much
simpler. =]
llvm-svn: 158656
This patch causes problems when both dynamic stack realignment and
dynamic allocas combine in the same function. With this patch, we no
longer build the epilog correctly, and silently restore registers from
the wrong position in the stack.
Thanks to Matt for tracking this down, and getting at least an initial
test case to Chad. I'm going to try to check a variation of that test
case in so we can easily track the fixes required.
llvm-svn: 158654
This cleans up the method used to find trip counts in order to form CTR loops on PPC.
This refactoring allows the pass to find loops which have a constant trip count but also
happen to end with a comparison to zero. This also adds explicit FIXMEs to mark two different
classes of loops that are currently ignored.
In addition, we now search through all potential induction operations instead of just the first.
Also, we check the predicate code on the conditional branch and abort the transformation if the
code is not EQ or NE, and we then make sure that the branch to be transformed matches the
condition register defined by the comparison (multiple possible comparisons will be considered).
llvm-svn: 158607
This patch will optimize abs(x-y)
FROM
sub, movs, rsbmi
TO
subs, rsbmi
For abs, we will use cmp instead of movs. This is necessary because we already
have an existing peephole pass which optimizes away cmp following sub.
rdar: 11633193
llvm-svn: 158551
For store->load dependencies that may alias, we should always use
TrueMemOrderLatency, which may eventually become a subtarget hook. In
effect, we should guarantee at least TrueMemOrderLatency on at least
one DAG path from a store to a may-alias load.
This should fix the standard mode as well as -enable-aa-sched-mi".
llvm-svn: 158380
We turned off the CMN instruction because it had semantics which we weren't
getting correct. If we are comparing with an immediate, then it's okay to use
the CMN instruction.
<rdar://problem/7569620>
llvm-svn: 158302
Over the entire test-suite, this has an insignificantly negative average
performance impact, but reduces some of the worst slowdowns from the
anti-dep. change (r158294).
Largest speedups:
SingleSource/Benchmarks/Stanford/Quicksort - 28%
SingleSource/Benchmarks/Stanford/Towers - 24%
SingleSource/Benchmarks/Shootout-C++/matrix - 23%
MultiSource/Benchmarks/SciMark2-C/scimark2 - 19%
MultiSource/Benchmarks/MiBench/automotive-bitcount/automotive-bitcount - 15%
(matrix and automotive-bitcount were both in the top-5 slowdown list from the
anti-dep. change)
Largest slowdowns:
MultiSource/Benchmarks/McCat/03-testtrie/testtrie - 28%
MultiSource/Benchmarks/mediabench/gsm/toast/toast - 26%
MultiSource/Benchmarks/MiBench/automotive-susan/automotive-susan - 21%
SingleSource/Benchmarks/CoyoteBench/lpbench - 20%
MultiSource/Applications/d/make_dparser - 16%
llvm-svn: 158296
The PPC64 backend had patterns for i32 <-> i64 extensions and truncations that
would leave self-moves in the final assembly. Replacing those patterns with ones
based on the SUBREG builtins yields better-looking code.
Thanks to Jakob and Owen for their suggestions in this matter.
llvm-svn: 158283
Tail merging had been disabled on PPC because it would disturb bundling decisions
made during pre-RA scheduling on the 970 cores. Now, however, all bundling decisions
are made during post-RA scheduling, and tail merging is generally beneficial (the
average test-suite speedup is insignificantly positive).
Largest test-suite speedups:
MultiSource/Benchmarks/mediabench/gsm/toast/toast - 30%
MultiSource/Benchmarks/BitBench/uuencode/uuencode - 23%
SingleSource/Benchmarks/Shootout-C++/ary - 21%
SingleSource/Benchmarks/Stanford/Queens - 17%
Largest slowdowns:
MultiSource/Benchmarks/MiBench/security-sha/security-sha - 24%
MultiSource/Benchmarks/McCat/03-testtrie/testtrie - 22%
MultiSource/Applications/JM/ldecod/ldecod - 14%
MultiSource/Benchmarks/mediabench/g721/g721encode/encode - 9%
This is improved by using full (instead of just critical) anti-dependency breaking,
but doing so still causes miscompiles and so cannot yet be enabled by default.
llvm-svn: 158259
The fast register allocator is not supposed to work in the optimizing
pipeline. It doesn't make sense to compute live intervals, run full copy
coalescing, and then run RAFast.
Fast register allocation in the optimizing pipeline is better done by
RABasic.
llvm-svn: 158242
Thanks to Jakob's help, this now causes no new test suite failures!
Over the entire test suite, this gives an average 1% speedup. The largest speedups are:
SingleSource/Benchmarks/Misc/pi - 108%
SingleSource/Benchmarks/CoyoteBench/lpbench - 54%
MultiSource/Benchmarks/Prolangs-C/unix-smail/unix-smail - 50%
SingleSource/Benchmarks/Shootout/ary3 - 32%
SingleSource/Benchmarks/Shootout-C++/matrix - 30%
The largest slowdowns are:
MultiSource/Benchmarks/mediabench/gsm/toast/toast - -30%
MultiSource/Benchmarks/Prolangs-C/bison/mybison - -25%
MultiSource/Benchmarks/BitBench/uuencode/uuencode - -22%
MultiSource/Applications/d/make_dparser - -14%
SingleSource/Benchmarks/Shootout-C++/ary - -13%
In light of these slowdowns, additional profiling work is obviously needed!
llvm-svn: 158223
The pass itself works well, but the something in the Machine* infrastructure
does not understand terminators which define registers. Without the ability
to use the block-placement pass, etc. this causes performance regressions (and
so is turned off by default). Turning off the analysis turns off the problems
with the Machine* infrastructure.
llvm-svn: 158206
The code which tests for an induction operation cannot assume that any
ADDI instruction will have a register operand because the operand could
also be a frame index; for example:
%vreg16<def> = ADDI8 <fi#0>, 0; G8RC:%vreg16
llvm-svn: 158205
This pass is derived from the Hexagon HardwareLoops pass. The only significant enhancement over the Hexagon
pass is that PPCCTRLoops will also attempt to delete the replaced add and compare operations if they are
no longer otherwise used. Also, invalid preheader DebugLoc is not used.
llvm-svn: 158204
This patch will generate the following for integer ABS:
movl %edi, %eax
negl %eax
cmovll %edi, %eax
INSTEAD OF
movl %edi, %ecx
sarl $31, %ecx
leal (%rdi,%rcx), %eax
xorl %ecx, %eax
There exists a target-independent DAG combine for integer ABS, which converts
integer ABS to sar+add+xor. For X86, we match this pattern back to neg+cmov.
This is implemented in PerformXorCombine.
rdar://10695237
llvm-svn: 158175
This patch will optimize the following
movq %rdi, %rax
subq %rsi, %rax
cmovsq %rsi, %rdi
movq %rdi, %rax
to
cmpq %rsi, %rdi
cmovsq %rsi, %rdi
movq %rdi, %rax
Perform this optimization if the actual result of SUB is not used.
rdar: 11540023
llvm-svn: 158126
The commit is intended to fix rdar://11540023.
It is implemented as part of peephole optimization. We can actually implement
this in the SelectionDAG lowering phase.
llvm-svn: 158122
when a compile time constant is known. This occurs when implicitly zero
extending function arguments from 16 bits to 32 bits.
<rdar://problem/11481151>
llvm-svn: 157966
It seems that this no longer causes test suite failures on PPC64 (after r157159),
and often gives a performance benefit, so it can be enabled by default.
llvm-svn: 157911
This patch will optimize the following:
sub r1, r3
cmp r3, r1 or cmp r1, r3
bge L1
TO
sub r1, r3
bge L1 or ble L1
If the branch instruction can use flag from "sub", then we can eliminate
the "cmp" instruction.
llvm-svn: 157831
This implements codegen support for accesses to thread-local variables
using the local-dynamic model, and adds a clean-up pass so that the base
address for the TLS block can be re-used between local-dynamic access on
an execution path.
llvm-svn: 157818
types, as well as int<->ptr casts. This allows us to tailcall functions
with some trivial casts between the call and return (i.e. because the
return types disagree).
llvm-svn: 157798
This patch will optimize the following
movq %rdi, %rax
subq %rsi, %rax
cmovsq %rsi, %rdi
movq %rdi, %rax
to
cmpq %rsi, %rdi
cmovsq %rsi, %rdi
movq %rdi, %rax
Perform this optimization if the actual result of SUB is not used.
rdar: 11540023
llvm-svn: 157755
I disabled FMA3 autodetection, since the result may differ from expected for some benchmarks.
I added tests for GodeGen and intrinsics.
I did not change llvm.fma.f32/64 - it may be done later.
llvm-svn: 157737
It helps compile exotic inline asm. In the test case, normal GR32
virtual registers use up eax-edx so the final GR32_ABCD live range has
no registers left. Since all the live ranges were tiny, we had no way of
prioritizing the smaller register class.
This patch allows tiny unspillable live ranges to be evicted by tiny
unspillable live ranges from a smaller register class.
<rdar://problem/11542429>
llvm-svn: 157715
integer registers. This is already supported by the fastcc convention, but it doesn't
hurt to support it in the standard conventions as well.
In cases where we can cheat at the calling convention, this allows us to avoid returning
things through memory in more cases.
llvm-svn: 157698
Besides adding the new insertPass function, this patch uses it to
enhance the existing -print-machineinstrs so that the MachineInstrs
after a specific pass can be printed.
Patch by Bin Zeng!
llvm-svn: 157655
This required light surgery on the assembler and disassembler
because the instructions use an uncommon encoding. They are
the only two instructions in x86 that use register operands
and two immediates.
llvm-svn: 157634
SimplifyCFG tends to form a lot of 2-3 case switches when merging branches. Move
the most likely condition to the front so it is checked first and the others can
be skipped. This is currently not as effective as it could be because SimplifyCFG
destroys profiling metadata when merging branches and switches. Merging branch
weight metadata is tricky though.
This code touches at most 3 cases so I didn't use a proper sorting algorithm.
llvm-svn: 157521
Live ranges with a constrained register class may benefit from splitting
around individual uses. It allows the remaining live range to use a
larger register class where it may allocate. This is like spilling to a
different register class.
This is only attempted on constrained register classes.
<rdar://problem/11438902>
llvm-svn: 157354
Now that the coalescer keeps live intervals and machine code in sync at
all times, it needs to deal with identity copies differently.
When merging two virtual registers, all identity copies are removed
right away. This means that other identity copies must come from
somewhere else, and they are going to have a value number.
Deal with such copies by merging the value numbers before erasing the
copy instruction. Otherwise, we leave dangling value numbers in the live
interval.
This fixes PR12927.
llvm-svn: 157340
may be RAUW'd by the recursive call to LegalizeOps; instead, retrieve
the other operands when calling UpdateNodeOperands. Fixes PR12889.
llvm-svn: 157162
Dead code elimination during coalescing could cause a virtual register
to be split into connected components. The following rewriting would be
confused about the already joined copies present in the code, but
without a corresponding value number in the live range.
Erase all joined copies instantly when joining intervals such that the
MI and LiveInterval representations are always in sync.
llvm-svn: 157135
The late dead code elimination is no longer necessary.
The test changes are cause by a register hint that can be either %rdi or
%rax. The choice depends on the use list order, which this patch changes.
llvm-svn: 157131
Use a dedicated MachO load command to annotate data-in-code regions.
This is the same format the linker produces for final executable images,
allowing consistency of representation and use of introspection tools
for both object and executable files.
Data-in-code regions are annotated via ".data_region"/".end_data_region"
directive pairs, with an optional region type.
data_region_directive := ".data_region" { region_type }
region_type := "jt8" | "jt16" | "jt32" | "jta32"
end_data_region_directive := ".end_data_region"
The previous handling of ARM-style "$d.*" labels was broken and has
been removed. Specifically, it didn't handle ARM vs. Thumb mode when
marking the end of the section.
rdar://11459456
llvm-svn: 157062
non-profitable commute using outdated info. The test case would still fail
because of poor pre-RA schedule. That will be fixed by MI scheduler.
rdar://11472010
llvm-svn: 157038
This is the same as the other tests: Clever tricks are required to make
the arguments and return value line up in a single-instruction function.
It rarely happens in real life.
We have plenty other examples of this behavior.
llvm-svn: 157030
This option has been disabled for a while, and it is going away so I can
clean up the coalescer code.
The tests that required physreg joining to be enabled were almost all of
the form "tiny function with interference between arguments and return
value". Such functions are usually inlined in the real world.
The problem exposed by phys_subreg_coalesce-3.ll is real, but fairly
rare.
llvm-svn: 157027
It is now possible to coalesce weird skewed sub-register copies by
picking a super-register class larger than both original registers. The
included test case produces code like this:
vld2.32 {d16, d17, d18, d19}, [r0]!
vst2.32 {d18, d19, d20, d21}, [r0]
We still perform interference checking as if it were a normal full copy
join, so this is still quite conservative. In particular, the f1 and f2
functions in the included test case still have remaining copies because
of false interference.
llvm-svn: 156878
- Remove code which lowers pseudo SETGP01.
- Fix LowerSETGP01. The first two of the three instructions that are emitted to
initialize the global pointer register now use register $2.
- Stop emitting .cpload directive.
llvm-svn: 156689
pointer register.
This is the first of the series of patches which clean up the way global pointer
register is used. The patches will make the following improvements:
- Make $gp an allocatable temporary register rather than reserving it.
- Use a virtual register as the global pointer register and let the register
allocator decide which register to assign to it or whether spill/reloads are
needed.
- Make sure $gp is valid at the entry of a called function, which is necessary
for functions using lazy binding.
- Remove the need for emitting .cprestore and .cpload directives.
llvm-svn: 156671
This patch will optimize the following cases:
sub r1, r3 | sub r1, imm
cmp r3, r1 or cmp r1, r3 | cmp r1, imm
bge L1
TO
subs r1, r3
bge L1 or ble L1
If the branch instruction can use flag from "sub", then we can replace
"sub" with "subs" and eliminate the "cmp" instruction.
rdar: 10734411
llvm-svn: 156599
This patch will optimize the following cases:
sub r1, r3 | sub r1, imm
cmp r3, r1 or cmp r1, r3 | cmp r1, imm
bge L1
TO
subs r1, r3
bge L1 or ble L1
If the branch instruction can use flag from "sub", then we can replace
"sub" with "subs" and eliminate the "cmp" instruction.
rdar: 10734411
llvm-svn: 156550
This patch will optimize -(x != 0) on X86
FROM
cmpl $0x01,%edi
sbbl %eax,%eax
notl %eax
TO
negl %edi
sbbl %eax %eax
In order to generate negl, I added patterns in Target/X86/X86InstrCompiler.td:
def : Pat<(X86sub_flag 0, GR32:$src), (NEG32r GR32:$src)>;
rdar: 10961709
llvm-svn: 156312
The primitive conservative heuristic seems to give a slight overall
improvement while not regressing stuff. Make it available to wider
testing. If you notice any speed regressions (or significant code
size regressions) let me know!
llvm-svn: 156258
This came up when a change in block placement formed a cmov and slowed down a
hot loop by 50%:
ucomisd (%rdi), %xmm0
cmovbel %edx, %esi
cmov is a really bad choice in this context because it doesn't get branch
prediction. If we emit it as a branch, an out-of-order CPU can do a better job
(if the branch is predicted right) and avoid waiting for the slow load+compare
instruction to finish. Of course it won't help if the branch is unpredictable,
but those are really rare in practice.
This patch uses a dumb conservative heuristic, it turns all cmovs that have one
use and a direct memory operand into branches. cmovs usually save some code
size, so we disable the transform in -Os mode. In-Order architectures are
unlikely to benefit as well, those are included in the
"predictableSelectIsExpensive" flag.
It would be better to reuse branch probability info here, but BPI doesn't
support select instructions currently. It would make sense to use the same
heuristics as the if-converter pass, which does the opposite direction of this
transform.
Test suite shows a small improvement here and there on corei7-level machines,
but the actual results depend a lot on the used microarchitecture. The
transformation is currently disabled by default and available by passing the
-enable-cgp-select2branch flag to the code generator.
Thanks to Chandler for the initial test case to him and Evan Cheng for providing
me with comments and test-suite numbers that were more stable than mine :)
llvm-svn: 156234
The new target machines are:
nvptx (old ptx32) => 32-bit PTX
nvptx64 (old ptx64) => 64-bit PTX
The sources are based on the internal NVIDIA NVPTX back-end, and
contain more functionality than the current PTX back-end currently
provides.
NV_CONTRIB
llvm-svn: 156196
to catch cases like:
%reg1024<def> = MOV r1
%reg1025<def> = MOV r0
%reg1026<def> = ADD %reg1024, %reg1025
r0 = MOV %reg1026
By commuting ADD, it let coalescer eliminate all of the copies. However, there
was a bug in the heuristics where it ended up commuting the ADD in:
%reg1024<def> = MOV r0
%reg1025<def> = MOV 0
%reg1026<def> = ADD %reg1024, %reg1025
r0 = MOV %reg1026
That did no benefit but rather ensure the last MOV would not be coalesced.
rdar://11355268
llvm-svn: 156048
This patch will optimize the following cases on X86
(a > b) ? (a-b) : 0
(a >= b) ? (a-b) : 0
(b < a) ? (a-b) : 0
(b <= a) ? (a-b) : 0
FROM
movl %edi, %ecx
subl %esi, %ecx
cmpl %edi, %esi
movl $0, %eax
cmovll %ecx, %eax
TO
xorl %eax, %eax
subl %esi, %edi
cmovll %eax, %edi
movl %edi, %eax
rdar: 10734411
llvm-svn: 155919
On x86-32, structure return via sret lets the callee pop the hidden
pointer argument off the stack, which the caller then re-pushes.
However if the calling convention is fastcc, then a register is used
instead, and the caller should not adjust the stack. This is
implemented with a check of IsTailCallConvention
X86TargetLowering::LowerCall but is now checked properly in
X86FastISel::DoSelectCall.
(this time, actually commit what was reviewed!)
llvm-svn: 155825
ARM BUILD_VECTORs created after type legalization cannot use i8 or i16
operands, since those types are not legal. Instead use i32 operands, which
will be implicitly truncated by the BUILD_VECTOR to match the element type.
llvm-svn: 155824
This time, also fix the caller of AddGlue to properly handle
incomplete chains. AddGlue had failure modes, but shamefully hid them
from its caller. It's luck ran out.
Fixes rdar://11314175: BuildSchedUnits assert.
llvm-svn: 155749
On x86-32, structure return via sret lets the callee pop the hidden
pointer argument off the stack, which the caller then re-pushes.
However if the calling convention is fastcc, then a register is used
instead, and the caller should not adjust the stack. This is
implemented with a check of IsTailCallConvention
X86TargetLowering::LowerCall but is now checked properly in
X86FastISel::DoSelectCall.
llvm-svn: 155745
Owen Anderson