In getOffsetRange, Max can be set to 0 to force the extender replacement
to be at or below the original value. This would cause the new offset to
be non-negative, which is preferred for memory instructions (to reduce
the likelihood of it getting constant-extended due to predication). The
problem happens when the range is shifted by an offset (present in the
instruction being examined) and the offset is negative. The entire range
for the allowable deviation will then be strictly negative. This creates
a problem, since 0 is assumed to be a valid deviation.
llvm-svn: 316601
This updates the MIRPrinter to include the regclass when printing
virtual register defs, which is already valid syntax for the
parser. That is, given 64 bit %0 and %1 in a "gpr" regbank,
%1(s64) = COPY %0(s64)
would now be written as
%1:gpr(s64) = COPY %0(s64)
While this change alone introduces a bit of redundancy with the
registers block, it allows us to update the tests to be more concise
and understandable and brings us closer to being able to remove the
registers block completely.
Note: We generally only print the class in defs, but there is one
exception. If there are uses without any defs whatsoever, we'll print
the class on all uses. I'm not completely convinced this comes up in
meaningful machine IR, but for now the MIRParser and MachineVerifier
both accept that kind of stuff, so we don't want to have a situation
where we can print something we can't parse.
llvm-svn: 316479
In HexagonISelLowering, there is code to handle the case when
a function returns an i1 type. In this case, we need to generate
extra nodes to copy the result from R0 to a predicate register.
The code was returning the wrong value for the chain edge which
caused an assert "Wrong topological sorting" when converting the
instructions to MIs.
This patch fixes the problem by returning the chain for the final
copy.
Patch by Brendon Cahoon.
llvm-svn: 316367
Normally, if the registers holding the induction variable's bounds
are redefined inside of the loop's body, the loop cannot be converted
to a hardware loop. However, if the redefining instruction is actually
loading an immediate value into the register, this conversion is both
possible and legal (since the immediate itself will be used in the
loop setup in the preheader).
llvm-svn: 316218
This patch lets the llvm tools handle the new HVX target features that
are added by frontend (clang). The target-features are of the form
"hvx-length64b" for 64 Byte HVX mode, "hvx-length128b" for 128 Byte mode HVX.
"hvx-double" is an alias to "hvx-length128b" and is soon will be deprecated.
The hvx version target feature is upgated form "+hvx" to "+hvxv{version_number}.
Eg: "+hvxv62"
For the correct HVX code generation, the user must use the following
target features.
For 64B mode: "+hvxv62" "+hvx-length64b"
For 128B mode: "+hvxv62" "+hvx-length128b"
Clang picks a default length if none is specified. If for some reason,
no hvx-length is specified to llvm, the compilation will bail out.
There is a corresponding clang patch.
Differential Revision: https://reviews.llvm.org/D38851
llvm-svn: 316101
Each constant extender requires an extra instruction, which adds to the
code size and also reduces the number of available slots in an instruction
packet. In most cases, the value of a repeated constant extender could be
loaded into a register, and the instructions using the extender could be
replaced with their counterparts that use that register instead.
This patch adds a pass that tries to reduce the number of constant
extenders, including extenders which differ only in an immediate offset
known at compile time, e.g. @global and @global+12.
llvm-svn: 315735
The pipeliner is generating a serial sequence that causes poor
register allocation when a post-increment instruction appears
prior to the use of the post-increment register. This occurs when
there is a circular set of dependences involved with a sequence
of instructions in the same cycle. In this case, there is no
serialization of the parallel semantics that will not cause an
additional register to be allocated.
This patch fixes the problem by changing the instructions so that
the post-increment instruction is used by the subsequent
instruction, which enables the register allocator to make a
better decision and not require another register.
Patch by Brendon Cahoon.
llvm-svn: 315466
If the two instructions being compared for equivalence have corresponding operands
that are integer constants, then check their values to determine equivalence.
Patch by Suyog Sarda!
llvm-svn: 314642
This patch produces a crash and hexagon_vector_loop_carried_reuse_constant.ll test fails on Windows (llvm-clang-x86_64-expensive-checks-win build bot).
llvm-svn: 314361
If the two instructions being compared for equivalence have corresponding operands
that are integer constants, then check their values to determine equivalence.
Patch by Suyog Sarda!
llvm-svn: 313993
Tail merging can convert an undef use into a normal one when creating a
common tail. Doing so can make the register live out from a block which
previously contained the undef use. To keep the liveness up-to-date,
insert IMPLICIT_DEFs in such blocks when necessary.
To enable this patch the computeLiveIns() function which used to
compute live-ins for a block and set them immediately is split into new
functions:
- computeLiveIns() just computes the live-ins in a LivePhysRegs set.
- addLiveIns() applies the live-ins to a block live-in list.
- computeAndAddLiveIns() is a convenience function combining the other
two functions and behaving like computeLiveIns() before this patch.
Based on a patch by Krzysztof Parzyszek <kparzysz@codeaurora.org>
Differential Revision: https://reviews.llvm.org/D37034
llvm-svn: 312668
When if-converting a diamond, two separate blocks will be placed back
to back to form a straight line code. To ensure correctness of the
liveness information, any registers that are live in the second block
should not be killed in the first block, even if they were in the
original code.
Additionally, when the two blocks share common instructions at the
beginning, these instructions will not be duplicated, but only placed
once, before both of the blocks. Since the function "isIdenticalTo"
(as used here) ignores kill flags, the common initial code in one
block may have a kill flag for a register that is live in the other
block.
Because the code that removes kill flags only runs for the non-common
parts of the predicated blocks, a kill flag mismatch in the common
code could still lead to a live register being killed prematurely.
llvm-svn: 312654
The check (assuming positive stride) for validity of memmove should be
(a) the destination is at a lower address than the source, or
(b) the distance between the source and destination is greater than or
equal the number of bytes copied.
For the second part it is sufficient to assume that the destination
is at a higher address, since the opposite case is covered by (a).
The distance calculation was previously done by subtracting the
pointers in the wrong order.
llvm-svn: 311650
isLegalAddressingMode() has recently gained the extra optional Instruction*
parameter, and therefore it can now do the job that previously only
isFoldableMemAccess() could do.
The SystemZ implementation of isLegalAddressingMode() has gained the
functionality of checking for offsets, which used to be done with
isFoldableMemAccess().
The isFoldableMemAccess() hook has been removed everywhere.
Review: Quentin Colombet, Ulrich Weigand
https://reviews.llvm.org/D35933
llvm-svn: 310463
Certain operations require vector of i1 values. However, for Hexagon
architecture compatibility, they need to be represented as vector of i8.
Patch by Suyog Sarda.
llvm-svn: 309677
The flag "-hexagon-emit-lut-text" (defaulted to false) is added to decide
on where to keep the switch generated lookup table.
Differential Revision: https://reviews.llvm.org/D34818
llvm-svn: 308316
The target-independent lowering works fine, except concatenating 32-bit
words. Add a pattern to generate A2_combinew instead of 64-bit asl/or.
llvm-svn: 308186
This is the LLVM part, adding definitions for
void @llvm.hexagon.Y2.dccleana(i8*)
void @llvm.hexagon.Y2.dccleaninva(i8*)
void @llvm.hexagon.Y2.dcinva(i8*)
void @llvm.hexagon.Y2.dczeroa(i8*)
void @llvm.hexagon.Y4.l2fetch(i8*, i32)
void @llvm.hexagon.Y5.l2fetch(i8*, i64)
The clang part will follow.
llvm-svn: 308032
This patch adds a new LLVM flag -hexagon-emit-jt-text which is defaulted to
"false". The value "true" emits the switch generated jump tables in text section.
Differential Revision: https://reviews.llvm.org/D34820
llvm-svn: 306872
The llvm flag "-hexagon-emit-lookup-tables" guards the generation
of lookup table from a switch statement.
Differential Revision: https://reviews.llvm.org/D34819
llvm-svn: 306869
The feeder instruction will be moved to right before the compare, so
the updating code should not be looking for kills past the compare.
llvm-svn: 306059
The second part of r305300: when placing the mux at the later location,
make sure that it won't use any register that was killed between the
two original instructions. Remove any such kills and transfer them to
the mux.
llvm-svn: 305553
Store-immediate instructions have a non-extendable offset. Since the
actual offset for a stack object is not known until much later, only
generate these stores when the stack size (at the time of instruction
selection) is small.
llvm-svn: 305305
When a mux instruction is created from a pair of complementary conditional
transfers, it can be placed at the location of either the earlier or the
later of the transfers. Since it will use the operands of the original
transfers, putting it in the earlier location may hoist a kill of a source
register that was originally further down. Make sure the kill flag is
removed if the register is still used afterwards.
llvm-svn: 305300
The initial assumption was that the simplification would converge to a
fixed point relatvely quickly. Turns out that there are legitimate situa-
tions where the complexity of the code causes it to take a large number
of iterations.
Two main changes:
- Instead of aborting upon hitting the limit, simply return nullptr.
- Reduce the limit to 10,000 from 100,000.
llvm-svn: 304441
For multiplications of 64-bit values (giving 64-bit result), detect
cases where the arguments are sign-extended 32-bit values, on a per-
operand basis. This will allow few patterns to match a wider variety
of combinations in which extensions can occur.
llvm-svn: 304223
Rewrite fixupKills() to use the LivePhysRegs class. Simplifies the code
and fixes a bug where the CSR registers in return blocks where missed
leading to invalid kill flags. Also remove the unnecessary rule that we
wouldn't set kill flags on tied operands.
No tests as I have an upcoming commit improving MachineVerifier checks
to catch these cases in multiple existing lit tests.
llvm-svn: 304055
Using arguments with attribute inalloca creates problems for verification
of machine representation. This attribute instructs the backend that the
argument is prepared in stack prior to CALLSEQ_START..CALLSEQ_END
sequence (see http://llvm.org/docs/InAlloca.htm for details). Frame size
stored in CALLSEQ_START in this case does not count the size of this
argument. However CALLSEQ_END still keeps total frame size, as caller can
be responsible for cleanup of entire frame. So CALLSEQ_START and
CALLSEQ_END keep different frame size and the difference is treated by
MachineVerifier as stack error. Currently there is no way to distinguish
this case from actual errors.
This patch adds additional argument to CALLSEQ_START and its
target-specific counterparts to keep size of stack that is set up prior to
the call frame sequence. This argument allows MachineVerifier to calculate
actual frame size associated with frame setup instruction and correctly
process the case of inalloca arguments.
The changes made by the patch are:
- Frame setup instructions get the second mandatory argument. It
affects all targets that use frame pseudo instructions and touched many
files although the changes are uniform.
- Access to frame properties are implemented using special instructions
rather than calls getOperand(N).getImm(). For X86 and ARM such
replacement was made previously.
- Changes that reflect appearance of additional argument of frame setup
instruction. These involve proper instruction initialization and
methods that access instruction arguments.
- MachineVerifier retrieves frame size using method, which reports sum of
frame parts initialized inside frame instruction pair and outside it.
The patch implements approach proposed by Quentin Colombet in
https://bugs.llvm.org/show_bug.cgi?id=27481#c1.
It fixes 9 tests failed with machine verifier enabled and listed
in PR27481.
Differential Revision: https://reviews.llvm.org/D32394
llvm-svn: 302527
Hoisting common code can cause registers that live-in in the successor
blocks to no longer be live-in. The live-in information needs to be
updated to reflect this, or otherwise incorrect code can be generated
later on.
Differential Revision: https://reviews.llvm.org/D32661
llvm-svn: 302228
Allocframe and the following stores on the stack have a latency of 2 cycles
when not in the same packet. This happens because R29 is needed early by the
store instruction. Since one of such stores can be packetized along with
allocframe and use old value of R29, we can assign it 0 cycle latency
while leaving latency of other stores to the default value of 2 cycles.
Patch by Jyotsna Verma.
llvm-svn: 302034
The compiler was generating code that ends up ignoring a multiple
latency dependence between two instructions by scheduling the
intructions in back-to-back packets.
The packetizer needs to end a packet if the latency of the current
current insruction and the source in the previous packet is
greater than 1 cycle. This case occurs when there is still room in
the current packet, but scheduling the instruction causes a stall.
Instead, the packetizer should start a new packet. Also, if the
current packet already contains a stall, then it is okay to add
another instruction to the packet that also causes a stall. This
occurs when there are no instructions that can be scheduled in
between the producer and consumer instructions.
This patch changes the latency for loads to 2 cycles from 3 cycles.
This change refects that a load only needs to be separated by
one extra packet to eliminate the stall.
Patch by Ikhlas Ajbar.
llvm-svn: 301954
When a PHI operand has a subregister, create a COPY instead of simply
replacing the PHI output with the input it.
Differential Revision: https://reviews.llvm.org/D32650
llvm-svn: 301699
Also, make a few changes to allow using the pass in .mir testcases.
Among other things, change the abbreviation from opt-amode to amode-opt,
because otherwise lit would expand the "opt" part to the full path to
the opt binary.
llvm-svn: 300707
- Avoid explosive growth of the simplification queue by not queuing
expressions that are alredy in it.
- Add an iteration counter and abort after a sufficiently large number
of iterations (assuming that it's a symptom of an infinite loop).
llvm-svn: 298655
[Hexagon] Recognize polynomial-modulo loop idiom again
Regain the ability to recognize loops calculating polynomial modulo
operation. This ability has been lost due to some changes in the
preceding optimizations. Add code to preprocess the IR to a form
that the pattern matching code can recognize.
llvm-svn: 298400
Regain the ability to recognize loops calculating polynomial modulo
operation. This ability has been lost due to some changes in the
preceding optimizations. Add code to preprocess the IR to a form
that the pattern matching code can recognize.
llvm-svn: 298282
- Fix the insertion point, which occasionally could have been incorrect.
- Avoid creating multiple bitsplits with the same operands, if an old one
could be reused.
llvm-svn: 297414
When extracting a bitfield from the high register in a register pair,
the final offset should be relative to the high register (for 32-bit
extracts).
llvm-svn: 297288
If a block has non-analyzable branches, the listed successors don't need
to add up to one. For example, if a block has a conditional tail call,
that tail call will not have a corresponding successor in the successor
list, but will still be a possible branch.
Differential Revision: https://reviews.llvm.org/D30556
llvm-svn: 297054
Merge the tail block into the loop in cases where the main loop body
exits early, subject to profitability constraints. This will coalesce
the loop body into fewer blocks.
For example:
loop: loop:
// loop body // loop body
if (...) jump exit --> // more body
more: if (...) jump exit
// more body jump loop
jump loop
llvm-svn: 297033
The code in updateDeadFlags removed unnecessary <dead> flags, but there
can be cases where such a flag is not set, and yet a register has become
dead. For example, if a mux with identical inputs is replaced with a COPY,
the predicate register may no longer be used after that.
llvm-svn: 297032
On Hexagon, values of type i1 are passed in registers of type i32,
even though i1 is not a legal value for these registers. This is a
special case and needs special handling to maintain consistency of
the lowering information.
This fixes PR32089.
llvm-svn: 296645
The motivation for filling out these select-of-constants cases goes back to D24480,
where we discussed removing an IR fold from add(zext) --> select. And that goes back to:
https://reviews.llvm.org/rL75531https://reviews.llvm.org/rL159230
The idea is that we should always canonicalize patterns like this to a select-of-constants
in IR because that's the smallest IR and the best for value tracking. Note that we currently
do the opposite in some cases (like the cases in *this* patch). Ie, the proposed folds in
this patch already exist in InstCombine today:
https://github.com/llvm-mirror/llvm/blob/master/lib/Transforms/InstCombine/InstCombineSelect.cpp#L1151
As this patch shows, most targets generate better machine code for simple ext/add/not ops
rather than a select of constants. So the follow-up steps to make this less of a patchwork
of special-case folds and missing IR canonicalization:
1. Have DAGCombiner convert any select of constants into ext/add/not ops.
2 Have InstCombine canonicalize in the other direction (create more selects).
Differential Revision: https://reviews.llvm.org/D30180
llvm-svn: 296137
When the pipeliner is renaming phi values, it may need to iterate through
the phi operands to check for other phis. However, the pipeliner should
stop once it reaches a phi that is outside the pipelined loop.
Also, when the generateExistingPhis code is unable to reuse an existing
phi, the default code that computes the PhiOp2 is only to be used when
the pipeliner is generating the kernel. Otherwise, the phi may be a value
computed earlier in the same epilog.
Patch by Brendon Cahoon.
llvm-svn: 290355
test/CodeGen/MIR should contain tests that intent to test the MIR
printing or parsing. Tests that test something else should be in
test/CodeGen/TargetName even when they are written in .mir.
As a rule of thumb, only tests using "llc -run-pass none" should be in
test/CodeGen/MIR.
llvm-svn: 289254
The Stack slot coloring pass removes a store that is followed by a load
that deal with the same stack slot. The function isLoadFromStackSlot
is supposed to consider the loads that have no side-effects. This
patch fixed the issue by removing the unsafe loads from this function
Eg:
%vreg0<def> = L2_loadruh_io <fi#15>, 0
S2_storeri_io <fi#15>, 0, %vreg0
In this case, we load an unsigned extended half word and store this in to
the same stack slot. The Stack slot coloring pass considers safe to remove
the store. This patch marked all the non-vector byte and half word loads as
unsafe.
llvm-svn: 286843
For pairs of 32-bit registers: isub_lo, isub_hi.
For pairs of vector registers: vsub_lo, vsub_hi.
Add generic subreg indices: ps_sub_lo, ps_sub_hi, and a function
HexagonRegisterInfo::getHexagonSubRegIndex(RegClass, GenericSubreg)
that returns the appropriate subreg index for RegClass.
llvm-svn: 286377
When LivePhysRegs adds live-in registers, it recognizes ~0 as a special
lane mask indicating the entire register. If the lane mask is not ~0,
it will only add the subregisters that overlap the specified lane mask.
The problem is that if a live-in register does not have subregisters,
and the lane mask is not ~0, it will not be added to the live set.
(The given lane mask may simply be the lane mask of its register class.)
If a register does not have subregisters, add it to the live set if
the lane mask is non-zero.
Differential Revision: https://reviews.llvm.org/D26094
llvm-svn: 285440
Do not use LiveIntervals to recalculate kills, because that cannot be
done accurately without implicit uses on predicated instructions.
llvm-svn: 285409
After register allocation it is possible to have a spill of a register
that is only partially defined. That in itself it fine, but creates a
problem for double vector registers. Stores of such registers are pseudo
instructions that are expanded into pairs of individual vector stores,
and in case of a partially defined source, one of the stores may use
an entirely undefined register. To avoid this, track the defined parts
and only generate actual stores for those.
llvm-svn: 284841
This required reengineering of some of the part of liveness calculation,
including fixing some issues caused by the limitations of the previous
approach. The current code is not necessarily the fastest, but it should
be functionally correct (at least more so than before). The compile-time
performance will be addressed in the future.
llvm-svn: 284609
Branch folder removes implicit defs if they are the only non-branching
instructions in a block, and the branches do not use the defined registers.
The problem is that in some cases these implicit defs are required for
the liveness information to be correct.
Differential Revision: https://reviews.llvm.org/D25478
llvm-svn: 284036
This avoids llc using the hosts OS/vendor as defaults and triggering
unwanted behaviour in the tests. This should deal with the buildbot
breakages on windows after r283140.
llvm-svn: 283149
Each shadow only represents data flow that is restricted to its reaching
def. Propagating more than that could lead to spurious register liveness,
resulting in extra (incorrectly) block live-ins.
llvm-svn: 283143
Normally, if conversion would add implicit uses for redefined registers,
e.g. R0<def> = add_if ..., R0<imp-use>. However, if only subregisters of
R0 are known to be live but not R0 itself, such implicit uses will not be
added, causing prior definitions of such subregisters and R0 itself to
become dead.
llvm-svn: 282626
descriptions now tag add instructions, and the Hexagon backend is using this to
identify loop induction statements.
Patch by Sam Parker and Sjoerd Meijer.
Differential Revision: https://reviews.llvm.org/D23601
llvm-svn: 281304
Shadow uses need to be analyzed together, since each individual shadow
will only have a partial reaching def. All shadows together may cover
a given register ref, while each individual shadow may not.
llvm-svn: 280855
This reverts commit r280268, it causes all MSVC 2013 to ICE. This
appears to have been fixed in a later MSVC 2013 update, because I cannot
reproduce it locally. That said, all upstream LLVM bots are broken right
now, so I am reverting.
Also reverts dependent change r280275, "[Hexagon] Deal with undefs when
extending live intervals".
llvm-svn: 280301
This bug shows up with diamonds that share unpredicable, unanalyzable branches.
There's an included test case from Hexagon. What was happening was that we were
attempting to predicate the branch instruction despite the fact that it was
checked to be the same. Now for unanalyzable branches we skip over the branch
instructions when predicating the block.
Differential Revision: https://reviews.llvm.org/D23939
llvm-svn: 279985
Rename AllVRegsAllocated to NoVRegs. This avoids the connotation of
running after register and simply describes that no vregs are used in
a machine function. With that we can simply compute the property and do
not need to dump/parse it in .mir files.
Differential Revision: http://reviews.llvm.org/D23850
llvm-svn: 279698
The register allocator can split a live interval of a register into a set
of smaller intervals. After the allocation of registers is complete, the
rewriter will modify the IR to replace virtual registers with the corres-
ponding physical registers. At this stage, if a register corresponding
to a subregister of a virtual register is used, the rewriter will check
if that subregister is undefined, and if so, it will add the <undef> flag
to the machine operand. The function verifying liveness of the subregis-
ter would assume that it is undefined, unless any of the subranges of the
live interval proves otherwise.
The problem is that the live intervals created during splitting do not
have any subranges, even if the original parent interval did. This could
result in the <undef> flag placed on a register that is actually defined.
Differential Revision: http://reviews.llvm.org/D21189
llvm-svn: 279625
Improved handling of fma, floating point min/max, additional load/store
instructions for floating point types.
Patch by Jyotsna Verma.
llvm-svn: 279239
The pipeliner was generating an invalid Phi name for an operand
in the epilog block, which caused an assert in the live variable
analysis pass. The fix is to the code that generates new Phis
in the epilog block. In this case, there is an existing Phi that
needs to be reused rather than creating a new Phi instruction.
Differential Revision: https://reviews.llvm.org/D23513
llvm-svn: 278805
If a loop is not rotated (for example when optimizing for size), the latch is not the backedge. If we promote an expression to post-inc form, we not only increase register pressure and add a COPY for that IV expression but for all IVs!
Motivating testcase:
void f(float *a, float *b, float *c, int n) {
while (n-- > 0)
*c++ = *a++ + *b++;
}
It's imperative that the pointer increments be located in the latch block and not the header block; if not, we cannot use post-increment loads and stores and we have to keep both the post-inc and pre-inc values around until the end of the latch which bloats register usage.
llvm-svn: 278658
LowerTargetConstantPool is not properly setting the TargetFlag to indicate
desired relocation. Coding error, the offset parameter was omitted, so the
TargetFlag was used as the offset, and the TargetFlag defaulted to zero.
This only affects -fpic compilation, and only those items created in a
Constant Pool, for example a vector of constants. Halide ran into this issue.
llvm-svn: 278614
From the point of view of register assignment, byval parameters are
ignored: a byval parameter is not going to be assigned to a register,
and it will not affect the assignments of subsequent parameters.
When matching registers with parameters in the bit tracker, make sure
to skip byval parameters before advancing the registers.
llvm-svn: 278375
This change makes it possible for tail-duplication and tail-merging to
be disjoint. By being less aggressive when merging during layout, there are no
overlapping cases between tail-duplication and tail-merging, provided the
thresholds are disjoint.
There is a remaining TODO to benchmark the succ_size() test for non-layout tail
merging.
llvm-svn: 278265
Floating point instructions use general purpose registers, so the few
instructions that can put floating point immediates into registers are,
in fact, integer instruction. Use them explicitly instead of having
pseudo-instructions specifically for dealing with floating point values.
Simplify the constant loading instructions (from sdata) to have only two:
one for 32-bit values and one for 64-bit values: CONST32 and CONST64.
llvm-svn: 278244
When the same base address is used to load two different data types, LSR
would assume a memory type of "void". This type is not sized and has no
alignment information. Checking for it causes a crash.
llvm-svn: 277601
Identify patterns where the address is aligned to an 8-byte boundary,
but both the base address and the constant offset are both proper
multiples of 4. In such cases, extract Base+4 into a separate instruc-
tion, and use S2_storerd_io, instead of using S4_storerd_rr.
llvm-svn: 277497
Scavenging slots were only reserved when pseudo-instruction expansion in
frame lowering created new virtual registers. It is possible to still
need a scavenging slot even if no virtual registers were created, in cases
where the stack is large enough to overflow instruction offsets.
llvm-svn: 277355
The DAG combiner will try to merge consecutive stores into a bigger
store, unless the resulting store is not fast. Misaligned vector stores
are allowed on Hexagon, but are not fast. Add a testcase to make sure
this type of merging does not occur.
Patch by Pranav Bhandarkar.
llvm-svn: 277182
The DAG combiner tries to merge stores to adjacent vector wide memory
locations by creating stores which are integral multiples of the vector
width. Discourage this by informing it that this is slow. This should
not affect legalization passes, because all of them ignore the "Fast"
argument.
Patch by Pranav Bhandarkar.
llvm-svn: 277178
Software pipelining is an optimization for improving ILP by
overlapping loop iterations. Swing Modulo Scheduling (SMS) is
an implementation of software pipelining that attempts to
reduce register pressure and generate efficient pipelines with
a low compile-time cost.
This implementaion of SMS is a target-independent back-end pass.
When enabled, the pass should run just prior to the register
allocation pass, while the machine IR is in SSA form. If the pass
is successful, then the original loop is replaced by the optimized
loop. The optimized loop contains one or more prolog blocks, the
pipelined kernel, and one or more epilog blocks.
This pass is enabled for Hexagon only. To enable for other targets,
a couple of target specific hooks must be implemented, and the
pass needs to be called from the target's TargetMachine
implementation.
Differential Review: http://reviews.llvm.org/D16829
llvm-svn: 277169
If the mask of a vector shuffle has alternating odd or even numbers
starting with 1 or 0 respectively up to the largest possible index
for the given type in the given HVX mode (single of double) we can
generate vpacko or vpacke instruction respectively.
E.g.
%42 = shufflevector <32 x i16> %37, <32 x i16> %41,
<32 x i32> <i32 1, i32 3, ..., i32 63>
is %42.h = vpacko(%41.w, %37.w)
Patch by Pranav Bhandarkar.
llvm-svn: 277168
Rebalances address calculation trees and applies Hexagon-specific
optimizations to the trees to improve instruction selection.
Patch by Tobias Edler von Koch.
llvm-svn: 277151
Normally, CFI instructions should be inserted after allocframe, but
if allocframe is in the same packet with a call, the CFI instructions
should be inserted before that packet.
llvm-svn: 277020
Before adding a new preheader block, check if there is a candidate block
where the loop setup could be placed speculatively. This will be off by
default.
llvm-svn: 276919
Consider this case:
vreg1 = A2_zxth vreg0 (1)
...
vreg2 = A2_zxth vreg1 (2)
Redundant instruction elimination could delete the instruction (1)
because the user (2) only cares about the low 16 bits. Then it could
delete (2) because the input is already zero-extended. The problem
is that the properties allowing each individual instruction to be
deleted depend on the existence of the other instruction, so either
one can be deleted, but not both.
The existing check for this situation in RIE was insufficient. The
fix is to update all dependent cells when an instruction is removed
(replaced via COPY) in RIE.
llvm-svn: 276792
Schedule a load and its use in the same packet in MISched. Previously,
isResourceAvailable was returning false for dependences in the same
packet, which prevented MISched from packetizing a load and its use in
the same packet for v60.
Patch by Ikhlas Ajbar.
llvm-svn: 275804
The machine scheduler needs to account for available resources
more accurately in order to avoid scheduling an instruction that
forces a new packet to be created.
This occurs in two ways: First, an instruction without an available
resource may have a large priority due to other metrics and be
scheduled when there are other instructions with available resources.
Second, an instruction with a non-zero latency may become available
prematurely. In both these cases, we attempt change the priority
in order to allow a better instruction to be scheduled.
Patch by Brendon Cahoon.
llvm-svn: 275793
An instruction may have multiple predecessors that are candidates
for using .cur. However, only one of them can use .cur in the
packet. When this case occurs, we need to make sure that only
one of the dependences gets a 0 latency value.
Patch by Brendon Cahoon.
llvm-svn: 275790
- Treat bitwise OR with a frame index as an ADD wherever possible, fold it
into addressing mode.
- Extend patterns for memops to allow memops with frame indexes as address
operands.
llvm-svn: 275569
On Hexagon is it legal to packetize the instructions setting up call
arguments with the call instruction itself. This was already done,
except for tail calls. Make sure tail calls are handled as well.
llvm-svn: 275458
Transform: (store ch addr (add x (add (shl y c) e)))
to: (store ch addr (add x (shl (add y d) c))),
where e = (shl d c) for some integer d.
The purpose of this is to enable generation of loads/stores with
shifted addressing mode, i.e. mem(x+y<<#c). For that, the shift
value c must be 0, 1 or 2.
llvm-svn: 273466
The aggressive anti-dependency breaker can rename the restored callee-
saved registers. To prevent this, mark these registers are live on all
paths to the return/tail-call instructions, and add implicit use operands
for them to these instructions.
llvm-svn: 270898
When looking for an available spill slot, the register scavenger would stop
after finding the first one with no register assigned to it. That slot may
have size and alignment that do not meet the requirements of the register
that is to be spilled. Instead, find an available slot that is the closest
in size and alignment to one that is needed to spill a register from RC.
Differential Revision: http://reviews.llvm.org/D20295
llvm-svn: 269969
Recent changes to the instruction selection code exposed a problem where
a dead node was not removed on time. This node had both input and output
chains, which lead to an apparent cycle.
llvm-svn: 269458
When generating .cfi_offset instructions, make sure that the offset is
calculated with respect to the register used to define the CFA (which is
currently always FP+8).
llvm-svn: 269191
An example from Hexagon where things went wrong:
%R0<def> = L2_loadrigp <ga:@fp04> ; load function address
J2_callr %R0<kill>, ..., %R0<imp-def> ; call *R0, return value in R0
ScheduleDAGInstrs::buildSchedGraph would visit all instructions going
backwards, and in each instruction it would visit all operands in their
order on the operand list. In the case of this call, it visited the use
of R0 first, then removed it from the set Uses after it visited the def.
This caused the DAG to be missing the data dependence edge on R0 between
the load and the call.
Differential Revision: http://reviews.llvm.org/D20102
llvm-svn: 269076
ScheduleDAGMI::initQueues changes the RegionBegin to the first non-debug
instruction. Since it does not track register pressure, it does not affect
any RP trackers. ScheduleDAGMILive inherits initQueues from ScheduleDAGMI,
and it does reset the TopTPTracker in its schedule method. Any derived,
target-specific scheduler will need to do it as well, but the TopRPTracker
is only exposed as a "const" object to derived classes. Without the ability
to modify the tracker directly, this leaves a derived scheduler with a
potential of having the TopRPTracker out-of-sync with the CurrentTop.
The symptom of the problem:
void llvm::ScheduleDAGMILive::scheduleMI(llvm::SUnit *, bool):
Assertion `TopRPTracker.getPos() == CurrentTop && "out of sync"' failed.
Differential Revision: http://reviews.llvm.org/D19438
llvm-svn: 267918
When a block is tail-duplicated, the PHI nodes from that block are
replaced with appropriate COPY instructions. When those PHI nodes
contained use operands with subregisters, the subregisters were
dropped from the COPY instructions, resulting in incorrect code.
Keep track of the subregister information and use this information
when remapping instructions from the duplicated block.
Differential Revision: http://reviews.llvm.org/D19337
llvm-svn: 267583
Krzysztof Parzyszek