Issue found by llvm-isel-fuzzer on OSS fuzz, https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=3725
If anyone actually cares about > 64 bit arithmetic, there's a lot more to do in this area. There's a bunch of obviously wrong code in the same function. I don't have the time to fix all of them and am just using this to understand what the workflow for fixing fuzzer cases might look like.
llvm-svn: 316967
- Targets that want to support memcmp expansions now return the list of
supported load sizes.
- Expansion codegen does not assume that all power-of-two load sizes
smaller than the max load size are valid. For examples, this is not the
case for x86(32bit)+sse2.
Fixes PR34887.
llvm-svn: 316905
Summary:
Currently we skip merging when extra moves may be added in the header of switch instead of the case block, if the case block is used as an incoming
block of a PHI. If all the incoming values of the PHIs are non-constants and the destination block is dominated by the switch block then extra moves are likely not added by ISel, so there is no need to skip merging in this case.
Reviewers: efriedma, junbuml, davidxl, hfinkel, qcolombet
Reviewed By: efriedma
Subscribers: dberlin, kuhar, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D37343
llvm-svn: 316711
Compute the actual decomposition only after deciding whether to expand
of not. Else, it's easy to make the compiler OOM with:
`memcpy(dst, src, 0xffffffffffffffff);`, which typically happens if
someone mistakenly passes a negative value. Add a test.
This reverts commit f8fc02fbd4ab33383c010d33675acf9763d0bd44.
llvm-svn: 316567
This reverts commit r316417, which causes internal compiles to OOM.
I don't unfortunately have a self-contained test case but will follow up
with courbet.
llvm-svn: 316497
Refactor ExpandMemcmp:
- Stop duplicating the logic for computation of the sequence of loads to
generate (thsi was done in three different places), this is now done
only once in MemCmpExpansion::MemCmpExpansion().
- Add a FIXME to expose a bug with the computation of the number of loads
when not all sizes are loadable. For example, on X86-32 + SSE, possible
loads are {16,4,2,1} bytes. The current code considers that all loads
starting at MaxLoadSize are possible. This is not an issue right now as
vector loads are not enabled, so I'm not fixing the issue here to keep
the change as small as possible. I'm going to address this in a
subsequent revision, where I enable vector loads.
See https://bugs.llvm.org/show_bug.cgi?id=34887
Differential Revision: https://reviews.llvm.org/D38498
llvm-svn: 316417
Summary:
Add LLVM_FORCE_ENABLE_DUMP cmake option, and use it along with
LLVM_ENABLE_ASSERTIONS to set LLVM_ENABLE_DUMP.
Remove NDEBUG and only use LLVM_ENABLE_DUMP to enable dump methods.
Move definition of LLVM_ENABLE_DUMP from config.h to llvm-config.h so
it'll be picked up by public headers.
Differential Revision: https://reviews.llvm.org/D38406
llvm-svn: 315590
Currently optimizeMemoryInst requires that all of the AddrModes it sees are
identical. This patch makes it capable of tracking multiple AddrModes, so long
as they differ in at most one field.
This patch does nothing by itself, but later patches will make use of it to
insert or reuse phi or select instructions for the differing fields.
Differential Revision: https://reviews.llvm.org/D38278
llvm-svn: 314795
This lets us optimize away selects that perform the same address computation in
two different ways and is also the first step towards being able to handle
selects between two different, but compatible, address computations.
Differential Revision: https://reviews.llvm.org/D38242
llvm-svn: 314794
Summary:
Right now there are two functions with the same name, one does the work
and the other one returns true if expansion is needed. Rename
TargetTransformInfo::expandMemCmp to make it more consistent with other
members of TargetTransformInfo.
Remove the unused Instruction* parameter.
Differential Revision: https://reviews.llvm.org/D38165
llvm-svn: 314096
Summary:
GEP merging can sometimes increase the number of live values and register
pressure across control edges and cause performance problems particularly if the
increased register pressure results in spills.
This change implements GEP unmerging around an IndirectBr in certain cases to
mitigate the issue. This is in the CodeGenPrepare pass (after all the GEP
merging has happened.)
With this patch, the Python interpreter loop runs faster by ~5%.
Reviewers: sanjoy, hfinkel
Reviewed By: hfinkel
Subscribers: eastig, junbuml, llvm-commits
Differential Revision: https://reviews.llvm.org/D36772
llvm-svn: 312930
As suggested in D37121, here's a wrapper for removeFromParent() + insertAfter(),
but implemented using moveBefore() for symmetry/efficiency.
Differential Revision: https://reviews.llvm.org/D37239
llvm-svn: 312001
If we want to substitute the relocation of derived pointer with gep of base then
we must ensure that relocation of base dominates the relocation of derived pointer.
Currently only check for basic block is present. However it is possible that both
relocation are in the same basic block but relocation of derived pointer is defined
earlier.
The patch moves the relocation of base pointer right before relocation of derived
pointer in this case.
Reviewers: sanjoy,artagnon,igor-laevsky,reames
Reviewed By: reames
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D36462
llvm-svn: 311067
As noted in the code comment, transforming this in the other direction might require
a separate transform here in CGP given the block-at-a-time DAG constraint.
Besides that theoretical motivation, there are 2 practical motivations for the
subtract-of-cmps form:
1. The codegen for both x86 and PPC is better for this IR (though PPC could be better still).
There is discussion about canonicalizing IR to the select form
( http://lists.llvm.org/pipermail/llvm-dev/2017-July/114885.html ),
so we probably need to add DAG transforms for those patterns anyway, but this improves the
memcmp output without waiting for that step.
2. If we allow vector-sized chunks for the load and compare, x86 is better prepared to convert
that to optimal code when using subtract-of-cmps, so another prerequisite patch is avoided
if we choose to enable that.
Differential Revision: https://reviews.llvm.org/D34904
llvm-svn: 309597
Summary:
Since r293359, most dump() function are only defined when
`!defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)` holds. print() functions
only used by dump() functions are now unused in release builds,
generating lots of warnings. This patch only defines some print()
functions if they are used.
Reviewers: MatzeB
Reviewed By: MatzeB
Subscribers: arsenm, mzolotukhin, nhaehnle, llvm-commits
Differential Revision: https://reviews.llvm.org/D35949
llvm-svn: 309553
This avoids excessive compile time. The case I'm looking at is
Function.cpp from an old version of LLVM that still had the giant memcmp
string matcher in it. Before r308322 this compiled in about 2 minutes,
after it, clang takes infinite* time to compile it. With this patch
we're at 5 min, which is still bad but this is a pathological case.
The cut off at 20 uses was chosen by looking at other cut-offs in LLVM
for user scanning. It's probably too high, but does the job and is very
unlikely to regress anything.
Fixes PR33900.
* I'm impatient and aborted after 15 minutes, on the bug report it was
killed after 2h.
llvm-svn: 308891
Allowing cycles in Phi traversal increases the scope of optimize memory instruction
in case we are in loop.
The added test shows an example of enabling optimization inside a loop.
Reviewers: loladiro, spatel, efriedma
Reviewed By: efriedma
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D35294
llvm-svn: 308419
optimizeMemoryInst contains a vector AddrModeInsts.
The only use of this vector is to check that all instructions are in the same
block as memory instruction. This check is guarded by PhiSeen flag,
so if we traversed through phi node then we do not need to keep information
in AddrModeInsts. AddModeInsts is set first time we found some addressing mode
and updated if we found new one later.
We can find next addressing mode only if we traverse phi node so all code
related to update of AddModeInsts can be safely removed.
Reviewers: loladiro, spatel, efriedma
Reviewed By: efriedma
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D35291
llvm-svn: 308265
Now, getUserCost() only checks the src and dst types of EXT to decide it is free
or not. This change first checks the types, then calls isExtFreeImpl(), and
check if EXT can form ExtLoad at last. Currently, only AArch64 has customized
implementation of isExtFreeImpl() to check if EXT can be folded into its use.
Differential Revision: https://reviews.llvm.org/D34458
llvm-svn: 308076
When we fail to sink an instruction, we must make sure not to modify
the function; otherwise, we end up in an infinite loop because
CodeGenPrepare iterates until it doesn't make any changes.
Fixes https://bugs.llvm.org/show_bug.cgi?id=33608 .
llvm-svn: 307866
CodeGenPrepare::optimizeMemoryInst contains a check that we do nothing
if all instructions combining the address for memory instruction is in the same
block as memory instruction itself.
However if any of these instruction are placed after memory instruction then
address calculation will not be folded to memory instruction.
The added test case shows an example.
Reviewers: loladiro, spatel, efriedma
Reviewed By: efriedma
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34862
llvm-svn: 307628
Summary:
Arguably non-integral pointers probably shouldn't show up here at all,
but since the backend doesn't complain and this takes valid (according
to the Verifier) IR and makes it invalid, make sure not to introduce
any inttoptr instructions if we're dealing with non-integral pointers.
Reviewed By: sanjoy
Differential Revision: https://reviews.llvm.org/D33110
llvm-svn: 306737
As noted in D34071, there are some IR optimization opportunities that could be
handled by normal IR passes if this expansion wasn't happening so late in CGP.
Regardless of that, it seems wasteful to knowingly produce suboptimal IR here,
so I'm proposing this change:
%s = sub i32 %x, %y
%r = icmp ne %s, 0
=>
%r = icmp ne %x, %y
Changing the predicate to 'eq' mimics what InstCombine would do, so that's just
an efficiency improvement if we decide this expansion should happen sooner.
The fact that the PowerPC backend doesn't eliminate the 'subf.' might be
something for PPC folks to investigate separately.
Differential Revision: https://reviews.llvm.org/D34416
llvm-svn: 306471
This is the last step needed to avoid regressions for x86 before we flip the switch to allow
expansion of the smallest set of memcpy() via CGP. The DAG version checks for constant strings,
so we need to do that here too.
FWIW, the 2 constant test is not handled by LibCallSimplifier::optimizeMemCmp() because that
code is limited to 8-bit constant arrays. LibCallSimplifier will also fail to optimize some 1
constant tests because its alignment requirements are too strict (shouldn't require alignment
for a constant operand).
Differential Revision: https://reviews.llvm.org/D34071
llvm-svn: 305734
Summary:
At present, `-profile-guided-section-prefix` is a `cl::Optional` option, which means it demands to be passed exactly zero or one times. Our build system makes it pretty tricky to guarantee this. We often accidentally pass the flag more than once (but always with the same "false" value) which results in an error, after which compilation fails:
```
clang (LLVM option parsing): for the -profile-guided-section-prefix option: may only occur zero or one times!
```
While we work on improving our build system, it also seems reasonable just to allow `-profile-guided-section-prefix` to be passed more than once, by to `cl::ZeroOrMore`. Quoting [[ http://llvm.org/docs/CommandLine.html#controlling-the-number-of-occurrences-required-and-allowed | the documentation ]]:
> The cl::ZeroOrMore modifier ... indicates that your program will allow the option to be specified zero or more times.
> ...
> If an option is specified multiple times for an option of the cl::opt class, only the last value will be retained.
Reviewers: danielcdh
Reviewed By: danielcdh
Subscribers: twoh, david2050, llvm-commits
Differential Revision: https://reviews.llvm.org/D34219
llvm-svn: 305413
We're currently passing endian-ness around as a param (and not uniformly),
so this eliminates the need for that. I'd like to add a constant fold
call too, and that requires a DL.
llvm-svn: 305129
The test diff for PowerPC shows we can better optimize if this case is one block.
For x86, there's would be a substantial difference if CGP expansion was enabled because branches are assumed
cheap and SDAG can't optimize across blocks.
Instead of this:
_cmp_eq8:
movq (%rdi), %rax
cmpq (%rsi), %rax
je LBB23_1
## BB#2: ## %res_block
movl $1, %ecx
jmp LBB23_3
LBB23_1:
xorl %ecx, %ecx
LBB23_3: ## %endblock
xorl %eax, %eax
testl %ecx, %ecx
sete %al
retq
We get this:
cmp_eq8:
movq (%rdi), %rcx
xorl %eax, %eax
cmpq (%rsi), %rcx
sete %al
retq
And that matches the optimal codegen that we get from the current expansion in SelectionDAGBuilder::visitMemCmpCall().
If this looks right, then I just need to confirm that vector-sized expansion will work from here, and we can enable
CGP memcmp() expansion for x86. Ie, we'll bypass the power-of-2 special cases currently optimized in SDAG because we
can lower the IR produced here optimally.
Differential Revision: https://reviews.llvm.org/D34005
llvm-svn: 304987
This could be viewed as another shortcoming of the DAGCombiner:
when both operands of a compare are zexted from the same source
type, we should be able to compare the original types.
The effect on PowerPC perf is likely unnoticeable, but there's a
visible regression for x86 if we feed the suboptimal IR for memcmp
expansion to the DAG:
_cmp_eq4_zexted_to_i64:
movl (%rdi), %ecx
movl (%rsi), %edx
xorl %eax, %eax
cmpq %rdx, %rcx
sete %al
_cmp_eq4_better:
movl (%rdi), %ecx
xorl %eax, %eax
cmpl (%rsi), %ecx
sete %al
llvm-svn: 304923
In the special (but also the likely common) case, we can avoid
the multi-block complexity of the general algorithm, so moving
this part off on its own will make it re-usable.
llvm-svn: 304908
I'd like to enable CGP memcmp expansion for x86, but the output from CGP would regress the
special cases (memcmp(x,y,N) != 0 for N=1,2,4,8,16,32 bytes) that we already handle.
I'm not sure if we'll actually be able to produce the optimal code given the block-at-a-time
limitation in the DAG. We might have to just avoid those special-cases here in CGP. But
regardless of that, I think this is a win for the more general cases.
http://rise4fun.com/Alive/cbQ
Differential Revision: https://reviews.llvm.org/D33963
llvm-svn: 304849
I did this a long time ago with a janky python script, but now
clang-format has built-in support for this. I fed clang-format every
line with a #include and let it re-sort things according to the precise
LLVM rules for include ordering baked into clang-format these days.
I've reverted a number of files where the results of sorting includes
isn't healthy. Either places where we have legacy code relying on
particular include ordering (where possible, I'll fix these separately)
or where we have particular formatting around #include lines that
I didn't want to disturb in this patch.
This patch is *entirely* mechanical. If you get merge conflicts or
anything, just ignore the changes in this patch and run clang-format
over your #include lines in the files.
Sorry for any noise here, but it is important to keep these things
stable. I was seeing an increasing number of patches with irrelevant
re-ordering of #include lines because clang-format was used. This patch
at least isolates that churn, makes it easy to skip when resolving
conflicts, and gets us to a clean baseline (again).
llvm-svn: 304787
This patch does an inline expansion of memcmp.
It changes the memcmp library call into an inline expansion when the size is
known at compile time and is under a target specified threshold.
This expansion is implemented in CodeGenPrepare and expands into straight line
code. The target specifies a maximum load size and the expansion works by using
this size to load the two sources, compare, and exit early if a difference is
found. It also has a special case when the memcmp result is used in a compare
to zero equality.
Differential Revision: https://reviews.llvm.org/D28637
llvm-svn: 304313
Rename the DEBUG_TYPE to match the names of corresponding passes where
it makes sense. Also establish the pattern of simply referencing
DEBUG_TYPE instead of repeating the passname where possible.
llvm-svn: 303921
Summary:
Implements PR889
Removing the virtual table pointer from Value saves 1% of RSS when doing
LTO of llc on Linux. The impact on time was positive, but too noisy to
conclusively say that performance improved. Here is a link to the
spreadsheet with the original data:
https://docs.google.com/spreadsheets/d/1F4FHir0qYnV0MEp2sYYp_BuvnJgWlWPhWOwZ6LbW7W4/edit?usp=sharing
This change makes it invalid to directly delete a Value, User, or
Instruction pointer. Instead, such code can be rewritten to a null check
and a call Value::deleteValue(). Value objects tend to have their
lifetimes managed through iplist, so for the most part, this isn't a big
deal. However, there are some places where LLVM deletes values, and
those places had to be migrated to deleteValue. I have also created
llvm::unique_value, which has a custom deleter, so it can be used in
place of std::unique_ptr<Value>.
I had to add the "DerivedUser" Deleter escape hatch for MemorySSA, which
derives from User outside of lib/IR. Code in IR cannot include MemorySSA
headers or call the MemoryAccess object destructors without introducing
a circular dependency, so we need some level of indirection.
Unfortunately, no class derived from User may have any virtual methods,
because adding a virtual method would break User::getHungOffOperands(),
which assumes that it can find the use list immediately prior to the
User object. I've added a static_assert to the appropriate OperandTraits
templates to help people avoid this trap.
Reviewers: chandlerc, mehdi_amini, pete, dberlin, george.burgess.iv
Reviewed By: chandlerc
Subscribers: krytarowski, eraman, george.burgess.iv, mzolotukhin, Prazek, nlewycky, hans, inglorion, pcc, tejohnson, dberlin, llvm-commits
Differential Revision: https://reviews.llvm.org/D31261
llvm-svn: 303362
This provides a new way to access the TargetMachine through
TargetPassConfig, as a dependency.
The patterns replaced here are:
* Passes handling a null TargetMachine call
`getAnalysisIfAvailable<TargetPassConfig>`.
* Passes not handling a null TargetMachine
`addRequired<TargetPassConfig>` and call
`getAnalysis<TargetPassConfig>`.
* MachineFunctionPasses now use MF.getTarget().
* Remove all the TargetMachine constructors.
* Remove INITIALIZE_TM_PASS.
This fixes a crash when running `llc -start-before prologepilog`.
PEI needs StackProtector, which gets constructed without a TargetMachine
by the pass manager. The StackProtector pass doesn't handle the case
where there is no TargetMachine, so it segfaults.
Related to PR30324.
Differential Revision: https://reviews.llvm.org/D33222
llvm-svn: 303360
Currently, when masked load, store, gather or scatter intrinsics are used, we check in CodeGenPrepare pass if the subtarget support these intrinsics, if not we replace them with scalar code - this is a functional transformation not an optimization (not optional).
CodeGenPrepare pass does not run when the optimization level is set to CodeGenOpt::None (-O0).
Functional transformation should run with all optimization levels, so here I created a new pass which runs on all optimization levels and does no more than this transformation.
Differential Revision: https://reviews.llvm.org/D32487
llvm-svn: 303050
Summary:
r284533 added hot and cold section prefixes based on profile
information, to enable grouping of hot/cold functions at link time.
However, it used "cold" as the prefix for cold sections, but gold only
recognizes "unlikely" (which is used by gcc for cold sections).
Therefore, cold sections were not properly being grouped. Switch to
using "unlikely"
Reviewers: danielcdh, davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32983
llvm-svn: 302502
Commits were:
"Use WeakVH instead of WeakTrackingVH in AliasSetTracker's UnkownInsts"
"Add a new WeakVH value handle; NFC"
"Rename WeakVH to WeakTrackingVH; NFC"
The changes assumed pointers are 8 byte aligned on all architectures.
llvm-svn: 301429
Summary:
I plan to use WeakVH to mean "nulls itself out on deletion, but does
not track RAUW" in a subsequent commit.
Reviewers: dblaikie, davide
Reviewed By: davide
Subscribers: arsenm, mehdi_amini, mcrosier, mzolotukhin, jfb, llvm-commits, nhaehnle
Differential Revision: https://reviews.llvm.org/D32266
llvm-svn: 301424
This patch uses lshrInPlace to replace code where the object that lshr is called on is being overwritten with the result.
This adds an lshrInPlace(const APInt &) version as well.
Differential Revision: https://reviews.llvm.org/D32155
llvm-svn: 300566
The splitIndirectCriticalEdges function generates and invalid CFG when the
'Target' basic block is a loop to itself. When this occurs, the code that
updates the predecessor terminator needs to update the terminator in the split
basic block.
This occurs when there is an edge from block D back to D. Since D is split in
to D0 and D1, the code needs to update the terminator in D1. But D1 is not in
the OtherPreds vector, so it was not getting updated.
Differential Revision: https://reviews.llvm.org/D32126
llvm-svn: 300480
and to expose a handle to represent the actual case rather than having
the iterator return a reference to itself.
All of this allows the iterator to be used with common STL facilities,
standard algorithms, etc.
Doing this exposed some missing facilities in the iterator facade that
I've fixed and required some work to the actual iterator to fully
support the necessary API.
Differential Revision: https://reviews.llvm.org/D31548
llvm-svn: 300032
The new codepath has been in the tree for years, and there isn't any
reason to use two codepaths here.
Differential Revision: https://reviews.llvm.org/D30596
llvm-svn: 299723
Summary:
Move the aarch64-type-promotion pass within the existing type promotion framework in CGP.
This change also support forking sexts when a new sext is required for promotion.
Note that change is based on D27853 and I am submitting this out early to provide a better idea on D27853.
Reviewers: jmolloy, mcrosier, javed.absar, qcolombet
Reviewed By: qcolombet
Subscribers: llvm-commits, aemerson, rengolin, mcrosier
Differential Revision: https://reviews.llvm.org/D28680
llvm-svn: 299379
This moves the isMask and isShiftedMask functions to be class methods. They now use the MathExtras.h function for single word size and leading/trailing zeros/ones or countPopulation for the multiword size. The previous implementation made multiple temorary memory allocations to do the bitwise arithmetic operations to match the MathExtras.h implementation.
Differential Revision: https://reviews.llvm.org/D31565
llvm-svn: 299362
Summary: The current prefix based function layout algorithm only looks at function's entry count, which is not sufficient. A function should be grouped together if its entry count or any call edge count is hot.
Reviewers: davidxl, eraman
Reviewed By: eraman
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D31225
llvm-svn: 298656
Summary:
This class is a list of AttributeSetNodes corresponding the function
prototype of a call or function declaration. This class used to be
called ParamAttrListPtr, then AttrListPtr, then AttributeSet. It is
typically accessed by parameter and return value index, so
"AttributeList" seems like a more intuitive name.
Rename AttributeSetImpl to AttributeListImpl to follow suit.
It's useful to rename this class so that we can rename AttributeSetNode
to AttributeSet later. AttributeSet is the set of attributes that apply
to a single function, argument, or return value.
Reviewers: sanjoy, javed.absar, chandlerc, pete
Reviewed By: pete
Subscribers: pete, jholewinski, arsenm, dschuff, mehdi_amini, jfb, nhaehnle, sbc100, void, llvm-commits
Differential Revision: https://reviews.llvm.org/D31102
llvm-svn: 298393
Summary:
Instead of just looking for a load which is mergable with Ext to form ExtLoad, trying to promote Exts as long as the cost is acceptable. This change is not a NFC as it continue promoting Exts even after finding a load during promotions; the change in arm64-codegen-prepare-extload.ll described in 2.b might show the case.
This change was motivated from D26524. Based on this change, I will move the transformation performed in aarch64-type-promotion into CGP.
Reviewers: jmolloy, qcolombet, mcrosier, javed.absar
Reviewed By: qcolombet
Subscribers: rengolin, llvm-commits, aemerson
Differential Revision: https://reviews.llvm.org/D27853
llvm-svn: 298114
Splitting critical edges when one of the source edges is an indirectbr
is hard in general (because it requires changing the memory the indirectbr
reads). But if a block only has a single indirectbr predecessor (which is
the common case), we can simulate splitting that edge by splitting
the destination block, and retargeting the *direct* branches.
This is motivated by the use of computed gotos in python 2.7: PyEval_EvalFrame()
ends up using an indirect branch with ~100 successors, and passing a constant to
each of those. Since MachineSink can't break indirect critical edges on demand
(and doing this in MIR doesn't look feasible), this causes us to emit about ~100
defs of registers containing constants, which we in the predecessor block, where
only one of those constants is used in each successor. So, at each computed goto,
we needlessly spill about a 100 constants to stack. The end result is that a
clang-compiled python interpreter can be about ~2.5x slower on a simple python
reduction loop than a gcc-compiled interpreter.
Differential Revision: https://reviews.llvm.org/D29916
llvm-svn: 296416
When we construct addressing modes, we use isNoopAddrSpaceCast to ignore
addrspacecast instructions. Make sure we insert the correct addrspacecast
when we reconstruct the addressing mode.
Differential Revision: https://reviews.llvm.org/D30114
llvm-svn: 296167
Splitting critical edges when one of the source edges is an indirectbr
is hard in general (because it requires changing the memory the indirectbr
reads). But if a block only has a single indirectbr predecessor (which is
the common case), we can simulate splitting that edge by splitting
the destination block, and retargeting the *direct* branches.
This is motivated by the use of computed gotos in python 2.7: PyEval_EvalFrame()
ends up using an indirect branch with ~100 successors, and passing a constant to
each of those. Since MachineSink can't break indirect critical edges on demand
(and doing this in MIR doesn't look feasible), this causes us to emit about ~100
defs of registers containing constants, which we in the predecessor block, where
only one of those constants is used in each successor. So, at each computed goto,
we needlessly spill about a 100 constants to stack. The end result is that a
clang-compiled python interpreter can be about ~2.5x slower on a simple python
reduction loop than a gcc-compiled interpreter.
Differential Revision: https://reviews.llvm.org/D29916
llvm-svn: 296149
Splitting critical edges when one of the source edges is an indirectbr
is hard in general (because it requires changing the memory the indirectbr
reads). But if a block only has a single indirectbr predecessor (which is
the common case), we can simulate splitting that edge by splitting
the destination block, and retargeting the *direct* branches.
This is motivated by the use of computed gotos in python 2.7: PyEval_EvalFrame()
ends up using an indirect branch with ~100 successors, and passing a constant to
each of those. Since MachineSink can't break indirect critical edges on demand
(and doing this in MIR doesn't look feasible), this causes us to emit about ~100
defs of registers containing constants, which we in the predecessor block, where
only one of those constants is used in each successor. So, at each computed goto,
we needlessly spill about a 100 constants to stack. The end result is that a
clang-compiled python interpreter can be about ~2.5x slower on a simple python
reduction loop than a gcc-compiled interpreter.
Differential Revision: https://reviews.llvm.org/D29916
llvm-svn: 296060
Summary:
Rework the code that was sinking/duplicating (icmp and, 0) sequences
into blocks where they were being used by conditional branches to form
more tbz instructions on AArch64. The new code is more general in that
it just looks for 'and's that have all icmp 0's as users, with a target
hook used to select which subset of 'and' instructions to consider.
This change also enables 'and' sinking for X86, where it is more widely
beneficial than on AArch64.
The 'and' sinking/duplicating code is moved into the optimizeInst phase
of CodeGenPrepare, where it can take advantage of the fact the
OptimizeCmpExpression has already sunk/duplicated any icmps into the
blocks where they are used. One minor complication from this change is
that optimizeLoadExt needed to be updated to always mark 'and's it has
determined should be in the same block as their feeding load in the
InsertedInsts set to avoid an infinite loop of hoisting and sinking the
same 'and'.
This change fixes a regression on X86 in the tsan runtime caused by
moving GVNHoist to a later place in the optimization pipeline (see
PR31382).
Reviewers: t.p.northover, qcolombet, MatzeB
Subscribers: aemerson, mcrosier, sebpop, llvm-commits
Differential Revision: https://reviews.llvm.org/D28813
llvm-svn: 295746
Summary: This change prevent the signed value of cost from being negative as the value is passed as an unsigned argument.
Reviewers: mcrosier, jmolloy, qcolombet, javed.absar
Reviewed By: mcrosier, qcolombet
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28871
llvm-svn: 293307
This is a succeeding patch of https://reviews.llvm.org/D22840 to address the
issue when a value to be merged into an int64 pair is in a different BB. Redoing
the store splitting in CodeGenPrepare so we can match the pattern across multiple
BBs and move some instructions into the same BB. We still keep the code in dag
combine so that we can catch cases that show up after DAG combining runs.
Differential Revision: https://reviews.llvm.org/D25914
llvm-svn: 290365
We're currently doing nearly the same thing for @llvm.objectsize in
three different places: two of them are missing checks for overflow,
and one of them could subtly break if InstCombine gets much smarter
about removing alloc sites. Seems like a good idea to not do that.
llvm-svn: 290214
This is recommit of r287553 after fixing the invalid loop info after eliminating an empty block and unit test failures in AVR and WebAssembly :
Summary: Merging an empty case block into the header block of switch could cause ISel to add COPY instructions in the header of switch, instead of the case block, if the case block is used as an incoming block of a PHI. This could potentially increase dynamic instructions, especially when the switch is in a loop. I added a test case which was reduced from the benchmark I was targetting.
Reviewers: t.p.northover, mcrosier, manmanren, wmi, joerg, davidxl
Subscribers: joerg, qcolombet, danielcdh, hfinkel, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D22696
llvm-svn: 289988
`dropUnknownNonDebugMetadata` takes a list of "known" metadata IDs. The
only reason it worked at all is that `getMetadataID` returns something
unrelated -- it returns the subclass ID of the receiver (which is used
in `dyn_cast` etc.). That does not numerically match
`LLVMContext::MD_invariant_group` and ends up dropping `invariant_group`
along with every other metadata that does not numerically match
`LLVMContext::MD_invariant_group`.
llvm-svn: 289973
This is recommit of r287553 after fixing the invalid loop info after eliminating an empty block:
Summary: Merging an empty case block into the header block of switch could cause ISel to add COPY instructions in the header of switch, instead of the case block, if the case block is used as an incoming block of a PHI. This could potentially increase dynamic instructions, especially when the switch is in a loop. I added a test case which was reduced from the benchmark I was targetting.
Reviewers: t.p.northover, mcrosier, manmanren, wmi, joerg, davidxl
Subscribers: joerg, qcolombet, danielcdh, hfinkel, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D22696
llvm-svn: 289951
Instead, expose whether the current type is an array or a struct, if an array
what the upper bound is, and if a struct the struct type itself. This is
in preparation for a later change which will make PointerType derive from
Type rather than SequentialType.
Differential Revision: https://reviews.llvm.org/D26594
llvm-svn: 288458
Summary:
Previously, CGP would unconditionally sink addrspacecast instructions,
even going so far as to sink them into a loop.
Now we check that the cast is "cheap", as defined by TLI.
We introduce a new "is-cheap" function to TLI rather than using
isNopAddrSpaceCast because some GPU platforms want the ability to ask
for non-nop casts to be sunk.
Reviewers: arsenm, tra
Subscribers: jholewinski, wdng, llvm-commits
Differential Revision: https://reviews.llvm.org/D26923
llvm-svn: 287591
Summary: Merging an empty case block into the header block of switch could cause
ISel to add COPY instructions in the header of switch, instead of the case
block, if the case block is used as an incoming block of a PHI. This could
potentially increase dynamic instructions, especially when the switch is in a
loop. I added a test case which was reduced from the benchmark I was targetting.
Reviewers: t.p.northover, mcrosier, manmanren, wmi, davidxl
Subscribers: qcolombet, danielcdh, hfinkel, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D22696
llvm-svn: 287553
of which that is hidden inside a separate function call) and helpfully
before building expensive transaction infrastructure. This will avoid
crashing when running CGP in a generic mode if we ever managed to hit
this case.
Note that I spent some time looking at alternatives. CGP is actually
used without a TM or TLI in order to do some target-independent testing.
Further, all of the neighboring optimization techniques actually have
some paths that are effective even in the absence of TLI so this seemed
the correct scope at which to check and bypass logic. It still isn't
clear that long-term support for missing TM/TLI is the right
cost/benefit tradeoff for CGP -- we seem to get relatively little for it
and the code is just littered with checks (and assumptions which
I suspect are still missing some checks).
This at least fixes the potential bug in this code spotted by
PVS-Studio, so we've got that going for us. ;]
llvm-svn: 285987
Summary:
The original implementation is in r261607, which was reverted in r269726 to accomendate the ProfileSummaryInfo analysis pass. The new implementation:
1. add a new metadata for function section prefix
2. query against ProfileSummaryInfo in CGP to set the correct section prefix for each function
3. output the section prefix set by CGP
Reviewers: davidxl, eraman
Subscribers: vsk, llvm-commits
Differential Revision: https://reviews.llvm.org/D24989
llvm-svn: 284533
CodeGenPrepare knows how to move a zext of a load into the same basic block
where the load lives. The goal is to help ISel match a zero-extending load
instead of two separated instructions.
CGP attempts to move a zext computation even if it lives in a basic block that
does not post-dominate the load's basic block. That means, the hoisted zext may
be speculated. Preserving the zext location would hurt the debugging experience
and the quality of sample pgo.
With this patch, when moving a zext near to its associated load, CGP no longer
propagates the zext's debug location. Instead, CGP conservatively reuses the
same debug location for the load and the zext.
An alternative approach would be to assign an artificial line-0 location to the
zext. However we don't want to over-use the 'line-0' for this particular case
because it would have a size cost in the line-table section for no additional
benefit.
Differential Revision: https://reviews.llvm.org/D25611
llvm-svn: 284377
Summary: If consecutive select instructions are lowered separately in CGP, it will introduce redundant condition check and branches that cannot be removed by later optimization phases. This patch lowers all consecutive select instructions at the same to to avoid inefficent code as demonstrated in https://llvm.org/bugs/show_bug.cgi?id=29095
Reviewers: davidxl
Subscribers: vsk, llvm-commits
Differential Revision: https://reviews.llvm.org/D24147
llvm-svn: 281252
CGP tail-duplicates rets into blocks that end with a call that feed the ret.
This puts the call in tail position, potentially allowing the DAG builder to
lower it as a tail call. To avoid tail duplication in cases where we won't
form the tail call, CGP tried to predict whether this is going to be possible,
and avoids doing it when lowering as a tail call will definitely fail.
However, it was being too conservative by always throwing away calls to
functions with a signext/zeroext attribute on the return type.
Instead, we can use the same logic the builder uses to determine whether the
attributes work out.
Differential Revision: https://reviews.llvm.org/D24315
llvm-svn: 280894
CGP currently drops select's MD_prof profile data when
generating conditional branch which can lead to bad
code layout. The patch fixes the issue.
Differential Revision: http://reviews.llvm.org/D24169
llvm-svn: 280600
Elsewhere (particularly computeKnownBits) we assume that a global will be
aligned to the value returned by Value::getPointerAlignment. This is used to
boost the alignment on memcpy/memset, so any target-specific request can only
increase that value.
llvm-svn: 275866
Also, rename recognizeBitReverseOrBSwapIdiom to recognizeBSwapOrBitReverseIdiom,
so the ordering of the MatchBSwaps and MatchBitReversals arguments are
consistent with the function name.
llvm-svn: 270715
The sink cast machinery is supposed to sink casts as close to their user
as possible. However, an EH pad is the first instruction in it's basic
block. Don't sink if the user is an EH pad.
This fixes PR27536.
llvm-svn: 267767
This is part of solving PR27344:
https://llvm.org/bugs/show_bug.cgi?id=27344
CGP should undo the SimplifyCFG transform for the same reason that earlier patches have used this
same mechanism: it's possible that passes between SimplifyCFG and CGP may be able to optimize the
IR further with a select in place.
For the TLI hook default, >99% taken or not taken is chosen as the default threshold for a highly
predictable branch. Even the most limited HW branch predictors will be correct on this branch almost
all the time, so even a massive mispredict penalty perf loss would be overcome by the win from all
the times the branch was predicted correctly.
As a follow-up, we could make the default target hook less conservative by using the SchedMachineModel's
MispredictPenalty. Or we could just let targets override the default by implementing the hook with that
and other target-specific options. Note that trying to statically determine mispredict rates for
close-to-balanced profile weight data is generally impossible if the HW is sufficiently advanced. Ie,
50/50 taken/not-taken might still be 100% predictable.
Finally, note that this patch as-is will not solve PR27344 because the current __builtin_unpredictable()
branch weight default values are 4 and 64. A proposal to change that is in D19435.
Differential Revision: http://reviews.llvm.org/D19488
llvm-svn: 267572
The original commit was reverted because of a buildbot problem with LazyCallGraph::SCC handling (not related to the OptBisect handling).
Differential Revision: http://reviews.llvm.org/D19172
llvm-svn: 267231
This patch implements a optimization bisect feature, which will allow optimizations to be selectively disabled at compile time in order to track down test failures that are caused by incorrect optimizations.
The bisection is enabled using a new command line option (-opt-bisect-limit). Individual passes that may be skipped call the OptBisect object (via an LLVMContext) to see if they should be skipped based on the bisect limit. A finer level of control (disabling individual transformations) can be managed through an addition OptBisect method, but this is not yet used.
The skip checking in this implementation is based on (and replaces) the skipOptnoneFunction check. Where that check was being called, a new call has been inserted in its place which checks the bisect limit and the optnone attribute. A new function call has been added for module and SCC passes that behaves in a similar way.
Differential Revision: http://reviews.llvm.org/D19172
llvm-svn: 267022
This patch fixes calculating of builtin_object_size if it depends on a
condition. Before this patch compiler did not know how to calculate the
object size when it finds a condition that cannot be eliminated.
This patch enables calculating of builtin_object_size even in case when
condition cannot be eliminated by choosing minimum or maximum value as a
result from condition. Choosing minimum or maximum value from condition
is based on the second argument of __builtin_object_size function.
Patch by Strahinja Petrovic.
Differential Revision: http://reviews.llvm.org/D18438
llvm-svn: 266193
Presently, CodeGenPrepare deletes all nearly empty (only phi and branch)
basic blocks. This pass can delete loop preheaders which frequently creates
critical edges. A preheader can be a convenient place to spill registers to
the stack. If the entrance to a loop body is a critical edge, then spills
may occur in the loop body rather than immediately before it. This patch
protects loop preheaders from deletion in CodeGenPrepare even if they are
nearly empty.
Since the patch alters the CFG, it affects a large number of test cases.
In most cases, the changes are merely cosmetic (basic blocks have different
names or instruction orders change slightly). I am somewhat concerned about
the test/CodeGen/Mips/brdelayslot.ll test case. If the loop preheader is not
deleted, then the MIPS backend does not take advantage of a branch delay
slot. Consequently, I would like some close review by a MIPS expert.
The patch also partially subsumes D16893 from George Burgess IV. George
correctly notes that CodeGenPrepare does not actually preserve the dominator
tree. I think the dominator tree was usually not valid when CodeGenPrepare
ran, but I am using LoopInfo to mark preheaders, so the dominator tree is
now always valid before CodeGenPrepare.
Author: Tom Jablin (tjablin)
Reviewers: hfinkel george.burgess.iv vkalintiris dsanders kbarton cycheng
http://reviews.llvm.org/D16984
llvm-svn: 265397
Sinking comparisons in CGP can undo the job of hoisting them done
earlier by LICM, and soft-FP makes this an expensive mistake.
A common pattern that produces floating point comparisons uniform
over a loop is an explicit check for division by zero. If the divisor
is hoisted out of the loop, the comparison can also be, but hoisting
the function that unwinds is never legal, since it may cause side
effects in the loop body prior to the unwinding to not be executed.
Differential Revision: http://reviews.llvm.org/D18744
llvm-svn: 265264
CGP modifies the domtree in some cases, so saying that it preserves the
domtree is a lie. We'll be able to selectively preserve it with the new
pass manager.
Differential Revision: http://reviews.llvm.org/D16893
llvm-svn: 264099
This patch teaches CGP to duplicate addressing mode computations into cold paths (detected via explicit cold attribute on calls) if required to let addressing mode be safely sunk into the basic block containing each load and store.
In general, duplicating code into cold blocks may result in code growth, but should not effect performance. In this case, it's better to duplicate some code than to put extra pressure on the register allocator by making it keep the address through the entirely of the fast path.
This patch only handles addressing computations, but in principal, we could implement a more general cold cold scheduling heuristic which tries to reduce register pressure in the fast path by duplicating code into the cold path. Getting the profitability of the general case right seemed likely to be challenging, so I stuck to the existing case (addressing computation) we already had.
Differential Revision: http://reviews.llvm.org/D17652
llvm-svn: 263074
Summary:
Both the hardware and LLVM have changed since 2012.
Now, load-based heuristic don't show big differences any more on OoO cores.
There is no notable regressons and improvements on spec2000/2006. (Cortex-A57, Core i5).
Reviewers: spatel, zansari
Differential Revision: http://reviews.llvm.org/D16836
llvm-svn: 261809
Stop using `getNodePtrUnchecked()` when building IR. Eventually a
dereference will be required to get at the downcast node, since the
iterator will only store an `ilist_node_base` of some sort.
This should have no functionality change for now, but is a path towards
removing some more UB from ilist.
llvm-svn: 261495
`ilist_iterator<NodeTy>::getNodePtrUnchecked()` is documented as being
for internal use only, but CodeGenPrepare was using it anyway. This
code relies on pulling out the `Value*` pointer even after the lifetime
of the iterator is over. But having this pointer available in
ilist_iterator depends on UB in the first place.
Instead, safely pull out the `Value*` when the iterator is alive and
stop using the internal-only API.
There should be no functionality change here.
llvm-svn: 261493
platforms.
With ELF, the alignment of a global variable in a shared library will
get copied into an executables linked against it, if the executable even
accesss the variable. So, it's not possible to implicitly increase
alignment based on access patterns, or you'll break existing binaries.
This happened to affect libc++'s std::cout symbol, for example. See
thread: http://thread.gmane.org/gmane.comp.compilers.clang.devel/45311
(This is a re-commit of r257719, without the bug reported in
PR26144. I've tweaked the code to not assert-fail in
enforceKnownAlignment when computeKnownBits doesn't recurse far enough
to find the underlying Alloca/GlobalObject value.)
Differential Revision: http://reviews.llvm.org/D16145
llvm-svn: 257902
There are several requirements that ended up with this design;
1. Matching bitreversals is too heavyweight for InstCombine and doesn't really need to be done so early.
2. Bitreversals and byteswaps are very related in their matching logic.
3. We want to implement support for matching more advanced bswap/bitreverse patterns like partial bswaps/bitreverses.
4. Bswaps are best matched early in InstCombine.
The result of these is that a new utility function is created in Transforms/Utils/Local.h that can be configured to search for bswaps, bitreverses or both. InstCombine uses it to find only bswaps, CGP uses it to find only bitreversals.
We can then extend the matching logic in one place only.
llvm-svn: 257875
platforms.
With ELF, the alignment of a global variable in a shared library will
get copied into an executables linked against it, if the executable even
accesss the variable. So, it's not possible to implicitly increase
alignment based on access patterns, or you'll break existing binaries.
This happened to affect libc++'s std::cout symbol, for example. See
thread: http://thread.gmane.org/gmane.comp.compilers.clang.devel/45311
llvm-svn: 257719
Summary:
This commit renames GCRelocateOperands to GCRelocateInst and makes it an
intrinsic wrapper, similar to e.g. MemCpyInst. Also, all users of
GCRelocateOperands were changed to use the new intrinsic wrapper instead.
Reviewers: sanjoy, reames
Subscribers: reames, sanjoy, llvm-commits
Differential Revision: http://reviews.llvm.org/D15762
llvm-svn: 256811
Update some comments to be more explicit.
Change bypassSlowDivision and the functions it calls so that they take
BasicBlock*s and Instruction*s, rather than Function::iterator&s and
BasicBlock::iterator&s.
Change the APIs so that the caller is responsible for updating the
iterator, rather than the callee. This makes control flow much easier
to follow.
Patch by Justin Lebar!
llvm-svn: 256789
Summary:
Add and instructions immediately after loads that only have their low
bits used, assuming that the (and (load x) c) will be matched as a
extload and the ands/truncs fed by the extload will be removed by isel.
Reviewers: mcrosier, qcolombet, ab
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D14584
llvm-svn: 253722
This is another step towards allowing SimplifyCFG to speculate harder, but then have
CGP clean things up if the target doesn't like it.
Previous patches in this series:
http://reviews.llvm.org/D12882http://reviews.llvm.org/D13297
D13297 should catch most expensive ops, but speculation of cttz/ctlz requires special
handling because of weirdness in the intrinsic definition for handling a zero input
(that definition can probably be blamed on x86).
For example, if we have the usual speculated-by-select expensive op pattern like this:
%tobool = icmp eq i64 %A, 0
%0 = tail call i64 @llvm.cttz.i64(i64 %A, i1 true) ; is_zero_undef == true
%cond = select i1 %tobool, i64 64, i64 %0
ret i64 %cond
There's an instcombine that will turn it into:
%0 = tail call i64 @llvm.cttz.i64(i64 %A, i1 false) ; is_zero_undef == false
This CGP patch is looking for that case and despeculating it back into:
entry:
%tobool = icmp eq i64 %A, 0
br i1 %tobool, label %cond.end, label %cond.true
cond.true:
%0 = tail call i64 @llvm.cttz.i64(i64 %A, i1 true) ; is_zero_undef == true
br label %cond.end
cond.end:
%cond = phi i64 [ %0, %cond.true ], [ 64, %entry ]
ret i64 %cond
This unfortunately may lead to poorer codegen (see the changes in the existing x86 test),
but if we increase speculation in SimplifyCFG (the next step in this patch series), then
we should avoid those kinds of cases in the first place.
The need for this patch was originally mentioned here:
http://reviews.llvm.org/D7506
with follow-up here:
http://reviews.llvm.org/D7554
Differential Revision: http://reviews.llvm.org/D14630
llvm-svn: 253573
Note, this was reviewed (and more details are in) http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
These intrinsics currently have an explicit alignment argument which is
required to be a constant integer. It represents the alignment of the
source and dest, and so must be the minimum of those.
This change allows source and dest to each have their own alignments
by using the alignment attribute on their arguments. The alignment
argument itself is removed.
There are a few places in the code for which the code needs to be
checked by an expert as to whether using only src/dest alignment is
safe. For those places, they currently take the minimum of src/dest
alignments which matches the current behaviour.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 8, i1 false)
will now read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 8 %dest, i8* align 8 %src, i32 500, i1 false)
For out of tree owners, I was able to strip alignment from calls using sed by replacing:
(call.*llvm\.memset.*)i32\ [0-9]*\,\ i1 false\)
with:
$1i1 false)
and similarly for memmove and memcpy.
I then added back in alignment to test cases which needed it.
A similar commit will be made to clang which actually has many differences in alignment as now
IRBuilder can generate different source/dest alignments on calls.
In IRBuilder itself, a new argument was added. Instead of calling:
CreateMemCpy(Dst, Src, getInt64(Size), DstAlign, /* isVolatile */ false)
you now call
CreateMemCpy(Dst, Src, getInt64(Size), DstAlign, SrcAlign, /* isVolatile */ false)
There is a temporary class (IntegerAlignment) which takes the source alignment and rejects
implicit conversion from bool. This is to prevent isVolatile here from passing its default
parameter to the source alignment.
Note, changes in future can now be made to codegen. I didn't change anything here, but this
change should enable better memcpy code sequences.
Reviewed by Hal Finkel.
llvm-svn: 253511
This is a redo of r251849 except the tests have been split into arch-specific folders
to hopefully make the bots happy.
This is a follow-up from the discussion in D12965. The block-at-a-time limitation of
SelectionDAG also came up in D13297.
Without the InstCombine change from D12965, I don't expect this patch to make any
difference in the real world because InstCombine does not shrink cases like this in
visitSwitchInst(). But we need to have this CGP safety harness in place before
proceeding with any shrinkage in D12965, so we won't generate extra extends for compares.
I've opted for IR regression tests in the patch because that seems like a clearer way to
test the transform, but PowerPC CodeGen for an i16 widening test is shown below. x86
will need more work to solve: https://llvm.org/bugs/show_bug.cgi?id=22473
Before:
BB#0:
mr 4, 3
extsh. 3, 4
ble 0, .LBB0_5
BB#1:
cmpwi 3, 99
bgt 0, .LBB0_9
BB#2:
rlwinm 4, 4, 0, 16, 31 <--- 32-bit mask/extend
li 3, 0
cmplwi 4, 1
beqlr 0
BB#3:
cmplwi 4, 10
bne 0, .LBB0_12
BB#4:
li 3, 1
blr
.LBB0_5:
rlwinm 3, 4, 0, 16, 31 <--- 32-bit mask/extend
cmplwi 3, 65436
beq 0, .LBB0_13
BB#6:
cmplwi 3, 65526
beq 0, .LBB0_15
BB#7:
cmplwi 3, 65535
bne 0, .LBB0_12
BB#8:
li 3, 4
blr
.LBB0_9:
rlwinm 3, 4, 0, 16, 31 <--- 32-bit mask/extend
cmplwi 3, 100
beq 0, .LBB0_14
...
After:
BB#0:
rlwinm 4, 3, 0, 16, 31 <--- mask/extend to 32-bit and then use that for comparisons
cmpwi 4, 999
ble 0, .LBB0_5
BB#1:
lis 3, 0
ori 3, 3, 65525
cmpw 4, 3
bgt 0, .LBB0_9
BB#2:
cmplwi 4, 1000
beq 0, .LBB0_14
BB#3:
cmplwi 4, 65436
bne 0, .LBB0_13
BB#4:
li 3, 6
blr
.LBB0_5:
li 3, 0
cmplwi 4, 1
beqlr 0
BB#6:
cmplwi 4, 10
beq 0, .LBB0_12
BB#7:
cmplwi 4, 100
bne 0, .LBB0_13
BB#8:
li 3, 2
blr
.LBB0_9:
cmplwi 4, 65526
beq 0, .LBB0_15
BB#10:
cmplwi 4, 65535
bne 0, .LBB0_13
...
Differential Revision: http://reviews.llvm.org/D13532
llvm-svn: 251857
This is a follow-up from the discussion in D12965. The block-at-a-time limitation of
SelectionDAG also came up in D13297.
Without the InstCombine change from D12965, I don't expect this patch to make any
difference in the real world because InstCombine does not shrink cases like this in
visitSwitchInst(). But we need to have this CGP safety harness in place before
proceeding with any shrinkage in D12965, so we won't generate extra extends for compares.
I've opted for IR regression tests in the patch because that seems like a clearer way to
test the transform, but PowerPC CodeGen for an i16 widening test is shown below. x86
will need more work to solve: https://llvm.org/bugs/show_bug.cgi?id=22473
Before:
BB#0:
mr 4, 3
extsh. 3, 4
ble 0, .LBB0_5
BB#1:
cmpwi 3, 99
bgt 0, .LBB0_9
BB#2:
rlwinm 4, 4, 0, 16, 31 <--- 32-bit mask/extend
li 3, 0
cmplwi 4, 1
beqlr 0
BB#3:
cmplwi 4, 10
bne 0, .LBB0_12
BB#4:
li 3, 1
blr
.LBB0_5:
rlwinm 3, 4, 0, 16, 31 <--- 32-bit mask/extend
cmplwi 3, 65436
beq 0, .LBB0_13
BB#6:
cmplwi 3, 65526
beq 0, .LBB0_15
BB#7:
cmplwi 3, 65535
bne 0, .LBB0_12
BB#8:
li 3, 4
blr
.LBB0_9:
rlwinm 3, 4, 0, 16, 31 <--- 32-bit mask/extend
cmplwi 3, 100
beq 0, .LBB0_14
...
After:
BB#0:
rlwinm 4, 3, 0, 16, 31 <--- mask/extend to 32-bit and then use that for comparisons
cmpwi 4, 999
ble 0, .LBB0_5
BB#1:
lis 3, 0
ori 3, 3, 65525
cmpw 4, 3
bgt 0, .LBB0_9
BB#2:
cmplwi 4, 1000
beq 0, .LBB0_14
BB#3:
cmplwi 4, 65436
bne 0, .LBB0_13
BB#4:
li 3, 6
blr
.LBB0_5:
li 3, 0
cmplwi 4, 1
beqlr 0
BB#6:
cmplwi 4, 10
beq 0, .LBB0_12
BB#7:
cmplwi 4, 100
bne 0, .LBB0_13
BB#8:
li 3, 2
blr
.LBB0_9:
cmplwi 4, 65526
beq 0, .LBB0_15
BB#10:
cmplwi 4, 65535
bne 0, .LBB0_13
...
Differential Revision: http://reviews.llvm.org/D13532
llvm-svn: 251849
When the target does not support these intrinsics they should be converted to a chain of scalar load or store operations.
If the mask is not constant, the scalarizer will build a chain of conditional basic blocks.
I added isLegalMaskedGather() isLegalMaskedScatter() APIs.
Differential Revision: http://reviews.llvm.org/D13722
llvm-svn: 251237
When we have to convert the masked.load, masked.store to scalar code, we generate a chain of conditional basic blocks.
I added optimization for constant mask vector.
Differential Revision: http://reviews.llvm.org/D13855
llvm-svn: 250893
This was originally checked in at r250527, but reverted at r250570 because of PR25222.
There were at least 2 problems:
1. The cost check was checking for an instruction with an exact cost of TCC_Expensive;
that should have been >=.
2. The cause of the clang stage 1 failures was illegally sinking 'call' instructions;
we can't sink instructions that may have side effects / are not safe to execute speculatively.
Fixed those conditions in sinkSelectOperand() and added test cases.
Original commit message:
This is a follow-up to the discussion in D12882.
Ideally, we would like SimplifyCFG to be able to form select instructions even when the operands
are expensive (as defined by the TTI cost model) because that may expose further optimizations.
However, we would then like a later pass like CodeGenPrepare to undo that transformation if the
target would likely benefit from not speculatively executing an expensive op (this patch).
Once we have this safety mechanism in place, we can adjust SimplifyCFG to restore its
select-formation behavior that changed with r248439.
Differential Revision: http://reviews.llvm.org/D13297
llvm-svn: 250743
Originally I planned to use the same interface for masked gather/scatter and set isConsecutive to "false" in this case.
Now I'm implementing masked gather/scatter and see that the interface is inconvenient. I want to add interfaces isLegalMaskedGather() / isLegalMaskedScatter() instead of using the "Consecutive" parameter in the existing interfaces.
Differential Revision: http://reviews.llvm.org/D13850
llvm-svn: 250686
Ideally, we would like SimplifyCFG to be able to form select instructions even when the operands
are expensive (as defined by the TTI cost model) because that may expose further optimizations.
However, we would then like a later pass like CodeGenPrepare to undo that transformation if the
target would likely benefit from not speculatively executing an expensive op (this patch).
Once we have this safety mechanism in place, we can adjust SimplifyCFG to restore its
select-formation behavior that changed with r248439.
Differential Revision: http://reviews.llvm.org/D13297
llvm-svn: 250527
This patch uses the metadata defined in D12341 to avoid creating an unpredictable branch.
Differential Revision: http://reviews.llvm.org/D12342
llvm-svn: 246692
Create wrapper methods in the Function class for the OptimizeForSize and MinSize
attributes. We want to hide the logic of "or'ing" them together when optimizing
just for size (-Os).
Currently, we are not consistent about this and rely on a front-end to always set
OptimizeForSize (-Os) if MinSize (-Oz) is on. Thus, there are 18 FIXME changes here
that should be added as follow-on patches with regression tests.
This patch is NFC-intended: it just replaces existing direct accesses of the attributes
by the equivalent wrapper call.
Differential Revision: http://reviews.llvm.org/D11734
llvm-svn: 243994
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.
Reviewers: echristo
Subscribers: jholewinski, llvm-commits, rafael, yaron.keren
Differential Revision: http://reviews.llvm.org/D11040
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241778
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.
Reviewers: echristo
Subscribers: jholewinski, ted, yaron.keren, rafael, llvm-commits
Differential Revision: http://reviews.llvm.org/D11028
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241775
Summary:
SelectionDAG itself is not invoking directly the DataLayout in the
TargetMachine, but the "TargetLowering" class is still using it. I'll
address it in a following commit.
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.
Reviewers: echristo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D11000
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241618
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.
Reviewers: echristo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D10986
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241614
The patch is generated using this command:
tools/clang/tools/extra/clang-tidy/tool/run-clang-tidy.py -fix \
-checks=-*,llvm-namespace-comment -header-filter='llvm/.*|clang/.*' \
llvm/lib/
Thanks to Eugene Kosov for the original patch!
llvm-svn: 240137
It's been used before to avoid infinite loops caused by separate CGP
optimizations undoing one another. We found one more such issue
caused by r238054. To avoid it, generalize the "InsertedTruncs"
set to any inst, and use it to avoid touching those again.
llvm-svn: 239938
The usual CodeGenPrepare trickery, on a target-specific intrinsic.
Without this, the expansion of atomics will usually have the zext
be hoisted out of the loop, defeating the various patterns we have
to catch this precise case.
Differential Revision: http://reviews.llvm.org/D9930
llvm-svn: 238054
We already had a method to iterate over all the incoming values of a PHI. This just changes all eligible code to use it.
Ineligible code included anything which cared about the index, or was also trying to get the i'th incoming BB.
llvm-svn: 237169
Summary:
The original code inserted new instructions by following a
Create->Remove->ReInsert flow. This patch removes the unnecessary
Remove->ReInsert part by setting up the InsertPoint correctly at the
very beginning. This change does not introduce any functionality change.
Patch by Chen Li!
Reviewers: reames, AndyAyers, sanjoy
Reviewed By: sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D9687
llvm-svn: 237070
Summary:
In RewriteStatepointsForGC pass, we create a gc_relocate intrinsic for
each relocated pointer, and the gc_relocate has the same type with the
pointer. During the creation of gc_relocate intrinsic, llvm requires to
mangle its type. However, llvm does not support mangling of all possible
types. RewriteStatepointsForGC will hit an assertion failure when it
tries to create a gc_relocate for pointer to vector of pointers because
mangling for vector of pointers is not supported.
This patch changes the way RewriteStatepointsForGC pass creates
gc_relocate. For each relocated pointer, we erase the type of pointers
and create an unified gc_relocate of type i8 addrspace(1)*. Then a
bitcast is inserted to convert the gc_relocate to the correct type. In
this way, gc_relocate does not need to deal with different types of
pointers and the unsupported type mangling is no longer a problem. This
change would also ease further merge when LLVM erases types of pointers
and introduces an unified pointer type.
Some minor changes are also introduced to gc_relocate related part in
InstCombineCalls, CodeGenPrepare, and Verifier accordingly.
Patch by Chen Li!
Reviewers: reames, AndyAyers, sanjoy
Reviewed By: sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D9592
llvm-svn: 237009
Fill in the TODO in CodeGenPrepare::OptimizeCallInst so that global
variables that are passed to memory intrinsics are aligned in the same
way that allocas are.
Differential Revision: http://reviews.llvm.org/D8421
llvm-svn: 234735
The patch is generated using clang-tidy misc-use-override check.
This command was used:
tools/clang/tools/extra/clang-tidy/tool/run-clang-tidy.py \
-checks='-*,misc-use-override' -header-filter='llvm|clang' \
-j=32 -fix -format
http://reviews.llvm.org/D8925
llvm-svn: 234679
r234638 chained another transform below which was tripping over the
deleted instruction. Use after free found by asan in many regression
tests.
llvm-svn: 234654
Summary:
This change moves creating calls to `llvm.uadd.with.overflow` from
InstCombine to CodeGenPrep. Combining overflow check patterns into
calls to the said intrinsic in InstCombine inhibits optimization because
it introduces an intrinsic call that not all other transforms and
analyses understand.
Depends on D8888.
Reviewers: majnemer, atrick
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D8889
llvm-svn: 234638
The changes to InstCombine do seem a bit silly - it doesn't make
anything obviously better to have the caller access the pointers element
type (the thing I'm trying to remove) than the GEP itself, but it's a
helpful migration step. This will allow me to more obviously lock down
GEP (& Load, etc) API usage, then fix all the code that accesses pointer
element types except the places that need to be removed (most of the
InstCombines) anyway - at which point I'll need to just remove all that
code because it won't be meaningful anymore (there will be no pointer
types, so no bitcasts to combine)
llvm-svn: 233126
Memcpy, and other memory intrinsics, typically tries to use LDM/STM if
the source and target addresses are 4-byte aligned. In CodeGenPrepare
look for calls to memory intrinsics and, if the object is on the
stack, 4-byte align it if it's large enough that we expect that memcpy
would want to use LDM/STM to copy it.
Differential Revision: http://reviews.llvm.org/D7908
llvm-svn: 232627
- Use TargetLowering to check for the actual cost of each extension.
- Provide a factorized method to check for the cost of an extension:
TargetLowering::isExtFree.
- Provide a virtual method TargetLowering::isExtFreeImpl for targets to be able
to tune the cost of non-free extensions.
This refactoring offers a better granularity to model what really happens on
different targets.
No performance changes and very few code differences.
Part of <rdar://problem/19267165>
llvm-svn: 231855
Summary:
Now that the DataLayout is a mandatory part of the module, let's start
cleaning the codebase. This patch is a first attempt at doing that.
This patch is not exactly NFC as for instance some places were passing
a nullptr instead of the DataLayout, possibly just because there was a
default value on the DataLayout argument to many functions in the API.
Even though it is not purely NFC, there is no change in the
validation.
I turned as many pointer to DataLayout to references, this helped
figuring out all the places where a nullptr could come up.
I had initially a local version of this patch broken into over 30
independant, commits but some later commit were cleaning the API and
touching part of the code modified in the previous commits, so it
seemed cleaner without the intermediate state.
Test Plan:
Reviewers: echristo
Subscribers: llvm-commits
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 231740
that is iterating over it
Inserting elements into a `DenseMap` invalidated iterators pointing
into the `DenseMap` instance.
Differential Revision: http://reviews.llvm.org/D7924
llvm-svn: 230719
a lookup, pass that in rather than use a naked call to getSubtargetImpl.
This involved passing down and around either a TargetMachine or
TargetRegisterInfo. Update all callers/definitions around the targets
and SelectionDAG.
llvm-svn: 230699
Canonicalize access to function attributes to use the simpler API.
getAttributes().getAttribute(AttributeSet::FunctionIndex, Kind)
=> getFnAttribute(Kind)
getAttributes().hasAttribute(AttributeSet::FunctionIndex, Kind)
=> hasFnAttribute(Kind)
Also, add `Function::getFnStackAlignment()`, and canonicalize:
getAttributes().getStackAlignment(AttributeSet::FunctionIndex)
=> getFnStackAlignment()
llvm-svn: 229208
SimplifyCFG now knows how to speculate calls to intrinsic cttz/ctlz that are
'cheap' for the target. Therefore, some of the logic in CodeGenPrepare
that was originally added at revision 224899 can now be removed.
This patch is basically a no functional change. It removes the duplicated
logic in CodeGenPrepare and converts all the existing target specific tests
for cttz/ctlz into SimplifyCFG tests.
Differential Revision: http://reviews.llvm.org/D7608
llvm-svn: 229105
getTTI method used to get an actual TTI object.
No functionality changed. This just threads the argument and ensures
code like the inliner can correctly look up the callee's TTI rather than
using a fixed one.
The next change will use this to implement per-function subtarget usage
by TTI. The changes after that should eliminate the need for FTTI as that
will have become the default.
llvm-svn: 227730
type erased interface and a single analysis pass rather than an
extremely complex analysis group.
The end result is that the TTI analysis can contain a type erased
implementation that supports the polymorphic TTI interface. We can build
one from a target-specific implementation or from a dummy one in the IR.
I've also factored all of the code into "mix-in"-able base classes,
including CRTP base classes to facilitate calling back up to the most
specialized form when delegating horizontally across the surface. These
aren't as clean as I would like and I'm planning to work on cleaning
some of this up, but I wanted to start by putting into the right form.
There are a number of reasons for this change, and this particular
design. The first and foremost reason is that an analysis group is
complete overkill, and the chaining delegation strategy was so opaque,
confusing, and high overhead that TTI was suffering greatly for it.
Several of the TTI functions had failed to be implemented in all places
because of the chaining-based delegation making there be no checking of
this. A few other functions were implemented with incorrect delegation.
The message to me was very clear working on this -- the delegation and
analysis group structure was too confusing to be useful here.
The other reason of course is that this is *much* more natural fit for
the new pass manager. This will lay the ground work for a type-erased
per-function info object that can look up the correct subtarget and even
cache it.
Yet another benefit is that this will significantly simplify the
interaction of the pass managers and the TargetMachine. See the future
work below.
The downside of this change is that it is very, very verbose. I'm going
to work to improve that, but it is somewhat an implementation necessity
in C++ to do type erasure. =/ I discussed this design really extensively
with Eric and Hal prior to going down this path, and afterward showed
them the result. No one was really thrilled with it, but there doesn't
seem to be a substantially better alternative. Using a base class and
virtual method dispatch would make the code much shorter, but as
discussed in the update to the programmer's manual and elsewhere,
a polymorphic interface feels like the more principled approach even if
this is perhaps the least compelling example of it. ;]
Ultimately, there is still a lot more to be done here, but this was the
huge chunk that I couldn't really split things out of because this was
the interface change to TTI. I've tried to minimize all the other parts
of this. The follow up work should include at least:
1) Improving the TargetMachine interface by having it directly return
a TTI object. Because we have a non-pass object with value semantics
and an internal type erasure mechanism, we can narrow the interface
of the TargetMachine to *just* do what we need: build and return
a TTI object that we can then insert into the pass pipeline.
2) Make the TTI object be fully specialized for a particular function.
This will include splitting off a minimal form of it which is
sufficient for the inliner and the old pass manager.
3) Add a new pass manager analysis which produces TTI objects from the
target machine for each function. This may actually be done as part
of #2 in order to use the new analysis to implement #2.
4) Work on narrowing the API between TTI and the targets so that it is
easier to understand and less verbose to type erase.
5) Work on narrowing the API between TTI and its clients so that it is
easier to understand and less verbose to forward.
6) Try to improve the CRTP-based delegation. I feel like this code is
just a bit messy and exacerbating the complexity of implementing
the TTI in each target.
Many thanks to Eric and Hal for their help here. I ended up blocked on
this somewhat more abruptly than I expected, and so I appreciate getting
it sorted out very quickly.
Differential Revision: http://reviews.llvm.org/D7293
llvm-svn: 227669
a DominatorTree argument as that is the analysis that it wants to
update.
This removes the last non-loop utility function in Utils/ which accepts
a raw Pass argument.
llvm-svn: 226537
The pass is really just a means of accessing a cached instance of the
TargetLibraryInfo object, and this way we can re-use that object for the
new pass manager as its result.
Lots of delta, but nothing interesting happening here. This is the
common pattern that is developing to allow analyses to live in both the
old and new pass manager -- a wrapper pass in the old pass manager
emulates the separation intrinsic to the new pass manager between the
result and pass for analyses.
llvm-svn: 226157
While the term "Target" is in the name, it doesn't really have to do
with the LLVM Target library -- this isn't an abstraction which LLVM
targets generally need to implement or extend. It has much more to do
with modeling the various runtime libraries on different OSes and with
different runtime environments. The "target" in this sense is the more
general sense of a target of cross compilation.
This is in preparation for porting this analysis to the new pass
manager.
No functionality changed, and updates inbound for Clang and Polly.
llvm-svn: 226078
The transform is somewhat involved, but the basic idea is simple: find
derived pointers that have been offset from the base pointer using gep
and replace the relocate of the derived pointer with a gep to the
relocated base pointer (with the same offset).
llvm-svn: 226060
type (in addition to the memory type).
The *LoadExt* legalization handling used to only have one type, the
memory type. This forced users to assume that as long as the extload
for the memory type was declared legal, and the result type was legal,
the whole extload was legal.
However, this isn't always the case. For instance, on X86, with AVX,
this is legal:
v4i32 load, zext from v4i8
but this isn't:
v4i64 load, zext from v4i8
Whereas v4i64 is (arguably) legal, even without AVX2.
Note that the same thing was done a while ago for truncstores (r46140),
but I assume no one needed it yet for extloads, so here we go.
Calls to getLoadExtAction were changed to add the value type, found
manually in the surrounding code.
Calls to setLoadExtAction were mechanically changed, by wrapping the
call in a loop, to match previous behavior. The loop iterates over
the MVT subrange corresponding to the memory type (FP vectors, etc...).
I also pulled neighboring setTruncStoreActions into some of the loops;
those shouldn't make a difference, as the additional types are illegal.
(e.g., i128->i1 truncstores on PPC.)
No functional change intended.
Differential Revision: http://reviews.llvm.org/D6532
llvm-svn: 225421
This patch improves the logic added at revision 224899 (see review D6728) that
teaches the backend when it is profitable to speculate calls to cttz/ctlz.
The original algorithm conservatively avoided speculating more than one
instruction from a basic block in a control flow grap modelling an if-statement.
In particular, the only allowed instruction (excluding the terminator) was a
call to cttz/ctlz. However, there are cases where we could be less conservative
and still be able to speculate a call to cttz/ctlz.
With this patch, CodeGenPrepare now tries to speculate a cttz/ctlz if the
result is zero extended/truncated in the same basic block, and the zext/trunc
instruction is "free" for the target.
Added new test cases to CodeGen/X86/cttz-ctlz.ll
Differential Revision: http://reviews.llvm.org/D6853
llvm-svn: 225274
If the control flow is modelling an if-statement where the only instruction in
the 'then' basic block (excluding the terminator) is a call to cttz/ctlz,
CodeGenPrepare can try to speculate the cttz/ctlz call and simplify the control
flow graph.
Example:
\code
entry:
%cmp = icmp eq i64 %val, 0
br i1 %cmp, label %end.bb, label %then.bb
then.bb:
%c = tail call i64 @llvm.cttz.i64(i64 %val, i1 true)
br label %end.bb
end.bb:
%cond = phi i64 [ %c, %then.bb ], [ 64, %entry]
\code
In this example, basic block %then.bb is taken if value %val is not zero.
Also, the phi node in %end.bb would propagate the size-of in bits of %val
only if %val is equal to zero.
With this patch, CodeGenPrepare will try to hoist the call to cttz from %then.bb
into basic block %entry only if cttz is cheap to speculate for the target.
Added two new hooks in TargetLowering.h to let targets customize the behavior
(i.e. decide whether it is cheap or not to speculate calls to cttz/ctlz). The
two new methods are 'isCheapToSpeculateCtlz' and 'isCheapToSpeculateCttz'.
By default, both methods return 'false'.
On X86, method 'isCheapToSpeculateCtlz' returns true only if the target has
LZCNT. Method 'isCheapToSpeculateCttz' only returns true if the target has BMI.
Differential Revision: http://reviews.llvm.org/D6728
llvm-svn: 224899
Masked vector intrinsics are a part of common LLVM IR, but they are really supported on AVX2 and AVX-512 targets. I added a code that translates masked intrinsic for all other targets. The masked vector intrinsic is converted to a chain of scalar operations inside conditional basic blocks.
http://reviews.llvm.org/D6436
llvm-svn: 224897
The type promotion helper does not support vector type, so when make
such it does not kick in in such cases.
Original commit message:
[CodeGenPrepare] Move sign/zero extensions near loads using type promotion.
This patch extends the optimization in CodeGenPrepare that moves a sign/zero
extension near a load when the target can combine them. The optimization may
promote any operations between the extension and the load to make that possible.
Although this optimization may be beneficial for all targets, in particular
AArch64, this is enabled for X86 only as I have not benchmarked it for other
targets yet.
** Context **
Most targets feature extended loads, i.e., loads that perform a zero or sign
extension for free. In that context it is interesting to expose such pattern in
CodeGenPrepare so that the instruction selection pass can form such loads.
Sometimes, this pattern is blocked because of instructions between the load and
the extension. When those instructions are promotable to the extended type, we
can expose this pattern.
** Motivating Example **
Let us consider an example:
define void @foo(i8* %addr1, i32* %addr2, i8 %a, i32 %b) {
%ld = load i8* %addr1
%zextld = zext i8 %ld to i32
%ld2 = load i32* %addr2
%add = add nsw i32 %ld2, %zextld
%sextadd = sext i32 %add to i64
%zexta = zext i8 %a to i32
%addza = add nsw i32 %zexta, %zextld
%sextaddza = sext i32 %addza to i64
%addb = add nsw i32 %b, %zextld
%sextaddb = sext i32 %addb to i64
call void @dummy(i64 %sextadd, i64 %sextaddza, i64 %sextaddb)
ret void
}
As it is, this IR generates the following assembly on x86_64:
[...]
movzbl (%rdi), %eax # zero-extended load
movl (%rsi), %es # plain load
addl %eax, %esi # 32-bit add
movslq %esi, %rdi # sign extend the result of add
movzbl %dl, %edx # zero extend the first argument
addl %eax, %edx # 32-bit add
movslq %edx, %rsi # sign extend the result of add
addl %eax, %ecx # 32-bit add
movslq %ecx, %rdx # sign extend the result of add
[...]
The throughput of this sequence is 7.45 cycles on Ivy Bridge according to IACA.
Now, by promoting the additions to form more extended loads we would generate:
[...]
movzbl (%rdi), %eax # zero-extended load
movslq (%rsi), %rdi # sign-extended load
addq %rax, %rdi # 64-bit add
movzbl %dl, %esi # zero extend the first argument
addq %rax, %rsi # 64-bit add
movslq %ecx, %rdx # sign extend the second argument
addq %rax, %rdx # 64-bit add
[...]
The throughput of this sequence is 6.15 cycles on Ivy Bridge according to IACA.
This kind of sequences happen a lot on code using 32-bit indexes on 64-bit
architectures.
Note: The throughput numbers are similar on Sandy Bridge and Haswell.
** Proposed Solution **
To avoid the penalty of all these sign/zero extensions, we merge them in the
loads at the beginning of the chain of computation by promoting all the chain of
computation on the extended type. The promotion is done if and only if we do not
introduce new extensions, i.e., if we do not degrade the code quality.
To achieve this, we extend the existing “move ext to load” optimization with the
promotion mechanism introduced to match larger patterns for addressing mode
(r200947).
The idea of this extension is to perform the following transformation:
ext(promotableInst1(...(promotableInstN(load))))
=>
promotedInst1(...(promotedInstN(ext(load))))
The promotion mechanism in that optimization is enabled by a new TargetLowering
switch, which is off by default. In other words, by default, the optimization
performs the “move ext to load” optimization as it was before this patch.
** Performance **
Configuration: x86_64: Ivy Bridge fixed at 2900MHz running OS X 10.10.
Tested Optimization Levels: O3/Os
Tests: llvm-testsuite + externals.
Results:
- No regression beside noise.
- Improvements:
CINT2006/473.astar: ~2%
Benchmarks/PAQ8p: ~2%
Misc/perlin: ~3%
The results are consistent for both O3 and Os.
<rdar://problem/18310086>
llvm-svn: 224402
This patch extends the optimization in CodeGenPrepare that moves a sign/zero
extension near a load when the target can combine them. The optimization may
promote any operations between the extension and the load to make that possible.
Although this optimization may be beneficial for all targets, in particular
AArch64, this is enabled for X86 only as I have not benchmarked it for other
targets yet.
** Context **
Most targets feature extended loads, i.e., loads that perform a zero or sign
extension for free. In that context it is interesting to expose such pattern in
CodeGenPrepare so that the instruction selection pass can form such loads.
Sometimes, this pattern is blocked because of instructions between the load and
the extension. When those instructions are promotable to the extended type, we
can expose this pattern.
** Motivating Example **
Let us consider an example:
define void @foo(i8* %addr1, i32* %addr2, i8 %a, i32 %b) {
%ld = load i8* %addr1
%zextld = zext i8 %ld to i32
%ld2 = load i32* %addr2
%add = add nsw i32 %ld2, %zextld
%sextadd = sext i32 %add to i64
%zexta = zext i8 %a to i32
%addza = add nsw i32 %zexta, %zextld
%sextaddza = sext i32 %addza to i64
%addb = add nsw i32 %b, %zextld
%sextaddb = sext i32 %addb to i64
call void @dummy(i64 %sextadd, i64 %sextaddza, i64 %sextaddb)
ret void
}
As it is, this IR generates the following assembly on x86_64:
[...]
movzbl (%rdi), %eax # zero-extended load
movl (%rsi), %es # plain load
addl %eax, %esi # 32-bit add
movslq %esi, %rdi # sign extend the result of add
movzbl %dl, %edx # zero extend the first argument
addl %eax, %edx # 32-bit add
movslq %edx, %rsi # sign extend the result of add
addl %eax, %ecx # 32-bit add
movslq %ecx, %rdx # sign extend the result of add
[...]
The throughput of this sequence is 7.45 cycles on Ivy Bridge according to IACA.
Now, by promoting the additions to form more extended loads we would generate:
[...]
movzbl (%rdi), %eax # zero-extended load
movslq (%rsi), %rdi # sign-extended load
addq %rax, %rdi # 64-bit add
movzbl %dl, %esi # zero extend the first argument
addq %rax, %rsi # 64-bit add
movslq %ecx, %rdx # sign extend the second argument
addq %rax, %rdx # 64-bit add
[...]
The throughput of this sequence is 6.15 cycles on Ivy Bridge according to IACA.
This kind of sequences happen a lot on code using 32-bit indexes on 64-bit
architectures.
Note: The throughput numbers are similar on Sandy Bridge and Haswell.
** Proposed Solution **
To avoid the penalty of all these sign/zero extensions, we merge them in the
loads at the beginning of the chain of computation by promoting all the chain of
computation on the extended type. The promotion is done if and only if we do not
introduce new extensions, i.e., if we do not degrade the code quality.
To achieve this, we extend the existing “move ext to load” optimization with the
promotion mechanism introduced to match larger patterns for addressing mode
(r200947).
The idea of this extension is to perform the following transformation:
ext(promotableInst1(...(promotableInstN(load))))
=>
promotedInst1(...(promotedInstN(ext(load))))
The promotion mechanism in that optimization is enabled by a new TargetLowering
switch, which is off by default. In other words, by default, the optimization
performs the “move ext to load” optimization as it was before this patch.
** Performance **
Configuration: x86_64: Ivy Bridge fixed at 2900MHz running OS X 10.10.
Tested Optimization Levels: O3/Os
Tests: llvm-testsuite + externals.
Results:
- No regression beside noise.
- Improvements:
CINT2006/473.astar: ~2%
Benchmarks/PAQ8p: ~2%
Misc/perlin: ~3%
The results are consistent for both O3 and Os.
<rdar://problem/18310086>
llvm-svn: 224351
Split `Metadata` away from the `Value` class hierarchy, as part of
PR21532. Assembly and bitcode changes are in the wings, but this is the
bulk of the change for the IR C++ API.
I have a follow-up patch prepared for `clang`. If this breaks other
sub-projects, I apologize in advance :(. Help me compile it on Darwin
I'll try to fix it. FWIW, the errors should be easy to fix, so it may
be simpler to just fix it yourself.
This breaks the build for all metadata-related code that's out-of-tree.
Rest assured the transition is mechanical and the compiler should catch
almost all of the problems.
Here's a quick guide for updating your code:
- `Metadata` is the root of a class hierarchy with three main classes:
`MDNode`, `MDString`, and `ValueAsMetadata`. It is distinct from
the `Value` class hierarchy. It is typeless -- i.e., instances do
*not* have a `Type`.
- `MDNode`'s operands are all `Metadata *` (instead of `Value *`).
- `TrackingVH<MDNode>` and `WeakVH` referring to metadata can be
replaced with `TrackingMDNodeRef` and `TrackingMDRef`, respectively.
If you're referring solely to resolved `MDNode`s -- post graph
construction -- just use `MDNode*`.
- `MDNode` (and the rest of `Metadata`) have only limited support for
`replaceAllUsesWith()`.
As long as an `MDNode` is pointing at a forward declaration -- the
result of `MDNode::getTemporary()` -- it maintains a side map of its
uses and can RAUW itself. Once the forward declarations are fully
resolved RAUW support is dropped on the ground. This means that
uniquing collisions on changing operands cause nodes to become
"distinct". (This already happened fairly commonly, whenever an
operand went to null.)
If you're constructing complex (non self-reference) `MDNode` cycles,
you need to call `MDNode::resolveCycles()` on each node (or on a
top-level node that somehow references all of the nodes). Also,
don't do that. Metadata cycles (and the RAUW machinery needed to
construct them) are expensive.
- An `MDNode` can only refer to a `Constant` through a bridge called
`ConstantAsMetadata` (one of the subclasses of `ValueAsMetadata`).
As a side effect, accessing an operand of an `MDNode` that is known
to be, e.g., `ConstantInt`, takes three steps: first, cast from
`Metadata` to `ConstantAsMetadata`; second, extract the `Constant`;
third, cast down to `ConstantInt`.
The eventual goal is to introduce `MDInt`/`MDFloat`/etc. and have
metadata schema owners transition away from using `Constant`s when
the type isn't important (and they don't care about referring to
`GlobalValue`s).
In the meantime, I've added transitional API to the `mdconst`
namespace that matches semantics with the old code, in order to
avoid adding the error-prone three-step equivalent to every call
site. If your old code was:
MDNode *N = foo();
bar(isa <ConstantInt>(N->getOperand(0)));
baz(cast <ConstantInt>(N->getOperand(1)));
bak(cast_or_null <ConstantInt>(N->getOperand(2)));
bat(dyn_cast <ConstantInt>(N->getOperand(3)));
bay(dyn_cast_or_null<ConstantInt>(N->getOperand(4)));
you can trivially match its semantics with:
MDNode *N = foo();
bar(mdconst::hasa <ConstantInt>(N->getOperand(0)));
baz(mdconst::extract <ConstantInt>(N->getOperand(1)));
bak(mdconst::extract_or_null <ConstantInt>(N->getOperand(2)));
bat(mdconst::dyn_extract <ConstantInt>(N->getOperand(3)));
bay(mdconst::dyn_extract_or_null<ConstantInt>(N->getOperand(4)));
and when you transition your metadata schema to `MDInt`:
MDNode *N = foo();
bar(isa <MDInt>(N->getOperand(0)));
baz(cast <MDInt>(N->getOperand(1)));
bak(cast_or_null <MDInt>(N->getOperand(2)));
bat(dyn_cast <MDInt>(N->getOperand(3)));
bay(dyn_cast_or_null<MDInt>(N->getOperand(4)));
- A `CallInst` -- specifically, intrinsic instructions -- can refer to
metadata through a bridge called `MetadataAsValue`. This is a
subclass of `Value` where `getType()->isMetadataTy()`.
`MetadataAsValue` is the *only* class that can legally refer to a
`LocalAsMetadata`, which is a bridged form of non-`Constant` values
like `Argument` and `Instruction`. It can also refer to any other
`Metadata` subclass.
(I'll break all your testcases in a follow-up commit, when I propagate
this change to assembly.)
llvm-svn: 223802
Rewrite the pattern match code to work also with Values instead with
Instructions only. Also remove the no longer need matcher (m_Instruction).
llvm-svn: 223797
This optimization transforms code like:
bb1:
%0 = icmp ne i32 %a, 0
%1 = icmp ne i32 %b, 0
%or.cond = or i1 %0, %1
br i1 %or.cond, label %TrueBB, label %FalseBB
into a multiple branch instructions like:
bb1:
%0 = icmp ne i32 %a, 0
br i1 %0, label %TrueBB, label %bb2
bb2:
%1 = icmp ne i32 %b, 0
br i1 %1, label %TrueBB, label %FalseBB
This optimization is already performed by SelectionDAG, but not by FastISel.
FastISel cannot perform this optimization, because it cannot generate new
MachineBasicBlocks.
Performing this optimization at CodeGenPrepare time makes it available to both -
SelectionDAG and FastISel - and the implementation in SelectiuonDAG could be
removed. There are currenty a few differences in codegen for X86 and PPC, so
this commmit only enables it for FastISel.
Reviewed by Jim Grosbach
This fixes rdar://problem/19034919.
llvm-svn: 223786
This is to be consistent with StringSet and ultimately with the standard
library's associative container insert function.
This lead to updating SmallSet::insert to return pair<iterator, bool>,
and then to update SmallPtrSet::insert to return pair<iterator, bool>,
and then to update all the existing users of those functions...
llvm-svn: 222334
Prior to this patch the TypePromotionHelper was promoting only sign extensions.
Supporting zero extensions changes:
- How constants are extended.
- How sign extensions, zero extensions, and truncate are composed together.
- How the type of the extended operation is recorded. Now we need to know the
kind of the extension as well as its type.
Each change is fairly small, unlike the diff.
Most of the diff are comments/variable renaming to say "extension" instead of
"sign extension".
The performance improvements on the test suite are within the noise.
Related to <rdar://problem/18310086>.
llvm-svn: 221851
r221820 fixed a problem (PR21548) where an iPTR was used in TLI legality checks,
which isn't valid and resulted in a failed assertion.
The solution was to lower pointer types into the correct target's VT, by
using TL::getValueType instead of EVT::getEVT.
This commit changes 3 other uses of EVT::getEVT, but without any tests:
- One of these non-lowered EVTs is passed to allowsMisalignedMemoryAccesses,
which goes into target's TL implementation and doesn't cause any problem (yet.)
- Two others are passed to TLI.isOperationLegalOrCustom:
- one only looks at extensions, so doesn't concern pointers.
- one only looks at binary operators, so also isn't a problem.
The latter might some day be exposed to pointers and cause the same assert as
the original PR, because there's a comment hinting at also supporting cast ops.
For consistency, update all of them and be done with it.
llvm-svn: 221827
This patch adds an optimization in CodeGenPrepare to move an extractelement
right before a store when the target can combine them.
The optimization may promote any scalar operations to vector operations in the
way to make that possible.
** Context **
Some targets use different register files for both vector and scalar operations.
This means that transitioning from one domain to another may incur copy from one
register file to another. These copies are not coalescable and may be expensive.
For example, according to the scheduling model, on cortex-A8 a vector to GPR
move is 20 cycles.
** Motivating Example **
Let us consider an example:
define void @foo(<2 x i32>* %addr1, i32* %dest) {
%in1 = load <2 x i32>* %addr1, align 8
%extract = extractelement <2 x i32> %in1, i32 1
%out = or i32 %extract, 1
store i32 %out, i32* %dest, align 4
ret void
}
As it is, this IR generates the following assembly on armv7:
vldr d16, [r0] @vector load
vmov.32 r0, d16[1] @ cross-register-file copy: 20 cycles
orr r0, r0, #1 @ scalar bitwise or
str r0, [r1] @ scalar store
bx lr
Whereas we could generate much faster code:
vldr d16, [r0] @ vector load
vorr.i32 d16, #0x1 @ vector bitwise or
vst1.32 {d16[1]}, [r1:32] @ vector extract + store
bx lr
Half of the computation made in the vector is useless, but this allows to get
rid of the expensive cross-register-file copy.
** Proposed Solution **
To avoid this cross-register-copy penalty, we promote the scalar operations to
vector operations. The penalty will be removed if we manage to promote the whole
chain of computation in the vector domain.
Currently, we do that only when the chain of computation ends by a store and the
target is able to combine an extract with a store.
Stores are the most likely candidates, because other instructions produce values
that would need to be promoted and so, extracted as some point[1]. Moreover,
this is customary that targets feature stores that perform a vector extract (see
AArch64 and X86 for instance).
The proposed implementation relies on the TargetTransformInfo to decide whether
or not it is beneficial to promote a chain of computation in the vector domain.
Unfortunately, this interface is rather inaccurate for this level of details and
although this optimization may be beneficial for X86 and AArch64, the inaccuracy
will lead to the optimization being too aggressive.
Basically in TargetTransformInfo, everything that is legal has a cost of 1,
whereas, even if a vector type is legal, usually a vector operation is slightly
more expensive than its scalar counterpart. That will lead to too many
promotions that may not be counter balanced by the saving of the
cross-register-file copy. For instance, on AArch64 this penalty is just 4
cycles.
For now, the optimization is just enabled for ARM prior than v8, since those
processors have a larger penalty on cross-register-file copies, and the scope is
limited to basic blocks. Because of these two factors, we limit the effects of
the inaccuracy. Indeed, I did not want to build up a fancy cost model with block
frequency and everything on top of that.
[1] We can imagine targets that can combine an extractelement with other
instructions than just stores. If we want to go into that direction, the current
interfaces must be augmented and, moreover, I think this becomes a global isel
problem.
Differential Revision: http://reviews.llvm.org/D5921
<rdar://problem/14170854>
llvm-svn: 220978
introduced in r217629.
We were returning the old sext instead of the new zext as the promoted instruction!
Thanks Joerg Sonnenberger for the test case.
llvm-svn: 217800
The test being performed is just an approximation anyway, so it really
shouldn't crash when things don't go entirely as expected.
Should fix PR20474.
llvm-svn: 214177
Summary: This patch introduces two new iterator ranges and updates existing code to use it. No functional change intended.
Test Plan: All tests (make check-all) still pass.
Reviewers: dblaikie
Reviewed By: dblaikie
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D4481
llvm-svn: 213474
Serguei Katkov