Commit Graph
195 Commits
Author SHA1 Message Date
Tomas Matheson a9e546cc71 [TableGen][NFC] convert TreePatternNode pointers to references (#81134)
Almost all uses of `*TreePatternNode` expect it to be non-null. There
was the occasional check that it wasn't, which I have removed. Making
them references makes it clear that they exist.

This was attempted in 2018 (1b465767d6)
for `TreePatternNode::getChild()` but that was reverted.
2024-02-09 13:35:42 +00:00
Pierre van Houtryve b9079baadd [NFC] clang-format utils/TableGen (#80973)
```
find llvm/utils/TableGen -iname "*.h" -o -iname "*.cpp" | xargs clang-format-16 -i
```

Split from #80847
2024-02-09 09:27:04 +01:00
Tomas Matheson 6d4a3e917c Revert "[TableGen] Use heap allocated arrays instead of vectors for TreePatternNode::Types and ResultPerm. NFC"
While working on DAGISelMatcherEmitter I've hit several runtime errors
caused by accessing TreePatternNode::Types out of bounds. These were
difficult to debug because the switch from std::vector to unique_ptr
removes bounds checking.

I don't think the slight reduction in class size is worth the extra
debugging and memory safety problems, so I suggest we revert this.

This reverts commit d34125a1a8.

Differential Revision: https://reviews.llvm.org/D154781
2023-07-09 10:14:48 +01:00
Tomas Matheson 4c66b03259 [NFC] make TypeInfer::expandOverloads const
Differential Revision: https://reviews.llvm.org/D154780
2023-07-09 09:19:06 +01:00
Tomas Matheson b25572137e [NFC] make TypeInfer::MergeInTypeInfo const
Differential Revision: https://reviews.llvm.org/D154779
2023-07-09 09:11:59 +01:00
Sergei Barannikov 60f815d241 [TableGen] Forward declare CodeGenRegister et al. (NFC) 2023-04-30 07:01:00 +03:00
Craig Topper 7a6dc3d24c [TableGen] Remove unused ForceMode and CodeGen fields from TypeInfer. NFC
As well as the ForceMode field in PatternToMatch.
2023-04-23 20:32:14 -07:00
Craig Topper d34125a1a8 [TableGen] Use heap allocated arrays instead of vectors for TreePatternNode::Types and ResultPerm. NFC
These vectors are resized in the constructor and never change size.
We can manually allocate two arrays instead.

This reduces the size of TreePatternNode by removing the
unneeded capacity end pointer fields from the std::vector.
2023-04-19 15:22:58 -07:00
Craig Topper 83507644ed [TableGen] Reduce the amount of storage space used for AddrSpaces in TypeSetByHwMode. NFC
We reserved 16 AddrSpaces in every TypeSetByHwMode. But we only ever
use the first one on targets that make use of the AddrSpace feature.

The vector was populated by pushing for each entry in the ArrayRef
passed to the TypeSetByHwMode constructor. Each entry is a
ValueTypeByHwMode that stores one VT for each HwMode.

The vector is accessed by a loop in TypeSetByHwMode::getValueTypeByHwMode.
That loop is over HwModes with in the TypeSetByHwMode. This is
unrelated to how the vector was created. The entries in the vector
don't represent HwModes.

The targets that use AddrSpace don't make use of HwModes so the
loop in getValueTypeByHwMode will only run 1 iteration. So we only
the first entry in the vector is meaningful used.

This patch simplifies things by storing only 1 AddrSpace in
TypeSetByMode. Reducing the memory used by TypeSetByHwMode.
More work will be needed to support HwModes with AddrSpace if we
need a different AddrSpace for each HwMode.

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D148194
2023-04-18 15:34:22 -07:00
Craig Topper 46b536f822 [TableGen] Const-correct TreePatternNode::getChild. NFC
Make the const version return a const TreePatterNode* and add a non-const
version to return a non-const TreePatternNode*.
2023-04-18 01:20:42 -07:00
Craig Topper 73925b3a0c [TableGen] Use PointerUnion for Operator and Val in TreePatternNode.
These fields are never set at the same time so we can store them
using a single PointerUnion.
2023-04-16 20:10:51 -07:00
Craig Topper f0c3dfdbf1 [TableGen] Avoid extra IntrusiveRefCntPtr when calling SimplifyTree. NFC
Add a helper to get modifiable access to the IntrusiveRefCntPtr
stored in the Children array. This avoids copying and overwriting the child.
2023-04-16 18:57:27 -07:00
Craig Topper 56fdac60aa [TableGen] Make InlinePatternFragments a member of TreePatternNode again. NFC
Now that we use IntrusiveRefCntPtr instead of std::shared_ptr
we don't need to pass a TreePatternNodePtr to this method.
2023-04-14 19:43:44 -07:00
Craig Topper 40e356d67f [TableGen] Replace std::shared_ptr with InstrusiveRefCntPtr for TreePatternNode. NFC
An intrusive reference counter uses less memory than the control
block of std::shared_ptr.

This should allow some additional code simplifications if we
don't need to pass around shared_ptr in order to create new
shared_ptrs.
2023-04-14 10:56:36 -07:00
Craig Topper 7df434e3ff [TableGen] Avoid vector copy in TreePatter::InlinePatternFragments. NFC
Instead of copying a vector and clearing the original, we can swap
with an empty vector.
2023-04-10 23:01:15 -07:00
Craig Topper dea417f75e [TableGen] Pass TreePatternNodePtr to InlinePatternFragments by const reference. NFC
Not all paths through this function require the reference count
to be incremened. Pass by reference so we don't increase the count
unnecessarily.
2023-04-10 15:04:24 -07:00
Craig Topper 551bfca86f [TableGen] Make TreePatternNode::InlinePatternFragments a static method. NFC
Previously we were passing 'this' and the std::shared_ptr version
of 'this'. This replaces all uses of 'this' with the shared_ptr and
makes the method static.
2023-04-10 00:17:29 -07:00
Craig Topper 2e6c72d341 [TableGen] Move vectors into DAGInstruction instead of copying them. NFC 2023-04-10 00:03:44 -07:00
Craig Topper ab84a93287 [TableGen] Remove TypeSetByHwMode::isDefaultOnly(). Use InfoByHwMode<T>::isSimple(). NFC
InfoByHwMode is the base class of TypeSetByHwMode. The two methods
did the same thing. No reason to have two ways to do it.

Also use the getSimple() access instead of Map.begin()->second.
2023-04-09 10:56:06 -07:00
Jay Foad a80b830e48 [TableGen] Enable "Type set is empty for each HW mode" error in non-debug builds
Differential Revision: https://reviews.llvm.org/D147127
2023-03-31 09:03:22 +01:00
Pierre van Houtryve 70924673af [RFC][GISel] Add a way to ignore COPY instructions in InstructionSelector
RFC to add a way to ignore COPY instructions when pattern-matching MIR in GISel.
    - Add a new "GISelFlags" class to TableGen. Both `Pattern`  and `PatFrags` defs can use it to alter matching behaviour.
    - Flags start at zero and are scoped: the setter returns a `SaveAndRestore` object so that when the current scope ends, the flags are restored to their previous values. This allows child patterns to modify the flags without affecting the parent pattern.
    - Child patterns always reuse the parent's pattern, but they can override its values. For more examples, see `GlobalISelEmitterFlags.td` tests.
    - [AMDGPU] Use the IgnoreCopies flag in BFI patterns, which are known to be bothered by cross-regbank copies.

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D136234
2023-02-10 08:37:42 +01:00
Kazu Hirata b3af04f880 [llvm] Use llvm::countr_zero instead of findFirstSet (NFC)
At each call to findFirstSet in this patch, the argument is known to
be nonzero, so we can safely switch to llvm::countr_zero, which will
become std::countr_zero once C++20 is available.
2023-01-23 23:07:41 -08:00
Kazu Hirata caa99a01f5 Use llvm::popcount instead of llvm::countPopulation(NFC) 2023-01-22 12:48:51 -08:00
Kazu Hirata 06b1e5fdc3 [llvm] Use std::underlying_type_t (NFC) 2022-09-25 23:14:15 -07:00
Kazu Hirata 8b1b0d1d81 Revert "Use std::is_same_v instead of std::is_same (NFC)"
This reverts commit c5da37e42d.

This patch seems to break builds with some versions of MSVC.
2022-08-20 23:00:39 -07:00
Kazu Hirata c5da37e42d Use std::is_same_v instead of std::is_same (NFC) 2022-08-20 22:36:26 -07:00
Abinav Puthan Purayil c42fe5bd7a [GlobalISel][SelectionDAG] Implement the HasNoUse builtin predicate
This change introduces the HasNoUse builtin predicate in PatFrags that
checks for the absence of use of the first result operand.
GlobalISelEmitter will allow source PatFrags with this predicate to be
matched with destination instructions with empty outs. This predicate is
required for selecting the no-return variant of atomic instructions in
AMDGPU.

Differential Revision: https://reviews.llvm.org/D125212
2022-07-08 09:47:33 +05:30
Krzysztof Parzyszek 91fe9e6ed3 [TableGen] Move printing to stream directly to MachineValueTypeSet 2022-07-07 09:28:28 -07:00
serge-sans-paille ed98c1b376 Cleanup includes: DebugInfo & CodeGen
Discourse thread: https://discourse.llvm.org/t/include-what-you-use-include-cleanup
Differential Revision: https://reviews.llvm.org/D121332
2022-03-12 17:26:40 +01:00
Craig Topper aefaf16758 [TableGen] Fix both sides of '&&' are same
The operand of the second any_of in EnforceSmallerThan should be
B not S like the FP code in the if below.

Unfortunately, fixing that causes an infinite loop in the build
of RISCV. So I've added a workaround for that as well.

Fixes PR44768.

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D111502
2021-10-12 09:19:20 -07:00
Craig Topper f391de8cb6 [TableGen] Store predicates in PatternToMatch as ListInit *. Add string for HwModeFeatures
This uses to be how predicates were handled prior to HwMode being
added. When the Predicates were converted to a std::vector it
significantly increased the cost of a compare in GenerateVariants.
Since ListInit's are uniquified by tablegen, we can use a simple
pointer comparison to check for identical lists.

In order to store the HwMode, we now add a separate string to
PatternToMatch. This will be appended separately to the predicate
string in getPredicateCheck. A new getPredicateRecords is added
to allow GlobalISel and getPredicateCheck to both get the sorted
list of Records. GlobalISel was ignoring any HwMode predicates
before and still is.

There is one slight change here, ListInits with different predicate
orders aren't sorted so the filtering in GenerateVariants might
fail to detect two isomorphic patterns with different predicate
orders. This doesn't seem to be happening in tree today.

My hope is this will allow us to remove all the BitVector tracking
in GenerateVariants that was making up for predicates beeing
expensive to compare. There's a decent amount of heap allocations
there on large targets like X86, AMDGPU, and RISCV.

Differential Revision: https://reviews.llvm.org/D100691
2021-04-28 12:05:49 -07:00
Craig Topper 6b446310be [TableGen] Predicate::operator== needs to compare the HwMode feature name in addition to the HwMode flag.
This was causing GenerateVariants to lose some variants since
HwMode is expanded first. We were mistakenly thinking the HwMode
predicate matched and finding the variant was isomorphic to a
pattern in another HwMode and discarding it.

Found while investigating it if would be better to generate
variants before expanding HwModes to improve RISCV build time.
I noticed an increase in the number of Opc_MorphNodeTo in the table
which indicated that the number of patterns had changed.
2021-04-16 13:22:28 -07:00
Craig Topper 07edd78993 [TableGen] Reduce the number of map lookups in TypeSetByHwMode::getOrCreate. NFCI
hasMode was looking up the map once. Then we'd either call get which
would look up again, or we'd insert into the map which requires
walking the map to find the insertion point.

I believe the hasMode was needed because get has a special case
to look for DefaultMode if the mode being asked for doesn't exist.
We don't want that here so we were using hasMode to make sure we
wouldn't hit that case.

Simplify to a regular operator[] access which will default
construct a SetType if the lookup fails.
2021-04-15 12:32:21 -07:00
Craig Topper 5f8ed1b220 [TableGen] Make all the fields in PatternToMatch private. NFCI
Add the few missing accessor methods. Use accessor methdods
everywhere.
2021-02-06 22:34:01 -08:00
Craig Topper 973c35d338 [TableGen] Make CodeGenDAGPatterns::getSDNodeNamed take a StringRef instead of const std::string &.
All callers use a string literal and the getDef method the string
is passed to already takes a StringRef.
2021-01-07 14:20:16 -08:00
Fangrui Song a2f922140f [TableGen] Delete 11 unused declarations 2020-12-06 13:21:07 -08:00
Simon Pilgrim a0ed0e3fac CodeGenDAGPatterns.h - remove unnecessary ComplexPattern forward declaration. NFCI.
This is defined in CodeGenTarget.h which we have to explicitly include already.
2020-07-20 15:17:19 +01:00
Simon Pilgrim 93c338fd0f CodeGenDAGPatterns.h - remove unused CodeGenHwModes.h include. NFCI. 2020-07-20 15:17:18 +01:00
Nadav Rotem b3417d80ae [TableGen] Change std::vector to SmallVector
The size of VTList that is pushed into this container is usually 1, but
often 6 or 7. Change the vector to SmallVector to eliminate frequent
mallocs.  This happens hundreds of thousands of times in each tablegen
execution during the LLVM/clang build.

https://reviews.llvm.org/D83849
2020-07-16 11:32:44 -07:00
Benjamin Kramer adcd026838 Make llvm::StringRef to std::string conversions explicit.
This is how it should've been and brings it more in line with
std::string_view. There should be no functional change here.

This is mostly mechanical from a custom clang-tidy check, with a lot of
manual fixups. It uncovers a lot of minor inefficiencies.

This doesn't actually modify StringRef yet, I'll do that in a follow-up.
2020-01-28 23:25:25 +01:00
Matt Arsenault de08244d3d TableGen: Remove unused code 2020-01-09 10:32:57 -05:00
Reid Kleckner 72c68f1352 [TableGen] Remove unused target intrinsic generation logic
AMDGPU was the last in tree target to use this tablegen mode. I plan to
split up the global intrinsic enum similar to the way that clang
diagnostics are split up today. I don't plan to build on this mode.

Reviewers: arsenm, echristo, efriedma

Reviewed By: echristo

Differential Revision: https://reviews.llvm.org/D71318
2019-12-11 07:38:45 -08:00
Dávid Bolvanský 745b6deacc Reland 'Fixed -Wdeprecated-copy warnings. NFCI.'
Fixed hashtable copy ctor.
2019-11-23 23:09:39 +01:00
Dávid Bolvanský 111a4cffa0 Revert 'Fixed -Wdeprecated-copy warnings. NFCI.'
pdbutil's test is failing.
2019-11-23 21:31:27 +01:00
Dávid Bolvanský 7124b45bee Fixed -Wdeprecated-copy warnings. NFCI. 2019-11-23 21:05:07 +01:00
Matt Arsenault 52c262484f TableGen: Add MinAlignment predicate
AMDGPU uses some custom code predicates for testing alignments.

I'm still having trouble comprehending the behavior of predicate bits
in the PatFrag hierarchy. Any attempt to abstract these properties
unexpectdly fails to apply them.

llvm-svn: 367373
2019-07-31 00:14:43 +00:00
Matt Arsenault 57ef94fb06 AMDGPU: Avoid emitting "true" predicates
Empty condition strings are considerde always true. This removes a lot
of clutter from the generated matcher tables.

This shrinks the source size of AMDGPUGenDAGISel.inc from 7.3M to
6.1M.

llvm-svn: 367326
2019-07-30 15:56:43 +00:00
Matt Arsenault d00d857801 TableGen: Add address space to matchers
Currently AMDGPU uses a CodePatPred to check address spaces from the
MachineMemOperand. Introduce a new first class property so that the
existing patterns can be easily modified to uses the new generated
predicate, which will also be handled for GlobalISel.

I would prefer these to match against the pointer type of the
instruction, but that would be difficult to get working with
SelectionDAG compatbility. This is much easier for now and will avoid
a painful tablegen rewrite for all the loads and stores.

I'm also not sure if there's a better way to encode multiple address
spaces in the table, rather than putting the number to expect.

llvm-svn: 366128
2019-07-15 20:59:42 +00:00
Simon Tatham c74322a11b [TableGen] Allow DAG isel patterns to override default operands.
When a Tablegen instruction description uses `OperandWithDefaultOps`,
isel patterns for that instruction don't have to fill in the default
value for the operand in question. But the flip side is that they
actually //can't// override the defaults even if they want to.

This will be very inconvenient for the Arm backend, when we start
wanting to write isel patterns that generate the many MVE predicated
vector instructions, in the form with predication actually enabled. So
this small Tablegen fix makes it possible to write an isel pattern
either with or without values for a defaulted operand, and have the
default values filled in only if they are not overridden.

If all the defaulted operands come at the end of the instruction's
operand list, there's a natural way to match them up to the arguments
supplied in the pattern: consume pattern arguments until you run out,
then fill in any missing instruction operands with their default
values. But if defaulted and non-defaulted operands are interleaved,
it's less clear what to do. This does happen in existing targets (the
first example I came across was KILLGT, in the AMDGPU/R600 backend),
and of course they expect the previous behaviour (that the default for
those operands is used and a pattern argument is not consumed), so for
backwards compatibility I've stuck with that.

Reviewers: nhaehnle, hfinkel, dmgreen

Subscribers: mehdi_amini, javed.absar, tpr, kristof.beyls, steven_wu, dexonsmith, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D63814

llvm-svn: 365114
2019-07-04 08:43:20 +00:00
Tom Stellard 9ad714f7d1 Add support for pointer types in patterns
Summary:
This adds support for defining patterns for global isel using pointer
types, for example:

def : Pat<(load GPR32:$src),
          (p1 (LOAD GPR32:$src))>;

DAGISelEmitter will ignore the pointer information and treat these
types as integers with the same bit-width as the pointer type.

Reviewers: dsanders, rtereshin, arsenm

Reviewed By: arsenm

Subscribers: Petar.Avramovic, wdng, rovka, kristof.beyls, jfb, volkan, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D57065

llvm-svn: 354510
2019-02-20 19:43:47 +00:00