Commit Graph

93 Commits

Author SHA1 Message Date
Tim Northover
ee8d5c393e TableGen: permit non-leaf ComplexPattern uses
This allows the results of a ComplexPattern check to be distributed to separate
named Operands, instead of the current system where all results must apply (and
match perfectly) with a single Operand.

For example, if "some_addrmode" is a ComplexPattern producing two results, you
can write:

   def : Pat<(load (some_addrmode GPR64:$base, imm:$offset)),
             (INST GPR64:$base, imm:$offset)>;

This should allow neater instruction definitions in TableGen that don't put all
possible aspects of addressing into a single operand, but are still usable with
relatively simple C++ CodeGen idioms.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@209206 91177308-0d34-0410-b5e6-96231b3b80d8
2014-05-20 11:52:46 +00:00
Craig Topper
095734c578 [C++11] More 'nullptr' conversion. In some cases just using a boolean check instead of comparing to nullptr.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@206254 91177308-0d34-0410-b5e6-96231b3b80d8
2014-04-15 07:20:03 +00:00
Craig Topper
858594edb0 Fix typo in commment tyep->type.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@200089 91177308-0d34-0410-b5e6-96231b3b80d8
2014-01-25 17:34:23 +00:00
Craig Topper
8f74ea3fa1 Use ArrayRef to simplify some code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199712 91177308-0d34-0410-b5e6-96231b3b80d8
2014-01-21 07:20:05 +00:00
Jakob Stoklund Olesen
8e3cb3ee0c Allow types to be omitted in output patterns.
This syntax is now preferred:

  def : Pat<(subc i32:$b, i32:$c), (SUBCCrr $b, $c)>;

There is no reason to repeat the types in the output pattern.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177844 91177308-0d34-0410-b5e6-96231b3b80d8
2013-03-24 19:37:00 +00:00
Jakob Stoklund Olesen
f0a804df49 Allow direct value types in pattern definitions.
Just like register classes, value types can be used in two ways in
patterns:

  (sext_inreg i32:$src, i16)

In a named leaf node like i32:$src, the value type simply provides the
type of the node directly. This simplifies type inference a lot compared
to the current practice of specifiying types indirectly with register
classes.

As an unnamed leaf node, like i16 above, the value type represents
itself as an MVT::Other immediate.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177828 91177308-0d34-0410-b5e6-96231b3b80d8
2013-03-23 20:35:01 +00:00
Ulrich Weigand
ec8d1a5b72 Extend TableGen instruction selection matcher to improve handling
of complex instruction operands (e.g. address modes).

Currently, if a Pat pattern creates an instruction that has a complex
operand (i.e. one that consists of multiple sub-operands at the MI
level), this operand must match a ComplexPattern DAG pattern with the
correct number of output operands.

This commit extends TableGen to alternatively allow match a complex
operands against multiple separate operands at the DAG level.

This allows using Pat patterns to match pre-increment nodes like
pre_store (which must have separate operands at the DAG level) onto
an instruction pattern that uses a multi-operand memory operand,
like the following example on PowerPC (will be committed as a
follow-on patch):

  def STWU  : DForm_1<37, (outs ptr_rc:$ea_res), (ins GPRC:$rS, memri:$dst),
                    "stwu $rS, $dst", LdStStoreUpd, []>,
                    RegConstraint<"$dst.reg = $ea_res">, NoEncode<"$ea_res">;

  def : Pat<(pre_store GPRC:$rS, ptr_rc:$ptrreg, iaddroff:$ptroff),
            (STWU GPRC:$rS, iaddroff:$ptroff, ptr_rc:$ptrreg)>;

Here, the pair of "ptroff" and "ptrreg" operands is matched onto the
complex operand "dst" of class "memri" in the "STWU" instruction.

Approved by Jakob Stoklund Olesen.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177428 91177308-0d34-0410-b5e6-96231b3b80d8
2013-03-19 19:51:09 +00:00
Chandler Carruth
4ffd89fa4d Sort the #include lines for utils/...
I've tried to find main moudle headers where possible, but the TableGen
stuff may warrant someone else looking at it.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@169251 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-04 10:37:14 +00:00
Joerg Sonnenberger
61131ab15f Remove exception handling usage from tblgen.
Most places can use PrintFatalError as the unwinding mechanism was not
used for anything other than printing the error. The single exception
was CodeGenDAGPatterns.cpp, where intermediate errors during type
resolution were ignored to simplify incremental platform development.
This use is replaced by an error flag in TreePattern and bailout earlier
in various places if it is set. 


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@166712 91177308-0d34-0410-b5e6-96231b3b80d8
2012-10-25 20:33:17 +00:00
Sean Silva
6cfc806a6b tblgen: Mechanically move dynamic_cast<> to dyn_cast<>.
Some of these dyn_cast<>'s would be better phrased as isa<> or cast<>.
That will happen in a future patch.

There are also two dyn_cast_or_null<>'s slipped in instead of
dyn_cast<>'s, since they were causing crashes with just dyn_cast<>.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@165646 91177308-0d34-0410-b5e6-96231b3b80d8
2012-10-10 20:24:43 +00:00
Tom Stellard
6d3d765653 Tablegen: Add OperandWithDefaultOps Operand type
This Operand type takes a default argument, and is initialized to
this value if it does not appear in a patter.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163315 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-06 14:15:52 +00:00
Tim Northover
ca578e7ba6 Teach TableGen to put chains on more instructions
When generating selection tables for Pat instances, TableGen relied on
an output Instruction's Pattern field being set to infer whether a
chain should be added.

This patch adds additional logic to check various flag fields so that
correct code can be generated even if Pattern is unset.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@159217 91177308-0d34-0410-b5e6-96231b3b80d8
2012-06-26 18:46:28 +00:00
Chris Lattner
ef18cd381c fix a failure path to print the right thing, part of PR12357
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@153457 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-26 19:11:51 +00:00
Peter Collingbourne
7c78888887 Move TableGen's parser and entry point into a library
This is the first step towards splitting LLVM and Clang's tblgen executables.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@140951 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-01 16:41:13 +00:00
Jakob Stoklund Olesen
29f018cee6 Switch to ArrayRef<CodeGenRegisterClass*>.
This makes it possible to allocate CodeGenRegisterClass instances
dynamically and reorder them.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@140816 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-29 22:28:37 +00:00
David Greene
05bce0beee Unconstify Inits
Remove const qualifiers from Init references, per Chris' request.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136531 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-29 22:43:06 +00:00
David Greene
f37dd02f77 [AVX] Constify Inits
Make references to Inits const everywhere.  This is the final step
before making them unique.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136485 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-29 19:07:05 +00:00
Eric Christopher
d568b3f552 Revert r134921, 134917, 134908 and 134907. They're causing failures
in multiple buildbots.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134936 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-11 23:06:52 +00:00
David Greene
d4a9066c93 [AVX] Make Inits Foldable
Manage Inits in a FoldingSet.  This provides several benefits:

- Memory for Inits is properly managed

- Duplicate Inits are folded into Flyweights, saving memory

- It enforces const-correctness, protecting against certain classes
  of bugs

The above benefits allow Inits to be used in more contexts, which in
turn provides more dynamism to TableGen.  This enhanced capability
will be used by the AVX code generator to a fold common patterns
together.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134907 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-11 18:25:51 +00:00
Owen Anderson
bea6f615ee Add support for alternative register names, useful for instructions whose operands are logically equivalent to existing registers, but happen to be printed specially. For example, an instruciton that prints d0[0] instead of s0.
Patch by Jim Grosbach.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133940 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-27 21:06:21 +00:00
Jakob Stoklund Olesen
abdbc84b4e Store CodeGenRegisters as pointers so they won't be reallocated.
Reuse the CodeGenRegBank DenseMap in a few places that would build their
own or use linear search.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133333 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-18 04:26:06 +00:00
Jakob Stoklund Olesen
ae1920b1ef Give CodeGenRegisterClass a real sorted member set.
Make the Elements vector private and expose an ArrayRef through
getOrder() instead. getOrder will eventually provide multiple
user-specified allocation orders.

Use the sorted member set for member and subclass tests. Clean up a lot
of ad hoc searches.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133040 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-15 04:50:36 +00:00
Cameron Zwarich
3a2d255ce6 Fix PR9947 by placing OPFL_MemRefs on the node using memory operands rather than
the root if there is only one such node. This leaves only 2 verifier failures in
the entire test suite when running "make check".

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131677 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 21:13:30 +00:00
Jim Grosbach
4a6d735105 Teach TableGen to pre-calculate register enum values when creating the
CodeGenRegister entries. Use this information to more intelligently build
the literal register entires in the DAGISel matcher table. Specifically,
use a single-byte OPC_EmitRegister entry for registers with a value of
less than 256 and OPC_EmitRegister2 entry for registers with a larger value.

rdar://9066491



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127456 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-11 02:19:02 +00:00
Chris Lattner
036609bd7d Flag -> Glue, the ongoing saga
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122513 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-23 18:28:41 +00:00
Chris Lattner
8950bcaa5a continue renaming flag -> glue.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122506 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-23 17:03:20 +00:00
Jim Grosbach
fbadcd0826 Tidy up a bit. Trailing whitespace, hard tabs and 80-columns.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122337 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-21 16:16:00 +00:00
Chris Lattner
f1b4eafbfe rename MVT::Flag to MVT::Glue. "Flag" is a terrible name for
something that just glues two nodes together, even if it is
sometimes used for flags.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122310 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-21 02:38:05 +00:00
Chris Lattner
c240bb0ede factor the operand list (and related fields/operations) out of
CodeGenInstruction into its own helper class.  No functionality change.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117893 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-01 04:03:32 +00:00
Chris Lattner
718045889e zap dead code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113071 91177308-0d34-0410-b5e6-96231b3b80d8
2010-09-04 17:45:58 +00:00
Eric Christopher
ae321ed287 We already have this as OperandNode.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@110748 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-10 23:46:20 +00:00
Jakob Stoklund Olesen
73ea7bf450 Add the SubRegIndex TableGen class.
This is the beginning of purely symbolic subregister indices, but we need a bit
of jiggling before the explicit numeric indices can be completely removed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104492 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-24 14:48:12 +00:00
Chris Lattner
b4a52b04bd fix a long standing fixme, which required fixing a bunch of other
issues to get here.  We now trim the result type list of the
CompleteMatch or MorphNodeTo operation to be the same size as the
thing we're matching.  this means that if you match (add GPR, GPR)
with an instruction that produces a normal result and a flag that
we now trim the result in tblgen instead of having to do it 
dynamically.  This exposed a bunch of inconsistencies in result
counting that happened to be getting lucky since the days of the
old isel.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99728 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-27 20:45:15 +00:00
Chris Lattner
9414ae5291 hoist some funky logic into CodeGenInstruction
from two places in CodeGenDAGPatterns.cpp, and
use it in DAGISelMatcherGen.cpp instead of using
an incorrect predicate that happened to get lucky
on our current targets.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99726 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-27 20:09:24 +00:00
Chris Lattner
0be6fe7139 continue pushing tblgen's support for nodes with multiple
results forward.  We can now handle an instruction that 
produces one implicit def and one result instead of one or
the other when not at the root of the pattern.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99725 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-27 19:15:02 +00:00
Chris Lattner
084df627c8 add plumbing for handling multiple result nodes
in some more places.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99366 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-24 00:41:19 +00:00
Chris Lattner
d734919465 major surgery on tblgen: generalize TreePatternNode
to maintain a list of types (one for each result of
the node) instead of a single type.  There are liberal
hacks added to emulate the old behavior in various
situations, but they can start disolving now.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98999 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-19 21:37:09 +00:00
Chris Lattner
1e50631675 resolve fixme: we now infer the instruction-level 'isvariadic' bit
from the pattern if present, and we use it instead of the bit.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98938 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-19 05:34:15 +00:00
Chris Lattner
e8cabf3c2e add a new SDNPVariadic SDNP node flag, and use it in
dag isel gen instead of instruction properties.  This
allows the oh-so-useful behavior of matching a variadic
non-root node.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98934 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-19 05:07:09 +00:00
Chris Lattner
f30187a363 look up instructions by record, not by name.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98904 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-19 00:07:20 +00:00
Chris Lattner
928795387f expand tblgen's support for instructions with implicit defs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98900 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-18 23:57:40 +00:00
Chris Lattner
2cacec55f9 Completely rewrite tblgen's type inference mechanism,
changing the primary datastructure from being a 
"std::vector<unsigned char>" to being a new TypeSet class
that actually has (gasp) invariants!

This changes more things than I remember, but one major
innovation here is that it enforces that named input 
values agree in type with their output values.

This also eliminates code that transparently assumes (in 
some cases) that SDNodeXForm input/output types are the
same, because this is wrong in many case.

This also eliminates a bug which caused a lot of ambiguous
patterns to go undetected, where a register class would
sometimes pick the first possible type, causing an
ambiguous pattern to get arbitrary results.

With all the recent target changes, this causes no 
functionality change!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98534 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-15 06:00:16 +00:00
Chris Lattner
6fd326b7ff more factoring.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97911 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-07 07:20:49 +00:00
Chris Lattner
6b4ae71725 zap fixme.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97705 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-04 01:26:00 +00:00
Chris Lattner
9752fb10a4 now that complexpatterns are all emitted at the end of the match
sequence, just emit instruction predicates right before them.  This
exposes yet more factoring opportunitites, shrinking the X86 table 
to 79144 bytes.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97704 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-04 01:25:36 +00:00
Chris Lattner
57bf8a483e change the new isel matcher to emit ComplexPattern matches
as the very last thing before node emission.  This should
dramatically reduce the number of times we do 'MatchAddress'
on X86, speeding up compile time.  This also improves comments
in the tables and shrinks the table a bit, now down to 
80506 bytes for x86.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97703 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-04 01:23:08 +00:00
Chris Lattner
d1aca7c8b4 enhance comment output to specify what recorded slot
numbers a ComplexPat will match into.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97696 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-04 00:28:05 +00:00
Chris Lattner
c6d7ad3c7d Rewrite chain handling validation and input TokenFactor handling
stuff now that we don't care about emulating the old broken 
behavior of the old isel.  This eliminates the 
'CheckChainCompatible' check (along with IsChainCompatible) which
did an incorrect and inefficient scan *up* the chain nodes which
happened as the pattern was being formed and does the validation
at the end in HandleMergeInputChains when it forms a structural 
pattern.  This scans "down" the graph, which means that it is
quickly bounded by nodes already selected.  This also handles
token factors that get "trapped" in the dag.

Removing the CheckChainCompatible nodes also shrinks the 
generated tables by about 6K for X86 (down to 83K).

There are two pieces remaining before I can nuke PreprocessRMW:
1. I xfailed a test because we're now producing worse code in a 
   case that has nothing to do with the change: it turns out that
   our use of MorphNodeTo will leave dead nodes in the graph
   which (depending on how the graph is walked) end up causing
   bogus uses of chains and blocking matches.  This is really 
   bad for other reasons, so I'll fix this in a follow-up patch.

2. CheckFoldableChainNode needs to be improved to handle the TF.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97539 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-02 02:22:10 +00:00
Chris Lattner
67ea6a7111 resolve some fixmes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97515 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-01 22:49:06 +00:00
Chris Lattner
01bcd947bf resolve a fixme and simplify code by moving insertion of the
EmitMergeInputChainsMatcher node up into EmitResultCode.  This
doesn't have much of an effect on the generated code, the X86
table is exactly the same size.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97514 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-01 22:46:42 +00:00