Commit Graph

61 Commits

Author SHA1 Message Date
Tom Stellard
8d56c59235 R600: Implement TargetInstrInfo::isLegalToSplitMBBAt()
Splitting a basic block will create a new ALU clause, so we need to make
sure we aren't moving uses of registers that are local to their
current clause into a new one.

I had a test case for this, but unfortunately unrelated schedule changes
invalidated it, and I wasn't been able to come up with another one.

NOTE: This is a candidate for the 3.4 branch.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195399 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-22 00:41:08 +00:00
Juergen Ributzka
354362524a [weak vtables] Remove a bunch of weak vtables
This patch removes most of the trivial cases of weak vtables by pinning them to
a single object file. The memory leaks in this version have been fixed. Thanks
Alexey for pointing them out.

Differential Revision: http://llvm-reviews.chandlerc.com/D2068

Reviewed by Andy

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195064 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-19 00:57:56 +00:00
Alexey Samsonov
b21ab43cfc Revert r194865 and r194874.
This change is incorrect. If you delete virtual destructor of both a base class
and a subclass, then the following code:
  Base *foo = new Child();
  delete foo;
will not cause the destructor for members of Child class. As a result, I observe
plently of memory leaks. Notable examples I investigated are:
ObjectBuffer and ObjectBufferStream, AttributeImpl and StringSAttributeImpl.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194997 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-18 09:31:53 +00:00
Vincent Lejeune
4110797853 R600: Make dot_4 instructions predicable
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194927 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-16 16:24:41 +00:00
Juergen Ributzka
5a364c5561 [weak vtables] Remove a bunch of weak vtables
This patch removes most of the trivial cases of weak vtables by pinning them to
a single object file.

Differential Revision: http://llvm-reviews.chandlerc.com/D2068

Reviewed by Andy

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194865 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-15 22:34:48 +00:00
Tom Stellard
19a99df130 R600: Fix scheduling of instructions that use the LDS output queue
The LDS output queue is accessed via the OQAP register.  The OQAP
register cannot be live across clauses, so if value is written to the
output queue, it must be retrieved before the end of the clause.
With the machine scheduler, we cannot statisfy this constraint, because
it lacks proper alias analysis and it will mark some LDS accesses as
having a chain dependency on vertex fetches.  Since vertex fetches
require a new clauses, the dependency may end up spiltting OQAP uses and
defs so the end up in different clauses.  See the lds-output-queue.ll
test for a more detailed explanation.

To work around this issue, we now combine the LDS read and the OQAP
copy into one instruction and expand it after register allocation.

This patch also adds some checks to the EmitClauseMarker pass, so that
it doesn't end a clause with a value still in the output queue and
removes AR.X and OQAP handling from the scheduler (AR.X uses and defs
were already being expanded post-RA, so the scheduler will never see
them).

Reviewed-by: Vincent Lejeune <vljn at ovi.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194755 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-15 00:12:45 +00:00
Tom Stellard
a2b4eb6d15 R600/SI: Add support for private address space load/store
Private address space is emulated using the register file with
MOVRELS and MOVRELD instructions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194626 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-13 23:36:50 +00:00
Tom Stellard
04c559569f R600: Simplify handling of private address space
The AMDGPUIndirectAddressing pass was previously responsible for
lowering private loads and stores to indirect addressing instructions.
However, this pass was buggy and way too complicated.  The only
advantage it had over the new simplified code was that it saved one
instruction per direct write to private memory.  This optimization
likely has a minimal impact on performance, and we may be able
to duplicate it using some other transformation.

For the private address space, we now:
1. Lower private loads/store to Register(Load|Store) instructions
2. Reserve part of the register file as 'private memory'
3. After regalloc lower the Register(Load|Store) instructions to
   MOV instructions that use indirect addressing.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193179 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-22 18:19:10 +00:00
Tom Stellard
34adeaf8b9 R600: Remove unused InstrInfo::getMovImmInstr() function
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193178 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-22 18:19:01 +00:00
Vincent Lejeune
dfef7cbfc6 R600: add a pass that merges clauses.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191790 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-01 19:32:58 +00:00
Vincent Lejeune
5b00e833fa R600: Enable -verify-machineinstrs in some tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191788 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-01 19:32:38 +00:00
Arnold Schwaighofer
d42730dc71 IfConverter: Use TargetSchedule for instruction latencies
For targets that have instruction itineraries this means no change. Targets
that move over to the new schedule model will use be able the new schedule
module for instruction latencies in the if-converter (the logic is such that if
there is no itineary we will use the new sched model for the latencies).

Before, we queried "TTI->getInstructionLatency()" for the instruction latency
and the extra prediction cost. Now, we query the TargetSchedule abstraction for
the instruction latency and TargetInstrInfo for the extra predictation cost. The
TargetSchedule abstraction will internally call "TTI->getInstructionLatency" if
an itinerary exists, otherwise it will use the new schedule model.

ATTENTION: Out of tree targets!

(I will also send out an email later to LLVMDev)

This means, if your target implements

 unsigned getInstrLatency(const InstrItineraryData *ItinData,
                          const MachineInstr *MI,
                          unsigned *PredCost);

and returns a value for "PredCost", you now also need to implement

 unsigned getPredictationCost(const MachineInstr *MI);

(if your target uses the IfConversion.cpp pass)

radar://15077010

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191671 91177308-0d34-0410-b5e6-96231b3b80d8
2013-09-30 15:28:56 +00:00
Tom Stellard
ac779b8494 R600: Don't use trans slot for instructions that read LDS source registers
This fixes some regressions in the piglit local memory store tests
introduced by recent commits which made the scheduler aware of the trans
slot.

It's not possible to test this using lit, because there is no way to
determine from the assembly dumps whether or not an instruction is in
the trans slot.

Even if this were possible, the test would be highly sensitive to
changes in the scheduler and might generate confusing false negatives.

Reviewed-by: Vincent Lejeune<vljn at ovi.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190574 91177308-0d34-0410-b5e6-96231b3b80d8
2013-09-12 02:55:06 +00:00
Vincent Lejeune
f94eea9e11 R600: Use shared op optimization when checking cycle compatibility
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189981 91177308-0d34-0410-b5e6-96231b3b80d8
2013-09-04 19:53:54 +00:00
Vincent Lejeune
bb25a01d23 R600: Non vector only instruction can be scheduled on trans unit
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189980 91177308-0d34-0410-b5e6-96231b3b80d8
2013-09-04 19:53:46 +00:00
Vincent Lejeune
b3df27d440 R600: Use SchedModel enum for is{Trans,Vector}Only functions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189979 91177308-0d34-0410-b5e6-96231b3b80d8
2013-09-04 19:53:30 +00:00
Tom Stellard
8e78012457 R600: Add support for i8 and i16 local memory stores
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189223 91177308-0d34-0410-b5e6-96231b3b80d8
2013-08-26 15:05:49 +00:00
Tom Stellard
e7ac2ed1c2 R600: Add IsExport bit to TableGen instruction definitions
Tested-by: Aaron Watry <awatry@gmail.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188516 91177308-0d34-0410-b5e6-96231b3b80d8
2013-08-16 01:11:51 +00:00
Tom Stellard
692ee102eb R600: Add 64-bit float load/store support
* Added R600_Reg64 class
* Added T#Index#.XY registers definition
* Added v2i32 register reads from parameter and global space
* Added f32 and i32 elements extraction from v2f32 and v2i32
* Added v2i32 -> v2f32 conversions

Tom Stellard:
  - Mark vec2 operations as expand.  The addition of a vec2 register
    class made them all legal.

Patch by: Dmitry Cherkassov

Signed-off-by: Dmitry Cherkassov <dcherkassov@gmail.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@187582 91177308-0d34-0410-b5e6-96231b3b80d8
2013-08-01 15:23:42 +00:00
Tom Stellard
6b3f6a744a Revert "R600: Non vector only instruction can be scheduled on trans unit"
This reverts commit 98ce62780e.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@187526 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-31 20:43:27 +00:00
Tom Stellard
af2ea2a4fb Revert "R600: Use SchedModel enum for is{Trans,Vector}Only functions"
This reverts commit 3f1de26cb5.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@187524 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-31 20:43:03 +00:00
Vincent Lejeune
8e37705a73 R600: Avoid more than 4 literals in the same instruction group at scheduling
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@187515 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-31 19:32:07 +00:00
Vincent Lejeune
98ce62780e R600: Non vector only instruction can be scheduled on trans unit
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@187514 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-31 19:31:56 +00:00
Vincent Lejeune
3f1de26cb5 R600: Use SchedModel enum for is{Trans,Vector}Only functions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@187512 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-31 19:31:35 +00:00
Tom Stellard
58d3335cb9 R600: Move CONST_ADDRESS folding into AMDGPUDAGToDAGISel::Select()
This increases the number of opportunites we have for folding.  With the
previous implementation we were unable to fold into any instructions
other than the first when multiple instructions were selected from a
single SDNode.

Reviewed-by: Vincent Lejeune <vljn at ovi.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@186919 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-23 01:48:24 +00:00
Craig Topper
a0ec3f9b7b Use SmallVectorImpl& instead of SmallVector to avoid repeating small vector size.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@186274 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-14 04:42:23 +00:00
Aaron Ballman
80095e5f8e Replacing an empty switch with its moral equivalent. No functional changes intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@186017 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-10 17:19:22 +00:00
Vincent Lejeune
f2cfef8172 R600: Do not predicated basic block with multiple alu clause
Test is not included as it is several 1000 lines long.
To test this functionnality, a test case must generate at least 2 ALU clauses,
where an ALU clause is ~110 instructions long.

NOTE: This is a candidate for the stable branch.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185943 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-09 15:03:33 +00:00
Vincent Lejeune
c36a8d2b3d R600: Fix an unitialized variable in R600InstrInfo.cpp
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185294 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-30 21:44:06 +00:00
Benjamin Kramer
97daabf318 R600: Unbreak GCC build.
operator++ on an enum is not legal. clang happens to accept it anyways, I think
that's a known bug.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185269 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-29 20:04:19 +00:00
Vincent Lejeune
8f9fbd67c3 R600: Support schedule and packetization of trans-only inst
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185268 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-29 19:32:43 +00:00
Vincent Lejeune
7d1a0d4e3e R600: Bank Swizzle now display SCL equivalent
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185267 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-29 19:32:29 +00:00
Tom Stellard
e3d4cbc7d2 R600: Add local memory support via LDS
Reviewed-by: Vincent Lejeune<vljn at ovi.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185162 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-28 15:47:08 +00:00
Tom Stellard
cedcfee405 R600: Add support for GROUP_BARRIER instruction
Reviewed-by: Vincent Lejeune<vljn at ovi.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185161 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-28 15:46:59 +00:00
Tom Stellard
7e9381951e R600: Add ALUInst bit to tablegen definitions v2
v2:
  - Remove functions left over from a previous rebase.

Reviewed-by: Vincent Lejeune<vljn at ovi.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185160 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-28 15:46:53 +00:00
Tom Stellard
5e48a0e9ae R600: Use new getNamedOperandIdx function generated by TableGen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@184880 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-25 21:22:18 +00:00
Vincent Lejeune
f7c9b95f94 R600: PV stores Reg id, not index
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@184117 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-17 20:16:40 +00:00
Tom Stellard
3ff0abfaab R600: Rework subtarget info and remove AMDILDevice classes
This should simplify the subtarget definitions and make it easier to
add new ones.

Reviewed-by: Vincent Lejeune <vljn@ovi.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183566 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-07 20:37:48 +00:00
Bill Wendling
b5632b5b45 Don't cache the instruction and register info from the TargetMachine, because
the internals of TargetMachine could change.

No functionality change intended.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183561 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-07 20:28:55 +00:00
Tom Stellard
ad7ecc65b1 R600: Make sure to schedule AR register uses and defs in the same clause
Reviewed-by: vljn at ovi.com

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183294 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-05 03:43:06 +00:00
Vincent Lejeune
e67a4afb5d R600: Const/Neg/Abs can be folded to dot4
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183278 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-04 23:17:15 +00:00
Benjamin Kramer
5c35290fa3 Move passes from namespace llvm into anonymous namespaces. Sort includes while there.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182594 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-23 17:10:37 +00:00
Vincent Lejeune
4ed9917147 R600: Relax some vector constraints on Dot4.
Dot4 now uses 8 scalar operands instead of 2 vectors one which allows register
coalescer to remove some unneeded COPY.
This patch also defines some structures/functions that can be used to handle
every vector instructions (CUBE, Cayman special instructions...) in a similar
fashion.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182126 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-17 16:50:32 +00:00
Vincent Lejeune
25c209e9a2 R600: Some factorization
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182123 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-17 16:50:02 +00:00
Tom Stellard
32c76107d0 R600: Remove dead code from the CodeEmitter v2
v2:
  - Replace switch statement with TSFlags query

Reviewed-by: Vincent Lejeune <vljn@ovi.com>
Tested-By: Aaron Watry <awatry@gmail.com>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181229 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-06 17:50:57 +00:00
Vincent Lejeune
4fb224e3bd R600: Always use texture cache for compute shaders
This will improve the performance of memory reads.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@180762 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-30 00:14:44 +00:00
Vincent Lejeune
25f259cde2 R600: Packetize instructions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@180760 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-30 00:14:27 +00:00
Vincent Lejeune
abcde265b1 R600: Rework Scheduling to handle difference between VLIW4 and VLIW5 chips
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@180759 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-30 00:14:17 +00:00
Vincent Lejeune
e332e3559b R600: Add a Bank Swizzle operand
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@180758 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-30 00:14:08 +00:00
Vincent Lejeune
631591e6f3 R600: Add FetchInst bit to instruction defs to denote vertex/tex instructions
v2[Vincent Lejeune]: Split FetchInst into usesTextureCache/usesVertexCache

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@180755 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-30 00:13:39 +00:00