Introduce llvm.loop.parallel_accesses and llvm.access.group metadata.
The current llvm.mem.parallel_loop_access metadata has a problem in that
it uses LoopIDs. LoopID unfortunately is not loop identifier. It is
neither unique (there's even a regression test assigning the some LoopID
to multiple loops; can otherwise happen if passes such as LoopVersioning
make copies of entire loops) nor persistent (every time a property is
removed/added from a LoopID's MDNode, it will also receive a new LoopID;
this happens e.g. when calling Loop::setLoopAlreadyUnrolled()).
Since most loop transformation passes change the loop attributes (even
if it just to mark that a loop should not be processed again as
llvm.loop.isvectorized does, for the versioned and unversioned loop),
the parallel access information is lost for any subsequent pass.
This patch unlinks LoopIDs and parallel accesses.
llvm.mem.parallel_loop_access metadata on instruction is replaced by
llvm.access.group metadata. llvm.access.group points to a distinct
MDNode with no operands (avoiding the problem to ever need to add/remove
operands), called "access group". Alternatively, it can point to a list
of access groups. The LoopID then has an attribute
llvm.loop.parallel_accesses with all the access groups that are parallel
(no dependencies carries by this loop).
This intentionally avoid any kind of "ID". Loops that are clones/have
their attributes modifies retain the llvm.loop.parallel_accesses
attribute. Access instructions that a cloned point to the same access
group. It is not necessary for each access to have it's own "ID" MDNode,
but those memory access instructions with the same behavior can be
grouped together.
The behavior of llvm.mem.parallel_loop_access is not changed by this
patch, but should be considered deprecated.
Differential Revision: https://reviews.llvm.org/D52116
llvm-svn: 349725
diff --git a/llvm/lib/Transforms/Scalar/GVNHoist.cpp b/llvm/lib/Transforms/Scalar/GVNHoist.cpp
index 0797ce9..76a42d7 100644
--- a/llvm/lib/Transforms/Scalar/GVNHoist.cpp
+++ b/llvm/lib/Transforms/Scalar/GVNHoist.cpp
@@ -246,7 +246,7 @@
LLVMContext::MD_tbaa, LLVMContext::MD_alias_scope,
LLVMContext::MD_noalias, LLVMContext::MD_range,
LLVMContext::MD_fpmath, LLVMContext::MD_invariant_load,
- LLVMContext::MD_invariant_group};
+ LLVMContext::MD_invariant_group, LLVMContext::MD_access_group};
combineMetadata(ReplInst, I, KnownIDs, true);
}
diff --git a/llvm/lib/Transforms/Scalar/LoopVersioningLICM.cpp b/llvm/lib/Transforms/Scalar/LoopVersioningLICM.cpp
index c0c59d2..f48d3cc 100644
--- a/llvm/lib/Transforms/Scalar/LoopVersioningLICM.cpp
+++ b/llvm/lib/Transforms/Scalar/LoopVersioningLICM.cpp
@@ -633,6 +633,8 @@
// Set Loop Versioning metaData for version loop.
addStringMetadataToLoop(LVer.getVersionedLoop(), LICMVersioningMetaData);
// Set "llvm.mem.parallel_loop_access" metaData to versioned loop.
+ // FIXME: "llvm.mem.parallel_loop_access" annotates memory access
+ // instructions, not loops.
addStringMetadataToLoop(LVer.getVersionedLoop(),
"llvm.mem.parallel_loop_access");
// Update version loop with aggressive aliasing assumption.
diff --git a/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp b/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
index 8756a1a..e93f56da 100644
--- a/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
+++ b/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
@@ -996,7 +996,8 @@
// handled here, but combineMetadata doesn't support them yet
unsigned KnownIDs[] = {LLVMContext::MD_tbaa, LLVMContext::MD_alias_scope,
LLVMContext::MD_noalias,
- LLVMContext::MD_invariant_group};
+ LLVMContext::MD_invariant_group,
+ LLVMContext::MD_access_group};
combineMetadata(C, cpy, KnownIDs, true);
// Remove the memcpy.
diff --git a/llvm/lib/Transforms/Scalar/SROA.cpp b/llvm/lib/Transforms/Scalar/SROA.cpp
index 81eea0d..d1a2dc0 100644
--- a/llvm/lib/Transforms/Scalar/SROA.cpp
+++ b/llvm/lib/Transforms/Scalar/SROA.cpp
@@ -2593,7 +2593,8 @@
}
V = convertValue(DL, IRB, V, NewAllocaTy);
StoreInst *Store = IRB.CreateAlignedStore(V, &NewAI, NewAI.getAlignment());
- Store->copyMetadata(SI, LLVMContext::MD_mem_parallel_loop_access);
+ Store->copyMetadata(SI, {LLVMContext::MD_mem_parallel_loop_access,
+ LLVMContext::MD_access_group});
if (AATags)
Store->setAAMetadata(AATags);
Pass.DeadInsts.insert(&SI);
@@ -2662,7 +2663,8 @@
NewSI = IRB.CreateAlignedStore(V, NewPtr, getSliceAlign(V->getType()),
SI.isVolatile());
}
- NewSI->copyMetadata(SI, LLVMContext::MD_mem_parallel_loop_access);
+ NewSI->copyMetadata(SI, {LLVMContext::MD_mem_parallel_loop_access,
+ LLVMContext::MD_access_group});
if (AATags)
NewSI->setAAMetadata(AATags);
if (SI.isVolatile())
@@ -3799,7 +3801,8 @@
PartPtrTy, BasePtr->getName() + "."),
getAdjustedAlignment(LI, PartOffset, DL), /*IsVolatile*/ false,
LI->getName());
- PLoad->copyMetadata(*LI, LLVMContext::MD_mem_parallel_loop_access);
+ PLoad->copyMetadata(*LI, {LLVMContext::MD_mem_parallel_loop_access,
+ LLVMContext::MD_access_group});
// Append this load onto the list of split loads so we can find it later
// to rewrite the stores.
@@ -3855,7 +3858,8 @@
APInt(DL.getIndexSizeInBits(AS), PartOffset),
PartPtrTy, StoreBasePtr->getName() + "."),
getAdjustedAlignment(SI, PartOffset, DL), /*IsVolatile*/ false);
- PStore->copyMetadata(*LI, LLVMContext::MD_mem_parallel_loop_access);
+ PStore->copyMetadata(*LI, {LLVMContext::MD_mem_parallel_loop_access,
+ LLVMContext::MD_access_group});
LLVM_DEBUG(dbgs() << " +" << PartOffset << ":" << *PStore << "\n");
}
diff --git a/llvm/lib/Transforms/Scalar/Scalarizer.cpp b/llvm/lib/Transforms/Scalar/Scalarizer.cpp
index 3a6f8e6..5eb3fda 100644
--- a/llvm/lib/Transforms/Scalar/Scalarizer.cpp
+++ b/llvm/lib/Transforms/Scalar/Scalarizer.cpp
@@ -379,7 +379,8 @@
|| Tag == LLVMContext::MD_invariant_load
|| Tag == LLVMContext::MD_alias_scope
|| Tag == LLVMContext::MD_noalias
- || Tag == ParallelLoopAccessMDKind);
+ || Tag == ParallelLoopAccessMDKind
+ || Tag == LLVMContext::MD_access_group);
}
// Transfer metadata from Op to the instructions in CV if it is known