[llvm] r244821 - [RewriteStatepointsForGC] Avoid using unrelocated pointers after safepoints

Philip Reames via llvm-commits llvm-commits at lists.llvm.org
Wed Aug 12 15:11:46 PDT 2015


Author: reames
Date: Wed Aug 12 17:11:45 2015
New Revision: 244821

URL: http://llvm.org/viewvc/llvm-project?rev=244821&view=rev
Log:
[RewriteStatepointsForGC] Avoid using unrelocated pointers after safepoints

To be clear: this is an *optimization* not a correctness change.

CodeGenPrep likes to duplicate icmps feeding branch instructions to take advantage of x86's ability to fuze many comparison/branch patterns into a single micro-op and to reduce the need for materializing i1s into general registers. PlaceSafepoints likes to place safepoint polls right at the end of basic blocks (immediately before terminators) when inserting entry and backedge safepoints. These two heuristics interact in a somewhat unfortunate way where the branch terminating the original block will be controlled by a condition driven by unrelocated pointers. This forces the register allocator to keep both the relocated and unrelocated values of the pointers feeding the icmp alive over the safepoint poll.

One simple fix would have been to just adjust PlaceSafepoints to move one back in the basic block, but you can reach similar cases as a result of LICM or other hoisting passes. As a result, doing a post insertion fixup seems to be more robust.

I considered doing this in CodeGenPrep itself, but having to update the live sets of already rewritten safepoints gets complicated fast. In particular, you can't just use def/use information because by moving the icmp, we're extending the live range of it's inputs potentially.

Instead, this patch teaches RewriteStatepointsForGC to make the required adjustments before making the relocations explicit in the IR. This change really highlights the fact that RSForGC is a CodeGenPrep-like pass which is performing target specific lowering. In the long run, we may even want to combine the two though this would require a lot more smarts to be integrated into RSForGC first. We currently rely on being able to run a set of cleanup passes post rewriting because the IR RSForGC generates is pretty damn ugly.

Differential Revision: http://reviews.llvm.org/D11819


Added:
    llvm/trunk/test/Transforms/RewriteStatepointsForGC/codegen-cond.ll
Modified:
    llvm/trunk/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp

Modified: llvm/trunk/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp?rev=244821&r1=244820&r2=244821&view=diff
==============================================================================
--- llvm/trunk/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp (original)
+++ llvm/trunk/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp Wed Aug 12 17:11:45 2015
@@ -2452,6 +2452,37 @@ bool RewriteStatepointsForGC::runOnFunct
       FoldSingleEntryPHINodes(&BB);
     }
 
+  // Before we start introducing relocations, we want to tweak the IR a bit to
+  // avoid unfortunate code generation effects.  The main example is that we 
+  // want to try to make sure the comparison feeding a branch is after any
+  // safepoints.  Otherwise, we end up with a comparison of pre-relocation
+  // values feeding a branch after relocation.  This is semantically correct,
+  // but results in extra register pressure since both the pre-relocation and
+  // post-relocation copies must be available in registers.  For code without
+  // relocations this is handled elsewhere, but teaching the scheduler to
+  // reverse the transform we're about to do would be slightly complex.
+  // Note: This may extend the live range of the inputs to the icmp and thus
+  // increase the liveset of any statepoint we move over.  This is profitable
+  // as long as all statepoints are in rare blocks.  If we had in-register
+  // lowering for live values this would be a much safer transform.
+  auto getConditionInst = [](TerminatorInst *TI) -> Instruction* {
+    if (auto *BI = dyn_cast<BranchInst>(TI))
+      if (BI->isConditional())
+        return dyn_cast<Instruction>(BI->getCondition());
+    // TODO: Extend this to handle switches
+    return nullptr;
+  };
+  for (BasicBlock &BB : F) {
+    TerminatorInst *TI = BB.getTerminator();
+    if (auto *Cond = getConditionInst(TI))
+      // TODO: Handle more than just ICmps here.  We should be able to move
+      // most instructions without side effects or memory access.  
+      if (isa<ICmpInst>(Cond) && Cond->hasOneUse()) {
+        MadeChange = true;
+        Cond->moveBefore(TI);
+      }
+  }
+
   MadeChange |= insertParsePoints(F, DT, this, ParsePointNeeded);
   return MadeChange;
 }

Added: llvm/trunk/test/Transforms/RewriteStatepointsForGC/codegen-cond.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/RewriteStatepointsForGC/codegen-cond.ll?rev=244821&view=auto
==============================================================================
--- llvm/trunk/test/Transforms/RewriteStatepointsForGC/codegen-cond.ll (added)
+++ llvm/trunk/test/Transforms/RewriteStatepointsForGC/codegen-cond.ll Wed Aug 12 17:11:45 2015
@@ -0,0 +1,74 @@
+; RUN: opt -rewrite-statepoints-for-gc -S < %s | FileCheck %s
+
+; A null test of a single value
+define i1 @test(i8 addrspace(1)* %p, i1 %rare) gc "statepoint-example" {
+; CHECK-LABEL: @test
+entry:
+   %cond = icmp eq i8 addrspace(1)* %p, null
+   br i1 %rare, label %safepoint, label %continue, !prof !0
+safepoint:
+   call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 2882400000, i32 0, void ()* @safepoint, i32 0, i32 0, i32 0, i32 0)
+   br label %continue
+continue:
+; CHECK-LABEL: continue:
+; CHECK: phi
+; CHECK-DAG: [ %p.relocated, %safepoint ]
+; CHECK-DAG [ %p, %entry ]
+; CHECK: %cond = icmp
+; CHECK: br i1 %cond
+   br i1 %cond, label %taken, label %untaken
+taken:
+   ret i1 true
+untaken:
+   ret i1 false
+}
+
+; Comparing two pointers
+define i1 @test2(i8 addrspace(1)* %p, i8 addrspace(1)* %q, i1 %rare) 
+    gc "statepoint-example" {
+; CHECK-LABEL: @test2
+entry:   
+   %cond = icmp eq i8 addrspace(1)* %p, %q
+   br i1 %rare, label %safepoint, label %continue, !prof !0
+safepoint:
+   call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 2882400000, i32 0, void ()* @safepoint, i32 0, i32 0, i32 0, i32 0)
+   br label %continue
+continue:
+; CHECK-LABEL: continue:
+; CHECK: phi
+; CHECK-DAG: [ %q.relocated, %safepoint ]
+; CHECK-DAG [ %q, %entry ]
+; CHECK: phi
+; CHECK-DAG: [ %p.relocated, %safepoint ]
+; CHECK-DAG [ %p, %entry ]
+; CHECK: %cond = icmp
+; CHECK: br i1 %cond
+   br i1 %cond, label %taken, label %untaken
+taken:
+   ret i1 true
+untaken:
+   ret i1 false
+}
+
+; Sanity check that nothing bad happens if already last instruction
+; before terminator
+define i1 @test3(i8 addrspace(1)* %p, i8 addrspace(1)* %q, i1 %rare) 
+    gc "statepoint-example" {
+; CHECK-LABEL: @test3
+entry:   
+   call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 2882400000, i32 0, void ()* @safepoint, i32 0, i32 0, i32 0, i32 0)
+; CHECK: gc.statepoint
+; CHECK: %cond = icmp
+; CHECK: br i1 %cond
+   %cond = icmp eq i8 addrspace(1)* %p, %q
+   br i1 %cond, label %taken, label %untaken
+taken:
+   ret i1 true
+untaken:
+   ret i1 false
+}
+
+declare void @safepoint()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
+
+!0 = !{!"branch_weights", i32 1, i32 10000}




More information about the llvm-commits mailing list