[llvm] r256028 - Enhance BranchProbabilityInfo::calcUnreachableHeuristics for InvokeInst
Jun Bum Lim via llvm-commits
llvm-commits at lists.llvm.org
Fri Dec 18 13:30:42 PST 2015
Thanks for reverting this for me. I will fix it soon.
-----Original Message-----
From: Rafael EspĂndola [mailto:rafael.espindola at gmail.com]
Sent: Friday, December 18, 2015 4:28 PM
To: Jun Bum Lim
Cc: llvm-commits
Subject: Re: [llvm] r256028 - Enhance BranchProbabilityInfo::calcUnreachableHeuristics for InvokeInst
Sorry, reverted in 256032 since it broke some tests.
Cheers,
Rafael
On 18 December 2015 at 15:53, Jun Bum Lim via llvm-commits <llvm-commits at lists.llvm.org> wrote:
> Author: junbuml
> Date: Fri Dec 18 14:53:47 2015
> New Revision: 256028
>
> URL: http://llvm.org/viewvc/llvm-project?rev=256028&view=rev
> Log:
> Enhance BranchProbabilityInfo::calcUnreachableHeuristics for
> InvokeInst
>
> When identifying blocks post-dominated by an unreachable-terminated
> block in BranchProbabilityInfo, consider only the edge to the normal
> destination block if the terminator is InvokeInst and let
> calcInvokeHeuristics() decide edge weights for the InvokeInst.
>
> Modified:
> llvm/trunk/lib/Analysis/BranchProbabilityInfo.cpp
> llvm/trunk/test/Analysis/BranchProbabilityInfo/noreturn.ll
> llvm/trunk/test/CodeGen/X86/catchret-empty-fallthrough.ll
>
> Modified: llvm/trunk/lib/Analysis/BranchProbabilityInfo.cpp
> URL:
> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Analysis/BranchProb
> abilityInfo.cpp?rev=256028&r1=256027&r2=256028&view=diff
> ======================================================================
> ========
> --- llvm/trunk/lib/Analysis/BranchProbabilityInfo.cpp (original)
> +++ llvm/trunk/lib/Analysis/BranchProbabilityInfo.cpp Fri Dec 18
> +++ 14:53:47 2015
> @@ -147,6 +147,16 @@ bool BranchProbabilityInfo::calcUnreacha
> if (TI->getNumSuccessors() == 1 || UnreachableEdges.empty())
> return false;
>
> + // If the terminator is an InvokeInst, check only the normal
> + destination block // as the unwind edge of InvokeInst is also very unlikely taken.
> + if (auto *II = dyn_cast<InvokeInst>(TI))
> + if (PostDominatedByUnreachable.count(II->getNormalDest())) {
> + PostDominatedByUnreachable.insert(BB);
> + // Return false here so that edge weights for InvokeInst could be decided
> + // in calcInvokeHeuristics().
> + return false;
> + }
> +
> uint32_t UnreachableWeight =
> std::max(UR_TAKEN_WEIGHT / (unsigned)UnreachableEdges.size(), MIN_WEIGHT);
> for (SmallVectorImpl<unsigned>::iterator I =
> UnreachableEdges.begin(),
>
> Modified: llvm/trunk/test/Analysis/BranchProbabilityInfo/noreturn.ll
> URL:
> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Analysis/BranchPro
> babilityInfo/noreturn.ll?rev=256028&r1=256027&r2=256028&view=diff
> ======================================================================
> ========
> --- llvm/trunk/test/Analysis/BranchProbabilityInfo/noreturn.ll
> (original)
> +++ llvm/trunk/test/Analysis/BranchProbabilityInfo/noreturn.ll Fri Dec
> +++ 18 14:53:47 2015
> @@ -77,3 +77,48 @@ abort:
> exit:
> ret i32 %b
> }
> +
> + at _ZTIi = external global i8*
> +
> +; CHECK-LABEL: throwSmallException
> +; CHECK-NOT: invoke i32 @smallFunction define i32
> + at throwSmallException(i32 %idx, i32 %limit) #0 personality i8* bitcast
> +(i32 (...)* @__gxx_personality_v0 to i8*) {
> +entry:
> + %cmp = icmp sge i32 %idx, %limit
> + br i1 %cmp, label %if.then, label %if.end ; CHECK: edge entry ->
> +if.then probability is 0x00000800 / 0x80000000 = 0.00% ; CHECK: edge
> +entry -> if.end probability is 0x7ffff800 / 0x80000000 = 100.00% [HOT
> +edge]
> +
> +if.then: ; preds = %entry
> + %exception = call i8* @__cxa_allocate_exception(i64 1) #0
> + invoke i32 @smallFunction(i32 %idx)
> + to label %invoke.cont unwind label %lpad ; CHECK: edge
> +if.then -> invoke.cont probability is 0x7ffff800 / 0x80000000 =
> +100.00% [HOT edge] ; CHECK: edge if.then -> lpad probability is
> +0x00000800 / 0x80000000 = 0.00%
> +
> +invoke.cont: ; preds = %if.then
> + call void @__cxa_throw(i8* %exception, i8* bitcast (i8** @_ZTIi to
> +i8*), i8* null) #1
> + unreachable
> +
> +lpad: ; preds = %if.then
> + %ll = landingpad { i8*, i32 }
> + cleanup
> + ret i32 %idx
> +
> +if.end: ; preds = %entry
> + ret i32 %idx
> +}
> +
> + at a = global i32 4
> +define i32 @smallFunction(i32 %a) {
> +entry:
> + %r = load volatile i32, i32* @a
> + ret i32 %r
> +}
> +
> +attributes #0 = { nounwind }
> +attributes #1 = { noreturn }
> +
> +declare i8* @__cxa_allocate_exception(i64) declare i32
> + at __gxx_personality_v0(...) declare void @__cxa_throw(i8*, i8*, i8*)
>
> Modified: llvm/trunk/test/CodeGen/X86/catchret-empty-fallthrough.ll
> URL:
> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/catchr
> et-empty-fallthrough.ll?rev=256028&r1=256027&r2=256028&view=diff
> ======================================================================
> ========
> --- llvm/trunk/test/CodeGen/X86/catchret-empty-fallthrough.ll
> (original)
> +++ llvm/trunk/test/CodeGen/X86/catchret-empty-fallthrough.ll Fri Dec
> +++ 18 14:53:47 2015
> @@ -34,12 +34,14 @@ return:
>
> ; CHECK-LABEL: foo: # @foo
> ; CHECK: testb $1, %cl
> -; CHECK: jne .LBB0_[[return:[0-9]+]]
> +; CHECK: je .LBB0_[[try:[0-9]+]]
> +; CHECK: .LBB0_[[return:[0-9]+]]:
> +; CHECK: retq
> +; CHECK: .LBB0_[[try]]:
> ; CHECK: .Ltmp0:
> ; CHECK: callq bar
> ; CHECK: .Ltmp1:
> ; CHECK: .LBB0_[[catch:[0-9]+]]:
> -; CHECK: .LBB0_[[return]]:
>
> ; CHECK: .seh_handlerdata
> ; CHECK-NEXT: .Lfoo$parent_frame_offset = 32
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
More information about the llvm-commits
mailing list