llvm dev - Apr 2012 - [LLVMdev] Should repetitive basic blocks be removed in the results of LoopBase::getExitBlocks()?

On Apr 23, 2012, at 10:31 PM, Andrew Trick <atrick at apple.com> wrote:
> 
> On Apr 19, 2012, at 10:43 PM, xiaoming gu <xiaoming.gu at gmail.com>
wrote:
> 
>> Hi, all. I'm using void LoopBase::getExitBlocks
(SmallVectorImpl< BlockT * > &ExitBlocks) const to get all
>> exit blocks for a loop. The problem I find with this API is that it
returns repetitive basic blocks in certain
>> situations. Should repetitive basic blocks be removed?
> 
> Users generally expect a unique set of exit blocks, but don't make any
strong assumption. The worst that can happen is missed optimization or redundant
analysis. In most cases, the loop is in LoopSimplifyForm, so it's probably
not a problem in practice.
Another thing I should mention. The iteration order of ExitBlocks is important.
In llvm, generating unique sets is a pain because values are not numbered and
iteration order needs to be reproducible. We would need to keep a SmallPtrSet
for membership checking while populating the result vector.

-Andy
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20120423/74bcabc8/attachment.html>

hi,

> Users generally expect a unique set of exit blocks, but don't make any
> strong assumption. The worst that can happen is missed optimization or
> redundant analysis. In most cases, the loop is in LoopSimplifyForm, so
it's
> probably not a problem in practice.
>
>
> Another thing I should mention. The iteration order of ExitBlocks is
> important. In llvm, generating unique sets is a pain because values are not
> numbered and iteration order needs to be reproducible. We would need to
keep
> a SmallPtrSet for membership checking while populating the result vector.
What about use the SetVector[1] as container? As its comment say:

This adapter class provides a way to keep a set of things that also
has the property of a deterministic iteration order. The order of
iteration is the order of insertion.

best regards
ether
[1]http://www.llvm.org/doxygen/classllvm_1_1SetVector.html
>
> -Andy
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>

On Apr 23, 2012, at 11:22 PM, Hongbin Zheng <etherzhhb at gmail.com>
wrote:
> hi,
> 
> 
>> Users generally expect a unique set of exit blocks, but don't make
any
>> strong assumption. The worst that can happen is missed optimization or
>> redundant analysis. In most cases, the loop is in LoopSimplifyForm, so
it's
>> probably not a problem in practice.
>> 
>> 
>> Another thing I should mention. The iteration order of ExitBlocks is
>> important. In llvm, generating unique sets is a pain because values are
not
>> numbered and iteration order needs to be reproducible. We would need to
keep
>> a SmallPtrSet for membership checking while populating the result
vector.
> What about use the SetVector[1] as container? As its comment say:
> 
> This adapter class provides a way to keep a set of things that also
> has the property of a deterministic iteration order. The order of
> iteration is the order of insertion.
Sure, you can wrap the SmallPtrSet within the SetVector. You need to change the
API then.
-Andy