[LLVMdev] Fwd: Regarding scope information for variable declaration.

[David Blaikie]

On Mon, Jul 15, 2013 at 2:40 AM, Eric Lu <eirc.lew at gmail.com> wrote:
>
> Thank your reply. Pankaj.
>
> Actually, I have done it very similar to yours. But I think for my demand,
> it is better to implement in Front End. Maybe I will re-implement it later
> in clang.

Depends what you're trying to do with this scope information... if
it's purely for optimization purposes then it should just be in LLVM,
with the exception of something like LLVM's lifetime intrinsics, used
as a hint from the frontend.

> ---------- Forwarded message ----------
> From: Pankaj Gode [via LLVM] <ml-node+s1065342n59345h22 at n5.nabble.com>
> Date: Mon, Jul 15, 2013 at 2:35 PM
> Subject: Re: Regarding scope information for variable declaration.
> To: "eric.lew" <eirc.lew at gmail.com>
>
>
> Hi Eric,
>
> I was considering machine instructions to get scope information. And
> variable declaration does not correspond to machine instruction, hence the
> problem i.e. no scope associated with it.
> If 'i' is initialized in the 'if-scope' then we get 'variable i' mapped to
> correct scope as corresponding machine instruction is generated for this.
> This is not a problem as we can't expect variable declaration in a machine
> instruction, I thought.
>
> S
> o instead of using machine instructions to collect scope information, (as
> used by LexicalScope pass),
> I had written code to collect scope information based on LLVM Instructions.
> I did this by iterating over 'Function->BasicBlock' instead of
> 'MachineFunction->MachineBasicBlock'.
> const Function *F1 =  MF->getFunction();
> for(Function::const_iterator BB = F1->begin(), E = F1->end();
>                                 BB != E; ++BB)
>   {
>      for(BasicBlock::const_iterator ii = BB->begin(), ie = BB->end();
>         ii != ie; ++ii)
>      {
>         const Instruction *I = ii;  //I->dump();//debug
>         DebugLoc MIDB = I->getDebugLoc();
>       }
>    }
> Though this is an overhead as scope information exists,
> but I need to collect specific information such as 'start line, end line,
> start column, end column'
> (End line information should be derived as is not obvious).
> Collecting information this way allowed me to get correct scope information,
> and hence I was able to map the variable declaration to the scope. It worked
> for me this way.
>
>
> Regards,
> Pankaj
>
>
> From: eric.lew <[hidden email]>
> To: [hidden email]
> Sent: Friday, July 12, 2013 8:30 AM
> Subject: Re: [LLVMdev] Regarding scope information for variable declaration.
>
> I have the same demand. Have you resolved this problems? if so, would you
> share me the solution?
>
> Best Regards.
>
> Eric
>
>
>
> --
> View this message in context:
> http://llvm.1065342.n5.nabble.com/Regarding-scope-information-for-variable-declaration-tp47707p59268.html
> Sent from the LLVM - Dev mailing list archive at Nabble.com.
> _______________________________________________
> LLVM Developers mailing list
> [hidden email]        http://llvm.cs.uiuc.edu/
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>
>
>
> _______________________________________________
> LLVM Developers mailing list
> [hidden email]         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>
>
> ________________________________
> If you reply to this email, your message will be added to the discussion
> below:
> http://llvm.1065342.n5.nabble.com/Regarding-scope-information-for-variable-declaration-tp47707p59345.html
> To unsubscribe from Regarding scope information for variable declaration.,
> click here.
> NAML
>
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>