[LLVMdev] which LLVM transforms can optimize this code?

[NAKAMURA Takumi]

In principle, transform passes are not responsible to inspect @yyy is
not aliased to @XXX.

I guess "opt -basicaa -{any-transform-passes}" would help you.
You may also try "opt -{any-analysis-passes} -aa-eval}"

Note, opt is the tool not for users but for developers. He does not
invoke unspecified passes automatically.

...Takumi

2012/8/11 Jun Koi <junkoi2004 at gmail.com>:
> On Sat, Aug 11, 2012 at 7:42 PM, Jun Koi <junkoi2004 at gmail.com> wrote:
>> hi,
>>
>> i am trying to optimize the below code, in one of my LLVM functions:
>>
>> ....
>>   store i32 96, i32* @XXX, align 4     ; (1)
>>   store i32 117, i32* @yyy, align 4    ; (2)
>>   store i32 31, i32* @XXX, align 4     ; (3)
>> ....
>>
>> naturally, optimize passes should remove the line (1), because later
>> line (3) overwrites the value of the same global variable XXX.
>> i run this code via "opt" with option "-O3", and indeed line (1) was removed.
>>
>> but i want to do that in my code, without having to call "opt".
>> however, after trying a lot of transforms (at FunctionPassManager
>> level), line (1) is still there.
>>
>> so i am wondering which passes (in the list
>> http://llvm.org/docs/Passes.html#transforms) can remove line (1)
>> above?
>
> i have some doubts here:
>
> - i only tried to optimize code with transforms (see the list
> http://llvm.org/docs/Passes.html#transforms). do i need to combine
> that with analysis passes (http://llvm.org/docs/Passes.html#analyses),
> or only transforms will do? if combination is necessary, how to
> combine?
>
> - i guess another reason that line (1) was not removed because i run
> the tranforms only once. do i need to run transforms in many rounds,
> and in different order to have better optimized code?
>
> many thanks,
> Jun
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev