llvm dev - Feb 2015 - [LLVMdev] question about licm

hi,

   I applied licm with basicaa on the following codes:

   int j = atoi(argc[1]);
   int lower[] = {10, 9, 8, 7, 6, 5, 4, 3, 2, 1};
   int upper[] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};

   for(i = lower[j]; a[i] < 100 && i < upper[j]; i ++);

   I notice that upper[j] is not hoisted out from the loop. Is this 
because j could be larger than 10?

   Thanks a lot!

   Best,

                                              Linhai

Hi,

LICM can only hoist instructions which dominates all loop exit blocks.
In this case 'upper[j]' is not dominating exit block as its appearing in
second operand of logical AND operator.

Prior to hoisting it check for condition in 'isGuaranteedToExecute' and
it decide not to hoist it.

<File: LICM.cpp>
666 bool LICM::isGuaranteedToExecute(Instruction &Inst) {
667
668   // We have to check to make sure that the instruction dominates all
669   // of the exit blocks.  If it doesn't, then there is a path out of the
loop
670   // which does not execute this instruction, so we can't hoist it.

Hope this helps.

Regards,
Ashutosh

-----Original Message-----
From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu] On
Behalf Of songlh
Sent: Wednesday, February 11, 2015 10:24 AM
To: llvmdev at cs.uiuc.edu
Subject: [LLVMdev] question about licm

hi,

   I applied licm with basicaa on the following codes:

   int j = atoi(argc[1]);
   int lower[] = {10, 9, 8, 7, 6, 5, 4, 3, 2, 1};
   int upper[] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};

   for(i = lower[j]; a[i] < 100 && i < upper[j]; i ++);

   I notice that upper[j] is not hoisted out from the loop. Is this because j
could be larger than 10?

   Thanks a lot!

   Best,

                                              Linhai
_______________________________________________
LLVM Developers mailing list
LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev

----- Original Message -----
> From: "Ashutosh Nema" <Ashutosh.Nema at amd.com>
> To: "songlh" <songlh at cs.wisc.edu>, llvmdev at
cs.uiuc.edu
> Sent: Wednesday, February 11, 2015 3:20:27 AM
> Subject: Re: [LLVMdev] question about licm
> 
> Hi,
> 
> LICM can only hoist instructions which dominates all loop exit
> blocks.
> In this case 'upper[j]' is not dominating exit block as its
appearing
> in second operand of logical AND operator.
> 
> Prior to hoisting it check for condition in 'isGuaranteedToExecute'
> and it decide not to hoist it.
> 
> <File: LICM.cpp>
> 666 bool LICM::isGuaranteedToExecute(Instruction &Inst) {
> 667
> 668   // We have to check to make sure that the instruction dominates
> all
> 669   // of the exit blocks.  If it doesn't, then there is a path out
> of the loop
> 670   // which does not execute this instruction, so we can't hoist
> it.
Yes, and to add to this, the load is not safe to speculatively execute (because,
as you mention, j might be larger than 10). Now this is an interesting case
because the cases where upper[j] might be an out-of-bounds access are the same
as those when lower[j] might be an out-of-bounds access, and lower[j] dominates
upper[j]. As a result, we *could* notice this and hoist the load anyway (to
after lower[j] in the loop preheader), but we currently have no capability to do
this.

 -Hal
> 
> Hope this helps.
> 
> Regards,
> Ashutosh
> 
> -----Original Message-----
> From: llvmdev-bounces at cs.uiuc.edu
> [mailto:llvmdev-bounces at cs.uiuc.edu] On Behalf Of songlh
> Sent: Wednesday, February 11, 2015 10:24 AM
> To: llvmdev at cs.uiuc.edu
> Subject: [LLVMdev] question about licm
> 
> hi,
> 
>    I applied licm with basicaa on the following codes:
> 
>    int j = atoi(argc[1]);
>    int lower[] = {10, 9, 8, 7, 6, 5, 4, 3, 2, 1};
>    int upper[] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
> 
>    for(i = lower[j]; a[i] < 100 && i < upper[j]; i ++);
> 
>    I notice that upper[j] is not hoisted out from the loop. Is this
>    because j could be larger than 10?
> 
>    Thanks a lot!
> 
>    Best,
> 
>                                               Linhai
>                                              
_______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
> 
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
> 
-- 
Hal Finkel
Assistant Computational Scientist
Leadership Computing Facility
Argonne National Laboratory