Displaying 5 results from an estimated 5 matches for "curr_array".
2011 Jul 19
0
[LLVMdev] speculative parallelization in LLVM
On 19 July 2011 10:12, Jimborean Alexandra <xinfinity_a at yahoo.com> wrote:
> %curr_array = alloca [10 x %struct.linked], align 8
>
> while..
> %tmp16 = getelementptr inbounds [10 x %struct.linked]* %curr_array, i32 0,
> i32 1
Hi Alexandra,
Can you guarantee that the linked list will be allocated in contiguous memory?
cheers,
--renato
2011 Jul 19
4
[LLVMdev] speculative parallelization in LLVM
...rr->next ;
}
This becomes in LLVM IR:
%curr = alloca %struct.linked*, align 8
while..
%tmp15 = load %struct.linked** %curr, align 8
%tmp16 = getelementptr inbounds %struct.linked* %tmp15, i32 0, i32 1 ( curr
= curr->next ;)
And I want to transform this into a SCoP like this:
%curr_array = alloca [10 x %struct.linked], align 8
while..
%tmp16 = getelementptr inbounds [10 x %struct.linked]* %curr_array, i32 0, i32
1
(replace all pointers similarly)
I only want Polly to accept the code (although incorrect at this point) and to
generate optimized code. Next I will replace back...
2011 Jul 19
3
[LLVMdev] speculative parallelization in LLVM
...lexandra <xinfinity_a at yahoo.com>
Cc: Tobias Grosser <tobias at grosser.es>; llvmdev at cs.uiuc.edu
Sent: Tue, July 19, 2011 11:39:02 AM
Subject: Re: [LLVMdev] speculative parallelization in LLVM
On 19 July 2011 10:12, Jimborean Alexandra <xinfinity_a at yahoo.com> wrote:
> %curr_array = alloca [10 x %struct.linked], align 8
>
> while..
> %tmp16 = getelementptr inbounds [10 x %struct.linked]* %curr_array, i32 0,
> i32 1
Hi Alexandra,
Can you guarantee that the linked list will be allocated in contiguous memory?
cheers,
--renato
-------------- next part -----------...
2011 Jul 19
0
[LLVMdev] speculative parallelization in LLVM
On 07/18/2011 07:03 PM, Jimborean Alexandra wrote:
> Hi,
>
> I plan to do some speculative parallelization in LLVM using Polly and I
> target loops that contain pointers and indirect references. As far as I
> know, Polly generates optimized code starting from the SCoPs, therefore
> I plan to replace all pointer accesses with array accesses, such that
> Polly will accept the
2011 Jul 18
3
[LLVMdev] speculative parallelization in LLVM
Hi,
I plan to do some speculative parallelization in LLVM using Polly and I target
loops that contain pointers and indirect references. As far as I know, Polly
generates optimized code starting from the SCoPs, therefore I plan to replace
all pointer accesses with array accesses, such that Polly will accept the code.
Each array access should use a liner function of the enclosing loops indices.