Displaying 6 results from an estimated 6 matches for "isl_int_init".
2011 Nov 19
0
[LLVMdev] How to make Polly ignore some non-affine memory accesses
...VM-IR
> - // as something like A[i * elementsize]. This hides the fact that two
> - // subsequent values of 'i' index two values that are stored next to each
> - // other in memory. By this devision we make this characteristic obvious
> - // again.
> - isl_int v;
> - isl_int_init(v);
> - isl_int_set_si(v, Access.getElemSizeInBytes());
> - Affine = isl_pw_aff_scale_down(Affine, v);
> - isl_int_clear(v);
> + isl_pw_aff *Affine = SCEVAffinator::getPwAff(Statement, Access.getOffset());
>
> - AccessRelation = isl_map_from_pw_aff(Affine);
> - AccessRe...
2011 Nov 18
5
[LLVMdev] How to make Polly ignore some non-affine memory accesses
2011/11/15 Marcello Maggioni <hayarms at gmail.com>:
> 2011/11/14 Tobias Grosser <tobias at grosser.es>:
>> On 11/14/2011 02:45 PM, Marcello Maggioni wrote:
>>>
>>> 2011/11/14 Tobias Grosser<tobias at grosser.es>:
>>>>
>>>> On 11/14/2011 01:24 AM, Marcello Maggioni wrote:
>>>>>
>>>>> Hi Tobias.
2011 Nov 20
2
[LLVMdev] How to make Polly ignore some non-affine memory accesses
...* elementsize]. This hides the fact that two
>> - // subsequent values of 'i' index two values that are stored next to
>> each
>> - // other in memory. By this devision we make this characteristic
>> obvious
>> - // again.
>> - isl_int v;
>> - isl_int_init(v);
>> - isl_int_set_si(v, Access.getElemSizeInBytes());
>> - Affine = isl_pw_aff_scale_down(Affine, v);
>> - isl_int_clear(v);
>> + isl_pw_aff *Affine = SCEVAffinator::getPwAff(Statement,
>> Access.getOffset());
>>
>> - AccessRelation = isl_map_from...
2013 Jul 28
0
[LLVMdev] [Polly] Analysis of the expensive compile-time overhead of Polly Dependence pass
...*c = cast<SCEVConstant>(AR->getOperand(0));
This is obviously a hack. The base is not always a constant.
You can probably just call use something like,
isl_pw_aff *BaseValue = visit(AR->getOperand(0))
Affine = isl_pw_aff_sum(Affine, BaseValue);
> + isl_int t;
> + isl_int_init(t);
> + isl_int_set_si(t, c->getValue()->getSExtValue());
We now use isl_val instead of isl_int.
> return isl_pw_aff_alloc(Domain, Affine);
> }
> @@ -717,6 +725,34 @@ void Scop::addParams(std::vector<const SCEV *> NewParameters) {
> if (Parame...
2011 Nov 21
0
[LLVMdev] How to get ELF section virtual starting address from MCSymbolRefExpr?
...* elementsize]. This hides the fact that two
>> - ?// subsequent values of 'i' index two values that are stored next to
>> each
>> - ?// other in memory. By this devision we make this characteristic
>> obvious
>> - ?// again.
>> - ?isl_int v;
>> - ?isl_int_init(v);
>> - ?isl_int_set_si(v, Access.getElemSizeInBytes());
>> - ?Affine = isl_pw_aff_scale_down(Affine, v);
>> - ?isl_int_clear(v);
>> + ? ?isl_pw_aff *Affine = SCEVAffinator::getPwAff(Statement,
>> Access.getOffset());
>>
>> - ?AccessRelation = isl_map_from...
2013 Jul 26
6
[LLVMdev] [Polly] Analysis of the expensive compile-time overhead of Polly Dependence pass
Hi Sebastian,
Recently, I found the "Polly - Calculate dependences" pass would lead to significant compile-time overhead when compiling some loop-intensive source code. Tobias told me you found similar problem as follows:
http://llvm.org/bugs/show_bug.cgi?id=14240
My evaluation shows that "Polly - Calculate dependences" pass consumes 96.4% of total compile-time overhead