[LLVMdev] Alias analysis issue with structs on PPC

[Daniel Berlin]

On Fri, Mar 13, 2015 at 2:54 PM Daniel Berlin <dberlin at dberlin.org> wrote:

> On Fri, Mar 13, 2015 at 2:39 PM Olivier H Sallenave <ohsallen at us.ibm.com>
> wrote:
>
>> Hi,
>>
>> I have the following C loop to vectorize:
>>
>> struct box {
>>     double* source;
>> };
>>
>> void test(double* restrict result, struct box my_struct, int len)
>> {
>>     for (int i=0 ; i<len; i++) {
>>         result[i] = my_struct.source[i] * my_struct.source[i];
>>     }
>> }
>>
>> There are two references in the loop, result[i] (restrict) and
>> my_struct.source[i] (readonly). The compiler should easily figure out that
>> they do not alias.
>>
>> Compiling for x86, the loop alias analysis works just fine:
>>   AST: Alias Set Tracker: 2 alias sets for 2 pointer values.
>>   AliasSet[0x7fd8e2f32290, 1] must alias, No access Pointers: (double*
>> %arrayidx5, 18446744073709551615)
>>   AliasSet[0x7fd8e2f322e0, 1] must alias, No access Pointers: (double*
>> %arrayidx, 18446744073709551615)
>>
>> Compiling for PPC with -target powerpc64le-ibm-linux-gnu, the two
>> addresses now alias:
>>   AST: Alias Set Tracker: 1 alias sets for 2 pointer values.
>>   AliasSet[0x7f931bd5bdc0, 2] may alias, No access Pointers: (double*
>> %arrayidx5, 18446744073709551615), (double* %arrayidx, 18446744073709551615)
>>
>> BasicAA is used for both targets by default. The difference is that in
>> PPC, the IR obtained from Clang takes an i64 as parameter instead of a
>> double* for my_struct.
>>
>
> I don't even want to know why this would be the case :)
>
>
>> This parameter is then coerced into double* using an inttoptr
>> instruction. The code in BasicAliasAnalysis.cpp which is triggered for x86
>> is the following:
>>
>>     // Function arguments can't alias with things that are known to be
>>     // unambigously identified at the function level.
>>     if ((isa<Argument>(O1) && isIdentifiedFunctionLocal(O2)) ||
>>         (isa<Argument>(O2) && isIdentifiedFunctionLocal(O1)))
>>       return NoAlias;
>>
>> isIdentifiedFunctionLocal(V) returns true for a noalias argument (such as
>> result), but the other address (my_struct) must be a function argument in
>> order to return NoAlias, which is not the case anymore for PPC (since
>> my_struct is now the result from an inttoptr instruction). If I understand,
>> the problem is that we cannot trust the fact that locals do not alias with
>> restrict parameters (because the compiler could generate some locals which
>> alias)?
>>
> Yes, because pointers *based on* the noalias'd argument are legal aliases.
>
> So if you don't know it's an argument or an identified local, it could be
> based on the restricted pointer, and thus, alias it.
>
>
> If someone has suggestions about this, that would help a lot.
>>
>
> The only way you could prove something in this case would be to walk the
> chain and prove the value comes directly from an argument with no
> modification.
>

Actually, you could do the opposite, too, pretty cheaply.

You could write a new pass or AA.
It traverses chains in the reverse direction (IE it goes from the
arguments, and walks down the immediate use chain, marking things as based
on arguments or not), and makes a lookup table of things it can prove are
also  unmolested identified objects.
(which would be the result of inttoptr in your case).

You can then use this simple lookup table to answer the isIdentifiedObject
question better.
(You'd have to make isIdentifiedObject part of the AA interface, or take an
optional table, blah blah blah)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20150313/63c5e267/attachment.html>