RFR: 8323220: Reassociate loop invariants involved in Cmps and Add/Subs [v9]

[Emanuel Peter]

On Thu, 8 Feb 2024 21:39:35 GMT, Joshua Cao <duke at openjdk.org> wrote:

>> // inv1 == (x + inv2) => ( inv1 - inv2 ) == x
>> // inv1 == (x - inv2) => ( inv1 + inv2 ) == x
>> // inv1 == (inv2 - x) => (-inv1 + inv2 ) == x
>> 
>> 
>> For example,
>> 
>> 
>> fn(inv1, inv2)
>>   while(...)
>>       x = foobar()
>>       if inv1 == x + inv2
>>            blackhole()
>> 
>> 
>> We can transform this into
>> 
>> 
>> fn(inv1, inv2)
>>   t = inv1 - inv2
>>   while(...)
>>       x = foobar()
>>       if t == x
>>            blackhole()
>> 
>> 
>> Here is an example: https://github.com/openjdk/jdk/blob/b78896b9aafcb15f453eaed6e154a5461581407b/src/java.base/share/classes/java/lang/invoke/LambdaFormEditor.java#L910. LHS `1` and RHS `pos` are both loop invariant
>> 
>> Passes tier1 locally on Linux machine. Passes GHA on my fork.
>
> Joshua Cao has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Add correctness test for some random tests with random inputs

https://github.com/openjdk/jdk/pull/17557
Is integrated. You need to merge with master now, and inspect all usages of `@Arguments`.

I imagine it like this:

public class InvariantCodeMotionReassociateCmp {
    int inv1Ref, inv2Ref, invDiff; // have fields to use in @Setup and @Check

    @Setup
    public Object[] setupArguments() {
        inv1Ref = RANDOM.nextInt();
        ... // setup your reference values in a smart way, so that half of the time the test returns -1, and half the time it returns i.
        return new Object[]{ inv1Ref, inv2Ref };
    }
    
    @Test
    @Arguments(setup = "setupArguments")
    @IR(counts = {IRNode.SUB_I, "1"})
    public int[] equalsAddInt(int inv1, int inv2) {
        for (int i = 0; i < size; ++i) {
            // Reassociate to `inv2 - inv1 == i`
            if (inv1 + i == inv2) {
                return i;
            }
        }
        return -1;
    }
    
    @Check(test = "equalsAddInt")
    public void checkEqualsAddInt(int returnValue) {
        // now check if the returnValue is as expected, given the field values
    }

This should also have fewer add/sub nodes in the test method, I think.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/17375#issuecomment-1935384310