Skip to content

Draft fix for a series of multiplications (#75413) #75414

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
SkiFoD wants to merge 1 commit into from
Closed

Draft fix for a series of multiplications (#75413) #75414

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 18 additions & 0 deletions src/coreclr/jit/morph.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12751,6 +12751,24 @@ GenTree* Compiler::fgOptimizeMultiply(GenTreeOp* mul)
fgUpdateConstTreeValueNumber(op2);
mul->ChangeOper(GT_LSH, GenTree::PRESERVE_VN);

if (mul->gtGetOp1()->OperIs(GT_LSH) && mul->gtGetOp1()->gtGetOp2()->IsIntegralConst())
{
GenTree* lsOp = mul->gtGetOp1();
GenTree* rsOp = mul->gtGetOp2();
GenTree* lsOpChOp1 = lsOp->gtGetOp1();
GenTree* lsOpChOp2 = lsOp->gtGetOp2();

int64_t child_val = lsOpChOp2->AsIntConCommon()->IntegralValue();
int64_t root_val = rsOp->AsIntConCommon()->IntegralValue();

mul->gtOp1 = lsOpChOp1;
mul->gtOp2 = lsOpChOp2;
mul->gtOp2->AsIntConCommon()->SetIntegralValue(root_val + child_val);

@Wraith2 Wraith2 Sep 11, 2022

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should you check if root_val+child_val can overflow here?

@SkiFoD SkiFoD Sep 11, 2022

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It seems there must be some checks, but I can't think of any right now.

The tricky part is that multiplications become shifts and the value transforms from a number to a count of shifts. I don't know if it actually possible to make
X >> 2,147,483,647

There is another problem with the PR which made me mark it as a draft. The problem is that the PR causes 9 regressions (along with some improvements) I need some help in addressing them.

It would be very nice if someone from the team became a reviewer.


DEBUG_DESTROY_NODE(lsOp);
DEBUG_DESTROY_NODE(rsOp);
}

return mul;
}
}
Expand Down