Commit Graph

6 Commits

Author SHA1 Message Date
Nikita Popov e97f3b1b43 [InstCombine] Fold abs of known negative operand
If we know that the abs operand is known negative, we can replace
it with a neg.

To avoid computing known bits twice, I've removed the fold for the
non-negative case from InstSimplify. Both the non-negative and the
negative case are handled by InstCombine now, with one known bits call.

Differential Revision: https://reviews.llvm.org/D87196
2020-09-08 20:14:35 +02:00
Nikita Popov ff218cbc84 [InstSimplify] Fold degenerate abs of abs form
This addresses the remaining issue from D87188. Due to a series of
folds, we may end up with abs-of-abs represented as
x == 0 ? -abs(x) : abs(x). Rather than recognizing this as a special
abs pattern and doing an abs-of-abs fold on it afterwards,
I'm directly folding this to one of the select operands in InstSimplify.

The general pattern falls into the "select with operand replaced"
category, but that fold is not powerful enough to recognize that
both hands of the select are the same for value zero.

Differential Revision: https://reviews.llvm.org/D87197
2020-09-06 09:43:08 +02:00
Nikita Popov 621b10ca18 [InstSimplify] Add tests for a peculiar abs of abs form (NFC)
This pattern shows up when canonicalizing to spf abs form to
intrinsic abs form.
2020-09-05 21:42:22 +02:00
Nikita Popov eba5f5f798 [ValueTracking] Add abs intrinsics support to computeConstantRange()
Implementation is the same as for SPF_ABS.
2020-08-12 22:28:46 +02:00
Nikita Popov 6446c11840 [InstSimplify] Add additional abs intrinsic icmp tests (NFC)
While abs >= 0 already folds, some variations thereon don't.
2020-08-12 22:28:46 +02:00
Nikita Popov 7397a019b8 [InstSimplify] Extract abs intrinsic tests into separate file (NFC)
Also move some tests from InstCombine to InstSimplify,
as they are already handled by InstSimplify.
2020-08-12 22:28:46 +02:00