On 7/22/2026 9:24 AM, [email protected] wrote:
From: Kyrylo Tkachov <[email protected]>
noce_convert_multiple_sets keys its dependency information by destination
register number. Hard registers can have overlapping definitions in
different modes. An exact RTL replacement cannot represent the value from a
partial or mode-changing definition.
Reject hard-register destinations in the block eligibility check. Assert
the resulting pseudo-only invariant when dependency information is
initialized.
Add an RTL execution test with mixed-mode hard-register definitions. It
verifies that the block remains unconverted.
Bootstrapped and regtested on aarch64-none-linux-gnu and
x86_64-pc-linux-gnu.
gcc/ChangeLog:
PR rtl-optimization/126184
* ifcvt.cc (init_noce_multiple_sets_info): Assert that destinations
are pseudos.
(bb_ok_for_noce_convert_multiple_sets): Reject hard-register
destinations.
gcc/testsuite/ChangeLog:
PR rtl-optimization/126184
* gcc.target/aarch64/pr126184-2.c: New test.
It may be over-aggressive in rejecting (imagine if we only have one hard
register assignment in the block within the set of multiple
assignments). But I wouldn't really want to chase that kind of case
without some evidence it matters.
So, OK.
Jeff
