This is a summary of discussions relative to the merge request created by Alex Coplan (acoplan) <[email protected]> titled aarch64: Fix up TME deprecation warning since its creation.
Description: This is a follow-up patch to Richard E's r17-2498-g17f084306c68c4 (#186 on the Forge). As Alice pointed out in her review: https://gcc.gnu.org/pipermail/gcc-patches/2026-July/724391.html the original patch missed things like +memtag+tme due to the string parsing approach. This instead just inspects the parsed feature flags from aarch64_parse_extension to determine whether to warn in the target attribute/pragma case. +tme on the command line is handled as per Richard's original patch (no change there). Bootstrapped/regtested on aarch64-linux-gnu, OK for trunk? Thanks, Alex gcc/ChangeLog: * config/aarch64/aarch64.cc (aarch64_handle_attr_isa_flags): Warn if +tme is enabled by the target attribute/pragma, relying on the feature flags from aarch64_parse_extension. (aarch64_process_target_attr): Drop +tme warning based on custom string parsing. gcc/testsuite/ChangeLog: * gcc.target/aarch64/deprecate-tme.c: New test. The full and up to date discussion can be found at https://forge.sourceware.org/gcc/gcc/pulls/195 The merge request has been closed without being merged directly on the forge repository. On 2026-07-20 13:57:07+00:00, Richard Earnshaw (rearnsha) <[email protected]> approved the changes: OK apart from this. > +++ gcc/testsuite/gcc.target/aarch64/deprecate-tme.c > @@ -0,0 +1,16 @@ > +/* { dg-do compile } */ Pedantically you might want to add an architecture spec here that lacks tme; otherwise the test will fail if run with, eg -march=armv8-a+tme. > +++ gcc/testsuite/gcc.target/aarch64/deprecate-tme.c > @@ -0,0 +1,16 @@ > +/* { dg-do compile } */ Hmm. I think we should probably just say that testing with +tme is unsupported. Forcing a baseline architecture with dg-options just reduces test coverage in valid configurations that we do care about, so I'm inclined to leave it as is. Another option would be dg-skip-if, but I'm not sure it's worth writing the dejagnu condition, given it's all going to be ripped out eventually anyway.
