================
@@ -164,6 +164,9 @@ bool X86TargetInfo::initFeatureMap(
for (const char *Sub :
{"egpr", "push2pop2", "ppx", "ndd", "ccmp", "nf", "zu", "jmpabs"})
setFeatureEnabled(Features, Sub, true);
+ // APX (EVEX) implies AVX-512, and WinCall expects every instruction to be
+ // EVEX-encoded so that no vzeroupper is needed at WinCall boundaries.
+ setFeatureEnabled(Features, "avx512f", true);
----------------
Andarwinux wrote:
Additionally personally, when defining the ISA baseline for x86_64apx or
"x86-64-v5", include GFNI and VPCLMULDQ will be useful for improving vector
bitwise operations lowering, see previous GFNI vXi8 and pext/pdep work.
https://github.com/llvm/llvm-project/pull/215231#issuecomment-5238798831
https://github.com/llvm/llvm-project/pull/215585
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
