On 5/8/25 02:54, Paolo Bonzini wrote:
Signed-off-by: Paolo Bonzini <pbonz...@redhat.com> --- target/i386/tcg/decode-new.c.inc | 36 ++++++++++++++++++++++++-------- 1 file changed, 27 insertions(+), 9 deletions(-)diff --git a/target/i386/tcg/decode-new.c.inc b/target/i386/tcg/decode-new.c.inc index cda32ee6784..55216e0d249 100644 --- a/target/i386/tcg/decode-new.c.inc +++ b/target/i386/tcg/decode-new.c.inc @@ -2542,7 +2542,13 @@ static void disas_insn(DisasContext *s, CPUState *cpu) s->has_modrm = false; s->prefix = 0;- next_byte:+ next_byte:; +#ifdef TARGET_X86_64 + /* clear any REX prefix followed by other prefixes. */ + int rex; + rex = -1; + next_byte_rex: +#endif b = x86_ldub_code(env, s);
I guess this is ok, with -1. It might be worthwhile rewriting without ifdefs, since we already let CODE64(s) and REX_PREFIX(s) expand to false without TARGET_X86_64.
The only irritating part would be actually setting s->rex_* at the end, because those members are themselves ifdef'ed over in DisasContext.
Reviewed-by: Richard Henderson <richard.hender...@linaro.org> r~
