On 22/10/2021 20:16, Richard Henderson wrote:
[E-MAIL EXTERNO] Não clique em links ou abra anexos, a menos que você
possa confirmar o remetente e saber que o conteúdo é seguro. Em caso de
e-mail suspeito entre imediatamente em contato com o DTI.
On 10/21/21 12:45 PM, matheus.fe...@eldorado.org.br wrote:
+uint64_t helper_CNTLZDM(uint64_t src, uint64_t mask)
+{
+ uint64_t sel_bit, count = 0;
+
+ while (mask != 0) {
+ sel_bit = 0x8000000000000000ULL >> clz64(mask);
+
+ if (src & sel_bit) {
+ break;
+ }
We need to count how many mask are set left of mask & src.
How about
sh = clz64(src & mask);
if (sh == 0) {
return 0;
}
return ctpop64(mask >> (64 - sh));
which could probably be implemented inline relatively easy.
Thanks for this suggestion Richard, we'll try to inline it.
+static bool trans_CNTLZDM(DisasContext *ctx, arg_X *a)
+{
+ REQUIRE_64BIT(ctx);
+ REQUIRE_INSNS_FLAGS2(ctx, ISA310);
+#if defined(TARGET_PPC64)
+ gen_helper_CNTLZDM(cpu_gpr[a->ra], cpu_gpr[a->rt], cpu_gpr[a->rb]);
+#else
+ qemu_build_not_reached();
+#endif
+ return true;
+}
Why the ifdef here? Oh, I see. You could just use target_long in the
helper to avoid
that. And if not, you should move the helper into an ifdef too.
That's the same case of cfuged. There is a vector version of this
instruction (vclzdm) that is not 64-bits only (at least on paper), so it
should receive i64 and cannot be inside an ifdef(TARGET_PPC64). I'll add
this info to the commit message.
If we dismiss the possibility of a future 32-bits implementation of
PowerISA v3.1, we can move the helper inside the ifdef and add
REQUIRE_64BITS in vclzdm/vctzdm (and vcfuged, vpdepd, vpextd, etc.)
--
Matheus K. Ferst
Instituto de Pesquisas ELDORADO <http://www.eldorado.org.br/>
Analista de Software
Aviso Legal - Disclaimer <https://www.eldorado.org.br/disclaimer.html>