| Message ID | 20220125122914.567599-1-nrb@linux.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [qemu] s390x: sck: load into a temporary not into in1 | expand |
On 25.01.22 13:29, Nico Boehr wrote: > We previously loaded into in1, but in1 is not filled during > disassembly and hence always zero. This leads to an assertion failure: > > qemu-system-s390x: /home/nrb/qemu/include/tcg/tcg.h:654: temp_idx: > Assertion `n >= 0 && n < tcg_ctx->nb_temps' failed.` > > Instead, load into a temporary and pass that to the helper. > > This fixes the SCK test I sent here under TCG: > <https://www.spinics.net/lists/kvm/msg265169.html> > > Signed-off-by: Nico Boehr <nrb@linux.ibm.com> > --- > target/s390x/tcg/translate.c | 11 +++++++++-- > 1 file changed, 9 insertions(+), 2 deletions(-) > > diff --git a/target/s390x/tcg/translate.c b/target/s390x/tcg/translate.c > index 46dea733571e..dc0baec5a5f4 100644 > --- a/target/s390x/tcg/translate.c > +++ b/target/s390x/tcg/translate.c > @@ -4290,9 +4290,16 @@ static DisasJumpType op_stcke(DisasContext *s, DisasOps *o) > #ifndef CONFIG_USER_ONLY > static DisasJumpType op_sck(DisasContext *s, DisasOps *o) > { > - tcg_gen_qemu_ld_i64(o->in1, o->addr1, get_mem_index(s), MO_TEUQ | MO_ALIGN); > - gen_helper_sck(cc_op, cpu_env, o->in1); > + TCGv_i64 t1; > + > + t1 = tcg_temp_new_i64(); > + > + tcg_gen_qemu_ld_i64(t1, o->addr1, get_mem_index(s), MO_TEUQ | MO_ALIGN); > + gen_helper_sck(cc_op, cpu_env, t1); > set_cc_static(s); > + > + tcg_temp_free_i64(t1); > + > return DISAS_NEXT; > } > I think you can actually just reuse in2_m2_64a, similar as we handle SCKC diff --git a/target/s390x/tcg/insn-data.def b/target/s390x/tcg/insn-data.def index f0af458aee..3a78606866 100644 --- a/target/s390x/tcg/insn-data.def +++ b/target/s390x/tcg/insn-data.def @@ -1317,7 +1317,7 @@ /* SET ADDRESS SPACE CONTROL FAST */ F(0xb279, SACF, S, Z, 0, a2, 0, 0, sacf, 0, IF_PRIV) /* SET CLOCK */ - F(0xb204, SCK, S, Z, la2, 0, 0, 0, sck, 0, IF_PRIV | IF_IO) + F(0xb204, SCK, S, Z, 0, m2_64a, 0, 0, sck, 0, IF_PRIV | IF_IO) /* SET CLOCK COMPARATOR */ F(0xb206, SCKC, S, Z, 0, m2_64a, 0, 0, sckc, 0, IF_PRIV | IF_IO) /* SET CLOCK PROGRAMMABLE FIELD */ diff --git a/target/s390x/tcg/translate.c b/target/s390x/tcg/translate.c index f180853e7a..c18011920e 100644 --- a/target/s390x/tcg/translate.c +++ b/target/s390x/tcg/translate.c @@ -4295,8 +4295,7 @@ static DisasJumpType op_stcke(DisasContext *s, DisasOps *o) #ifndef CONFIG_USER_ONLY static DisasJumpType op_sck(DisasContext *s, DisasOps *o) { - tcg_gen_qemu_ld_i64(o->in1, o->addr1, get_mem_index(s), MO_TEUQ | MO_ALIGN); - gen_helper_sck(cc_op, cpu_env, o->in1); + gen_helper_sck(cc_op, cpu_env, o->in2); set_cc_static(s); return DISAS_NEXT; }
On Tue, 2022-01-25 at 14:13 +0100, David Hildenbrand wrote: > I think you can actually just reuse in2_m2_64a, similar as we handle > SCKC I tried my SCK tests with your patch, it works just as well and seems much cleaner, thanks. Do you want to send this or should I make a v2 and add you as Suggested-by?
On 25/1/22 15:54, Nico Boehr wrote: > On Tue, 2022-01-25 at 14:13 +0100, David Hildenbrand wrote: >> I think you can actually just reuse in2_m2_64a, similar as we handle >> SCKC > > I tried my SCK tests with your patch, it works just as well and seems > much cleaner, thanks. > > Do you want to send this or should I make a v2 and add you as > Suggested-by? v2 ;)
On 25.01.22 17:42, Philippe Mathieu-Daudé wrote: > On 25/1/22 15:54, Nico Boehr wrote: >> On Tue, 2022-01-25 at 14:13 +0100, David Hildenbrand wrote: >>> I think you can actually just reuse in2_m2_64a, similar as we handle >>> SCKC >> >> I tried my SCK tests with your patch, it works just as well and seems >> much cleaner, thanks. >> >> Do you want to send this or should I make a v2 and add you as >> Suggested-by? > > v2 ;) Yes, v2 please :)
diff --git a/target/s390x/tcg/translate.c b/target/s390x/tcg/translate.c index 46dea733571e..dc0baec5a5f4 100644 --- a/target/s390x/tcg/translate.c +++ b/target/s390x/tcg/translate.c @@ -4290,9 +4290,16 @@ static DisasJumpType op_stcke(DisasContext *s, DisasOps *o) #ifndef CONFIG_USER_ONLY static DisasJumpType op_sck(DisasContext *s, DisasOps *o) { - tcg_gen_qemu_ld_i64(o->in1, o->addr1, get_mem_index(s), MO_TEUQ | MO_ALIGN); - gen_helper_sck(cc_op, cpu_env, o->in1); + TCGv_i64 t1; + + t1 = tcg_temp_new_i64(); + + tcg_gen_qemu_ld_i64(t1, o->addr1, get_mem_index(s), MO_TEUQ | MO_ALIGN); + gen_helper_sck(cc_op, cpu_env, t1); set_cc_static(s); + + tcg_temp_free_i64(t1); + return DISAS_NEXT; }
We previously loaded into in1, but in1 is not filled during disassembly and hence always zero. This leads to an assertion failure: qemu-system-s390x: /home/nrb/qemu/include/tcg/tcg.h:654: temp_idx: Assertion `n >= 0 && n < tcg_ctx->nb_temps' failed.` Instead, load into a temporary and pass that to the helper. This fixes the SCK test I sent here under TCG: <https://www.spinics.net/lists/kvm/msg265169.html> Signed-off-by: Nico Boehr <nrb@linux.ibm.com> --- target/s390x/tcg/translate.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)