| Message ID | 20191127175046.4911-5-frankja@linux.ibm.com |
|---|---|
| State | New |
| Headers | show |
| Series | s390x: Cleanup | expand |
On 27.11.19 18:50, Janosch Frank wrote: > Let's also move the clear reset function into the reset handler. > > Signed-off-by: Janosch Frank <frankja@linux.ibm.com> > --- [...] Gave the first 4 patches a tcg test. Looks good. Reviewed-by: David Hildenbrand <david@redhat.com>
On 11/27/19 7:15 PM, David Hildenbrand wrote: > On 27.11.19 18:50, Janosch Frank wrote: >> Let's also move the clear reset function into the reset handler. >> >> Signed-off-by: Janosch Frank <frankja@linux.ibm.com> >> --- > [...] > > Gave the first 4 patches a tcg test. Looks good. Great to hear :) > > Reviewed-by: David Hildenbrand <david@redhat.com> Thanks
On 27/11/2019 18.50, Janosch Frank wrote: > Let's also move the clear reset function into the reset handler. > > Signed-off-by: Janosch Frank <frankja@linux.ibm.com> > --- > target/s390x/cpu-qom.h | 1 + > target/s390x/cpu.c | 58 +++++++++++++----------------------------- > 2 files changed, 18 insertions(+), 41 deletions(-) I really wonder why we had so much duplicated code here before and nobody dared to clean it up in all those years... Thanks for doing that now! Reviewed-by: Thomas Huth <thuth@redhat.com>
On 11/28/19 8:09 AM, Thomas Huth wrote: > On 27/11/2019 18.50, Janosch Frank wrote: >> Let's also move the clear reset function into the reset handler. >> >> Signed-off-by: Janosch Frank <frankja@linux.ibm.com> >> --- >> target/s390x/cpu-qom.h | 1 + >> target/s390x/cpu.c | 58 +++++++++++++----------------------------- >> 2 files changed, 18 insertions(+), 41 deletions(-) > > I really wonder why we had so much duplicated code here before and > nobody dared to clean it up in all those years... Thanks for doing that now! Looking back at my old patch series, I most often start with a cleanup before I change or add anything... I'd guess that QEMU hasn't seen some love for a longer time. > > Reviewed-by: Thomas Huth <thuth@redhat.com> > Thanks!
On 28.11.19 09:35, Janosch Frank wrote: > On 11/28/19 8:09 AM, Thomas Huth wrote: >> On 27/11/2019 18.50, Janosch Frank wrote: >>> Let's also move the clear reset function into the reset handler. >>> >>> Signed-off-by: Janosch Frank <frankja@linux.ibm.com> >>> --- >>> target/s390x/cpu-qom.h | 1 + >>> target/s390x/cpu.c | 58 +++++++++++++----------------------------- >>> 2 files changed, 18 insertions(+), 41 deletions(-) >> >> I really wonder why we had so much duplicated code here before and >> nobody dared to clean it up in all those years... Thanks for doing that now! > > Looking back at my old patch series, I most often start with a cleanup > before I change or add anything... > > I'd guess that QEMU hasn't seen some love for a longer time. Oh, I give QEMU some love quite often ... I just can't love every part of it equally well ;) I'd like to second the "thanks" for doing this cleanup :)
diff --git a/target/s390x/cpu-qom.h b/target/s390x/cpu-qom.h index 6f0a12042e..dbe5346ec9 100644 --- a/target/s390x/cpu-qom.h +++ b/target/s390x/cpu-qom.h @@ -37,6 +37,7 @@ typedef struct S390CPUDef S390CPUDef; typedef enum cpu_reset_type { S390_CPU_RESET_NORMAL, S390_CPU_RESET_INITIAL, + S390_CPU_RESET_CLEAR, } cpu_reset_type; /** diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c index 55e2d1fe7b..3e745e8ecc 100644 --- a/target/s390x/cpu.c +++ b/target/s390x/cpu.c @@ -94,6 +94,9 @@ static void s390_cpu_reset(CPUState *s, cpu_reset_type type) s390_cpu_set_state(S390_CPU_STATE_STOPPED, cpu); switch (type) { + case S390_CPU_RESET_CLEAR: + memset(env, 0, offsetof(CPUS390XState, start_initial_reset_fields)); + /* fall through */ case S390_CPU_RESET_INITIAL: /* initial reset does not clear everything! */ memset(&env->start_initial_reset_fields, 0, @@ -107,6 +110,14 @@ static void s390_cpu_reset(CPUState *s, cpu_reset_type type) env->cregs[0] = CR0_RESET; env->cregs[14] = CR14_RESET; +#if defined(CONFIG_USER_ONLY) + /* user mode should always be allowed to use the full FPU */ + env->cregs[0] |= CR0_AFP; + if (s390_has_feat(S390_FEAT_VECTOR)) { + env->cregs[0] |= CR0_VECTOR; + } +#endif + /* tininess for underflow is detected before rounding */ set_float_detect_tininess(float_tininess_before_rounding, &env->fpu_status); @@ -125,46 +136,6 @@ static void s390_cpu_reset(CPUState *s, cpu_reset_type type) } } -/* CPUClass:reset() */ -static void s390_cpu_full_reset(CPUState *s) -{ - S390CPU *cpu = S390_CPU(s); - S390CPUClass *scc = S390_CPU_GET_CLASS(cpu); - CPUS390XState *env = &cpu->env; - - scc->parent_reset(s); - cpu->env.sigp_order = 0; - s390_cpu_set_state(S390_CPU_STATE_STOPPED, cpu); - - memset(env, 0, offsetof(CPUS390XState, end_reset_fields)); - - /* architectured initial values for CR 0 and 14 */ - env->cregs[0] = CR0_RESET; - env->cregs[14] = CR14_RESET; - -#if defined(CONFIG_USER_ONLY) - /* user mode should always be allowed to use the full FPU */ - env->cregs[0] |= CR0_AFP; - if (s390_has_feat(S390_FEAT_VECTOR)) { - env->cregs[0] |= CR0_VECTOR; - } -#endif - - /* architectured initial value for Breaking-Event-Address register */ - env->gbea = 1; - - env->pfault_token = -1UL; - - /* tininess for underflow is detected before rounding */ - set_float_detect_tininess(float_tininess_before_rounding, - &env->fpu_status); - - /* Reset state inside the kernel that we cannot access yet from QEMU. */ - if (kvm_enabled()) { - kvm_s390_reset_vcpu(cpu); - } -} - #if !defined(CONFIG_USER_ONLY) static void s390_cpu_machine_reset_cb(void *opaque) { @@ -456,6 +427,11 @@ static Property s390x_cpu_properties[] = { DEFINE_PROP_END_OF_LIST() }; +static void s390_cpu_reset_full(CPUState *s) +{ + return s390_cpu_reset(s, S390_CPU_RESET_CLEAR); +} + static void s390_cpu_class_init(ObjectClass *oc, void *data) { S390CPUClass *scc = S390_CPU_CLASS(oc); @@ -472,7 +448,7 @@ static void s390_cpu_class_init(ObjectClass *oc, void *data) scc->load_normal = s390_cpu_load_normal; #endif scc->reset = s390_cpu_reset; - cc->reset = s390_cpu_full_reset; + cc->reset = s390_cpu_reset_full; cc->class_by_name = s390_cpu_class_by_name, cc->has_work = s390_cpu_has_work; #ifdef CONFIG_TCG
Let's also move the clear reset function into the reset handler. Signed-off-by: Janosch Frank <frankja@linux.ibm.com> --- target/s390x/cpu-qom.h | 1 + target/s390x/cpu.c | 58 +++++++++++++----------------------------- 2 files changed, 18 insertions(+), 41 deletions(-)