From: "Alex Bennée" <alex.bennee@linaro.org>
To: Claudio Fontana <cfontana@suse.de>
Cc: "Peter Maydell" <peter.maydell@linaro.org>,
"Eduardo Habkost" <ehabkost@redhat.com>,
"Richard Henderson" <richard.henderson@linaro.org>,
qemu-devel@nongnu.org, "Roman Bolshakov" <r.bolshakov@yadro.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Philippe Mathieu-Daudé" <philmd@redhat.com>
Subject: Re: [RFC v10 39/49] target/arm: add tcg cpu accel class
Date: Tue, 23 Mar 2021 19:26:16 +0000 [thread overview]
Message-ID: <87h7l1isb0.fsf@linaro.org> (raw)
In-Reply-To: <20210322140206.9513-40-cfontana@suse.de>
Claudio Fontana <cfontana@suse.de> writes:
> move init, realizefn and reset code into it.
w.r.t my testing this is fingered by bisect for causing:
Running test qtest-aarch64/migration-test
ERROR qtest-aarch64/migration-test - Bail out! ERROR:../../target/arm/tcg/tcg-cpu.c:233:tcg_arm_init_accel_cpu: assertion failed: (object_class_by_name(ACCEL_CPU_NAME("tcg")) == OBJECT_CLASS(accel_cpu))
Broken pipe
make: *** [Makefile.mtest:272: run-test-32] Error 1
The bisect log:
git bisect start
# bad: [474677f44c0d581381db57ff6030a6553f16db95] XXX target/arm: experiment refactoring cpu "max"
git bisect bad 474677f44c0d581381db57ff6030a6553f16db95
# good: [5ca634afcf83215a9a54ca6e66032325b5ffb5f6] Merge remote-tracking branch 'remotes/philmd/tags/sdmmc-20210322' into staging
git bisect good 5ca634afcf83215a9a54ca6e66032325b5ffb5f6
# good: [bf31455e28246f6b9b732dc56f89e61895f6f4f0] i386: split svm_helper into sysemu and stub-only user
git bisect good bf31455e28246f6b9b732dc56f89e61895f6f4f0
# good: [903f6e1bdc20939d023d6b577875370023bcac5f] target/arm: move arm_cpu_list to common_cpu
git bisect good 903f6e1bdc20939d023d6b577875370023bcac5f
# good: [1999c2ce6cc82200a39f6e41041f304eba5d4e7e] tests: device-introspect-test: cope with ARM TCG-only devices
git bisect good 1999c2ce6cc82200a39f6e41041f304eba5d4e7e
# bad: [fa37cc7fbabcf4de1cc963530c75b0ea7a73139a] target/arm: cpu-sve: new module
git bisect bad fa37cc7fbabcf4de1cc963530c75b0ea7a73139a
# good: [d0c4b7a7d2aa12537a06527692f10066b7acbed9] target/arm: create kvm cpu accel class
git bisect good d0c4b7a7d2aa12537a06527692f10066b7acbed9
# bad: [3288c8a8cfc8dd9999e7b5908eaebb561f0169eb] target/arm: add tcg cpu accel class
git bisect bad 3288c8a8cfc8dd9999e7b5908eaebb561f0169eb
# good: [683b948d40bc81f1eaed27bef8631ab3f8a937d6] target/arm: move kvm post init initialization to kvm cpu accel
git bisect good 683b948d40bc81f1eaed27bef8631ab3f8a937d6
# first bad commit: [3288c8a8cfc8dd9999e7b5908eaebb561f0169eb] target/arm: add tcg cpu accel class
3288c8a8cfc8dd9999e7b5908eaebb561f0169eb is the first bad commit
commit 3288c8a8cfc8dd9999e7b5908eaebb561f0169eb
Author: Claudio Fontana <cfontana@suse.de>
Date: Mon Mar 22 15:01:56 2021 +0100
target/arm: add tcg cpu accel class
move init, realizefn and reset code into it.
Signed-off-by: Claudio Fontana <cfontana@suse.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20210322140206.9513-40-cfontana@suse.de>
target/arm/cpu.c | 44 +++----------------------------
target/arm/tcg/sysemu/tcg-cpu.c | 27 +++++++++++++++++++
target/arm/tcg/tcg-cpu-models.c | 11 +++++---
target/arm/tcg/tcg-cpu.c | 57 +++++++++++++++++++++++++++++++++++++++--
target/arm/tcg/tcg-cpu.h | 4 ++-
5 files changed, 96 insertions(+), 47 deletions(-)
bisect run success
>
> Signed-off-by: Claudio Fontana <cfontana@suse.de>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> ---
> target/arm/tcg/tcg-cpu.h | 4 ++-
> target/arm/cpu.c | 44 ++-----------------------
> target/arm/tcg/sysemu/tcg-cpu.c | 27 ++++++++++++++++
> target/arm/tcg/tcg-cpu-models.c | 11 +++++--
> target/arm/tcg/tcg-cpu.c | 57 +++++++++++++++++++++++++++++++--
> 5 files changed, 96 insertions(+), 47 deletions(-)
>
> diff --git a/target/arm/tcg/tcg-cpu.h b/target/arm/tcg/tcg-cpu.h
> index d93c6a6749..dd08587949 100644
> --- a/target/arm/tcg/tcg-cpu.h
> +++ b/target/arm/tcg/tcg-cpu.h
> @@ -22,15 +22,17 @@
>
> #include "cpu.h"
> #include "hw/core/tcg-cpu-ops.h"
> +#include "hw/core/accel-cpu.h"
>
> void arm_cpu_synchronize_from_tb(CPUState *cs,
> const TranslationBlock *tb);
>
> -extern struct TCGCPUOps arm_tcg_ops;
> +void tcg_arm_init_accel_cpu(AccelCPUClass *accel_cpu, CPUClass *cc);
>
> #ifndef CONFIG_USER_ONLY
> /* Do semihosting call and set the appropriate return value. */
> void tcg_handle_semihosting(CPUState *cs);
> +bool tcg_cpu_realizefn(CPUState *cs, Error **errp);
>
> #endif /* !CONFIG_USER_ONLY */
>
> diff --git a/target/arm/cpu.c b/target/arm/cpu.c
> index 5e0f6bd01d..9248e096df 100644
> --- a/target/arm/cpu.c
> +++ b/target/arm/cpu.c
> @@ -410,12 +410,6 @@ static void arm_cpu_reset(DeviceState *dev)
> &env->vfp.fp_status_f16);
> set_float_detect_tininess(float_tininess_before_rounding,
> &env->vfp.standard_fp_status_f16);
> -
> - if (tcg_enabled()) {
> - hw_breakpoint_update_all(cpu);
> - hw_watchpoint_update_all(cpu);
> - arm_rebuild_hflags(env);
> - }
> }
>
> void arm_cpu_update_virq(ARMCPU *cpu)
> @@ -576,10 +570,6 @@ static void arm_cpu_initfn(Object *obj)
> cpu->dtb_compatible = "qemu,unknown";
> cpu->psci_version = 1; /* By default assume PSCI v0.1 */
> cpu->kvm_target = QEMU_KVM_ARM_TARGET_NONE;
> -
> - if (tcg_enabled()) {
> - cpu->psci_version = 2; /* TCG implements PSCI 0.2 */
> - }
> }
>
> static Property arm_cpu_gt_cntfrq_property =
> @@ -868,34 +858,7 @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp)
> Error *local_err = NULL;
> bool no_aa32 = false;
>
> - /*
> - * If we needed to query the host kernel for the CPU features
> - * then it's possible that might have failed in the initfn, but
> - * this is the first point where we can report it.
> - */
> - if (cpu->host_cpu_probe_failed) {
> - error_setg(errp, "The 'host' CPU type can only be used with KVM");
> - return;
> - }
> -
> -#ifndef CONFIG_USER_ONLY
> - /* The NVIC and M-profile CPU are two halves of a single piece of
> - * hardware; trying to use one without the other is a command line
> - * error and will result in segfaults if not caught here.
> - */
> - if (arm_feature(env, ARM_FEATURE_M)) {
> - if (!env->nvic) {
> - error_setg(errp, "This board cannot be used with Cortex-M CPUs");
> - return;
> - }
> - } else {
> - if (env->nvic) {
> - error_setg(errp, "This board can only be used with Cortex-M CPUs");
> - return;
> - }
> - }
> -
> -#ifdef CONFIG_TCG
> +#if defined(CONFIG_TCG) && !defined(CONFIG_USER_ONLY)
> {
> uint64_t scale;
>
> @@ -921,8 +884,7 @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp)
> cpu->gt_timer[GTIMER_HYPVIRT] = timer_new(QEMU_CLOCK_VIRTUAL, scale,
> arm_gt_hvtimer_cb, cpu);
> }
> -#endif /* CONFIG_TCG */
> -#endif /* !CONFIG_USER_ONLY */
> +#endif /* CONFIG_TCG && !CONFIG_USER_ONLY */
>
> cpu_exec_realizefn(cs, &local_err);
> if (local_err != NULL) {
> @@ -1458,7 +1420,7 @@ static void arm_cpu_class_init(ObjectClass *oc, void *data)
> cc->disas_set_info = arm_disas_set_info;
>
> #ifdef CONFIG_TCG
> - cc->tcg_ops = &arm_tcg_ops;
> + cc->init_accel_cpu = tcg_arm_init_accel_cpu;
> #endif /* CONFIG_TCG */
>
> arm32_cpu_class_init(oc, data);
> diff --git a/target/arm/tcg/sysemu/tcg-cpu.c b/target/arm/tcg/sysemu/tcg-cpu.c
> index 327b2a5073..115ac523dc 100644
> --- a/target/arm/tcg/sysemu/tcg-cpu.c
> +++ b/target/arm/tcg/sysemu/tcg-cpu.c
> @@ -19,10 +19,13 @@
> */
>
> #include "qemu/osdep.h"
> +#include "qapi/error.h"
> +#include "qemu/timer.h"
> #include "cpu.h"
> #include "semihosting/common-semi.h"
> #include "qemu/log.h"
> #include "tcg/tcg-cpu.h"
> +#include "internals.h"
>
> /*
> * Do semihosting call and set the appropriate return value. All the
> @@ -50,3 +53,27 @@ void tcg_handle_semihosting(CPUState *cs)
> env->regs[15] += env->thumb ? 2 : 4;
> }
> }
> +
> +bool tcg_cpu_realizefn(CPUState *cs, Error **errp)
> +{
> + ARMCPU *cpu = ARM_CPU(cs);
> + CPUARMState *env = &cpu->env;
> +
> + /*
> + * The NVIC and M-profile CPU are two halves of a single piece of
> + * hardware; trying to use one without the other is a command line
> + * error and will result in segfaults if not caught here.
> + */
> + if (arm_feature(env, ARM_FEATURE_M)) {
> + if (!env->nvic) {
> + error_setg(errp, "This board cannot be used with Cortex-M CPUs");
> + return false;
> + }
> + } else {
> + if (env->nvic) {
> + error_setg(errp, "This board can only be used with Cortex-M CPUs");
> + return false;
> + }
> + }
> + return true;
> +}
> diff --git a/target/arm/tcg/tcg-cpu-models.c b/target/arm/tcg/tcg-cpu-models.c
> index 16ab5d5364..2f44fd1b41 100644
> --- a/target/arm/tcg/tcg-cpu-models.c
> +++ b/target/arm/tcg/tcg-cpu-models.c
> @@ -844,15 +844,20 @@ static struct TCGCPUOps arm_v7m_tcg_ops = {
> #endif /* !CONFIG_USER_ONLY */
> };
>
> +static void arm_v7m_init_accel_cpu(AccelCPUClass *accel_cpu, CPUClass *cc)
> +{
> + g_assert(object_class_by_name(ACCEL_CPU_NAME("tcg")) == OBJECT_CLASS(accel_cpu));
> +
> + cc->tcg_ops = &arm_v7m_tcg_ops;
> +}
> +
> static void arm_v7m_class_init(ObjectClass *oc, void *data)
> {
> ARMCPUClass *acc = ARM_CPU_CLASS(oc);
> CPUClass *cc = CPU_CLASS(oc);
>
> acc->info = data;
> -#ifdef CONFIG_TCG
> - cc->tcg_ops = &arm_v7m_tcg_ops;
> -#endif /* CONFIG_TCG */
> + cc->init_accel_cpu = arm_v7m_init_accel_cpu;
>
> cc->gdb_core_xml_file = "arm-m-profile.xml";
> }
> diff --git a/target/arm/tcg/tcg-cpu.c b/target/arm/tcg/tcg-cpu.c
> index 9fd996d908..d6c3a0ba41 100644
> --- a/target/arm/tcg/tcg-cpu.c
> +++ b/target/arm/tcg/tcg-cpu.c
> @@ -20,8 +20,8 @@
>
> #include "qemu/osdep.h"
> #include "cpu.h"
> +#include "qapi/error.h"
> #include "tcg-cpu.h"
> -#include "hw/core/tcg-cpu-ops.h"
> #include "cpregs.h"
> #include "internals.h"
> #include "exec/exec-all.h"
> @@ -212,7 +212,7 @@ static bool arm_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
> return true;
> }
>
> -struct TCGCPUOps arm_tcg_ops = {
> +static struct TCGCPUOps arm_tcg_ops = {
> .initialize = arm_translate_init,
> .synchronize_from_tb = arm_cpu_synchronize_from_tb,
> .cpu_exec_interrupt = arm_cpu_exec_interrupt,
> @@ -227,3 +227,56 @@ struct TCGCPUOps arm_tcg_ops = {
> .debug_check_watchpoint = arm_debug_check_watchpoint,
> #endif /* !CONFIG_USER_ONLY */
> };
> +
> +void tcg_arm_init_accel_cpu(AccelCPUClass *accel_cpu, CPUClass *cc)
> +{
> + g_assert(object_class_by_name(ACCEL_CPU_NAME("tcg")) == OBJECT_CLASS(accel_cpu));
> +
> + cc->tcg_ops = &arm_tcg_ops;
> +}
> +
> +static void tcg_cpu_instance_init(CPUState *cs)
> +{
> + ARMCPU *cpu = ARM_CPU(cs);
> +
> + /*
> + * this would be the place to move TCG-specific props
> + * in future refactoring of cpu properties.
> + */
> +
> + cpu->psci_version = 2; /* TCG implements PSCI 0.2 */
> +}
> +
> +static void tcg_cpu_reset(CPUState *cs)
> +{
> + ARMCPU *cpu = ARM_CPU(cs);
> + CPUARMState *env = &cpu->env;
> +
> + hw_breakpoint_update_all(cpu);
> + hw_watchpoint_update_all(cpu);
> + arm_rebuild_hflags(env);
> +}
> +
> +static void tcg_cpu_accel_class_init(ObjectClass *oc, void *data)
> +{
> + AccelCPUClass *acc = ACCEL_CPU_CLASS(oc);
> +
> +#ifndef CONFIG_USER_ONLY
> + acc->cpu_realizefn = tcg_cpu_realizefn;
> +#endif /* CONFIG_USER_ONLY */
> +
> + acc->cpu_instance_init = tcg_cpu_instance_init;
> + acc->cpu_reset = tcg_cpu_reset;
> +}
> +static const TypeInfo tcg_cpu_accel_type_info = {
> + .name = ACCEL_CPU_NAME("tcg"),
> +
> + .parent = TYPE_ACCEL_CPU,
> + .class_init = tcg_cpu_accel_class_init,
> + .abstract = true,
> +};
> +static void tcg_cpu_accel_register_types(void)
> +{
> + type_register_static(&tcg_cpu_accel_type_info);
> +}
> +type_init(tcg_cpu_accel_register_types);
--
Alex Bennée
next prev parent reply other threads:[~2021-03-23 19:29 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-22 14:01 [RFC v10 00/49] arm cleanup experiment for kvm-only build Claudio Fontana
2021-03-22 14:01 ` [RFC v10 01/49] target/arm: move translate modules to tcg/ Claudio Fontana
2021-03-22 16:27 ` Alex Bennée
2021-03-22 14:01 ` [RFC v10 02/49] target/arm: move helpers " Claudio Fontana
2021-03-22 16:29 ` Alex Bennée
2021-03-22 14:01 ` [RFC v10 03/49] arm: tcg: only build under CONFIG_TCG Claudio Fontana
2021-03-22 16:32 ` Alex Bennée
2021-03-22 14:01 ` [RFC v10 04/49] target/arm: tcg: add sysemu and user subsirs Claudio Fontana
2021-03-22 16:32 ` Alex Bennée
2021-03-22 14:01 ` [RFC v10 05/49] target/arm: only build psci for TCG Claudio Fontana
2021-03-22 14:01 ` [RFC v10 06/49] target/arm: split off cpu-sysemu.c Claudio Fontana
2021-03-22 17:17 ` Alex Bennée
2021-03-22 14:01 ` [RFC v10 07/49] target/arm: move physical address translation to cpu-mmu Claudio Fontana
2021-03-22 14:01 ` [RFC v10 08/49] target/arm: cpu-mmu: fix comment style Claudio Fontana
2021-03-22 14:01 ` [RFC v10 09/49] target/arm: split cpregs from tcg/helper.c Claudio Fontana
2021-03-22 14:01 ` [RFC v10 10/49] target/arm: cpregs: fix style (mostly just comments) Claudio Fontana
2021-03-22 14:01 ` [RFC v10 11/49] target/arm: move cpu definitions to common cpu module Claudio Fontana
2021-03-22 14:01 ` [RFC v10 12/49] target/arm: only perform TCG cpu and machine inits if TCG enabled Claudio Fontana
2021-03-22 14:01 ` [RFC v10 13/49] target/arm: kvm: add stubs for some helpers Claudio Fontana
2021-03-22 14:01 ` [RFC v10 14/49] target/arm: move cpsr_read, cpsr_write to cpu_common Claudio Fontana
2021-03-22 14:01 ` [RFC v10 15/49] target/arm: add temporary stub for arm_rebuild_hflags Claudio Fontana
2021-03-22 14:01 ` [RFC v10 16/49] target/arm: split vfp state setting from tcg helpers Claudio Fontana
2021-03-22 14:01 ` [RFC v10 17/49] target/arm: move arm_mmu_idx* to cpu-mmu Claudio Fontana
2021-03-22 14:01 ` [RFC v10 18/49] target/arm: move sve_zcr_len_for_el to common_cpu Claudio Fontana
2021-03-22 14:01 ` [RFC v10 19/49] target/arm: move arm_sctlr away from tcg helpers Claudio Fontana
2021-03-22 14:01 ` [RFC v10 20/49] target/arm: move arm_cpu_list to common_cpu Claudio Fontana
2021-03-22 14:01 ` [RFC v10 21/49] target/arm: move aarch64_sync_32_to_64 (and vv) to cpu code Claudio Fontana
2021-03-22 14:01 ` [RFC v10 22/49] target/arm: split a15 cpu model and 32bit class functions to cpu32.c Claudio Fontana
2021-03-22 14:01 ` [RFC v10 23/49] target/arm: move sve_exception_el out of TCG helpers Claudio Fontana
2021-03-22 14:01 ` [RFC v10 24/49] target/arm: refactor exception and cpu code Claudio Fontana
2021-03-22 14:01 ` [RFC v10 25/49] target/arm: cpu: fix style Claudio Fontana
2021-03-22 14:01 ` [RFC v10 26/49] target/arm: wrap call to aarch64_sve_change_el in tcg_enabled() Claudio Fontana
2021-03-22 14:01 ` [RFC v10 27/49] target/arm: remove kvm include file for PSCI and arm-powerctl Claudio Fontana
2021-03-22 14:01 ` [RFC v10 28/49] target/arm: move kvm-const.h, kvm.c, kvm64.c, kvm_arm.h to kvm/ Claudio Fontana
2021-03-22 14:01 ` [RFC v10 29/49] target/arm: cleanup cpu includes Claudio Fontana
2021-03-22 14:01 ` [RFC v10 30/49] target/arm: remove broad "else" statements when checking accels Claudio Fontana
2021-03-22 14:01 ` [RFC v10 31/49] tests/qtest: skip bios-tables-test test_acpi_oem_fields_virt for KVM Claudio Fontana
2021-03-22 14:01 ` [RFC v10 32/49] tests: restrict TCG-only arm-cpu-features tests to TCG builds Claudio Fontana
2021-03-22 14:01 ` [RFC v10 33/49] tests: do not run test-hmp on all machines for ARM KVM-only Claudio Fontana
2021-03-22 14:01 ` [RFC v10 34/49] tests: device-introspect-test: cope with ARM TCG-only devices Claudio Fontana
2021-03-22 14:01 ` [RFC v10 35/49] tests: do not run qom-test on all machines for ARM KVM-only Claudio Fontana
2021-03-22 14:01 ` [RFC v10 36/49] Revert "target/arm: Restrict v8M IDAU to TCG" Claudio Fontana
2021-03-22 14:01 ` [RFC v10 37/49] target/arm: create kvm cpu accel class Claudio Fontana
2021-03-22 14:01 ` [RFC v10 38/49] target/arm: move kvm post init initialization to kvm cpu accel Claudio Fontana
2021-03-22 14:01 ` [RFC v10 39/49] target/arm: add tcg cpu accel class Claudio Fontana
2021-03-23 19:26 ` Alex Bennée [this message]
2021-03-23 19:33 ` Claudio Fontana
2021-03-22 14:01 ` [RFC v10 40/49] target/arm: move TCG gt timer creation code in tcg/ Claudio Fontana
2021-03-22 14:01 ` [RFC v10 41/49] target/arm: cpu-sve: new module Claudio Fontana
2021-03-22 14:01 ` [RFC v10 42/49] target/arm: cpu-sve: split TCG and KVM functionality Claudio Fontana
2021-03-22 14:02 ` [RFC v10 43/49] target/arm: make is_aa64 and arm_el_is_aa64 a macro for !TARGET_AARCH64 Claudio Fontana
2021-03-22 14:02 ` [RFC v10 44/49] target/arm: arch_dump: restrict ELFCLASS64 to AArch64 Claudio Fontana
2021-03-22 14:02 ` [RFC v10 45/49] target/arm: cpu-exceptions: new module Claudio Fontana
2021-03-22 14:02 ` [RFC v10 46/49] target/arm: tcg: restrict ZCR cpregs to TARGET_AARCH64 Claudio Fontana
2021-03-22 14:02 ` [RFC v10 47/49] target/arm: cpu-pauth: new module for ARMv8.3 Pointer Authentication Claudio Fontana
2021-03-22 14:02 ` [RFC v10 48/49] target/arm: refactor arm_cpu_finalize_features into cpu64 Claudio Fontana
2021-03-22 14:02 ` [RFC v10 49/49] XXX target/arm: experiment refactoring cpu "max" Claudio Fontana
2021-03-23 18:35 ` [RFC v10 00/49] arm cleanup experiment for kvm-only build Alex Bennée
2021-03-23 19:20 ` Claudio Fontana
2021-03-23 19:38 ` Richard Henderson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87h7l1isb0.fsf@linaro.org \
--to=alex.bennee@linaro.org \
--cc=cfontana@suse.de \
--cc=ehabkost@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=philmd@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=r.bolshakov@yadro.com \
--cc=richard.henderson@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).