This is pure duplication now. Both bsd-user and linux-user have builtin strace support and we can also track syscalls via the plugins system. Reviewed-by: Warner Losh <imp(a)bsdimp.com&gt; Reviewed-by: Stefan Hajnoczi <stefanha(a)redhat.com&gt; Reviewed-by: Richard Henderson <richard.henderson(a)linaro.org&gt; Reviewed-by: Philippe Mathieu-Daudé <philmd(a)linaro.org&gt; Signed-off-by: Alex Bennée <alex.bennee(a)linaro.org&gt; Message-Id: <20230524133952.3971948-2-alex.bennee(a)linaro.org&gt; --- include/user/syscall-trace.h | 4 ---- bsd-user/freebsd/os-syscall.c | 2 -- trace-events | 19 ------------------- 3 files changed, 25 deletions(-) diff --git a/include/user/syscall-trace.h b/include/user/syscall-trace.h index 90bda7631c..557f881a79 100644 --- a/include/user/syscall-trace.h +++ b/include/user/syscall-trace.h @@ -26,9 +26,6 @@ static inline void record_syscall_start(void *cpu, int num, abi_long arg5, abi_long arg6, abi_long arg7, abi_long arg8) { - trace_guest_user_syscall(cpu, num, - arg1, arg2, arg3, arg4, - arg5, arg6, arg7, arg8); qemu_plugin_vcpu_syscall(cpu, num, arg1, arg2, arg3, arg4, arg5, arg6, arg7, arg8); @@ -36,7 +33,6 @@ static inline void record_syscall_start(void *cpu, int num, static inline void record_syscall_return(void *cpu, int num, abi_long ret) { - trace_guest_user_syscall_ret(cpu, num, ret); qemu_plugin_vcpu_syscall_ret(cpu, num, ret); } diff --git a/bsd-user/freebsd/os-syscall.c b/bsd-user/freebsd/os-syscall.c index c8f998ecec..b0ae43766f 100644 --- a/bsd-user/freebsd/os-syscall.c +++ b/bsd-user/freebsd/os-syscall.c @@ -531,7 +531,6 @@ abi_long do_freebsd_syscall(void *cpu_env, int num, abi_long arg1, CPUState *cpu = env_cpu(cpu_env); abi_long ret; - trace_guest_user_syscall(cpu, num, arg1, arg2, arg3, arg4, arg5, arg6, arg7, arg8); if (do_strace) { print_freebsd_syscall(num, arg1, arg2, arg3, arg4, arg5, arg6); } @@ -541,7 +540,6 @@ abi_long do_freebsd_syscall(void *cpu_env, int num, abi_long arg1, if (do_strace) { print_freebsd_syscall_ret(num, ret); } - trace_guest_user_syscall_ret(cpu, num, ret); return ret; } diff --git a/trace-events b/trace-events index b6b84b175e..691c3533e4 100644 --- a/trace-events +++ b/trace-events @@ -85,22 +85,3 @@ vcpu guest_cpu_exit(void) # Targets: all vcpu guest_cpu_reset(void) -# include/user/syscall-trace.h - -# @num: System call number. -# @arg*: System call argument value. -# -# Start executing a guest system call in syscall emulation mode. -# -# Mode: user -# Targets: TCG(all) -vcpu guest_user_syscall(uint64_t num, uint64_t arg1, uint64_t arg2, uint64_t arg3, uint64_t arg4, uint64_t arg5, uint64_t arg6, uint64_t arg7, uint64_t arg8) "num=0x%016"PRIx64" arg1=0x%016"PRIx64" arg2=0x%016"PRIx64" arg3=0x%016"PRIx64" arg4=0x%016"PRIx64" arg5=0x%016"PRIx64" arg6=0x%016"PRIx64" arg7=0x%016"PRIx64" arg8=0x%016"PRIx64 - -# @num: System call number. -# @ret: System call result value. -# -# Finish executing a guest system call in syscall emulation mode. -# -# Mode: user -# Targets: TCG(all) -vcpu guest_user_syscall_ret(uint64_t num, uint64_t ret) "num=0x%016"PRIx64" ret=0x%016"PRIx64 -- 2.39.2

This does involve temporarily stubbing out some helper functions before we excise the rest of the code. Reviewed-by: Philippe Mathieu-Daudé <philmd(a)linaro.org&gt; Reviewed-by: Stefan Hajnoczi <stefanha(a)redhat.com&gt; Reviewed-by: Richard Henderson <richard.henderson(a)linaro.org&gt; Signed-off-by: Alex Bennée <alex.bennee(a)linaro.org&gt; Message-Id: <20230524133952.3971948-4-alex.bennee(a)linaro.org&gt; --- trace/control-internal.h | 4 ++-- trace/event-internal.h | 2 -- trace/control.c | 10 ---------- scripts/tracetool/format/c.py | 6 ------ scripts/tracetool/format/h.py | 11 +---------- 5 files changed, 3 insertions(+), 30 deletions(-) diff --git a/trace/control-internal.h b/trace/control-internal.h index 8b2b50a7cf..0178121720 100644 --- a/trace/control-internal.h +++ b/trace/control-internal.h @@ -27,12 +27,12 @@ static inline uint32_t trace_event_get_id(TraceEvent *ev) static inline uint32_t trace_event_get_vcpu_id(TraceEvent *ev) { - return ev->vcpu_id; + return 0; } static inline bool trace_event_is_vcpu(TraceEvent *ev) { - return ev->vcpu_id != TRACE_VCPU_EVENT_NONE; + return false; } static inline const char * trace_event_get_name(TraceEvent *ev) diff --git a/trace/event-internal.h b/trace/event-internal.h index f63500b37e..0c24e01b52 100644 --- a/trace/event-internal.h +++ b/trace/event-internal.h @@ -19,7 +19,6 @@ /** * TraceEvent: * @id: Unique event identifier. - * @vcpu_id: Unique per-vCPU event identifier. * @name: Event name. * @sstate: Static tracing state. * @dstate: Dynamic tracing state @@ -33,7 +32,6 @@ */ typedef struct TraceEvent { uint32_t id; - uint32_t vcpu_id; const char * name; const bool sstate; uint16_t *dstate; diff --git a/trace/control.c b/trace/control.c index d24af91004..5dfb609954 100644 --- a/trace/control.c +++ b/trace/control.c @@ -68,16 +68,6 @@ void trace_event_register_group(TraceEvent **events) size_t i; for (i = 0; events[i] != NULL; i++) { events[i]->id = next_id++; - if (events[i]->vcpu_id == TRACE_VCPU_EVENT_NONE) { - continue; - } - - if (likely(next_vcpu_id < CPU_TRACE_DSTATE_MAX_EVENTS)) { - events[i]->vcpu_id = next_vcpu_id++; - } else { - warn_report("too many vcpu trace events; dropping '%s'", - events[i]->name); - } } event_groups = g_renew(TraceEventGroup, event_groups, nevent_groups + 1); event_groups[nevent_groups].events = events; diff --git a/scripts/tracetool/format/c.py b/scripts/tracetool/format/c.py index c390c1844a..69edf0d588 100644 --- a/scripts/tracetool/format/c.py +++ b/scripts/tracetool/format/c.py @@ -32,19 +32,13 @@ def generate(events, backend, group): out('uint16_t %s;' % e.api(e.QEMU_DSTATE)) for e in events: - if "vcpu" in e.properties: - vcpu_id = 0 - else: - vcpu_id = "TRACE_VCPU_EVENT_NONE" out('TraceEvent %(event)s = {', ' .id = 0,', - ' .vcpu_id = %(vcpu_id)s,', ' .name = \"%(name)s\",', ' .sstate = %(sstate)s,', ' .dstate = &%(dstate)s ', '};', event = e.api(e.QEMU_EVENT), - vcpu_id = vcpu_id, name = e.name, sstate = "TRACE_%s_ENABLED" % e.name.upper(), dstate = e.api(e.QEMU_DSTATE)) diff --git a/scripts/tracetool/format/h.py b/scripts/tracetool/format/h.py index e94f0be7da..285d7b03a9 100644 --- a/scripts/tracetool/format/h.py +++ b/scripts/tracetool/format/h.py @@ -74,16 +74,7 @@ def generate(events, backend, group): out('}') - # tracer wrapper with checks (per-vCPU tracing) - if "vcpu" in e.properties: - trace_cpu = next(iter(e.args))[1] - cond = "trace_event_get_vcpu_state(%(cpu)s,"\ - " TRACE_%(id)s)"\ - % dict( - cpu=trace_cpu, - id=e.name.upper()) - else: - cond = "true" + cond = "true" out('', 'static inline void %(api)s(%(args)s)', -- 2.39.2

Also rename the section to make the fact this is part of the management protocol even clearer. Suggested-by: Markus Armbruster <armbru(a)redhat.com&gt; Signed-off-by: Alex Bennée <alex.bennee(a)linaro.org&gt; --- docs/about/deprecated.rst | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/docs/about/deprecated.rst b/docs/about/deprecated.rst index e934e0a13a..7c45a64363 100644 --- a/docs/about/deprecated.rst +++ b/docs/about/deprecated.rst @@ -218,6 +218,15 @@ instruction per translated block" mode (which can be set on the command line or via the HMP, but not via QMP). The information remains available via the HMP 'info jit' command. +QEMU Machine Protocol (QMP) events +---------------------------------- + +``MEM_UNPLUG_ERROR`` (since 6.2) +'''''''''''''''''''''''''''''''''''''''''''''''''''''''' + +Use the more generic event ``DEVICE_UNPLUG_GUEST_ERROR`` instead. + + Human Monitor Protocol (HMP) commands ------------------------------------- @@ -251,15 +260,6 @@ it. Since all recent x86 hardware from the past >10 years is capable of the 64-bit x86 extensions, a corresponding 64-bit OS should be used instead. -QEMU API (QAPI) events ----------------------- - -``MEM_UNPLUG_ERROR`` (since 6.2) -'''''''''''''''''''''''''''''''''''''''''''''''''''''''' - -Use the more generic event ``DEVICE_UNPLUG_GUEST_ERROR`` instead. - - System emulator machines ------------------------ -- 2.39.2

Now we no longer have any events that are for vcpus we can start excising the code from the trace control. As the vcpu parameter is encoded as part of QMP we just stub out the has_vcpu/vcpu parameters rather than alter the API. Reviewed-by: Stefan Hajnoczi <stefanha(a)redhat.com&gt; Reviewed-by: Richard Henderson <richard.henderson(a)linaro.org&gt; Signed-off-by: Alex Bennée <alex.bennee(a)linaro.org&gt; Message-Id: <20230524133952.3971948-7-alex.bennee(a)linaro.org&gt; --- trace/control-internal.h | 10 ---- trace/control-vcpu.h | 16 ------ trace/control.h | 48 ----------------- hw/core/cpu-common.c | 2 - stubs/trace-control.c | 13 ----- trace/control-target.c | 108 ++++----------------------------------- trace/control.c | 16 ------ trace/qmp.c | 74 +++------------------------ trace/trace-hmp-cmds.c | 18 ++----- 9 files changed, 20 insertions(+), 285 deletions(-) diff --git a/trace/control-internal.h b/trace/control-internal.h index 0178121720..8d818d359b 100644 --- a/trace/control-internal.h +++ b/trace/control-internal.h @@ -25,16 +25,6 @@ static inline uint32_t trace_event_get_id(TraceEvent *ev) return ev->id; } -static inline uint32_t trace_event_get_vcpu_id(TraceEvent *ev) -{ - return 0; -} - -static inline bool trace_event_is_vcpu(TraceEvent *ev) -{ - return false; -} - static inline const char * trace_event_get_name(TraceEvent *ev) { assert(ev != NULL); diff --git a/trace/control-vcpu.h b/trace/control-vcpu.h index 0f98ebe7b5..800fc5a219 100644 --- a/trace/control-vcpu.h +++ b/trace/control-vcpu.h @@ -30,13 +30,6 @@ trace_event_get_vcpu_state_dynamic_by_vcpu_id( \ vcpu, _ ## id ## _EVENT.vcpu_id)) -/** - * trace_event_get_vcpu_state_dynamic: - * - * Get the dynamic tracing state of an event for the given vCPU. - */ -static bool trace_event_get_vcpu_state_dynamic(CPUState *vcpu, TraceEvent *ev); - #include "control-internal.h" static inline bool @@ -51,13 +44,4 @@ trace_event_get_vcpu_state_dynamic_by_vcpu_id(CPUState *vcpu, } } -static inline bool trace_event_get_vcpu_state_dynamic(CPUState *vcpu, - TraceEvent *ev) -{ - uint32_t vcpu_id; - assert(trace_event_is_vcpu(ev)); - vcpu_id = trace_event_get_vcpu_id(ev); - return trace_event_get_vcpu_state_dynamic_by_vcpu_id(vcpu, vcpu_id); -} - #endif diff --git a/trace/control.h b/trace/control.h index 23b8393b29..dfd209edd8 100644 --- a/trace/control.h +++ b/trace/control.h @@ -89,23 +89,6 @@ static bool trace_event_is_pattern(const char *str); */ static uint32_t trace_event_get_id(TraceEvent *ev); -/** - * trace_event_get_vcpu_id: - * - * Get the per-vCPU identifier of an event. - * - * Special value #TRACE_VCPU_EVENT_NONE means the event is not vCPU-specific - * (does not have the "vcpu" property). - */ -static uint32_t trace_event_get_vcpu_id(TraceEvent *ev); - -/** - * trace_event_is_vcpu: - * - * Whether this is a per-vCPU event. - */ -static bool trace_event_is_vcpu(TraceEvent *ev); - /** * trace_event_get_name: * @@ -172,21 +155,6 @@ static bool trace_event_get_state_dynamic(TraceEvent *ev); */ void trace_event_set_state_dynamic(TraceEvent *ev, bool state); -/** - * trace_event_set_vcpu_state_dynamic: - * - * Set the dynamic tracing state of an event for the given vCPU. - * - * Pre-condition: trace_event_get_vcpu_state_static(ev) == true - * - * Note: Changes for execution-time events with the 'tcg' property will not be - * propagated until the next TB is executed (iff executing in TCG mode). - */ -void trace_event_set_vcpu_state_dynamic(CPUState *vcpu, - TraceEvent *ev, bool state); - - - /** * trace_init_backends: * @@ -205,22 +173,6 @@ bool trace_init_backends(void); */ void trace_init_file(void); -/** - * trace_init_vcpu: - * @vcpu: Added vCPU. - * - * Set initial dynamic event state for a hot-plugged vCPU. - */ -void trace_init_vcpu(CPUState *vcpu); - -/** - * trace_fini_vcpu: - * @vcpu: Removed vCPU. - * - * Disable dynamic event state for a hot-unplugged vCPU. - */ -void trace_fini_vcpu(CPUState *vcpu); - /** * trace_list_events: * @f: Where to send output. diff --git a/hw/core/cpu-common.c b/hw/core/cpu-common.c index 951477a7fd..f4e51c8a1b 100644 --- a/hw/core/cpu-common.c +++ b/hw/core/cpu-common.c @@ -211,7 +211,6 @@ static void cpu_common_realizefn(DeviceState *dev, Error **errp) } /* NOTE: latest generic point where the cpu is fully realized */ - trace_init_vcpu(cpu); } static void cpu_common_unrealizefn(DeviceState *dev) @@ -219,7 +218,6 @@ static void cpu_common_unrealizefn(DeviceState *dev) CPUState *cpu = CPU(dev); /* NOTE: latest generic point before the cpu is fully unrealized */ - trace_fini_vcpu(cpu); cpu_exec_unrealizefn(cpu); } diff --git a/stubs/trace-control.c b/stubs/trace-control.c index 7f856e5c24..b428f34c87 100644 --- a/stubs/trace-control.c +++ b/stubs/trace-control.c @@ -36,16 +36,3 @@ void trace_event_set_state_dynamic(TraceEvent *ev, bool state) } } } - -void trace_event_set_vcpu_state_dynamic(CPUState *vcpu, - TraceEvent *ev, bool state) -{ - /* should never be called on non-target binaries */ - abort(); -} - -void trace_init_vcpu(CPUState *vcpu) -{ - /* should never be called on non-target binaries */ - abort(); -} diff --git a/trace/control-target.c b/trace/control-target.c index a10752924b..97f21e476d 100644 --- a/trace/control-target.c +++ b/trace/control-target.c @@ -36,112 +36,22 @@ void trace_event_set_state_dynamic_init(TraceEvent *ev, bool state) void trace_event_set_state_dynamic(TraceEvent *ev, bool state) { - CPUState *vcpu; assert(trace_event_get_state_static(ev)); - if (trace_event_is_vcpu(ev) && likely(first_cpu != NULL)) { - CPU_FOREACH(vcpu) { - trace_event_set_vcpu_state_dynamic(vcpu, ev, state); - } - } else { - /* - * Without the "vcpu" property, dstate can only be 1 or 0. With it, we - * haven't instantiated any vCPU yet, so we will set a global state - * instead, and trace_init_vcpu will reconcile it afterwards. - */ - bool state_pre = *ev->dstate; - if (state_pre != state) { - if (state) { - trace_events_enabled_count++; - *ev->dstate = 1; - } else { - trace_events_enabled_count--; - *ev->dstate = 0; - } - } - } -} -static void trace_event_synchronize_vcpu_state_dynamic( - CPUState *vcpu, run_on_cpu_data ignored) -{ - bitmap_copy(vcpu->trace_dstate, vcpu->trace_dstate_delayed, - CPU_TRACE_DSTATE_MAX_EVENTS); - tcg_flush_jmp_cache(vcpu); -} - -void trace_event_set_vcpu_state_dynamic(CPUState *vcpu, - TraceEvent *ev, bool state) -{ - uint32_t vcpu_id; - bool state_pre; - assert(trace_event_get_state_static(ev)); - assert(trace_event_is_vcpu(ev)); - vcpu_id = trace_event_get_vcpu_id(ev); - state_pre = test_bit(vcpu_id, vcpu->trace_dstate); + /* + * There is no longer a "vcpu" property, dstate can only be 1 or + * 0. With it, we haven't instantiated any vCPU yet, so we will + * set a global state instead, and trace_init_vcpu will reconcile + * it afterwards. + */ + bool state_pre = *ev->dstate; if (state_pre != state) { if (state) { trace_events_enabled_count++; - set_bit(vcpu_id, vcpu->trace_dstate_delayed); - (*ev->dstate)++; + *ev->dstate = 1; } else { trace_events_enabled_count--; - clear_bit(vcpu_id, vcpu->trace_dstate_delayed); - (*ev->dstate)--; - } - if (vcpu->created) { - /* - * Delay changes until next TB; we want all TBs to be built from a - * single set of dstate values to ensure consistency of generated - * tracing code. - */ - async_run_on_cpu(vcpu, trace_event_synchronize_vcpu_state_dynamic, - RUN_ON_CPU_NULL); - } else { - trace_event_synchronize_vcpu_state_dynamic(vcpu, RUN_ON_CPU_NULL); - } - } -} - -static bool adding_first_cpu1(void) -{ - CPUState *cpu; - size_t count = 0; - CPU_FOREACH(cpu) { - count++; - if (count > 1) { - return false; - } - } - return true; -} - -static bool adding_first_cpu(void) -{ - QEMU_LOCK_GUARD(&qemu_cpu_list_lock); - - return adding_first_cpu1(); -} - -void trace_init_vcpu(CPUState *vcpu) -{ - TraceEventIter iter; - TraceEvent *ev; - trace_event_iter_init_all(&iter); - while ((ev = trace_event_iter_next(&iter)) != NULL) { - if (trace_event_is_vcpu(ev) && - trace_event_get_state_static(ev) && - trace_event_get_state_dynamic(ev)) { - if (adding_first_cpu()) { - /* check preconditions */ - assert(*ev->dstate == 1); - /* disable early-init state ... */ - *ev->dstate = 0; - trace_events_enabled_count--; - /* ... and properly re-enable */ - trace_event_set_vcpu_state_dynamic(vcpu, ev, true); - } else { - trace_event_set_vcpu_state_dynamic(vcpu, ev, true); - } + *ev->dstate = 0; } } } diff --git a/trace/control.c b/trace/control.c index 5dfb609954..1a48a7e266 100644 --- a/trace/control.c +++ b/trace/control.c @@ -262,22 +262,6 @@ void trace_init_file(void) #endif } -void trace_fini_vcpu(CPUState *vcpu) -{ - TraceEventIter iter; - TraceEvent *ev; - - trace_event_iter_init_all(&iter); - while ((ev = trace_event_iter_next(&iter)) != NULL) { - if (trace_event_is_vcpu(ev) && - trace_event_get_state_static(ev) && - trace_event_get_vcpu_state_dynamic(vcpu, ev)) { - /* must disable to affect the global counter */ - trace_event_set_vcpu_state_dynamic(vcpu, ev, false); - } - } -} - bool trace_init_backends(void) { #ifdef CONFIG_TRACE_SIMPLE diff --git a/trace/qmp.c b/trace/qmp.c index 3b4f4702b4..aa760f1fc4 100644 --- a/trace/qmp.c +++ b/trace/qmp.c @@ -13,20 +13,7 @@ #include "control-vcpu.h" -static CPUState *get_cpu(bool has_vcpu, int vcpu, Error **errp) -{ - if (has_vcpu) { - CPUState *cpu = qemu_get_cpu(vcpu); - if (cpu == NULL) { - error_setg(errp, "invalid vCPU index %u", vcpu); - } - return cpu; - } else { - return NULL; - } -} - -static bool check_events(bool has_vcpu, bool ignore_unavailable, bool is_pattern, +static bool check_events(bool ignore_unavailable, bool is_pattern, const char *name, Error **errp) { if (!is_pattern) { @@ -38,12 +25,6 @@ static bool check_events(bool has_vcpu, bool ignore_unavailable, bool is_pattern return false; } - /* error for non-vcpu event */ - if (has_vcpu && !trace_event_is_vcpu(ev)) { - error_setg(errp, "event \"%s\" is not vCPU-specific", name); - return false; - } - /* error for unavailable event */ if (!ignore_unavailable && !trace_event_get_state_static(ev)) { error_setg(errp, "event \"%s\" is disabled", name); @@ -70,22 +51,13 @@ TraceEventInfoList *qmp_trace_event_get_state(const char *name, bool has_vcpu, int64_t vcpu, Error **errp) { - Error *err = NULL; TraceEventInfoList *events = NULL; TraceEventIter iter; TraceEvent *ev; bool is_pattern = trace_event_is_pattern(name); - CPUState *cpu; - - /* Check provided vcpu */ - cpu = get_cpu(has_vcpu, vcpu, &err); - if (err) { - error_propagate(errp, err); - return NULL; - } /* Check events */ - if (!check_events(has_vcpu, true, is_pattern, name, errp)) { + if (!check_events(true, is_pattern, name, errp)) { return NULL; } @@ -93,33 +65,17 @@ TraceEventInfoList *qmp_trace_event_get_state(const char *name, trace_event_iter_init_pattern(&iter, name); while ((ev = trace_event_iter_next(&iter)) != NULL) { TraceEventInfo *value; - bool is_vcpu = trace_event_is_vcpu(ev); - if (has_vcpu && !is_vcpu) { - continue; - } value = g_new(TraceEventInfo, 1); - value->vcpu = is_vcpu; value->name = g_strdup(trace_event_get_name(ev)); if (!trace_event_get_state_static(ev)) { value->state = TRACE_EVENT_STATE_UNAVAILABLE; } else { - if (has_vcpu) { - if (is_vcpu) { - if (trace_event_get_vcpu_state_dynamic(cpu, ev)) { - value->state = TRACE_EVENT_STATE_ENABLED; - } else { - value->state = TRACE_EVENT_STATE_DISABLED; - } - } - /* else: already skipped above */ + if (trace_event_get_state_dynamic(ev)) { + value->state = TRACE_EVENT_STATE_ENABLED; } else { - if (trace_event_get_state_dynamic(ev)) { - value->state = TRACE_EVENT_STATE_ENABLED; - } else { - value->state = TRACE_EVENT_STATE_DISABLED; - } + value->state = TRACE_EVENT_STATE_DISABLED; } } QAPI_LIST_PREPEND(events, value); @@ -133,21 +89,12 @@ void qmp_trace_event_set_state(const char *name, bool enable, bool has_vcpu, int64_t vcpu, Error **errp) { - Error *err = NULL; TraceEventIter iter; TraceEvent *ev; bool is_pattern = trace_event_is_pattern(name); - CPUState *cpu; - - /* Check provided vcpu */ - cpu = get_cpu(has_vcpu, vcpu, &err); - if (err) { - error_propagate(errp, err); - return; - } /* Check events */ - if (!check_events(has_vcpu, has_ignore_unavailable && ignore_unavailable, + if (!check_events(has_ignore_unavailable && ignore_unavailable, is_pattern, name, errp)) { return; } @@ -155,14 +102,9 @@ void qmp_trace_event_set_state(const char *name, bool enable, /* Apply changes (all errors checked above) */ trace_event_iter_init_pattern(&iter, name); while ((ev = trace_event_iter_next(&iter)) != NULL) { - if (!trace_event_get_state_static(ev) || - (has_vcpu && !trace_event_is_vcpu(ev))) { + if (!trace_event_get_state_static(ev)) { continue; } - if (has_vcpu) { - trace_event_set_vcpu_state_dynamic(cpu, ev, enable); - } else { - trace_event_set_state_dynamic(ev, enable); - } + trace_event_set_state_dynamic(ev, enable); } } diff --git a/trace/trace-hmp-cmds.c b/trace/trace-hmp-cmds.c index 792876c34a..86211fce27 100644 --- a/trace/trace-hmp-cmds.c +++ b/trace/trace-hmp-cmds.c @@ -37,16 +37,10 @@ void hmp_trace_event(Monitor *mon, const QDict *qdict) { const char *tp_name = qdict_get_str(qdict, "name"); bool new_state = qdict_get_bool(qdict, "option"); - bool has_vcpu = qdict_haskey(qdict, "vcpu"); - int vcpu = qdict_get_try_int(qdict, "vcpu", 0); Error *local_err = NULL; - if (vcpu < 0) { - monitor_printf(mon, "argument vcpu must be positive"); - return; - } - - qmp_trace_event_set_state(tp_name, new_state, true, true, has_vcpu, vcpu, &local_err); + qmp_trace_event_set_state(tp_name, new_state, + true, true, false, 0, &local_err); if (local_err) { error_report_err(local_err); } @@ -80,8 +74,6 @@ void hmp_trace_file(Monitor *mon, const QDict *qdict) void hmp_info_trace_events(Monitor *mon, const QDict *qdict) { const char *name = qdict_get_try_str(qdict, "name"); - bool has_vcpu = qdict_haskey(qdict, "vcpu"); - int vcpu = qdict_get_try_int(qdict, "vcpu", 0); TraceEventInfoList *events; TraceEventInfoList *elem; Error *local_err = NULL; @@ -89,12 +81,8 @@ void hmp_info_trace_events(Monitor *mon, const QDict *qdict) if (name == NULL) { name = "*"; } - if (vcpu < 0) { - monitor_printf(mon, "argument vcpu must be positive"); - return; - } - events = qmp_trace_event_get_state(name, has_vcpu, vcpu, &local_err); + events = qmp_trace_event_get_state(name, false, 0, &local_err); if (local_err) { error_report_err(local_err); return; -- 2.39.2

Now we no longer have dynamic state affecting things we can remove the additional fields in cpu.h and simplify the TB hash calculation. For the benchmark: hyperfine -w 2 -m 20 \ "./arm-softmmu/qemu-system-arm -cpu cortex-a15 \ -machine type=virt,highmem=off \ -display none -m 2048 \ -serial mon:stdio \ -netdev user,id=unet,hostfwd=tcp::2222-:22 \ -device virtio-net-pci,netdev=unet \ -device virtio-scsi-pci \ -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-armhf \ -device scsi-hd,drive=hd -smp 4 \ -kernel /home/alex/lsrc/linux.git/builds/arm/arch/arm/boot/zImage \ -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark.service' \ -snapshot" It has a marginal effect on runtime, before: Time (mean ± σ): 26.279 s ± 2.438 s [User: 41.113 s, System: 1.843 s] Range (min … max): 24.420 s … 32.565 s 20 runs after: Time (mean ± σ): 24.440 s ± 2.885 s [User: 34.474 s, System: 2.028 s] Range (min … max): 21.663 s … 29.937 s 20 runs Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1358 Reviewed-by: Stefan Hajnoczi <stefanha(a)redhat.com&gt; Reviewed-by: Richard Henderson <richard.henderson(a)linaro.org&gt; Reviewed-by: Philippe Mathieu-Daudé <philmd(a)linaro.org&gt; Signed-off-by: Alex Bennée <alex.bennee(a)linaro.org&gt; Message-Id: <20230524133952.3971948-9-alex.bennee(a)linaro.org&gt; --- accel/tcg/tb-hash.h | 6 +++--- include/exec/exec-all.h | 3 --- include/hw/core/cpu.h | 5 ----- accel/tcg/cpu-exec.c | 7 +------ accel/tcg/tb-maint.c | 5 ++--- accel/tcg/translate-all.c | 6 ------ 6 files changed, 6 insertions(+), 26 deletions(-) diff --git a/accel/tcg/tb-hash.h b/accel/tcg/tb-hash.h index 83dc610e4c..1d19c69caa 100644 --- a/accel/tcg/tb-hash.h +++ b/accel/tcg/tb-hash.h @@ -61,10 +61,10 @@ static inline unsigned int tb_jmp_cache_hash_func(target_ulong pc) #endif /* CONFIG_SOFTMMU */ static inline -uint32_t tb_hash_func(tb_page_addr_t phys_pc, target_ulong pc, uint32_t flags, - uint32_t cf_mask, uint32_t trace_vcpu_dstate) +uint32_t tb_hash_func(tb_page_addr_t phys_pc, target_ulong pc, + uint32_t flags, uint32_t cf_mask) { - return qemu_xxhash7(phys_pc, pc, flags, cf_mask, trace_vcpu_dstate); + return qemu_xxhash6(phys_pc, pc, flags, cf_mask); } #endif diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index 4d2b151986..3b1b57f6ad 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -545,9 +545,6 @@ struct TranslationBlock { #define CF_CLUSTER_MASK 0xff000000 /* Top 8 bits are cluster ID */ #define CF_CLUSTER_SHIFT 24 - /* Per-vCPU dynamic tracing state used to generate this TB */ - uint32_t trace_vcpu_dstate; - /* * Above fields used for comparing */ diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h index 39150cf8f8..383456d1b3 100644 --- a/include/hw/core/cpu.h +++ b/include/hw/core/cpu.h @@ -266,7 +266,6 @@ typedef void (*run_on_cpu_func)(CPUState *cpu, run_on_cpu_data data); struct qemu_work_item; #define CPU_UNSET_NUMA_NODE_ID -1 -#define CPU_TRACE_DSTATE_MAX_EVENTS 32 /** * CPUState: @@ -407,10 +406,6 @@ struct CPUState { /* Use by accel-block: CPU is executing an ioctl() */ QemuLockCnt in_ioctl_lock; - /* Used for events with 'vcpu' and *without* the 'disabled' properties */ - DECLARE_BITMAP(trace_dstate_delayed, CPU_TRACE_DSTATE_MAX_EVENTS); - DECLARE_BITMAP(trace_dstate, CPU_TRACE_DSTATE_MAX_EVENTS); - DECLARE_BITMAP(plugin_mask, QEMU_PLUGIN_EV_MAX); #ifdef CONFIG_PLUGIN diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index 0e741960da..4a1dce98ff 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -175,7 +175,6 @@ struct tb_desc { tb_page_addr_t page_addr0; uint32_t flags; uint32_t cflags; - uint32_t trace_vcpu_dstate; }; static bool tb_lookup_cmp(const void *p, const void *d) @@ -187,7 +186,6 @@ static bool tb_lookup_cmp(const void *p, const void *d) tb_page_addr0(tb) == desc->page_addr0 && tb->cs_base == desc->cs_base && tb->flags == desc->flags && - tb->trace_vcpu_dstate == desc->trace_vcpu_dstate && tb_cflags(tb) == desc->cflags) { /* check next page if needed */ tb_page_addr_t tb_phys_page1 = tb_page_addr1(tb); @@ -228,7 +226,6 @@ static TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, desc.cs_base = cs_base; desc.flags = flags; desc.cflags = cflags; - desc.trace_vcpu_dstate = *cpu->trace_dstate; desc.pc = pc; phys_pc = get_page_addr_code(desc.env, pc); if (phys_pc == -1) { @@ -236,7 +233,7 @@ static TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, } desc.page_addr0 = phys_pc; h = tb_hash_func(phys_pc, (cflags & CF_PCREL ? 0 : pc), - flags, cflags, *cpu->trace_dstate); + flags, cflags); return qht_lookup_custom(&tb_ctx.htable, &desc, h, tb_lookup_cmp); } @@ -263,7 +260,6 @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, jc->array[hash].pc == pc && tb->cs_base == cs_base && tb->flags == flags && - tb->trace_vcpu_dstate == *cpu->trace_dstate && tb_cflags(tb) == cflags)) { return tb; } @@ -282,7 +278,6 @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, tb->pc == pc && tb->cs_base == cs_base && tb->flags == flags && - tb->trace_vcpu_dstate == *cpu->trace_dstate && tb_cflags(tb) == cflags)) { return tb; } diff --git a/accel/tcg/tb-maint.c b/accel/tcg/tb-maint.c index 991746f80f..bc1961ea55 100644 --- a/accel/tcg/tb-maint.c +++ b/accel/tcg/tb-maint.c @@ -50,7 +50,6 @@ static bool tb_cmp(const void *ap, const void *bp) a->cs_base == b->cs_base && a->flags == b->flags && (tb_cflags(a) & ~CF_INVALID) == (tb_cflags(b) & ~CF_INVALID) && - a->trace_vcpu_dstate == b->trace_vcpu_dstate && tb_page_addr0(a) == tb_page_addr0(b) && tb_page_addr1(a) == tb_page_addr1(b)); } @@ -888,7 +887,7 @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) /* remove the TB from the hash list */ phys_pc = tb_page_addr0(tb); h = tb_hash_func(phys_pc, (orig_cflags & CF_PCREL ? 0 : tb->pc), - tb->flags, orig_cflags, tb->trace_vcpu_dstate); + tb->flags, orig_cflags); if (!qht_remove(&tb_ctx.htable, tb, h)) { return; } @@ -969,7 +968,7 @@ TranslationBlock *tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, /* add in the hash table */ h = tb_hash_func(phys_pc, (tb->cflags & CF_PCREL ? 0 : tb->pc), - tb->flags, tb->cflags, tb->trace_vcpu_dstate); + tb->flags, tb->cflags); qht_insert(&tb_ctx.htable, tb, h, &existing_tb); /* remove TB from the page(s) if we couldn't insert it */ diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index c87648b99e..bf814b9e81 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -65,11 +65,6 @@ #include "internal.h" #include "perf.h" -/* Make sure all possible CPU event bits fit in tb->trace_vcpu_dstate */ -QEMU_BUILD_BUG_ON(CPU_TRACE_DSTATE_MAX_EVENTS > - sizeof_field(TranslationBlock, trace_vcpu_dstate) - * BITS_PER_BYTE); - TBContext tb_ctx; /* @@ -352,7 +347,6 @@ TranslationBlock *tb_gen_code(CPUState *cpu, tb->cs_base = cs_base; tb->flags = flags; tb->cflags = cflags; - tb->trace_vcpu_dstate = *cpu->trace_dstate; tb_set_page_addr0(tb, phys_pc); tb_set_page_addr1(tb, -1); tcg_ctx->gen_tb = tb; -- 2.39.2

We weren't using cs_base in the hash calculations before. Since the arm front end moved a chunk of flags in a378206a20 (target/arm: Move mode specific TB flags to tb->cs_base) they comprise of an important part of the execution state. Widen the tb_hash_func to include cs_base and expand to qemu_xxhash8() to accommodate it. My initial benchmark shows very little difference in the runtime. Before: armhf ➜ hyperfine -w 2 -m 20 "./arm-softmmu/qemu-system-arm -cpu cortex-a15 -machine type=virt,highmem=off -display none -m 2048 -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-armhf -device scsi-hd,drive=hd -smp 4 -kernel /home/alex/lsrc/linux.git/builds/arm/arch/arm/boot/zImage -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark.service' -snapshot" Benchmark 1: ./arm-softmmu/qemu-system-arm -cpu cortex-a15 -machine type=virt,highmem=off -display none -m 2048 -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-armhf -device scsi-hd,drive=hd -smp 4 -kernel /home/alex/lsrc/linux.git/builds/arm/arch/arm/boot/zImage -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark.service' -snapshot Time (mean ± σ): 24.627 s ± 2.708 s [User: 34.309 s, System: 1.797 s] Range (min … max): 22.345 s … 29.864 s 20 runs arm64 ➜ hyperfine -w 2 -n 20 "./qemu-system-aarch64 -cpu max,pauth-impdef=on -machine type=virt,virtualization=on,gic-version=3 -display none -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22,hostfwd=tcp::1234-:1234 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-arm64 -device scsi-hd,drive=hd -smp 4 -kernel ~/lsrc/linux.git/builds/arm64/arch/arm64/boot/Image.gz -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark-pigz.service' -snapshot" Benchmark 1: 20 Time (mean ± σ): 62.559 s ± 2.917 s [User: 189.115 s, System: 4.089 s] Range (min … max): 59.997 s … 70.153 s 10 runs After: armhf Benchmark 1: ./arm-softmmu/qemu-system-arm -cpu cortex-a15 -machine type=virt,highmem=off -display none -m 2048 -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-armhf -device scsi-hd,drive=hd -smp 4 -kernel /home/alex/lsrc/linux.git/builds/arm/arch/arm/boot/zImage -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark.service' -snapshot Time (mean ± σ): 24.223 s ± 2.151 s [User: 34.284 s, System: 1.906 s] Range (min … max): 22.000 s … 28.476 s 20 runs arm64 hyperfine -w 2 -n 20 "./qemu-system-aarch64 -cpu max,pauth-impdef=on -machine type=virt,virtualization=on,gic-version=3 -display none -serial mon:stdio -netdev user,id=unet,hostfwd=tcp::2222-:22,hostfwd=tcp::1234-:1234 -device virtio-net-pci,netdev=unet -device virtio-scsi-pci -blockdev driver=raw,node-name=hd,discard=unmap,file.driver=host_device,file.filename=/dev/zen-disk/debian-bullseye-arm64 -device scsi-hd,drive=hd -smp 4 -kernel ~/lsrc/linux.git/builds/arm64/arch/arm64/boot/Image.gz -append 'console=ttyAMA0 root=/dev/sda2 systemd.unit=benchmark-pigz.service' -snapshot" Benchmark 1: 20 Time (mean ± σ): 62.769 s ± 1.978 s [User: 188.431 s, System: 5.269 s] Range (min … max): 60.285 s … 66.868 s 10 runs Signed-off-by: Alex Bennée <alex.bennee(a)linaro.org&gt; Reviewed-by: Richard Henderson <richard.henderson(a)linaro.org&gt; Message-Id: <20230524133952.3971948-11-alex.bennee(a)linaro.org&gt; --- accel/tcg/tb-hash.h | 4 ++-- include/qemu/xxhash.h | 23 +++++++++++++++++------ accel/tcg/cpu-exec.c | 2 +- accel/tcg/tb-maint.c | 4 ++-- util/qsp.c | 2 +- 5 files changed, 23 insertions(+), 12 deletions(-) diff --git a/accel/tcg/tb-hash.h b/accel/tcg/tb-hash.h index 1d19c69caa..2ba2193731 100644 --- a/accel/tcg/tb-hash.h +++ b/accel/tcg/tb-hash.h @@ -62,9 +62,9 @@ static inline unsigned int tb_jmp_cache_hash_func(target_ulong pc) static inline uint32_t tb_hash_func(tb_page_addr_t phys_pc, target_ulong pc, - uint32_t flags, uint32_t cf_mask) + uint32_t flags, uint64_t flags2, uint32_t cf_mask) { - return qemu_xxhash6(phys_pc, pc, flags, cf_mask); + return qemu_xxhash8(phys_pc, pc, flags2, flags, cf_mask); } #endif diff --git a/include/qemu/xxhash.h b/include/qemu/xxhash.h index c2dcccadbf..0259bbef18 100644 --- a/include/qemu/xxhash.h +++ b/include/qemu/xxhash.h @@ -48,8 +48,8 @@ * xxhash32, customized for input variables that are not guaranteed to be * contiguous in memory. */ -static inline uint32_t -qemu_xxhash7(uint64_t ab, uint64_t cd, uint32_t e, uint32_t f, uint32_t g) +static inline uint32_t qemu_xxhash8(uint64_t ab, uint64_t cd, uint64_t ef, + uint32_t g, uint32_t h) { uint32_t v1 = QEMU_XXHASH_SEED + PRIME32_1 + PRIME32_2; uint32_t v2 = QEMU_XXHASH_SEED + PRIME32_2; @@ -59,6 +59,8 @@ qemu_xxhash7(uint64_t ab, uint64_t cd, uint32_t e, uint32_t f, uint32_t g) uint32_t b = ab >> 32; uint32_t c = cd; uint32_t d = cd >> 32; + uint32_t e = ef; + uint32_t f = ef >> 32; uint32_t h32; v1 += a * PRIME32_2; @@ -89,6 +91,9 @@ qemu_xxhash7(uint64_t ab, uint64_t cd, uint32_t e, uint32_t f, uint32_t g) h32 += g * PRIME32_3; h32 = rol32(h32, 17) * PRIME32_4; + h32 += h * PRIME32_3; + h32 = rol32(h32, 17) * PRIME32_4; + h32 ^= h32 >> 15; h32 *= PRIME32_2; h32 ^= h32 >> 13; @@ -100,23 +105,29 @@ qemu_xxhash7(uint64_t ab, uint64_t cd, uint32_t e, uint32_t f, uint32_t g) static inline uint32_t qemu_xxhash2(uint64_t ab) { - return qemu_xxhash7(ab, 0, 0, 0, 0); + return qemu_xxhash8(ab, 0, 0, 0, 0); } static inline uint32_t qemu_xxhash4(uint64_t ab, uint64_t cd) { - return qemu_xxhash7(ab, cd, 0, 0, 0); + return qemu_xxhash8(ab, cd, 0, 0, 0); } static inline uint32_t qemu_xxhash5(uint64_t ab, uint64_t cd, uint32_t e) { - return qemu_xxhash7(ab, cd, e, 0, 0); + return qemu_xxhash8(ab, cd, 0, e, 0); } static inline uint32_t qemu_xxhash6(uint64_t ab, uint64_t cd, uint32_t e, uint32_t f) { - return qemu_xxhash7(ab, cd, e, f, 0); + return qemu_xxhash8(ab, cd, 0, e, f); +} + +static inline uint32_t qemu_xxhash7(uint64_t ab, uint64_t cd, uint64_t ef, + uint32_t g) +{ + return qemu_xxhash8(ab, cd, ef, g, 0); } /* diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index 4a1dce98ff..60ca9e229e 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -233,7 +233,7 @@ static TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, } desc.page_addr0 = phys_pc; h = tb_hash_func(phys_pc, (cflags & CF_PCREL ? 0 : pc), - flags, cflags); + flags, cs_base, cflags); return qht_lookup_custom(&tb_ctx.htable, &desc, h, tb_lookup_cmp); } diff --git a/accel/tcg/tb-maint.c b/accel/tcg/tb-maint.c index bc1961ea55..892eecda2d 100644 --- a/accel/tcg/tb-maint.c +++ b/accel/tcg/tb-maint.c @@ -887,7 +887,7 @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) /* remove the TB from the hash list */ phys_pc = tb_page_addr0(tb); h = tb_hash_func(phys_pc, (orig_cflags & CF_PCREL ? 0 : tb->pc), - tb->flags, orig_cflags); + tb->flags, tb->cs_base, orig_cflags); if (!qht_remove(&tb_ctx.htable, tb, h)) { return; } @@ -968,7 +968,7 @@ TranslationBlock *tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, /* add in the hash table */ h = tb_hash_func(phys_pc, (tb->cflags & CF_PCREL ? 0 : tb->pc), - tb->flags, tb->cflags); + tb->flags, tb->cs_base, tb->cflags); qht_insert(&tb_ctx.htable, tb, h, &existing_tb); /* remove TB from the page(s) if we couldn't insert it */ diff --git a/util/qsp.c b/util/qsp.c index 8562b14a87..2fe3764906 100644 --- a/util/qsp.c +++ b/util/qsp.c @@ -144,7 +144,7 @@ uint32_t do_qsp_callsite_hash(const QSPCallSite *callsite, uint64_t ab) uint32_t e = callsite->line; uint32_t f = callsite->type; - return qemu_xxhash6(ab, cd, e, f); + return qemu_xxhash8(ab, cd, 0, e, f); } static inline -- 2.39.2