linux-kernel - Re: [RFC v2] perf: Rewrite core context handling

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Fri, 17 Jun 2022 15:36:51 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Ravi Bangoria <ravi.bangoria@....com>
Cc:     acme@...nel.org, alexander.shishkin@...ux.intel.com,
        jolsa@...hat.com, namhyung@...nel.org, songliubraving@...com,
        eranian@...gle.com, alexey.budankov@...ux.intel.com,
        ak@...ux.intel.com, mark.rutland@....com, megha.dey@...el.com,
        frederic@...nel.org, maddy@...ux.ibm.com, irogers@...gle.com,
        kim.phillips@....com, linux-kernel@...r.kernel.org,
        santosh.shukla@....com
Subject: Re: [RFC v2] perf: Rewrite core context handling

On Mon, Jun 13, 2022 at 04:35:11PM +0200, Peter Zijlstra wrote:
> +/* XXX: No need of list now. Convert it to per-cpu variable */
>  static DEFINE_PER_CPU(struct list_head, cgrp_cpuctx_list);

Something like so I suppose...

---
 include/linux/perf_event.h |    1 
 kernel/events/core.c       |   70 ++++++++++++++-------------------------------
 2 files changed, 22 insertions(+), 49 deletions(-)

--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -936,7 +936,6 @@ struct perf_cpu_context {
 
 #ifdef CONFIG_CGROUP_PERF
 	struct perf_cgroup		*cgrp;
-	struct list_head		cgrp_cpuctx_entry;
 #endif
 
 	/*
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -829,55 +829,41 @@ perf_cgroup_set_timestamp(struct perf_cp
 	}
 }
 
-/* XXX: No need of list now. Convert it to per-cpu variable */
-static DEFINE_PER_CPU(struct list_head, cgrp_cpuctx_list);
-
 /*
  * reschedule events based on the cgroup constraint of task.
  */
 static void perf_cgroup_switch(struct task_struct *task)
 {
+	struct perf_cpu_context *cpuctx = this_cpu_ptr(&cpu_context);
 	struct perf_cgroup *cgrp;
-	struct perf_cpu_context *cpuctx, *tmp;
 	struct list_head *list;
 	unsigned long flags;
 
-	/*
-	 * Disable interrupts and preemption to avoid this CPU's
-	 * cgrp_cpuctx_entry to change under us.
-	 */
-	local_irq_save(flags);
-
 	cgrp = perf_cgroup_from_task(task, NULL);
 
-	list = this_cpu_ptr(&cgrp_cpuctx_list);
-	list_for_each_entry_safe(cpuctx, tmp, list, cgrp_cpuctx_entry) {
-		WARN_ON_ONCE(cpuctx->ctx.nr_cgroups == 0);
-		if (READ_ONCE(cpuctx->cgrp) == cgrp)
-			continue;
-
-		perf_ctx_lock(cpuctx, cpuctx->task_ctx);
-		perf_ctx_disable(&cpuctx->ctx);
+	WARN_ON_ONCE(cpuctx->ctx.nr_cgroups == 0);
+	if (READ_ONCE(cpuctx->cgrp) == cgrp)
+		continue;
 
-		ctx_sched_out(&cpuctx->ctx, EVENT_ALL);
-		/*
-		 * must not be done before ctxswout due
-		 * to update_cgrp_time_from_cpuctx() in
-		 * ctx_sched_out()
-		 */
-		cpuctx->cgrp = cgrp;
-		/*
-		 * set cgrp before ctxsw in to allow
-		 * perf_cgroup_set_timestamp() in ctx_sched_in()
-		 * to not have to pass task around
-		 */
-		ctx_sched_in(&cpuctx->ctx, EVENT_ALL);
+	perf_ctx_lock(cpuctx, cpuctx->task_ctx);
+	perf_ctx_disable(&cpuctx->ctx);
 
-		perf_ctx_enable(&cpuctx->ctx);
-		perf_ctx_unlock(cpuctx, cpuctx->task_ctx);
-	}
+	ctx_sched_out(&cpuctx->ctx, EVENT_ALL);
+	/*
+	 * must not be done before ctxswout due
+	 * to update_cgrp_time_from_cpuctx() in
+	 * ctx_sched_out()
+	 */
+	cpuctx->cgrp = cgrp;
+	/*
+	 * set cgrp before ctxsw in to allow
+	 * perf_cgroup_set_timestamp() in ctx_sched_in()
+	 * to not have to pass task around
+	 */
+	ctx_sched_in(&cpuctx->ctx, EVENT_ALL);
 
-	local_irq_restore(flags);
+	perf_ctx_enable(&cpuctx->ctx);
+	perf_ctx_unlock(cpuctx, cpuctx->task_ctx);
 }
 
 static int perf_cgroup_ensure_storage(struct perf_event *event,
@@ -979,8 +965,6 @@ perf_cgroup_event_enable(struct perf_eve
 		return;
 
 	cpuctx->cgrp = perf_cgroup_from_task(current, ctx);
-	list_add(&cpuctx->cgrp_cpuctx_entry,
-			per_cpu_ptr(&cgrp_cpuctx_list, event->cpu));
 }
 
 static inline void
@@ -1001,7 +985,6 @@ perf_cgroup_event_disable(struct perf_ev
 		return;
 
 	cpuctx->cgrp = NULL;
-	list_del(&cpuctx->cgrp_cpuctx_entry);
 }
 
 #else /* !CONFIG_CGROUP_PERF */
@@ -2372,11 +2355,7 @@ static void perf_remove_from_context(str
 	 * event_function_call() user.
 	 */
 	raw_spin_lock_irq(&ctx->lock);
-	/*
-	 * Cgroup events are per-cpu events, and must IPI because of
-	 * cgrp_cpuctx_list.
-	 */
-	if (!ctx->is_active && !is_cgroup_event(event)) {
+	if (!ctx->is_active) {
 		__perf_remove_from_context(event, this_cpu_ptr(&cpu_context),
 					   ctx, (void *)flags);
 		raw_spin_unlock_irq(&ctx->lock);
@@ -2807,8 +2786,6 @@ perf_install_in_context(struct perf_even
 	 * perf_event_attr::disabled events will not run and can be initialized
 	 * without IPI. Except when this is the first event for the context, in
 	 * that case we need the magic of the IPI to set ctx->is_active.
-	 * Similarly, cgroup events for the context also needs the IPI to
-	 * manipulate the cgrp_cpuctx_list.
 	 *
 	 * The IOC_ENABLE that is sure to follow the creation of a disabled
 	 * event will issue the IPI and reprogram the hardware.
@@ -13301,9 +13278,6 @@ static void __init perf_event_init_all_c
 		INIT_LIST_HEAD(&per_cpu(pmu_sb_events.list, cpu));
 		raw_spin_lock_init(&per_cpu(pmu_sb_events.lock, cpu));
 
-#ifdef CONFIG_CGROUP_PERF
-		INIT_LIST_HEAD(&per_cpu(cgrp_cpuctx_list, cpu));
-#endif
 		INIT_LIST_HEAD(&per_cpu(sched_cb_list, cpu));
 
 		cpuctx = per_cpu_ptr(&cpu_context, cpu);