From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 752E8253B61; Tue, 11 Feb 2025 17:28:18 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1739294899; cv=none; b=A/PnUN+qY59AEdKkJUce5kX1YhpBMFdmvVudROo+9DBPtFPBbD9nj1B7s88p9L5fkVm6Va+iAArnaHPlf3Uk2ifvnm3YMP/cUZAHvP3a9MJMsMLbfNJGw/Lo/WCU4g/6UZzpw1x6uEmm6BV3DKmQhfD2l0tseQAbPfZaJMs3WDs= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1739294899; c=relaxed/simple; bh=LD/O1EBHcgMEITUbuApU66HPZd2hSaN6Qz+lpvhgYLk=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version: Content-Type:Content-Disposition:In-Reply-To; b=sHG6GxyPkEtghFaF2nhLWKG/EnLPYfrCnT1TMl28lrbIKlGL94GG1h2pGo1G8PyuD+cc8w69EE/HhLaeCcsa/EM/Y/539SaYWX4U5zeTNMGJzwmkS6BF5ethdl69QlfRpyCksH2r1torTcSZISz+qEwyFyQs0L39odRRXvMM5yk= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=meqzADNx; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="meqzADNx" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 08833C4CEDD; Tue, 11 Feb 2025 17:28:17 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1739294898; bh=LD/O1EBHcgMEITUbuApU66HPZd2hSaN6Qz+lpvhgYLk=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=meqzADNxPyy8lDrn5qHUVqcRkkz99t3Vk0zXCnSBCgpLq011eZ/z/SGkvFSSCu34C WbsSKgXWHsgj3xRbkQq3Q5TTPxoinT+5CkjT2v3gHoWCtWP9EoI7fOsYXMPELm7Qz3 nFlynvCmFqFHwzc2PKQQeoyb6J2BAByPGvl4TTbMUEa0nZyuavi6Zp8AzcV8p65CfW UdGiLhenehDi4+QkNAKjyBmXgUcsQ05mKGaCAPkyp+zy2aU3W0XPGWiQxhfsUImREB t5sRluJ0A5zrRkuEiiaSzwa6Jwaw+J8oejSbQTEEmzdF7TFTgDNN+AjsmF9QIllesY GNXnvTKyNStxg== Date: Tue, 11 Feb 2025 18:28:15 +0100 From: Arnaldo Carvalho de Melo To: Howard Chu Cc: namhyung@kernel.org, mark.rutland@arm.com, alexander.shishkin@linux.intel.com, jolsa@kernel.org, irogers@google.com, adrian.hunter@intel.com, kan.liang@linux.intel.com, linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, Ingo Molnar , James Clark , Peter Zijlstra , Arnaldo Carvalho de Melo Subject: Re: [PATCH v14 04/10] perf record --off-cpu: Dump off-cpu samples in BPF Message-ID: References: <20241215181220.754822-1-howardchu95@gmail.com> <20241215181220.754822-5-howardchu95@gmail.com> Precedence: bulk X-Mailing-List: linux-perf-users@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20241215181220.754822-5-howardchu95@gmail.com> On Sun, Dec 15, 2024 at 10:12:14AM -0800, Howard Chu wrote: > Collect tid, period, callchain, and cgroup id and dump them when off-cpu > time threshold is reached. > > We don't collect the off-cpu time twice (the delta), it's either in > direct samples, or accumulated samples that are dumped at the end of > perf.data. > > Suggested-by: Namhyung Kim > Signed-off-by: Howard Chu > Cc: Adrian Hunter > Cc: Alexander Shishkin > Cc: Ian Rogers > Cc: Ingo Molnar > Cc: James Clark > Cc: Jiri Olsa > Cc: Kan Liang > Cc: Mark Rutland > Cc: Peter Zijlstra > Link: https://lore.kernel.org/r/20241108204137.2444151-6-howardchu95@gmail.com > Signed-off-by: Arnaldo Carvalho de Melo > --- > tools/perf/util/bpf_skel/off_cpu.bpf.c | 88 ++++++++++++++++++++++++-- > 1 file changed, 83 insertions(+), 5 deletions(-) > > diff --git a/tools/perf/util/bpf_skel/off_cpu.bpf.c b/tools/perf/util/bpf_skel/off_cpu.bpf.c > index 1cdd4d63ea92..77fdc9e81db3 100644 > --- a/tools/perf/util/bpf_skel/off_cpu.bpf.c > +++ b/tools/perf/util/bpf_skel/off_cpu.bpf.c > @@ -19,11 +19,17 @@ > #define MAX_ENTRIES 102400 > > #define MAX_CPUS 4096 > +#define MAX_OFFCPU_LEN 37 > + > +struct stack { > + u64 array[MAX_STACKS]; > +}; So I needed to rename the above as it fails 'make -C tools/perf build-test', the direct make command line to hit that: ⬢ [acme@toolbox perf-tools-next]$ rm -rf /tmp/build/$(basename $PWD)/ ; mkdir -p /tmp/build/$(basename $PWD)/ ⬢ [acme@toolbox perf-tools-next]$ alias m='rm -rf ~/libexec/perf-core/ ; make -k GEN_VMLINUX_H=1 CORESIGHT=1 O=/tmp/build/$(basename $PWD)/ -C tools/perf install-bin && perf test python && cat /tmp/build/perf-tools-next/feature/test-all.output' ⬢ [acme@toolbox perf-tools-next]$ m GENSKEL /tmp/build/perf-tools-next/util/bpf_skel/sample_filter.skel.h util/bpf_skel/off_cpu.bpf.c:24:8: error: redefinition of 'stack' 24 | struct stack { | ^ /tmp/build/perf-tools-next/util/bpf_skel/.tmp/../vmlinux.h:128096:8: note: previous definition is here 128096 | struct stack { | ^ util/bpf_skel/off_cpu.bpf.c:218:42: error: no member named 'array' in 'struct stack' 218 | for (int i = 0; i < MAX_STACKS && from->array[i]; ++i, ++len) | ~~~~ ^ util/bpf_skel/off_cpu.bpf.c:219:32: error: no member named 'array' in 'struct stack' 219 | to->array[n + 2 + i] = from->array[i]; | ~~~~ ^ 3 errors generated. make[2]: *** [Makefile.perf:1276: /tmp/build/perf-tools-next/util/bpf_skel/.tmp/off_cpu.bpf.o] Error 1 So for now I'm adding the patch below to get it going. IIRC there was some discussion about ditching GEN_VMLINUX_H=1, we can remove this hack later if/when we progress on that. - Arnaldo diff --git a/tools/perf/util/bpf_skel/off_cpu.bpf.c b/tools/perf/util/bpf_skel/off_cpu.bpf.c index 77fdc9e81db395d1..848a123e5610f17b 100644 --- a/tools/perf/util/bpf_skel/off_cpu.bpf.c +++ b/tools/perf/util/bpf_skel/off_cpu.bpf.c @@ -21,7 +21,8 @@ #define MAX_CPUS 4096 #define MAX_OFFCPU_LEN 37 -struct stack { +// We have a 'struct stack' in vmlinux.h when building with GEN_VMLINUX_H=1 +struct __stack { u64 array[MAX_STACKS]; }; @@ -29,7 +30,7 @@ struct tstamp_data { __u32 stack_id; __u32 state; __u64 timestamp; - struct stack stack; + struct __stack stack; }; struct offcpu_key { @@ -211,7 +212,7 @@ static inline int can_record(struct task_struct *t, int state) return 1; } -static inline int copy_stack(struct stack *from, struct offcpu_data *to, int n) +static inline int copy_stack(struct __stack *from, struct offcpu_data *to, int n) { int len = 0; @@ -231,7 +232,7 @@ static inline int copy_stack(struct stack *from, struct offcpu_data *to, int n) * information of the task, and dump it as a raw sample to perf ring buffer */ static int off_cpu_dump(void *ctx, struct offcpu_data *data, struct offcpu_key *key, - struct stack *stack, __u64 delta) + struct __stack *stack, __u64 delta) { int n = 0, len = 0; > struct tstamp_data { > __u32 stack_id; > __u32 state; > __u64 timestamp; > + struct stack stack; > }; > > struct offcpu_key { > @@ -41,6 +47,10 @@ struct { > __uint(max_entries, MAX_ENTRIES); > } stacks SEC(".maps"); > > +struct offcpu_data { > + u64 array[MAX_OFFCPU_LEN]; > +}; > + > struct { > __uint(type, BPF_MAP_TYPE_PERF_EVENT_ARRAY); > __uint(key_size, sizeof(__u32)); > @@ -48,6 +58,13 @@ struct { > __uint(max_entries, MAX_CPUS); > } offcpu_output SEC(".maps"); > > +struct { > + __uint(type, BPF_MAP_TYPE_PERCPU_ARRAY); > + __uint(key_size, sizeof(__u32)); > + __uint(value_size, sizeof(struct offcpu_data)); > + __uint(max_entries, 1); > +} offcpu_payload SEC(".maps"); > + > struct { > __uint(type, BPF_MAP_TYPE_TASK_STORAGE); > __uint(map_flags, BPF_F_NO_PREALLOC); > @@ -106,6 +123,8 @@ const volatile bool uses_cgroup_v1 = false; > > int perf_subsys_id = -1; > > +__u64 offcpu_thresh_ns = 500000000ull; > + > /* > * Old kernel used to call it task_struct->state and now it's '__state'. > * Use BPF CO-RE "ignored suffix rule" to deal with it like below: > @@ -192,6 +211,47 @@ static inline int can_record(struct task_struct *t, int state) > return 1; > } > > +static inline int copy_stack(struct stack *from, struct offcpu_data *to, int n) > +{ > + int len = 0; > + > + for (int i = 0; i < MAX_STACKS && from->array[i]; ++i, ++len) > + to->array[n + 2 + i] = from->array[i]; > + > + return len; > +} > + > +/** > + * off_cpu_dump - dump off-cpu samples to ring buffer > + * @data: payload for dumping off-cpu samples > + * @key: off-cpu data > + * @stack: stack trace of the task before being scheduled out > + * > + * If the threshold of off-cpu time is reached, acquire tid, period, callchain, and cgroup id > + * information of the task, and dump it as a raw sample to perf ring buffer > + */ > +static int off_cpu_dump(void *ctx, struct offcpu_data *data, struct offcpu_key *key, > + struct stack *stack, __u64 delta) > +{ > + int n = 0, len = 0; > + > + data->array[n++] = (u64)key->tgid << 32 | key->pid; > + data->array[n++] = delta; > + > + /* data->array[n] is callchain->nr (updated later) */ > + data->array[n + 1] = PERF_CONTEXT_USER; > + data->array[n + 2] = 0; > + len = copy_stack(stack, data, n); > + > + /* update length of callchain */ > + data->array[n] = len + 1; > + n += len + 2; > + > + data->array[n++] = key->cgroup_id; > + > + return bpf_perf_event_output(ctx, &offcpu_output, BPF_F_CURRENT_CPU, data, n * sizeof(u64)); > +} > + > static int off_cpu_stat(u64 *ctx, struct task_struct *prev, > struct task_struct *next, int state) > { > @@ -216,6 +276,16 @@ static int off_cpu_stat(u64 *ctx, struct task_struct *prev, > pelem->state = state; > pelem->stack_id = stack_id; > > + /* > + * If stacks are successfully collected by bpf_get_stackid(), collect them once more > + * in task_storage for direct off-cpu sample dumping > + */ > + if (stack_id > 0 && bpf_get_stack(ctx, &pelem->stack, MAX_STACKS * sizeof(u64), BPF_F_USER_STACK)) { > + /* > + * This empty if block is used to avoid 'result unused warning' from bpf_get_stack(). > + * If the collection fails, continue with the logic for the next task. > + */ > + } > next: > pelem = bpf_task_storage_get(&tstamp, next, NULL, 0); > > @@ -230,11 +300,19 @@ static int off_cpu_stat(u64 *ctx, struct task_struct *prev, > __u64 delta = ts - pelem->timestamp; > __u64 *total; > > - total = bpf_map_lookup_elem(&off_cpu, &key); > - if (total) > - *total += delta; > - else > - bpf_map_update_elem(&off_cpu, &key, &delta, BPF_ANY); > + if (delta >= offcpu_thresh_ns) { > + int zero = 0; > + struct offcpu_data *data = bpf_map_lookup_elem(&offcpu_payload, &zero); > + > + if (data) > + off_cpu_dump(ctx, data, &key, &pelem->stack, delta); > + } else { > + total = bpf_map_lookup_elem(&off_cpu, &key); > + if (total) > + *total += delta; > + else > + bpf_map_update_elem(&off_cpu, &key, &delta, BPF_ANY); > + } > > /* prevent to reuse the timestamp later */ > pelem->timestamp = 0; > -- > 2.43.0