From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753808AbeCYRJm (ORCPT ); Sun, 25 Mar 2018 13:09:42 -0400 Received: from orthanc.universe-factory.net ([104.238.176.138]:32772 "EHLO orthanc.universe-factory.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753758AbeCYRJl (ORCPT ); Sun, 25 Mar 2018 13:09:41 -0400 Subject: Re: [PATCH] ftrace: fix stddev calculation in function profiler (again) From: Matthias Schiffer To: rostedt@goodmis.org, mingo@redhat.com Cc: linux-kernel@vger.kernel.org References: Message-ID: <41987dff-d05f-168f-2662-ddd69eed11dc@universe-factory.net> Date: Sun, 25 Mar 2018 19:09:39 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.6.0 MIME-Version: 1.0 In-Reply-To: Content-Type: multipart/signed; micalg=pgp-sha512; protocol="application/pgp-signature"; boundary="q15pmqLmdL7Oc1MgbQD8WZkGld68p0wnD" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --q15pmqLmdL7Oc1MgbQD8WZkGld68p0wnD Content-Type: multipart/mixed; boundary="r3tWRYhNo8n8Qo5GrDBXfjn4eMIUUAoFw"; protected-headers="v1" From: Matthias Schiffer To: rostedt@goodmis.org, mingo@redhat.com Cc: linux-kernel@vger.kernel.org Message-ID: <41987dff-d05f-168f-2662-ddd69eed11dc@universe-factory.net> Subject: Re: [PATCH] ftrace: fix stddev calculation in function profiler (again) References: In-Reply-To: --r3tWRYhNo8n8Qo5GrDBXfjn4eMIUUAoFw Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: quoted-printable On 03/24/2018 11:03 PM, Matthias Schiffer wrote: > On 03/24/2018 05:26 PM, Matthias Schiffer wrote: >> It is well-known that it is not possible to accurately calculate varia= nces >> just by accumulating squared samples; in fact, such an approach can ev= en >> result in negative numbers. An earlier attempt to fix the calculation >> referred to Welford's method, but did not implement it correctly, lead= ing >> to meeaningless output like the following: >> >> nf_conntrack_proto_fini 50 373.523 us 7.470 us 32343159= 51 us >> >> Welford's method uses one do_div() in the tracing path; this cannot be= >> avoided. The average time is added to struct ftrace_profile, so only a= >> single division is required. I also considered the following alternati= ves: >> >> 1) Only keeping the avg field and removing the aggregated time would >> greatly lead to severe rounding errors in calculating the total time b= ased >> on counter and avg. >> >> 2) Calculating both the old and the new average in profile_graph_retur= n() >> instead of storing it in struct ftrace_profile would require a second >> division. >> >> 3) I managed to transform Welford's equations in a way that uses the t= otal >> time instead of the average and requires only a single division. >> Unfortunately, the divisor is counter^3 in this case, easily overflowi= ng >> even 64bit integers. >> >> Ruling out the above alternatives, I chose the present approach to fix= the >> issue. >> >> Fixes: e330b3bcd831 ("tracing: Show sample std dev in function profili= ng") >> Fixes: 52d85d763086 ("ftrace: Fix stddev calculation in function profi= ler") >> Signed-off-by: Matthias Schiffer >=20 >=20 > Hmm, in further testing, I'm seeing profile_graph_return() being called= > without rec->counter getting incremented a lot, completely throwing off= the > stddev calculation (and also slightly displacing the average). Is this > expected? >=20 > Should we possibly move the counter increment to profile_graph_return()= in > the CONFIG_FUNCTION_GRAPH_TRACER case, or introduce a second counter fo= r > returns? >=20 > Matthias profile_graph_return() being called without rec->counter getting incremented turned out to be another MIPS-specific bug, I'll send patch f= or that, too. I have two more questions/issues regarding the generic code: 1) There is a small window when enabling the profiler where rec->counter gets incremented, but profile_graph_return() does not record times yet. F= or functions that are called a lot, the effect is negligible, but for very infrequent calls, the shown average time and variance are way off. Moving= the rec->counter increment to profile_graph_return() when CONFIG_FUNCTION_GRAPH_TRACER is enabled would fix this. Does this sound reasonable? 2) I noticed that my patch does not handle rec->counter overflows correct= ly and will cause a zero division. I can think of two solutions for handling= overflows that keep the avg / stddev fields meaningful: - Reset time/avg/stddev to 0 on overflows - Stop updating the record at 0xffffffff, avoiding the overflow Matthias >=20 >=20 >=20 >> --- >> kernel/trace/ftrace.c | 33 +++++++++++++++++++-------------- >> 1 file changed, 19 insertions(+), 14 deletions(-) >> >> diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c >> index eac9ce2c57a2..16dce67b855a 100644 >> --- a/kernel/trace/ftrace.c >> +++ b/kernel/trace/ftrace.c >> @@ -460,7 +460,8 @@ struct ftrace_profile { >> unsigned long counter; >> #ifdef CONFIG_FUNCTION_GRAPH_TRACER >> unsigned long long time; >> - unsigned long long time_squared; >> + unsigned long long avg; >> + unsigned long long stddev; >> #endif >> }; >> =20 >> @@ -580,7 +581,6 @@ static int function_stat_show(struct seq_file *m, = void *v) >> int ret =3D 0; >> #ifdef CONFIG_FUNCTION_GRAPH_TRACER >> static struct trace_seq s; >> - unsigned long long avg; >> unsigned long long stddev; >> #endif >> mutex_lock(&ftrace_profile_lock); >> @@ -592,9 +592,7 @@ static int function_stat_show(struct seq_file *m, = void *v) >> } >> =20 >> #ifdef CONFIG_FUNCTION_GRAPH_TRACER >> - avg =3D rec->time; >> - do_div(avg, rec->counter); >> - if (tracing_thresh && (avg < tracing_thresh)) >> + if (tracing_thresh && (rec->avg < tracing_thresh)) >> goto out; >> #endif >> =20 >> @@ -608,24 +606,19 @@ static int function_stat_show(struct seq_file *m= , void *v) >> if (rec->counter <=3D 1) >> stddev =3D 0; >> else { >> - /* >> - * Apply Welford's method: >> - * s^2 =3D 1 / (n * (n-1)) * (n * \Sum (x_i)^2 - (\Sum x_i)^2) >> - */ >> - stddev =3D rec->counter * rec->time_squared - >> - rec->time * rec->time; >> + stddev =3D rec->stddev; >> =20 >> /* >> * Divide only 1000 for ns^2 -> us^2 conversion. >> * trace_print_graph_duration will divide 1000 again. >> */ >> - do_div(stddev, rec->counter * (rec->counter - 1) * 1000); >> + do_div(stddev, 1000 * (rec->counter - 1)); >> } >> =20 >> trace_seq_init(&s); >> trace_print_graph_duration(rec->time, &s); >> trace_seq_puts(&s, " "); >> - trace_print_graph_duration(avg, &s); >> + trace_print_graph_duration(rec->avg, &s); >> trace_seq_puts(&s, " "); >> trace_print_graph_duration(stddev, &s); >> trace_print_seq(m, &s); >> @@ -905,8 +898,20 @@ static void profile_graph_return(struct ftrace_gr= aph_ret *trace) >> =20 >> rec =3D ftrace_find_profiled_func(stat, trace->func); >> if (rec) { >> + unsigned long long avg, delta1, delta2; >> + >> rec->time +=3D calltime; >> - rec->time_squared +=3D calltime * calltime; >> + >> + /* Apply Welford's method */ >> + delta1 =3D calltime - rec->avg; >> + >> + avg =3D rec->time; >> + do_div(avg, rec->counter); >> + rec->avg =3D avg; >> + >> + delta2 =3D calltime - rec->avg; >> + >> + rec->stddev +=3D delta1 * delta2; >> } >> =20 >> out: >> >=20 >=20 --r3tWRYhNo8n8Qo5GrDBXfjn4eMIUUAoFw-- --q15pmqLmdL7Oc1MgbQD8WZkGld68p0wnD Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCgAdFiEEZmTnvaa2aYgexS51Fu8/ZMsgHZwFAlq319MACgkQFu8/ZMsg HZxTWhAAsLKg2Tjgxog8kka0I4aHy5fMN0rXOqWkL/fAQMnM0P5O9OgkOk0LyjX6 IFHyqFANUNv7J+bgpwW5awfI5DWFU0eR5Ijeu4hmZpppH/ASKFlyaYitid8QLOi6 Iq06DgbMQ43hvBdnFCNg9ZPDF45tKriiBIYiffSzPXNyzg3NhmdwZzEAZTmJCOq2 vM0TKIjWznfwer6k04UO7wWzDqA3D6IesirohYWQW9CwEFek22+X2rdCnjytyrYJ NKDYxFJC7Pu2vlCK0MrRkpl/n1Nynw0+j1OTKWuMSyLAdpGeZd/njW0zum4m63TN L7khkMi4CxCPQPXAg/5fxjxIsysgvzTjEKubnOigP4+pVDyD6I+ZRM5tht8Fk4JN 8TGt1+jfDa1b114KYZMeN7Hotf3achEuwg+BAJlELJxHjFU81jSx4dJhTvv9RdaO K5net6mmLjlarWRsQ3Hb2PLu4xKcYCvgo7G9qLn3sHVoPLtKYyK5gdn7qqM2QIPm PYp2S1Yu060z4jiyw6ZEdVKx8U4S911HPmc22GbjavPIAFPB6yY6dr09R1bHQCD8 LDq+7xSq98p/qm8OSTCtb7DTCgPqxtJgS5AMz/1dqgPQXAB4Ns/u83/Oc/HEWbFo Jz6AIuvTF8fialdjQuZ63BTbNkt4MdJ3ziR3aXI3mmoJNsbmXr0= =XC6Q -----END PGP SIGNATURE----- --q15pmqLmdL7Oc1MgbQD8WZkGld68p0wnD--