From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from simark.ca by simark.ca with LMTP id rEwiM1Bh3Wf3dRYAWB0awg (envelope-from ) for ; Fri, 21 Mar 2025 08:53:36 -0400 Authentication-Results: simark.ca; dkim=pass (2048-bit key; unprotected) header.d=lists.lttng.org header.i=@lists.lttng.org header.a=rsa-sha256 header.s=default header.b=sDwz4Mfd; dkim-atps=neutral Received: by simark.ca (Postfix, from userid 112) id BFC4A1E100; Fri, 21 Mar 2025 08:53:36 -0400 (EDT) X-Spam-Checker-Version: SpamAssassin 4.0.1 (2024-03-25) on simark.ca X-Spam-Level: X-Spam-Status: No, score=-2.8 required=5.0 tests=ARC_SIGNED,ARC_VALID,BAYES_00, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_REPLYTO_END_DIGIT, HTML_MESSAGE,MAILING_LIST_MULTI,RCVD_IN_DNSWL_NONE autolearn=ham autolearn_force=no version=4.0.1 Received: from lists.lttng.org (lists.lttng.org [158.69.130.29]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1) server-digest SHA256) (No client certificate requested) by simark.ca (Postfix) with ESMTPS id B2CA61E0C0 for ; Fri, 21 Mar 2025 08:53:33 -0400 (EDT) ARC-Filter: OpenARC Filter v1.2.1 lists.lttng.org 4ZK2SF2q5bz13x2 ARC-Seal: i=1; d=lists.lttng.org; s=arc1; a=rsa-sha256; cv=none; t=1742561613; b=jwIHuKD8Tc/U/MMEN60ti/5sGPGFNdISOUHD+poW5QKiBCCErMIAk2+HxFVv8a2/ETX0 kOVfKPwLv63o2wEohbPzgjQ88hLDRUlmc4qRKzFDpOHkpI8h0UPCJ7C3CtEH/QfkpHNEv OwzhniLq6UGt1MbQNKxv2cxoFMWwCumjjyhnDde8d0vIYG42d1GFNrIYGKboapttGqSon Sv6hj0g6tGTEPhPjNuehKGayuOqjGe4aIIdDIsEoKPtEbvGZq51c9d1Hs03iF/eV8FHw9 DYXGtnWhcfODX7NAOZ2SQlMRPi+NBTr1QQZ2vd16qR8MIMT9KYMJbqDQqWZuyi9gU9w== ARC-Message-Signature: i=1; d=lists.lttng.org; s=arc1; a=rsa-sha256; c=relaxed/simple; t=1742561613; h=DKIM-Signature:MIME-Version:Date:Message-ID:Subject:To:From:From; bh=9fAb0fPshTzmrf89EH6BBmflcjHQ6draDnMMYEfiwRQ=; b=RwtIsqlKFp0ewNDyxhj0Wt+eME/m+wyod1t7VL63q6jULqpSjaJD761gFsP428gPjGkK Pf/8jI/kqvyE/p1ZtECbKqgRjFyr+cBD8nhSHgh6Tr5j1p15HupRx056W7rqcW0KfL+x7 cPp8lDS/IqYE9K3oHYK5GdB2ftXug0R7k6+f7JpSnfWDYo+u/Nyz/Iw9BX9j6HRoQ3PF2 H8LfayqOHBj/kfwfRK2i4a6P+vkzIc+SDsOp17Gy6YF+xX4N6OG4MZxzxCx0l416I6Z6t h4m9mduxsLna/NSz/S1U/IRv9AkfYIugeRCsPEFUOkYke/I5y89UfmpUMfUbjuRvY0Q== ARC-Authentication-Results: i=1; lists.lttng.org; arc=none smtp.remote-ip="::1" DKIM-Filter: OpenDKIM Filter v2.11.0 lists.lttng.org 4ZK2SF2q5bz13x2 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=lists.lttng.org; s=default; t=1742561613; bh=9fAb0fPshTzmrf89EH6BBmflcjHQ6draDnMMYEfiwRQ=; h=References:In-Reply-To:Date:Subject:To:List-Id:List-Unsubscribe: List-Archive:List-Post:List-Help:List-Subscribe:From:Reply-To: From; b=sDwz4MfdfFXLzj9S8Kj8+GtbGM/Qhg0a6VJEKXf/wtTLSxxYgEIH8kbHAuTMuTEo1 0Tpf+n6XB9dzenZykz8sYMFRgRIi9+TLCWWA+Pq+2A+Rs0Ehi6/TBYc4j56tk45sxu 1hbEGHvFkKxv27+wlvY1ra6pbe2dn5L11ZDkB97pWO0uXrAd1txxrTzg8JVIaW/xrv p9dEqBMUzJ2OJcaJEvceVjchrXZ/CxV0E9LGGr/Qd1GhbqGGVT2sgMqT936i2kwTxZ 2OwRtr+JgnG6bW2oDb1OBYdNVRwUpACZPosyD7cnetI/K5+7hub/rNolP8nWXxLUNi TVeDJaBT+SxVg== Received: from lists-lttng01.efficios.com (localhost [IPv6:::1]) by lists.lttng.org (Postfix) with ESMTP id 4ZK2SF2q5bz13x2; Fri, 21 Mar 2025 08:53:33 -0400 (EDT) ARC-Filter: OpenARC Filter v1.2.1 lists.lttng.org 4ZK2SD1JdHz144s DKIM-Filter: OpenDKIM Filter v2.11.0 lists.lttng.org 4ZK2SD1JdHz144s Received: from mail-pj1-x1031.google.com (mail-pj1-x1031.google.com [IPv6:2607:f8b0:4864:20::1031]) by lists.lttng.org (Postfix) with ESMTPS id 4ZK2SD1JdHz144s for ; Fri, 21 Mar 2025 08:53:31 -0400 (EDT) Received: by mail-pj1-x1031.google.com with SMTP id 98e67ed59e1d1-301493f45aeso3191783a91.1 for ; Fri, 21 Mar 2025 05:53:31 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1742561611; x=1743166411; h=to:subject:message-id:date:from:in-reply-to:references:mime-version :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=qykdTWrI8WGBoA/CZClW4Bu44hoHYAQiVpP9gujB6+w=; b=B69c0u+oHut/2FjjyqfsC68SUahOfSKV+j0FJdo3VkQLqLfJVIXKH1hnrBL5wS34kX osdSrEnME5jOZliTTOw9WgotXJp+aeH5X8CrTaKgrNbmX+T8WRWxLUXG9en9Bot5F+aa dXul/pA/zvIsE+sxNDOrKOg0HtsyExsv+J3430J8F8zZHa3rYfOIL6pJSLAC4O2jUAUd HOMEzXT/4ZIzHP5dB7JXM5JWyRVx89On6atwHwRz/T9gvcQReD2Y4RTFneb3F3lAL5GF 7YJIwUB5K7R5a5fBMJDgb2DvkowBpBj/2yRibH0n+jqfnCiwVPEK6K/sOd31chlD6z0p nxpw== X-Forwarded-Encrypted: i=1; AJvYcCWaHNwNcFOGsficdIbIKJ3AMRBMcNTMtSVDXiVJzajqqgv/Z4o7Iedu/P8VwxKE3eWD82ZuVHFfl7Q=@lists.lttng.org X-Gm-Message-State: AOJu0YzVyHU1o9KZsX4Y0AUfvh6Hmz1SH25Uc8wk0YF0+PaVx0nLqXAq C5V+XGqsJ2WX6zx6zvHnSX9mAOpw7T5K8QvujB1En77kNdoFvd+lG8VzxYfDisn0i/AV/JSSbEp 6CjuYmzf2ntjvcRS7amT+dnpuWqJ8XfD6xMA= X-Gm-Gg: ASbGnctLwcnbIuHZCP95Hua8QAKX/bWqQLfHypudM1JStEkNUpsdpSTsWrsS4Zgmaqx ELMn5n+I1GrIbxD6ojb4l8B+9fuEc6INWJkOiValqaGr2DxqXJUVqA4YRLzX6ZpAelvwzg+pVsw RrZp81zjKQUgC/Er3rqzmKjqH9tcnt5V8E0Q7EouV09CBfJ1baIFi2A0SKq2c= X-Google-Smtp-Source: AGHT+IGaNDKG8NxqH5pVzlIw6kKNmClSde/BDw7IOHsXuJp8+vASYfkzhwSnpoaUU6WL4uZJJsbHEVdWIEmDXcwEaEY= X-Received: by 2002:a17:90b:2647:b0:2f8:49ad:4079 with SMTP id 98e67ed59e1d1-3030fe6e164mr4661319a91.6.1742561609979; Fri, 21 Mar 2025 05:53:29 -0700 (PDT) MIME-Version: 1.0 References: <605bbc18-bb08-4a72-b470-fc0c34e67acd@efficios.com> In-Reply-To: <605bbc18-bb08-4a72-b470-fc0c34e67acd@efficios.com> Date: Fri, 21 Mar 2025 18:22:52 +0530 X-Gm-Features: AQ5f1JrHA6f4d7qEUj0jHYoZ-HPJio_E4fvHfEHr_DhpXjx37q2J5I3__wImc_k Message-ID: Subject: Re: lttng-dev Digest, Vol 203, Issue 7 To: Kienan Stewart , lttng-dev@lists.lttng.org Content-Type: multipart/alternative; boundary="000000000000c31f030630d9bd43" X-BeenThere: lttng-dev@lists.lttng.org X-Mailman-Version: 2.1.39 Precedence: list List-Id: LTTng development list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , From: Gour DEV via lttng-dev Reply-To: Gour DEV Errors-To: lttng-dev-bounces@lists.lttng.org Sender: "lttng-dev" --000000000000c31f030630d9bd43 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Thanks Kienan for digging this and explaining the behaviour, Looks like this option is not present in the version I am using v2.13.5 and the MAP_POPULATE is enabled by default there /* memory_map: mmap */ memory_map =3D mmap(NULL, memory_map_size, PROT_READ | PROT_WRITE, MAP_SHARED | LTTNG_MAP_POPULATE, shmfd, 0); I have updated to the master branch and have verified by setting the LTTNG_UST_MAP_POPULATE_POLICY=3Dnone, and saw that now the RSS consumption = is not very high now. I think using LTTNG_UST_MAP_POPULATE_POLICY=3Dcpu_possible is better becaus= e here I will definitely know how much amount of memory lttng will use at start only, and allocate memory accordingly to my servers. Also, I thought using --buffers-global would be better for me as I could know how much memory will be required but looks like that option is only available for kernel tracking, it would be very good if this option could be available for user buffering also, but that could come later. But again, thanks for the help Kienan. Regards Lakshya. On Fri, Mar 21, 2025 at 2:21=E2=80=AFAM Kienan Stewart wrote: > Hi Lakshya, > > I did some digging around. What you are seeing is the result of the > switching to MAP_POPULATE by default in LTTng-UST 2.12[1] in commit > 4d4838b ("Use MAP_POPULATE to reduce pagefault when available"). > > The purpose of this change is to avoid taking page faults which tracing, > reducing first-event in a page latency. > > In the master branch, this feature has been made configurable for users > who don't want to pre-populate the pages and would rather take page > faults while tracing[2]. > > Here is an example from LTTng master with map populate per possible CPU: > > ``` > export LTTNG_UST_MAP_POPULATE_POLICY=3Dcpu_possible > > # Create session, channels, start tracing, and run test app > # top -n1 -b | grep -E '(MiB|COMMAND|lttng)' > MiB Mem : 32768.0 total, 21883.7 free, 1456.0 used, 9428.3 > buff/cache > > MiB Swap: 0.0 total, 0.0 free, 0.0 used. 31312.0 avail > Mem > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ > COMMAND > > 301117 debian 20 0 880176 2760 2760 S 0.0 0.0 0:00.04 > lttng-sessiond > > 301118 debian 20 0 43856 1376 1376 S 0.0 0.0 0:00.01 > lttng-runas > > 301133 debian 20 0 718616 263456 263456 S 0.0 0.8 0:00.17 > lttng-consumerd > > 301135 debian 20 0 6996 0 0 S 0.0 0.0 0:00.05 > lttng-runas > > # cat /proc/$(pgrep lttng-sessiond)/statm > lttng-sessiond: 220044 690 690 345 0 29900 0 > > > > # pmap $(pgrep lttng-sessiond) | grep total > > total 880176K > > # smem -P lttng-sessiond > > > PID User Command Swap USS PSS > RSS > > 301118 debian lttng-sessiond --daemonize 0 344 881 > 2236 > > 301117 debian lttng-sessiond --daemonize 0 5676 6683 > 9276 > > 301201 debian /usr/bin/python3 /usr/bin/s 0 8636 10086 > 12936 > > > # /proc/PID/statm for lttng-consumerd > > > > lttng-consumerd: 1749 0 0 129 0 130 0 > > # pmap lttng-consumerd-pid | grep total > total kB 6996 1700 472 > > > > # smem -P lttng-consumerd > PID User Command Swap USS PSS > RSS > > 301135 debian lttng-consumerd -u --consu 0 280 563 > 1700 > > 301211 debian /usr/bin/python3 /usr/bin/s 0 10048 11501 > 14404 > > 301133 debian lttng-consumerd -u --consu 0 262376 263177 > 265480 > > # smem -m | grep -i ust > > /dev/shm/lttng-ust-wait-8-1000 1 4 4 > /dev/shm/shm-ust-consumer-301133 1 260756 260756 > ``` > > When using the none policy: > > ``` > # export LTTNG_UST_MAP_POPULATE_POLICY=3Dnone > # as above > > Running test app UID 0 > > > procs -----------memory---------- ---swap-- -----io---- -system-- > ------cpu----- > > r b swpd free buff cache si so bi bo in cs us sy > id wa st > > 1 0 0 21875 0 9434 0 0 39 636 1105 2496 0 1 > 99 0 0 > > MiB Mem : 32768.0 total, 21875.0 free, 1458.2 used, 9434.7 > buff/cache > > MiB Swap: 0.0 total, 0.0 free, 0.0 used. 31309.8 avail > Mem > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ > COMMAND > > 301616 debian 20 0 880176 2756 2756 S 0.0 0.0 0:00.04 > lttng-sessiond > > 301617 debian 20 0 43856 1392 1392 S 0.0 0.0 0:00.01 > lttng-runas > > 301632 debian 20 0 718612 5416 5416 S 0.0 0.0 0:00.17 > lttng-consumerd > > 301634 debian 20 0 6992 0 0 S 0.0 0.0 0:00.05 > lttng-runas > > lttng-sessiond: 220044 689 689 345 0 29900 0 > > > total 880176K > > > PID User Command Swap USS PSS > RSS > 301617 debian lttng-sessiond --daemonize 0 344 862 > 2188 > 301616 debian lttng-sessiond --daemonize 0 5784 6759 > 9328 > 301700 debian /usr/bin/python3 /usr/bin/s 0 8632 10079 > 12928 > > lttng-consumerd: 1748 0 0 129 0 129 0 > total kB 6992 1580 468 > PID User Command Swap USS PSS > RSS > 301634 debian lttng-consumerd -u --consu 0 276 536 > 1580 > 301632 debian lttng-consumerd -u --consu 0 5672 6433 > 8652 > 301710 debian /usr/bin/python3 /usr/bin/s 0 9996 11449 > 14328 > > /dev/shm/lttng-ust-wait-8-1000 1 4 4 > /dev/shm/shm-ust-consumer-301632 1 4048 4048 > ``` > > thanks, > kienan > > [1]: > > https://github.com/lttng/lttng-ust/commit/4d4838bad480d48424bddc686f5ad00= 89e28ac94 > [2]: > > https://github.com/lttng/lttng-ust/commit/97572c0438845cee953ebd3e39615f7= 8bfa405a7 > > On 3/17/25 2:29 AM, Gour DEV wrote: > > Hi, Kienan > > > > Sorry for the late reply. > > > > Looks like in buster the memory is allocated by lttng-consumerd reserve= d > > > > I buster, the rss is less than the VIRT > > root@localhost:~# COLUMNS=3D500 top -b -n 1 | grep lttng > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ > COMMAND > > 4095 root 20 0 1003188 31256 4660 S 0.0 0.1 0:03.81 > > lttng-sessiond > > 4096 root 20 0 44260 796 0 S 0.0 0.0 0:00.01 > > lttng-runas > > 4440 root 20 0 5236020 10224 8756 S 0.0 0.0 2:56.25 > > lttng-consumerd -- here the VIRT is much more higher than RSS > > 4443 root 20 0 48048 540 0 S 0.0 0.0 0:00.12 > > lttng-runas > > > > > > > > In bookworm the VIRT and RES are nearly the same only. > > root@edgecore-40XKE-j2-101-32:~# COLUMNS=3D500 top -b -n 1 | grep lttn= g > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ > COMMAND > > 4382 root 20 0 1098824 42600 8436 S 0.0 0.1 0:08.87 > > lttng-sessiond > > 4403 root 20 0 48928 2116 996 S 0.0 0.0 0:00.00 > > lttng-runas > > 5171 root 20 0 9879764 8.9g 8.9g S 0.0 28.7 108:23.53 > > lttng-consumerd -- here the VRIT is nearly equal to RSS > > 5173 root 20 0 3680 1028 680 S 0.0 0.0 0:00.88 > > lttng-runas > > > > > > Looks like lttng consumerd is allocating and reserving those pages, whe= n > > any instrumented application starts. > > > > I am attaching the lttng status output in the mail, please do tell me i= f > > you need any more information regarding this. > > > > > > These is how we used to create the lttng channels and enable event whic= h > is > > same in both buster and bookworm, (number of channels might differ) > > > > def enable_channel(channels, session, subbuf_size, subbuf_num): > > for c in channels: > > call(['lttng', 'enable-channel', '-u', c, '-s', session, '--subbuf-size= ', > > str(subbuf_size), '--num-subbuf', str(subbuf_num),], > > stdout=3Ddevnull, stderr=3Dsubprocess.STDOUT) > > > > > > def enable_events(traces, session): > > for t in traces: > > if 'log-level-only' in t: > > log_opt =3D '--loglevel-only=3D' + t['log-level-only'] > > elif 'log-level' in t: > > log_opt =3D '--loglevel=3D' + t['log-level'] > > else: > > log_opt =3D '' > > > > else: > > call(['lttng', 'enable-event', '-u', t['name'], '-c', t['channel'], > > '-s', session], stdout=3Ddevnull, stderr=3Dsubprocess.STDOUT) > > > > > > Thank You. > > Lakshya > > > > > > > > > > On Wed, Mar 12, 2025 at 8:06=E2=80=AFPM > wrote: > > > >> Send lttng-dev mailing list submissions to > >> lttng-dev@lists.lttng.org > >> > >> To subscribe or unsubscribe via the World Wide Web, visit > >> https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev > >> or, via email, send a message with subject or body 'help' to > >> lttng-dev-request@lists.lttng.org > >> > >> You can reach the person managing the list at > >> lttng-dev-owner@lists.lttng.org > >> > >> When replying, please edit your Subject line so it is more specific > >> than "Re: Contents of lttng-dev digest..." > >> > >> > >> Today's Topics: > >> > >> 1. Re: Memory Consumption High After Upgrading to 2.13 from 2.10 > >> (Kienan Stewart) > >> 2. Re: Memory Consumption High After Upgrading to 2.13 from 2.10 > >> (Gour DEV) > >> 3. Re: Memory Consumption High After Upgrading to 2.13 from 2.10 > >> (Kienan Stewart) > >> > >> > >> ---------------------------------------------------------------------- > >> > >> Message: 1 > >> Date: Tue, 11 Mar 2025 14:55:21 -0400 > >> From: Kienan Stewart > >> To: Gour DEV , lttng-dev@lists.lttng.org > >> Subject: Re: Memory Consumption High After Upgrading to 2.13 from 2.10 > >> Message-ID: <38dab5ef-f106-4e57-9e36-b4b30015c019@efficios.com> > >> Content-Type: text/plain; charset=3DUTF-8; format=3Dflowed > >> > >> Hi Lakshya, > >> > >> On 3/11/25 12:25 PM, Gour DEV wrote: > >> > Hi, Kienan > >> > > >> > here is the requested output > >> > > >> > root@localhost:~# top -b -n 1 | grep lttng > >> > 4841 root 20 0 11.5g 11.0g 11.0g S 5.9 35.4 > 8:39.93 > >> > lttng-c+ > >> > 4824 root 20 0 1098824 26456 5380 S 0.0 0.1 > 0:07.25 > >> > lttng-s+ > >> > 4825 root 20 0 48872 2188 1012 S 0.0 0.0 > 0:00.00 > >> > lttng-r+ > >> > 4843 root 20 0 3680 1160 816 S 0.0 0.0 > 0:00.23 > >> > >> This top output for `localhost` seems very different than the output f= or > >> `localhost` in your previous message. > >> > >> > >> > lttng-r+ > >> > root@localhost:~# nrpco > >> > bash: nrpco: command not found > >> > root@localhost:~# nproc > >> > 16 > >> > root@localhost:~# cat /sys/devices/system/cpu/possible > >> > 0-15 > >> > > >> > >> You indicated the bookworm machine has 32 cores, this is showing 16. I= f > >> you're comparing a 16 core machine to a 32 core machine, it is very > >> normal that the memory usage is higher on the 32 core machine. > >> > >> > > >> > Most of the process are running as asorcs user but some are runnin= g > >> as root. > >> > >> So you have two users with instrumented applications. > >> > >> > >> Given the discrepancies in the information provided I'm finding it a b= it > >> hard to understand what you're looking at. > >> > >> > >> In general, a channel's shared memory footprint can be estimated > with[1]: > >> > >> (nSubbuf * subbufSize) * (nCPUs + 1 iff snapshot mode is enabled) = * > >> (nUIDs or nPIDs) > >> > >> Note that the sub-buffer sizes you are using get rounded to the neares= t > >> larger power of 2. See [2]. > >> > >> thanks, > >> kienan > >> > >> [1]: https://lttng.org/docs/v2.13/#doc-channel-buffering-schemes > >> [2]: > >> > https://lttng.org/man/1/lttng-enable-channel/v2.13/#doc-opt--subbuf-size > >> > >> > >> ------------------------------ > >> > >> Message: 2 > >> Date: Wed, 12 Mar 2025 14:49:07 +0530 > >> From: Gour DEV > >> To: Kienan Stewart > >> Cc: lttng-dev@lists.lttng.org > >> Subject: Re: Memory Consumption High After Upgrading to 2.13 from 2.10 > >> Message-ID: > >> >> nqOHNZMw@mail.gmail.com> > >> Content-Type: text/plain; charset=3D"utf-8" > >> > >> Hi, Kienan > >> > >> I am attaching an screen recording of the behaviour I am seeing in thi= s > >> mail. The behaviour is same irrespective of the device i use, sorry fo= r > >> miscommunication in the npocs output (I assumed it was 32), but other > than > >> that all outputs are same (except the hostname as there are multiple > >> devices with same lttng config but this memory cosumption is seen on a= ll > >> the devices). > >> > >> I had few question > >> > >> 1. Does lltng allocated all the memory it needs and mark it as dirty i= n > ram > >> when any process which links/uses lttng-ust runs? (here i tried with o= ne > >> process but it is same for any of my process) > >> 2. (nSubbuf * subbufSize) * (nCPUs + 1 iff snapshot mode is enabled) * > >> (nUIDs or nPIDs) > >> > >> How do we calculate uid in the system is it all uids in the system? is > it > >> equal to `cat /etc/passwd | wc -l` ? > >> > >> I will put my calculations according to the above estimate based on al= l > the > >> channel i am creating > >> > >> (4194304*4 + 262144*4 + 16384*4) * (16) * (30 if number user are equal > to > >> `cat /etc/passwd | wc -l`)B =3D 7.998046875 GB approx [this is based o= n > the > >> start_lttng.py please do correct me if am wrong here.] > >> > >> But since there are only two users which uses lttng i think the correc= t > >> estimate would be > >> (4194304*4 + 262144*4 + 16384*4) * (16) * (2)B =3D 546MB > >> > >> Please do correct me If I am wrong calculations here. > >> > >> Now, there are a few things here, according to my output lttng is usin= g > 11G > >> which is much more higher than the what is configured. > >> > >> I am attaching the lttng status and the file which is uses to create t= he > >> lttng sessions. > >> > >> > >> > >> Thank You. > >> > >> > >> > >> > https://drive.google.com/file/d/1tS_ZWEsXDpHZXfWzZHXmWcT0igiIOIaa/view?us= p=3Dsharing > >> -- recording of the behaviour which is seen > >> > >> > https://drive.google.com/file/d/1PrU31oyEw1n9tKETlUtmNGO50s6ywx7p/view?us= p=3Dsharing > >> -- the file which is used to create lttng sessions > >> > >> > >> On Wed, Mar 12, 2025 at 12:25?AM Kienan Stewart > >> wrote: > >> > >>> Hi Lakshya, > >>> > >>> On 3/11/25 12:25 PM, Gour DEV wrote: > >>> > Hi, Kienan > >>> > > >>> > here is the requested output > >>> > > >>> > root@localhost:~# top -b -n 1 | grep lttng > >>> > 4841 root 20 0 11.5g 11.0g 11.0g S 5.9 35.4 > >> 8:39.93 > >>> > lttng-c+ > >>> > 4824 root 20 0 1098824 26456 5380 S 0.0 0.1 > >> 0:07.25 > >>> > lttng-s+ > >>> > 4825 root 20 0 48872 2188 1012 S 0.0 0.0 > >> 0:00.00 > >>> > lttng-r+ > >>> > 4843 root 20 0 3680 1160 816 S 0.0 0.0 > >> 0:00.23 > >>> > >>> This top output for `localhost` seems very different than the output > for > >>> `localhost` in your previous message. > >>> > >>> > >>> > lttng-r+ > >>> > root@localhost:~# nrpco > >>> > bash: nrpco: command not found > >>> > root@localhost:~# nproc > >>> > 16 > >>> > root@localhost:~# cat /sys/devices/system/cpu/possible > >>> > 0-15 > >>> > > >>> > >>> You indicated the bookworm machine has 32 cores, this is showing 16. = If > >>> you're comparing a 16 core machine to a 32 core machine, it is very > >>> normal that the memory usage is higher on the 32 core machine. > >>> > >>> > > >>> > Most of the process are running as asorcs user but some are runni= ng > >>> as root. > >>> > >>> So you have two users with instrumented applications. > >>> > >>> > >>> Given the discrepancies in the information provided I'm finding it a > bit > >>> hard to understand what you're looking at. > >>> > >>> > >>> In general, a channel's shared memory footprint can be estimated > with[1]: > >>> > >>> (nSubbuf * subbufSize) * (nCPUs + 1 iff snapshot mode is enabled)= * > >>> (nUIDs or nPIDs) > >>> > >>> Note that the sub-buffer sizes you are using get rounded to the neare= st > >>> larger power of 2. See [2]. > >>> > >>> thanks, > >>> kienan > >>> > >>> [1]: https://lttng.org/docs/v2.13/#doc-channel-buffering-schemes > >>> [2]: > >>> > https://lttng.org/man/1/lttng-enable-channel/v2.13/#doc-opt--subbuf-size > >>> > >> -------------- next part -------------- > >> An HTML attachment was scrubbed... > >> URL: < > >> > https://lists.lttng.org/pipermail/lttng-dev/attachments/20250312/57f240d8= /attachment-0001.htm > >>> > >> > >> ------------------------------ > >> > >> Message: 3 > >> Date: Wed, 12 Mar 2025 10:36:28 -0400 > >> From: Kienan Stewart > >> To: Gour DEV , lttng-dev@lists.lttng.org > >> Subject: Re: Memory Consumption High After Upgrading to 2.13 from 2.10 > >> Message-ID: <0f819583-ea8e-468e-9102-e1410d886a6f@efficios.com> > >> Content-Type: text/plain; charset=3DUTF-8; format=3Dflowed > >> > >> Hi Lakshya, > >> > >> On 3/12/25 5:03 AM, Gour DEV wrote: > >>> Hi, Kienan > >>> > >>> I am attaching an screen recording of the behaviour I am seeing in th= is > >>> mail. The behaviour is same irrespective of the device i use, sorry f= or > >>> miscommunication in the npocs output (I assumed it was 32), but other > >> than > >>> that all outputs are same (except the hostname as there are multiple > >>> devices with same lttng config but this memory cosumption is seen on > all > >>> the devices). > >>> > >>> I had few question > >>> > >>> 1. Does lltng allocated all the memory it needs and mark it as dirty = in > >> ram > >>> when any process which links/uses lttng-ust runs? (here i tried with > one > >>> process but it is same for any of my process) > >> > >> I believe the shared memory for per-CPU data structures is allocated > >> when an instrumented application connects. There is no pre-allocation > >> for each possible UID on the system. > >> > >> You can run your instrumented applications with `LTTNG_UST_DEBUG=3D1` = to > >> see when the connection happens[1]. > >> > >>> 2. (nSubbuf * subbufSize) * (nCPUs + 1 iff snapshot mode is enabled) = * > >>> (nUIDs or nPIDs) > >>> > >>> How do we calculate uid in the system is it all uids in the system? i= s > it > >>> equal to `cat /etc/passwd | wc -l` ? > >> > >> nUIDs is the number of distinct UIDs running instrumented applications= . > >> > >>> > >>> I will put my calculations according to the above estimate based on a= ll > >> the > >>> channel i am creating > >>> > >>> (4194304*4 + 262144*4 + 16384*4) * (16) * (30 if number user are equa= l > to > >>> `cat /etc/passwd | wc -l`)B =3D 7.998046875 GB approx [this is based = on > the > >>> start_lttng.py please do correct me if am wrong here.] > >>> > >>> But since there are only two users which uses lttng i think the corre= ct > >>> estimate would be > >>> (4194304*4 + 262144*4 + 16384*4) * (16) * (2)B =3D 546MB > >> > >> The estimate I gave is per-channel. > >> > >> small channel: (0.015625 MiB * 4) * (16 + 1) =3D 1.0625 MiB per-channe= l > >> per-UID > >> medium channel: (0.250 MiB * 4) * (16 + 1) =3D 17.0 MiB per-channel > per-UID > >> large channel: (4 MiB * 4) * (16 + 1) =3D 27 2MiB per-channel per-UID > >> > >> Now, you said you have 0 small channels, 6 medium channels, and 16 lar= ge > >> channels in your session. (Note: I see your script differs from these > >> stated channel counts). > >> > >> small: 0 * 1.0625 MiB =3D 0 MiB per-UID > >> medium: 6 * 17 MiB =3D 102 MiB per-UID > >> large: 16 * 272 MiB =3D 4352 MiB per-UID > >> > >> And if you're running instrumented applications with 2 users: > >> > >> small: 0 MiB * 2 =3D 0 MiB with 2 UIDs > >> medium: 102 MiB * 2 =3D 204 MiB with 2 UIDs > >> large: 4352 MiB * 2 =3D 8704 MiB with 2 UIDs > >> > >> Now this is just an estimation for the per-CPU ring buffers only, and > >> you numbers aren't hugely off so without analyzing your specific syste= m > >> it doesn't seem to be that strange to me. > >> > >> If I take the number of channels I see in your script, it becomes: > >> > >> small: 0 MiB with 2 UIDs > >> medium: 136 MiB with 2 UIDs > >> large: 7616 MiB with 2 UIDs > >> > >> total: 7.57 GiB with 2 UIDs > >> > >>> > >>> Please do correct me If I am wrong calculations here. > >>> > >>> Now, there are a few things here, according to my output lttng is usi= ng > >> 11G > >>> which is much more higher than the what is configured. > >>> > >> > >> I have no idea what 'service start spyder' is doing. Maybe it's runnin= g > >> instrumented applications with an extra user that you didn't expect? I > >> can't help you with that aspect of your system. > >> > >> The above estimated 7.57 GiB with 2 UIDs would be 11.35 GiB with 3 UID= s > >> so maybe? > >> > >> I'd recommend you read your verbose sessiond log so see which > >> applications are connecting and with which UIDs. > >> > >>> I am attaching the lttng status and the file which is uses to create > the > >>> lttng sessions. > >>> > >>> > >>> > >>> Thank You. > >>> > >> > >> In any case, the information you have given to date hasn't demonstrate= d > >> to me in a tangible manner that you are seeing a difference related to > >> the version of LTTng being used. > >> > >> thanks, > >> kienan > >> > >> [1]: > https://lttng.org/man/3/lttng-ust/v2.13/#doc-_environment_variables > >> > >> > >> ------------------------------ > >> > >> Subject: Digest Footer > >> > >> _______________________________________________ > >> lttng-dev mailing list > >> lttng-dev@lists.lttng.org > >> https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev > >> > >> > >> ------------------------------ > >> > >> End of lttng-dev Digest, Vol 203, Issue 7 > >> ***************************************** > >> > > > > --000000000000c31f030630d9bd43 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Thanks=C2=A0Kienan for digging this and explaining the beh= aviour,

Looks like this option is not present in the version I am us= ing v2.13.5 and the MAP_POPULATE is enabled by default there
/* memory_m= ap: mmap */
memory_map =3D mmap(NULL, memory_map_size, PROT_READ | PROT= _WRITE,
=C2=A0MAP_SHARED | LTTNG_MAP_POPULATE, shmfd, 0);

I h= ave updated to the master branch and have verified by setting the LTTNG_UST= _MAP_POPULATE_POLICY=3Dnone, and saw that now the RSS consumption is not ve= ry high now.

I think using LTTNG_UST_MAP_POPULATE_POLICY=3Dcpu_possi= ble is better because here I will definitely=C2=A0know how much amount of m= emory lttng will use at start only, and allocate memory=C2=A0accordingly to= my servers.


Also, I thought using=C2=A0--buffers-global=C2=A0would be better for me as I could know=C2= =A0how much memory will be required but looks like that option is only avai= lable for kernel tracking, it would be very good if this option could be av= ailable for user buffering also, but that could come later.


But = again, thanks for the help Kienan.

Regards
Lakshya.


= =C2=A0



On Fri, Mar 21, 2025 at 2:21=E2=80= =AFAM Kienan Stewart <kstewart@= efficios.com> wrote:
Hi Lakshya,

I did some digging around. What you are seeing is the result of the
switching to MAP_POPULATE by default in LTTng-UST 2.12[1] in commit
4d4838b ("Use MAP_POPULATE to reduce pagefault when available").<= br>
The purpose of this change is to avoid taking page faults which tracing, reducing first-event in a page latency.

In the master branch, this feature has been made configurable for users who don't want to pre-populate the pages and would rather take page faults while tracing[2].

Here is an example from LTTng master with map populate per possible CPU:
```
export LTTNG_UST_MAP_POPULATE_POLICY=3Dcpu_possible

# Create session, channels, start tracing, and run test app
# top -n1 -b | grep -E '(MiB|COMMAND|lttng)'
MiB Mem :=C2=A0 32768.0 total,=C2=A0 21883.7 free,=C2=A0 =C2=A01456.0 used,= =C2=A0 =C2=A09428.3
buff/cache

MiB Swap:=C2=A0 =C2=A0 =C2=A0 0.0 total,=C2=A0 =C2=A0 =C2=A0 0.0 free,=C2= =A0 =C2=A0 =C2=A0 0.0 used.=C2=A0 31312.0 avail
Mem

=C2=A0 =C2=A0 =C2=A0PID USER=C2=A0 =C2=A0 =C2=A0 PR=C2=A0 NI=C2=A0 =C2=A0 V= IRT=C2=A0 =C2=A0 RES=C2=A0 =C2=A0 SHR S=C2=A0 %CPU=C2=A0 %MEM=C2=A0 =C2=A0 = =C2=A0TIME+
COMMAND

=C2=A0 301117 debian=C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 880176=C2=A0 =C2= =A02760=C2=A0 =C2=A02760 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.0=C2=A0 =C2=A00:00= .04
lttng-sessiond

=C2=A0 301118 debian=C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 =C2=A043856=C2=A0 = =C2=A01376=C2=A0 =C2=A01376 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.0=C2=A0 =C2=A00= :00.01
lttng-runas

=C2=A0 301133 debian=C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 718616 263456 2634= 56 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.8=C2=A0 =C2=A00:00.17
lttng-consumerd

=C2=A0 301135 debian=C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 =C2=A0 6996=C2=A0 = =C2=A0 =C2=A0 0=C2=A0 =C2=A0 =C2=A0 0 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.0=C2= =A0 =C2=A00:00.05
lttng-runas

# cat /proc/$(pgrep lttng-sessiond)/statm
lttng-sessiond: 220044 690 690 345 0 29900 0



# pmap $(pgrep lttng-sessiond) | grep total

=C2=A0 total=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0880176K

# smem -P lttng-sessiond


=C2=A0 =C2=A0PID User=C2=A0 =C2=A0 =C2=A0Command=C2=A0 =C2=A0 =C2=A0 =C2=A0= =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Swap=C2=A0 = =C2=A0 =C2=A0 USS=C2=A0 =C2=A0 =C2=A0 PSS
=C2=A0 =C2=A0 RSS

301118 debian=C2=A0 =C2=A0lttng-sessiond --daemonize=C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A00=C2=A0 =C2=A0 =C2=A0 344=C2=A0 =C2=A0 =C2=A0 881
=C2=A0 =C2=A02236

301117 debian=C2=A0 =C2=A0lttng-sessiond --daemonize=C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A00=C2=A0 =C2=A0 =C2=A05676=C2=A0 =C2=A0 =C2=A06683
=C2=A0 =C2=A09276

301201 debian=C2=A0 =C2=A0/usr/bin/python3 /usr/bin/s=C2=A0 =C2=A0 =C2=A0 = =C2=A0 0=C2=A0 =C2=A0 =C2=A08636=C2=A0 =C2=A0 10086
=C2=A0 12936


# /proc/PID/statm for lttng-consumerd



lttng-consumerd: 1749 0 0 129 0 130 0

# pmap lttng-consumerd-pid | grep total
total kB=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 6996=C2=A0 =C2=A0 1700=C2= =A0 =C2=A0 =C2=A0472



# smem -P lttng-consumerd
=C2=A0 =C2=A0PID User=C2=A0 =C2=A0 =C2=A0Command=C2=A0 =C2=A0 =C2=A0 =C2=A0= =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Swap=C2=A0 = =C2=A0 =C2=A0 USS=C2=A0 =C2=A0 =C2=A0 PSS
=C2=A0 =C2=A0 RSS

301135 debian=C2=A0 =C2=A0lttng-consumerd=C2=A0 -u --consu=C2=A0 =C2=A0 =C2= =A0 =C2=A0 0=C2=A0 =C2=A0 =C2=A0 280=C2=A0 =C2=A0 =C2=A0 563
=C2=A0 =C2=A01700

301211 debian=C2=A0 =C2=A0/usr/bin/python3 /usr/bin/s=C2=A0 =C2=A0 =C2=A0 = =C2=A0 0=C2=A0 =C2=A0 10048=C2=A0 =C2=A0 11501
=C2=A0 14404

301133 debian=C2=A0 =C2=A0lttng-consumerd=C2=A0 -u --consu=C2=A0 =C2=A0 =C2= =A0 =C2=A0 0=C2=A0 =C2=A0262376=C2=A0 =C2=A0263177
265480

# smem -m | grep -i ust

/dev/shm/lttng-ust-wait-8-1000=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A01=C2=A0 =C2=A0 =C2=A0 =C2=A0 4=C2=A0 =C2=A0 =C2=A0 =C2=A0 4
/dev/shm/shm-ust-consumer-301133=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A01=C2=A0 =C2=A0260756=C2=A0 =C2=A0260756
```

When using the none policy:

```
# export LTTNG_UST_MAP_POPULATE_POLICY=3Dnone
# as above

Running test app UID 0


procs -----------memory---------- ---swap-- -----io---- -system--
------cpu-----

=C2=A0 r=C2=A0 b=C2=A0 =C2=A0swpd=C2=A0 =C2=A0free=C2=A0 =C2=A0buff=C2=A0 c= ache=C2=A0 =C2=A0si=C2=A0 =C2=A0so=C2=A0 =C2=A0 bi=C2=A0 =C2=A0 bo=C2=A0 = =C2=A0in=C2=A0 =C2=A0cs us sy
id wa st

=C2=A0 1=C2=A0 0=C2=A0 =C2=A0 =C2=A0 0=C2=A0 21875=C2=A0 =C2=A0 =C2=A0 0=C2= =A0 =C2=A09434=C2=A0 =C2=A0 0=C2=A0 =C2=A0 0=C2=A0 =C2=A0 39=C2=A0 =C2=A063= 6 1105 2496=C2=A0 0=C2=A0 1
99=C2=A0 0=C2=A0 0

MiB Mem :=C2=A0 32768.0 total,=C2=A0 21875.0 free,=C2=A0 =C2=A01458.2 used,= =C2=A0 =C2=A09434.7
buff/cache

MiB Swap:=C2=A0 =C2=A0 =C2=A0 0.0 total,=C2=A0 =C2=A0 =C2=A0 0.0 free,=C2= =A0 =C2=A0 =C2=A0 0.0 used.=C2=A0 31309.8 avail
Mem

=C2=A0 =C2=A0 =C2=A0PID USER=C2=A0 =C2=A0 =C2=A0 PR=C2=A0 NI=C2=A0 =C2=A0 V= IRT=C2=A0 =C2=A0 RES=C2=A0 =C2=A0 SHR S=C2=A0 %CPU=C2=A0 %MEM=C2=A0 =C2=A0 = =C2=A0TIME+
COMMAND

=C2=A0 301616 debian=C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 880176=C2=A0 =C2= =A02756=C2=A0 =C2=A02756 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.0=C2=A0 =C2=A00:00= .04
lttng-sessiond

=C2=A0 301617 debian=C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 =C2=A043856=C2=A0 = =C2=A01392=C2=A0 =C2=A01392 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.0=C2=A0 =C2=A00= :00.01
lttng-runas

=C2=A0 301632 debian=C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 718612=C2=A0 =C2= =A05416=C2=A0 =C2=A05416 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.0=C2=A0 =C2=A00:00= .17
lttng-consumerd

=C2=A0 301634 debian=C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 =C2=A0 6992=C2=A0 = =C2=A0 =C2=A0 0=C2=A0 =C2=A0 =C2=A0 0 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.0=C2= =A0 =C2=A00:00.05
lttng-runas

lttng-sessiond: 220044 689 689 345 0 29900 0


=C2=A0 total=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0880176K


=C2=A0 =C2=A0PID User=C2=A0 =C2=A0 =C2=A0Command=C2=A0 =C2=A0 =C2=A0 =C2=A0= =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Swap=C2=A0 = =C2=A0 =C2=A0 USS=C2=A0 =C2=A0 =C2=A0 PSS
=C2=A0 =C2=A0 RSS
301617 debian=C2=A0 =C2=A0lttng-sessiond --daemonize=C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A00=C2=A0 =C2=A0 =C2=A0 344=C2=A0 =C2=A0 =C2=A0 862
=C2=A0 =C2=A02188
301616 debian=C2=A0 =C2=A0lttng-sessiond --daemonize=C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A00=C2=A0 =C2=A0 =C2=A05784=C2=A0 =C2=A0 =C2=A06759
=C2=A0 =C2=A09328
301700 debian=C2=A0 =C2=A0/usr/bin/python3 /usr/bin/s=C2=A0 =C2=A0 =C2=A0 = =C2=A0 0=C2=A0 =C2=A0 =C2=A08632=C2=A0 =C2=A0 10079
=C2=A0 12928

lttng-consumerd: 1748 0 0 129 0 129 0
total kB=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 6992=C2=A0 =C2=A0 1580=C2= =A0 =C2=A0 =C2=A0468
=C2=A0 =C2=A0PID User=C2=A0 =C2=A0 =C2=A0Command=C2=A0 =C2=A0 =C2=A0 =C2=A0= =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Swap=C2=A0 = =C2=A0 =C2=A0 USS=C2=A0 =C2=A0 =C2=A0 PSS
=C2=A0 =C2=A0 RSS
301634 debian=C2=A0 =C2=A0lttng-consumerd=C2=A0 -u --consu=C2=A0 =C2=A0 =C2= =A0 =C2=A0 0=C2=A0 =C2=A0 =C2=A0 276=C2=A0 =C2=A0 =C2=A0 536
=C2=A0 =C2=A01580
301632 debian=C2=A0 =C2=A0lttng-consumerd=C2=A0 -u --consu=C2=A0 =C2=A0 =C2= =A0 =C2=A0 0=C2=A0 =C2=A0 =C2=A05672=C2=A0 =C2=A0 =C2=A06433
=C2=A0 =C2=A08652
301710 debian=C2=A0 =C2=A0/usr/bin/python3 /usr/bin/s=C2=A0 =C2=A0 =C2=A0 = =C2=A0 0=C2=A0 =C2=A0 =C2=A09996=C2=A0 =C2=A0 11449
=C2=A0 14328

/dev/shm/lttng-ust-wait-8-1000=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A01=C2=A0 =C2=A0 =C2=A0 =C2=A0 4=C2=A0 =C2=A0 =C2=A0 =C2=A0 4
/dev/shm/shm-ust-consumer-301632=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A01=C2=A0 =C2=A0 =C2=A04048=C2=A0 =C2=A0 =C2=A04048
```

thanks,
kienan

[1]:
https://github.com= /lttng/lttng-ust/commit/4d4838bad480d48424bddc686f5ad0089e28ac94
[2]:
https://github.com= /lttng/lttng-ust/commit/97572c0438845cee953ebd3e39615f78bfa405a7

On 3/17/25 2:29 AM, Gour DEV wrote:
> Hi, Kienan
>
> Sorry for the late reply.
>
> Looks like in buster the memory is allocated by lttng-consumerd reserv= ed
>
> I buster, the rss is less than the VIRT
> root@localhost:~#=C2=A0 COLUMNS=3D500 top=C2=A0 -b -n 1 | grep lttng >=C2=A0 =C2=A0PID USER=C2=A0 =C2=A0 =C2=A0 PR=C2=A0 NI=C2=A0 =C2=A0 VIRT= =C2=A0 =C2=A0 RES=C2=A0 =C2=A0 SHR S=C2=A0 %CPU=C2=A0 %MEM=C2=A0 =C2=A0 =C2= =A0TIME+ COMMAND
>=C2=A0 =C2=A04095 root=C2=A0 =C2=A0 =C2=A0 20=C2=A0 =C2=A00 1003188=C2= =A0 31256=C2=A0 =C2=A04660 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.1=C2=A0 =C2=A00:= 03.81
> lttng-sessiond
>=C2=A0 =C2=A04096 root=C2=A0 =C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 =C2= =A044260=C2=A0 =C2=A0 796=C2=A0 =C2=A0 =C2=A0 0 S=C2=A0 =C2=A00.0=C2=A0 =C2= =A00.0=C2=A0 =C2=A00:00.01
> lttng-runas
>=C2=A0 =C2=A04440 root=C2=A0 =C2=A0 =C2=A0 20=C2=A0 =C2=A00 5236020=C2= =A0 10224=C2=A0 =C2=A08756 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.0=C2=A0 =C2=A02:= 56.25
> lttng-consumerd -- here the VIRT is much more higher than RSS
>=C2=A0 =C2=A04443 root=C2=A0 =C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2=A0 =C2= =A048048=C2=A0 =C2=A0 540=C2=A0 =C2=A0 =C2=A0 0 S=C2=A0 =C2=A00.0=C2=A0 =C2= =A00.0=C2=A0 =C2=A00:00.12
> lttng-runas
>
>
>
> In bookworm the VIRT and RES are nearly the same only.
> root@edgecore-40XKE-j2-101-32:~# COLUMNS=3D500 top=C2=A0 -b -n 1 | gre= p lttng
>=C2=A0 =C2=A0PID USER=C2=A0 =C2=A0 =C2=A0 PR=C2=A0 NI=C2=A0 =C2=A0 VIRT= =C2=A0 =C2=A0 RES=C2=A0 =C2=A0 SHR S=C2=A0 %CPU=C2=A0 %MEM=C2=A0 =C2=A0 =C2= =A0TIME+ COMMAND
>=C2=A0 =C2=A0 =C2=A04382 root=C2=A0 =C2=A0 =C2=A0 20=C2=A0 =C2=A00 1098= 824=C2=A0 42600=C2=A0 =C2=A08436 S=C2=A0 =C2=A00.0=C2=A0 =C2=A00.1=C2=A0 = =C2=A00:08.87
> lttng-sessiond
>=C2=A0 =C2=A0 =C2=A04403 root=C2=A0 =C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2= =A0 =C2=A048928=C2=A0 =C2=A02116=C2=A0 =C2=A0 996 S=C2=A0 =C2=A00.0=C2=A0 = =C2=A00.0=C2=A0 =C2=A00:00.00
> lttng-runas
>=C2=A0 =C2=A0 =C2=A05171 root=C2=A0 =C2=A0 =C2=A0 20=C2=A0 =C2=A00 9879= 764=C2=A0 =C2=A08.9g=C2=A0 =C2=A08.9g S=C2=A0 =C2=A00.0=C2=A0 28.7 108:23.5= 3
> lttng-consumerd -- here the VRIT is nearly equal to RSS
>=C2=A0 =C2=A0 =C2=A05173 root=C2=A0 =C2=A0 =C2=A0 20=C2=A0 =C2=A00=C2= =A0 =C2=A0 3680=C2=A0 =C2=A01028=C2=A0 =C2=A0 680 S=C2=A0 =C2=A00.0=C2=A0 = =C2=A00.0=C2=A0 =C2=A00:00.88
> lttng-runas
>
>
> Looks like lttng consumerd is allocating and reserving those pages, wh= en
> any instrumented application starts.
>
> I am attaching the lttng status output in the mail, please do tell me = if
> you need any more information regarding this.
>
>
> These is how we used to create the lttng channels and enable event whi= ch is
> same in both buster and bookworm, (number of channels might differ) >
> def enable_channel(channels, session, subbuf_size, subbuf_num):
> for c in channels:
> call(['lttng', 'enable-channel', '-u', c, '= ;-s', session, '--subbuf-size',
> str(subbuf_size), '--num-subbuf', str(subbuf_num),],
> stdout=3Ddevnull, stderr=3Dsubprocess.STDOUT)
>
>
> def enable_events(traces, session):
> for t in traces:
> if 'log-level-only' in t:
> log_opt =3D '--loglevel-only=3D' + t['log-level-only']=
> elif 'log-level' in t:
> log_opt =3D '--loglevel=3D' + t['log-level']
> else:
> log_opt =3D ''
>
> else:
> call(['lttng', 'enable-event', '-u', t['na= me'], '-c', t['channel'],
> '-s', session], stdout=3Ddevnull, stderr=3Dsubprocess.STDOUT)<= br> >
>
> Thank You.
> Lakshya
>
>
>
>
> On Wed, Mar 12, 2025 at 8:06=E2=80=AFPM <lttng-dev-request@lists.lttng.= org> wrote:
>
>> Send lttng-dev mailing list submissions to
>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 lttng-dev@lists.lttng.org
>>
>> To subscribe or unsubscribe via the World Wide Web, visit
>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
>> or, via email, send a message with subject or body 'help' = to
>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 lttng-dev-request@lists.lttng.org
>>
>> You can reach the person managing the list at
>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 lttng-dev-owner@lists.lttng.org >>
>> When replying, please edit your Subject line so it is more specifi= c
>> than "Re: Contents of lttng-dev digest..."
>>
>>
>> Today's Topics:
>>
>>=C2=A0 =C2=A0 =C2=A01. Re: Memory Consumption High After Upgrading = to 2.13 from 2.10
>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 (Kienan Stewart)
>>=C2=A0 =C2=A0 =C2=A02. Re: Memory Consumption High After Upgrading = to 2.13 from 2.10
>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 (Gour DEV)
>>=C2=A0 =C2=A0 =C2=A03. Re: Memory Consumption High After Upgrading = to 2.13 from 2.10
>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 (Kienan Stewart)
>>
>>
>> ------------------------------------------------------------------= ----
>>
>> Message: 1
>> Date: Tue, 11 Mar 2025 14:55:21 -0400
>> From: Kienan Stewart <kstewart@efficios.com>
>> To: Gour DEV <lakshyagour10@gmail.com>, lttng-dev@lists.lttng.org
>> Subject: Re: Memory Consumption High After Upgrading to 2.13 from = 2.10
>> Message-ID: <38dab5ef-f106-4e57-9e36-b4b30015c019@e= fficios.com>
>> Content-Type: text/plain; charset=3DUTF-8; format=3Dflowed
>>
>> Hi Lakshya,
>>
>> On 3/11/25 12:25 PM, Gour DEV wrote:
>>=C2=A0 =C2=A0> Hi, Kienan
>>=C2=A0 =C2=A0>
>>=C2=A0 =C2=A0> here is the requested output
>>=C2=A0 =C2=A0>
>>=C2=A0 =C2=A0> root@localhost:~# top -b -n 1 | grep=C2=A0 lttng<= br> >>=C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A04841 root=C2=A0 =C2=A0 =C2=A0 = 20=C2=A0 =C2=A00=C2=A0 =C2=A011.5g=C2=A0 11.0g=C2=A0 11.0g S=C2=A0 =C2=A05.= 9=C2=A0 35.4=C2=A0 =C2=A08:39.93
>>=C2=A0 =C2=A0> lttng-c+
>>=C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A04824 root=C2=A0 =C2=A0 =C2=A0 = 20=C2=A0 =C2=A00 1098824=C2=A0 26456=C2=A0 =C2=A05380 S=C2=A0 =C2=A00.0=C2= =A0 =C2=A00.1=C2=A0 =C2=A00:07.25
>>=C2=A0 =C2=A0> lttng-s+
>>=C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A04825 root=C2=A0 =C2=A0 =C2=A0 = 20=C2=A0 =C2=A00=C2=A0 =C2=A048872=C2=A0 =C2=A02188=C2=A0 =C2=A01012 S=C2= =A0 =C2=A00.0=C2=A0 =C2=A00.0=C2=A0 =C2=A00:00.00
>>=C2=A0 =C2=A0> lttng-r+
>>=C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A04843 root=C2=A0 =C2=A0 =C2=A0 = 20=C2=A0 =C2=A00=C2=A0 =C2=A0 3680=C2=A0 =C2=A01160=C2=A0 =C2=A0 816 S=C2= =A0 =C2=A00.0=C2=A0 =C2=A00.0=C2=A0 =C2=A00:00.23
>>
>> This top output for `localhost` seems very different than the outp= ut for
>> `localhost` in your previous message.
>>
>>
>>=C2=A0 =C2=A0> lttng-r+
>>=C2=A0 =C2=A0> root@localhost:~# nrpco
>>=C2=A0 =C2=A0> bash: nrpco: command not found
>>=C2=A0 =C2=A0> root@localhost:~# nproc
>>=C2=A0 =C2=A0> 16
>>=C2=A0 =C2=A0> root@localhost:~# cat /sys/devices/system/cpu/pos= sible
>>=C2=A0 =C2=A0> 0-15
>>=C2=A0 =C2=A0>
>>
>> You indicated the bookworm machine has 32 cores, this is showing 1= 6. If
>> you're comparing a 16 core machine to a 32 core machine, it is= very
>> normal that the memory usage is higher on the 32 core machine.
>>
>>=C2=A0 =C2=A0>
>>=C2=A0 =C2=A0> Most of the process are running as asorcs user bu= t some are running
>> as root.
>>
>> So you have two users with instrumented applications.
>>
>>
>> Given the discrepancies in the information provided I'm findin= g it a bit
>> hard to understand what you're looking at.
>>
>>
>> In general, a channel's shared memory footprint can be estimat= ed with[1]:
>>
>>=C2=A0 =C2=A0 =C2=A0(nSubbuf * subbufSize) * (nCPUs + 1 iff snapsho= t mode is enabled) *
>> (nUIDs or nPIDs)
>>
>> Note that the sub-buffer sizes you are using get rounded to the ne= arest
>> larger power of 2. See [2].
>>
>> thanks,
>> kienan
>>
>> [1]: https://lttng.org/docs/v2.1= 3/#doc-channel-buffering-schemes
>> [2]:
>> https://lttng.org/m= an/1/lttng-enable-channel/v2.13/#doc-opt--subbuf-size
>>
>>
>> ------------------------------
>>
>> Message: 2
>> Date: Wed, 12 Mar 2025 14:49:07 +0530
>> From: Gour DEV <lakshyagour10@gmail.com>
>> To: Kienan Stewart <kstewart@efficios.com>
>> Cc: lttng-dev@lists.lttng.org
>> Subject: Re: Memory Consumption High After Upgrading to 2.13 from = 2.10
>> Message-ID:
>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 <CAE9Jrzg7qsabhPiO-0=3DB1DY3b= Vo-3FYu2tiJR2Bmb=3D
>> nqOHN= ZMw@mail.gmail.com>
>> Content-Type: text/plain; charset=3D"utf-8"
>>
>> Hi, Kienan
>>
>> I am attaching an screen recording of the behaviour I am seeing in= this
>> mail. The behaviour is same irrespective of the device i use, sorr= y for
>> miscommunication in the npocs output (I assumed it was 32), but ot= her than
>> that all outputs are same (except the hostname as there are multip= le
>> devices with same lttng config but this memory cosumption is seen = on all
>> the devices).
>>
>> I had few question
>>
>> 1. Does lltng allocated all the memory it needs and mark it as dir= ty in ram
>> when any process which links/uses lttng-ust runs? (here i tried wi= th one
>> process but it is same for any of my process)
>> 2. (nSubbuf * subbufSize) * (nCPUs + 1 iff snapshot mode is enable= d) *
>> (nUIDs or nPIDs)
>>
>> How do we calculate uid in the system is it all uids in the system= ? is it
>> equal to `cat /etc/passwd | wc -l` ?
>>
>> I will put my calculations according to the above estimate based o= n all the
>> channel i am creating
>>
>> (4194304*4 + 262144*4 + 16384*4) * (16) * (30 if number user are e= qual to
>> `cat /etc/passwd | wc -l`)B =3D 7.998046875 GB approx [this is bas= ed on the
>> start_lttng.py please do correct me if am wrong here.]
>>
>> But since there are only two users which uses lttng i think the co= rrect
>> estimate would be
>> (4194304*4 + 262144*4 + 16384*4) * (16) * (2)B =3D 546MB
>>
>> Please do correct me If I am wrong calculations here.
>>
>> Now, there are a few things here, according to my output lttng is = using 11G
>> which is much more higher than the what is configured.
>>
>> I am attaching the lttng status and the file which is uses to crea= te the
>> lttng sessions.
>>
>>
>>
>> Thank You.
>>
>>
>>
>> https:/= /drive.google.com/file/d/1tS_ZWEsXDpHZXfWzZHXmWcT0igiIOIaa/view?usp=3Dshari= ng
>> -- recording of the behaviour which is seen
>>
>> https:/= /drive.google.com/file/d/1PrU31oyEw1n9tKETlUtmNGO50s6ywx7p/view?usp=3Dshari= ng
>> -- the file which is used to create lttng sessions
>>
>>
>> On Wed, Mar 12, 2025 at 12:25?AM Kienan Stewart <kstewart@efficios.com><= br> >> wrote:
>>
>>> Hi Lakshya,
>>>
>>> On 3/11/25 12:25 PM, Gour DEV wrote:
>>>=C2=A0 =C2=A0> Hi, Kienan
>>>=C2=A0 =C2=A0>
>>>=C2=A0 =C2=A0> here is the requested output
>>>=C2=A0 =C2=A0>
>>>=C2=A0 =C2=A0> root@localhost:~# top -b -n 1 | grep=C2=A0 lt= tng
>>>=C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A04841 root=C2=A0 =C2=A0 =C2= =A0 20=C2=A0 =C2=A00=C2=A0 =C2=A011.5g=C2=A0 11.0g=C2=A0 11.0g S=C2=A0 =C2= =A05.9=C2=A0 35.4
>>=C2=A0 =C2=A08:39.93
>>>=C2=A0 =C2=A0> lttng-c+
>>>=C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A04824 root=C2=A0 =C2=A0 =C2= =A0 20=C2=A0 =C2=A00 1098824=C2=A0 26456=C2=A0 =C2=A05380 S=C2=A0 =C2=A00.0= =C2=A0 =C2=A00.1
>>=C2=A0 =C2=A00:07.25
>>>=C2=A0 =C2=A0> lttng-s+
>>>=C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A04825 root=C2=A0 =C2=A0 =C2= =A0 20=C2=A0 =C2=A00=C2=A0 =C2=A048872=C2=A0 =C2=A02188=C2=A0 =C2=A01012 S= =C2=A0 =C2=A00.0=C2=A0 =C2=A00.0
>>=C2=A0 =C2=A00:00.00
>>>=C2=A0 =C2=A0> lttng-r+
>>>=C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A04843 root=C2=A0 =C2=A0 =C2= =A0 20=C2=A0 =C2=A00=C2=A0 =C2=A0 3680=C2=A0 =C2=A01160=C2=A0 =C2=A0 816 S= =C2=A0 =C2=A00.0=C2=A0 =C2=A00.0
>>=C2=A0 =C2=A00:00.23
>>>
>>> This top output for `localhost` seems very different than the = output for
>>> `localhost` in your previous message.
>>>
>>>
>>>=C2=A0 =C2=A0> lttng-r+
>>>=C2=A0 =C2=A0> root@localhost:~# nrpco
>>>=C2=A0 =C2=A0> bash: nrpco: command not found
>>>=C2=A0 =C2=A0> root@localhost:~# nproc
>>>=C2=A0 =C2=A0> 16
>>>=C2=A0 =C2=A0> root@localhost:~# cat /sys/devices/system/cpu= /possible
>>>=C2=A0 =C2=A0> 0-15
>>>=C2=A0 =C2=A0>
>>>
>>> You indicated the bookworm machine has 32 cores, this is showi= ng 16. If
>>> you're comparing a 16 core machine to a 32 core machine, i= t is very
>>> normal that the memory usage is higher on the 32 core machine.=
>>>
>>>=C2=A0 =C2=A0>
>>>=C2=A0 =C2=A0> Most of the process are running as asorcs use= r but some are running
>>> as root.
>>>
>>> So you have two users with instrumented applications.
>>>
>>>
>>> Given the discrepancies in the information provided I'm fi= nding it a bit
>>> hard to understand what you're looking at.
>>>
>>>
>>> In general, a channel's shared memory footprint can be est= imated with[1]:
>>>
>>>=C2=A0 =C2=A0 =C2=A0(nSubbuf * subbufSize) * (nCPUs + 1 iff sna= pshot mode is enabled) *
>>> (nUIDs or nPIDs)
>>>
>>> Note that the sub-buffer sizes you are using get rounded to th= e nearest
>>> larger power of 2. See [2].
>>>
>>> thanks,
>>> kienan
>>>
>>> [1]: https://lttng.org/docs/= v2.13/#doc-channel-buffering-schemes
>>> [2]:
>>> https://lttng.o= rg/man/1/lttng-enable-channel/v2.13/#doc-opt--subbuf-size
>>>
>> -------------- next part --------------
>> An HTML attachment was scrubbed...
>> URL: <
>> https://lists.lttng.org/pipermail/lttng-dev/attachments/20250312/57f240d8= /attachment-0001.htm
>>>
>>
>> ------------------------------
>>
>> Message: 3
>> Date: Wed, 12 Mar 2025 10:36:28 -0400
>> From: Kienan Stewart <kstewart@efficios.com>
>> To: Gour DEV <lakshyagour10@gmail.com>, lttng-dev@lists.lttng.org
>> Subject: Re: Memory Consumption High After Upgrading to 2.13 from = 2.10
>> Message-ID: <0f819583-ea8e-468e-9102-e1410d886a6f@e= fficios.com>
>> Content-Type: text/plain; charset=3DUTF-8; format=3Dflowed
>>
>> Hi Lakshya,
>>
>> On 3/12/25 5:03 AM, Gour DEV wrote:
>>> Hi, Kienan
>>>
>>> I am attaching an screen recording of the behaviour I am seein= g in this
>>> mail. The behaviour is same irrespective of the device i use, = sorry for
>>> miscommunication in the npocs output (I assumed it was 32), bu= t other
>> than
>>> that all outputs are same (except the hostname as there are mu= ltiple
>>> devices with same lttng config but this memory cosumption is s= een on all
>>> the devices).
>>>
>>> I had few question
>>>
>>> 1. Does lltng allocated all the memory it needs and mark it as= dirty in
>> ram
>>> when any process which links/uses lttng-ust runs? (here i trie= d with one
>>> process but it is same for any of my process)
>>
>> I believe the shared memory for per-CPU data structures is allocat= ed
>> when an instrumented application connects. There is no pre-allocat= ion
>> for each possible UID on the system.
>>
>> You can run your instrumented applications with `LTTNG_UST_DEBUG= =3D1` to
>> see when the connection happens[1].
>>
>>> 2. (nSubbuf * subbufSize) * (nCPUs + 1 iff snapshot mode is en= abled) *
>>> (nUIDs or nPIDs)
>>>
>>> How do we calculate uid in the system is it all uids in the sy= stem? is it
>>> equal to `cat /etc/passwd | wc -l` ?
>>
>> nUIDs is the number of distinct UIDs running instrumented applicat= ions.
>>
>>>
>>> I will put my calculations according to the above estimate bas= ed on all
>> the
>>> channel i am creating
>>>
>>> (4194304*4 + 262144*4 + 16384*4) * (16) * (30 if number user a= re equal to
>>> `cat /etc/passwd | wc -l`)B =3D 7.998046875 GB approx [this is= based on the
>>> start_lttng.py please do correct me if am wrong here.]
>>>
>>> But since there are only two users which uses lttng i think th= e correct
>>> estimate would be
>>> (4194304*4 + 262144*4 + 16384*4) * (16) * (2)B =3D 546MB
>>
>> The estimate I gave is per-channel.
>>
>> small channel: (0.015625 MiB * 4) * (16 + 1) =3D 1.0625 MiB per-ch= annel
>> per-UID
>> medium channel: (0.250 MiB * 4) * (16 + 1) =3D 17.0 MiB per-channe= l per-UID
>> large channel: (4 MiB * 4) * (16 + 1) =3D 27 2MiB per-channel per-= UID
>>
>> Now, you said you have 0 small channels, 6 medium channels, and 16= large
>> channels in your session. (Note: I see your script differs from th= ese
>> stated channel counts).
>>
>> small: 0 * 1.0625 MiB =3D 0 MiB per-UID
>> medium: 6 * 17 MiB =3D 102 MiB per-UID
>> large: 16 * 272 MiB =3D 4352 MiB per-UID
>>
>> And if you're running instrumented applications with 2 users:<= br> >>
>> small: 0 MiB * 2 =3D 0 MiB with 2 UIDs
>> medium: 102 MiB * 2 =3D 204 MiB with 2 UIDs
>> large: 4352 MiB * 2 =3D 8704 MiB with 2 UIDs
>>
>> Now this is just an estimation for the per-CPU ring buffers only, = and
>> you numbers aren't hugely off so without analyzing your specif= ic system
>> it doesn't seem to be that strange to me.
>>
>> If I take the number of channels I see in your script, it becomes:=
>>
>> small: 0 MiB with 2 UIDs
>> medium: 136 MiB with 2 UIDs
>> large: 7616 MiB with 2 UIDs
>>
>> total: 7.57 GiB with 2 UIDs
>>
>>>
>>> Please do correct me If I am wrong calculations here.
>>>
>>> Now, there are a few things here, according to my output lttng= is using
>> 11G
>>> which is much more higher than the what is configured.
>>>
>>
>> I have no idea what 'service start spyder' is doing. Maybe= it's running
>> instrumented applications with an extra user that you didn't e= xpect? I
>> can't help you with that aspect of your system.
>>
>> The above estimated 7.57 GiB with 2 UIDs would be 11.35 GiB with 3= UIDs
>> so maybe?
>>
>> I'd recommend you read your verbose sessiond log so see which<= br> >> applications are connecting and with which UIDs.
>>
>>> I am attaching the lttng status and the file which is uses to = create the
>>> lttng sessions.
>>>
>>>
>>>
>>> Thank You.
>>>
>>
>> In any case, the information you have given to date hasn't dem= onstrated
>> to me in a tangible manner that you are seeing a difference relate= d to
>> the version of LTTng being used.
>>
>> thanks,
>> kienan
>>
>> [1]: https://lttng.org/m= an/3/lttng-ust/v2.13/#doc-_environment_variables
>>
>>
>> ------------------------------
>>
>> Subject: Digest Footer
>>
>> _______________________________________________
>> lttng-dev mailing list
>> ltt= ng-dev@lists.lttng.org
>> https://lists.lttng.org/cgi-bin/m= ailman/listinfo/lttng-dev
>>
>>
>> ------------------------------
>>
>> End of lttng-dev Digest, Vol 203, Issue 7
>> *****************************************
>>
>

--000000000000c31f030630d9bd43--