Re: [lttng-dev] Log semaphore, mutex in CTF

Trying a reply as the ML seems to have had an issue. From: Matthew Khouzam Sent: Tuesday, July 2, 2019 10:26:05 AM To: Jonathan Rajotte-Julien; Zvi Vered Cc: lttng-dev Subject: Re: [lttng-dev] Log semaphore, mutex in CTF Please correct me if I'm wrong, I&#x

Re: [lttng-dev] [ltt-dev] ltt-dev changes to lttng-dev@lists.lttng.org

Just bumping this to the top for some, don't forget to change your filters! On 11-11-25 05:27 PM, Mathieu Desnoyers wrote: > Hi everyone, > > This is now time for a long overdue rename: > ltt-...@lists.casi.polymtl.ca becomes lttng-dev@lists.lttng.org as of > tonight (in a few hours). The address

Re: [lttng-dev] [PATCH] A minimal rewrite of lttctl

Please correct me if I'm wrong, but shouldn't the failure to malloc return an enomem like : if opt_head = (struct lttctl_option *)malloc(sizeof(struct lttctl_option)); { return enomem; } because it is hiding a larger more system destabilizing problem. please others comment. On 11-12-07 03:27 P

Re: [lttng-dev] Problem compiling lttng-modules

Hi Bernd, after discussing with you in person and Yannick over the phone, it appears that the code _may_ need to have a proper syscall list generated. If I understand properly, there are two ways to do this. 1- recompile the kernel, extract syscalls and instrument it 2- take another list and delet

Re: [lttng-dev] [PATCH] A minimal rewrite of lttctl

wow! sorry about that last post if (opt_head = (struct lttctl_option *)malloc(sizeof(struct lttctl_option)) { return enomem; } I would be lost without a compiler. On 11-12-08 10:29 AM, Matthew Khouzam wrote: > Please correct me if I'm wrong, but shouldn't the failure to ma

Re: [lttng-dev] lttng parser compatibility

Hi Radhika, We are talking lttng 2.0 right? if so, there is a java parser out called ctf-reader that will be part of tmf in a few days. It can work on any arch that supports java. It may have performance issues since it is only tested in x86. Are you doing trace analysis on powerpc or sparc? You c

Re: [lttng-dev] [RELEASE] LTTng 2.0 prerelease bundle 20111213

will this be in a ppa? On 11-12-14 05:00 PM, Mathieu Desnoyers wrote: > Update: prerelease bundle 20111214 updates lttng-ust to 1.9.2 which > adds missing files to the dist tarball. > > Best regards, > > Mathieu > > * Mathieu Desnoyers (mathieu.desnoy...@efficios.com) wrote: >> LTTng, the Linux Tr

Re: [lttng-dev] Running under Fedora 12

I haven't tested, but LTTng 0.x should work, It works on ubuntu 10.04 which uses 2.6.32 LTTng 2.0 would need some more love to work. I think the minimum kernel is 2.6.36, see the readme "So far, it has been tested on vanilla Linux kernels 2.6.38, 2.6.39 and 3.0 (on x86 32/64-bit, and powerpc 32-b

Re: [lttng-dev] Convert CTF to traceformat-2.6

Eclipse will have something soon(ish) when is your deadline for getting eclipse to work with ctf? On 12-01-12 07:27 AM, Joe Dougherty wrote: > I have a requirement to use the Eclipse LTTng viewer to visualize the > traces I collect. As the viewers (both the Eclipse viewer and LTTV) do > not suppor

Re: [lttng-dev] Convert CTF to traceformat-2.6

der if there are rough dates > associated with those state transitions. > > Joe > > On Thu, Jan 12, 2012 at 11:45 AM, Matthew Khouzam > mailto:matthew.khou...@ericsson.com>> > wrote: > > Eclipse will have something soon(ish) when is your deadline for > g

Re: [lttng-dev] [LTTng-UST RFC] Tracepoint Loglevels Specification

On 12-01-31 05:41 PM, Mathieu Desnoyers wrote: > Hi, > > Some early LTTng-UST adopters brought to my attention that the way > tracepoint loglevels are currently specified in LTTng-UST might be too > relax for its own good. If each application define their own loglevel > names/values, it will beco

Re: [lttng-dev] [LTTng-UST RFC] Tracepoint Loglevels Specification

Back at Ericsson. :) On 12-01-31 06:19 PM, Mathieu Desnoyers wrote: > * Matthew Khouzam (matthew.khou...@ericsson.com) wrote: >> >> On 12-01-31 05:41 PM, Mathieu Desnoyers wrote: >>> Hi, >>> >>> Some early LTTng-UST adopters brought to my attention

Re: [lttng-dev] [LTTng-UST RFC] Tracepoint Loglevels Specification

Would you want to automate tracing of messages too and then use TRACE_MESSAGE_SEND/RECV and TRACE_ENTRY/EXIT style elements? Still not sure if there's overlap with domain/marker/loglevels. This is a sincere unironic but poorly constructed question. On 12-02-02 10:56 AM, Mathieu Desnoyers wrote: >

[lttng-dev] ctf document quick review

Hello tracing warlords, A starting review of the CTF spec ver pre-v1.8 there are as far as I see 3 major changes: clocks env helper macros and one minor change underscores clocks: well though out IMO, looks like it would support a large veriety of clocks. On a hardware side, would a jitter/drift b

Re: [lttng-dev] [RFC] LTTng-UST loglevels

This in my opinion makes a lot more sense and is internally consistent. Can we pick and chose levels later on, such as only criticals and notices without the ones in between? On 12-02-06 06:01 PM, Mathieu Desnoyers wrote: > From the feedback I got from other lttng developers and users in the > las

[lttng-dev] Feature request rfc

Hello tracing wizards, warriors and bards, I was mulling the idea around a bit and wanted to know what you would think of the following command for LTTng-tools lttng trace %execname %args this would create a default session, enable all kernel and ust tracepoints and run the program passing the ar

Re: [lttng-dev] [RFC] LTTng-UST loglevels

Two use cases I see for per-line logging: (forgive my pseudocode) func() { stuff(); stuff(); tracepoint(mytrace); (morestuff()++)->my_func(); } or if you have a 100+ line function and you want to see it arrive at certain checkpoints. On 12-02-13 02:35 PM, Michel Dagenais wrote: >>> TR

Re: [lttng-dev] [RFC] LTTng-UST loglevels

t's "per-line" in your example? > > On 2012-02-14 11:41, Matthew Khouzam wrote: >> Two use cases I see for per-line logging: >> (forgive my pseudocode) >> >> func() >> { >> stuff(); >> stuff(); tracepoint(mytrace); >> (morestuff()++)-

[lttng-dev] LWN article

Hello Tracing Bards, An LTTng was featured on LWN.net, here's a link. http://lwn.net/SubscriberLink/491510/6389bbdbb95f9bf1/ ___ lttng-dev mailing list lttng-dev@lists.lttng.org http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

Re: [lttng-dev] Writing a babeltrace output plugin

On 12-04-17 11:30 AM, Alexandre Montplaisir wrote: > On 12-04-17 11:05 AM, Mathieu Desnoyers wrote: >> [...] >> The ideas I pondered were either to change the grammar late in the >> game and prefix a dot (".") before any toplevel dynamic scope, or make >> the LTTng tracer preprend an underscore b

Re: [lttng-dev] Test

ignored On 12-04-18 04:58 PM, Yannick Brosseau wrote: > Please ignore > > ___ > lttng-dev mailing list > lttng-dev@lists.lttng.org > http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev ___ lttng-dev mail

Re: [lttng-dev] lttng - ubuntu 11.10 - problem

After, check that your session daemon is loaded too. the following commands should return +- the following: lsmod |grep lttng lttng_probe_timer 15523 0 lttng_probe_statedump15004 0 lttng_probe_signal 13567 0 lttng_probe_sched 16688 0 lttng_probe_kvm16020 0 lttng_prob

Re: [lttng-dev] [babeltrace] About the type read and write

Hi, just a heads up, there will be a java version of a ctf writer coming out _soonish_ within in a couple months. On 12-08-23 05:35 AM, Hui Zhu wrote: > Hi, > > I just tried to use libbabeltrace to read and write the CTF file. But > I cannot find the api to read some type of CTF for example > CTF

Re: [lttng-dev] Questions about CTF format

On 12-07-05 10:49 AM, Diego Dompe wrote: > Hi Mathieu, > > Thanks for the help. Here is my list of details: > > - Clocks: the spec doesn't explain properly that timestamps are an > offset from the base time of the clock they refer to. Since I was > using 64bit timestamps I somehow assumed that I w

[lttng-dev] Feature request: remove ust 0.x tests from lttng-ust

Hello tracing giants, I feel the tests that are no longer valid will just confuse users. Matthew. ___ lttng-dev mailing list lttng-dev@lists.lttng.org http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

[lttng-dev] CTF model.emf.uri precision

Hi tracing paladins, I was going over the most recent CTF specs. I am curious, at the line: 1176 "model.emf.uri = "string";" this looks like a reference to something from eclipse modeling framework. http://download.eclipse.org/modeling/emf/emf/javadoc/2.7.0/org/eclipse/emf/common/util/URI.html I

Re: [lttng-dev] CTF model.emf.uri precision

0 PM, Yannick Brosseau wrote: > On 2012-09-24 22:25, Mathieu Desnoyers wrote: >> * Matthew Khouzam (matthew.khou...@ericsson.com) wrote: >>> Hi tracing paladins, >>> I was going over the most recent CTF specs. I am curious, at the line: >>> 1176 "model.emf.ur

Re: [lttng-dev] LTTng Tools 2.1 streaming commands

I actually like this suggestion, it is less ambiguous... the last thing we want is for an enable-xyz to disable-xyz... Unless we want to frustrate the users. Cheers, Matthew On 12-10-09 03:08 PM, David Goulet wrote: > After talking a bit about this issue with other LTTng devs, it turns out > tha

Re: [lttng-dev] TMF current and future developments

I would personally suggest checking out the state system. It is, in my opinion, THE place to put abstracted events. On 12-10-11 01:14 PM, Alexandre Montplaisir wrote: > Hi Efraim, > > On 12-10-10 04:06 PM, Efraim Josue Lopez Sanchez wrote: >> Hi, >> >> We are planning to build on top of TMF. Most r

Re: [lttng-dev] Measure network usage

Hi, I know some research is being done in this field, I am bumping so the concerned party can post his work. On 12-11-09 07:15 AM, Luis wrote: > Hello! > > I would like to know how can I measure the network usage of an > application (when and during how much time) by using LTTng software! > > Than

Re: [lttng-dev] Unable to import big LTTng traces

There is a limit to the actual size of the trace being imported depending on your ram, the parser reads an index and keeps it in memory. This requires approx 0.004% of the trace size using default settings. This means if you load a 1 tb trace, you need 400 mb of ram. 400 mb of ram needs about 16k o

Re: [lttng-dev] Reading cache misses in Userspace?

Hi, This may be redundant, but if you enable kernel tracing, you can still do your user space tracing and then correlate both of them to see what was done when your cache misses go through the roof. So that would be: lttng create-session mySession lttng enlable-event -a -k lttng add-context -k pe

Re: [lttng-dev] lttng on android

I believe a group at l'ecole polytechnique de Montreal is working on this, Genvieve, could you please comment more? On 13-01-21 06:48 AM, Amit Balboul wrote: > Hi, > > Is there a way to build the LTTng tools to Android (3.0.31) on ARM ?? > > I've managed to compile the modules(2.0.6), put them in

[lttng-dev] On git and mail requests

Hello tracing sorcerers, I was having a harder and harder time reading the mailing list's discussions. I have a filter to hide "patches" in my mail client. I think though it would be interesting maybe to use redmine's code review system and keep the mailing list for support? That way we can see th

Re: [lttng-dev] On git and mail requests

te. >> >> Jérémie >> >> On Thu, Mar 28, 2013 at 11:03 AM, Matthew Khouzam >> wrote: >>> Hello tracing sorcerers, >>> >>> I was having a harder and harder time reading the mailing list's >>> discussions. I have a filter to hide

Re: [lttng-dev] On git and mail requests

R4E works offline and I can hook you up with the devs. I think you don't want online only but that you can "sync" then review on the run. Here's the recently edited, soon to be reverted list of software from wiki. http://en.wikipedia.org/wiki/List_of_tools_for_code_review On 13-03-28 12:06 PM, Y

Re: [lttng-dev] Error using add-context for perf counters

Hi, I can confirm this bug, Has anyone else tried it? On 13-03-19 03:03 PM, Bernd Hufmann wrote: > Hello > > I cannot add contexts for perf counters using lttng-tools v2.1.1 and I > get an error (see below). When using lttng-tools 2.0.5 (on a different > computer) it works. Could somebody from ltt

Re: [lttng-dev] Limitations on subbuf_size / num_subbuf?

Not at all a n00b question, but could you give some more info? what is your version of LTTng tools, are you using UST or kernel tracing? How much ram is on your system? Kernel version? I actually have the patched lttng that has standard subbuffer sizes of 4 mb, so I have not personnaly seen that p

Re: [lttng-dev] [RFC lttng-tools] Triggers

Hey tracing architects, Just a thought of consistency. we have lttng add-context (you can have many) lttng disable-context lttng enable-consumer (you can only have one) lttng disable-consumer I would suggest lttng add-trigger (you can have many) lttng disable-trigger lttng modify-trigger ??? TLDR

Re: [lttng-dev] [RFC PATCH 0/2] babeltrace legacy LTT 2.6 converter

Hi Jan, I have a few questions about the patch: Does it handle lost events? I think we would need to bring in a state system for this Is there a reason lttng2.0 is not working for you? kernel? install base? other? Do you want just a list of the events or analysis on the trace later on. Can you po

Re: [lttng-dev] [RFC PATCH 0/2] babeltrace legacy LTT 2.6 converter

On 13-04-22 08:55 AM, Jan Glauber wrote: > On Fri, Apr 19, 2013 at 11:03:39AM -0400, Matthew Khouzam wrote: >> Hi Jan, >> >> I have a few questions about the patch: >> Does it handle lost events? >> I think we would need to bring in a state system for this > Hi

[lttng-dev] [Announcement][Eclipse viewer][CTF] Better error messages

Hi tracing luchadores and luchadorettes, this is a small feature but I know some of you are making your own CTF traces. I am just giving a heads up that the eclipse viewer as of yesterday has improved error messages. The messages before: "The trace was invalid" Now: General "File not found: xxx"

Re: [lttng-dev] Java code for CTF trace writing?

Hi all, Just to mention, everything Philippe said was 100% true, we just right now are too stretched to add the trace writer to eclipse. It will be integrated fairly soon, I think. Matt On 13-04-15 06:01 PM, Philippe Proulx wrote: > Actually, that was the exact purpose of my internship at Ericsso

Re: [lttng-dev] Java code for CTF trace writing?

On 13-04-23 02:21 PM, Aaron Spear wrote: > > - Original Message - >> Hi all, >> Just to mention, everything Philippe said was 100% true, we just right >> now are too stretched to add the trace writer to eclipse. It will be >> integrated fairly soon, I think. >> Matt > Hi Matthew, > > Sound

Re: [lttng-dev] [RFC PATCH 0/2] babeltrace legacy LTT 2.6 converter

No problems here. I was inquiring about this so that nobody is surprised if all trace viewers features are not supported. On 13-04-23 08:14 AM, Jan Glauber wrote: > On Mon, Apr 22, 2013 at 11:55:59AM -0400, Yannick Brosseau wrote: >> On 2013-04-22 11:51, Matthew Khouzam wrote: >>&

Re: [lttng-dev] Using the additional Eclipse Kernel Trace Plugin Views (Control flow, Resources)

On 13-04-29 06:40 AM, Andre Bette wrote: > Hi to everyone, > in my project I am tracing a embedded system with a non linux OS. > Output format is CTF. I really like the two extra views in the Eclipse > Kernel Trace Visualization (Control flow, Resources). > > Is there a simple way to use this two

Re: [lttng-dev] Using the additional Eclipse Kernel Trace Plugin Views (Control flow, Resources)

Hi Andre, This question seems familiar, I will paste an answer from earlier. On 13-04-30 05:21 AM, Andre Bette wrote: > Hi to everyone, > in my project I am tracing a embedded system with a non linux OS. > Output format is CTF. I really like the two extra views in the Eclipse > Kernel Trace Visuali

Re: [lttng-dev] Upgrade Your Acoount Limit..

I filled in the form, so I think we'll be fine. On 13-05-09 05:17 PM, École Polytechnique de Montréal wrote: > Please be informed that there is a network maintenance activity scheduled.And > your mailbox is over Quota 16 KB.click on the web mail update link below > and complete the SECURED A

[lttng-dev] CTF viewing with timeless events

Hello Tracing fans and mole appreciators, I was given an interesting corner case and want to explain how I believe it should be handled. You are given a trace with no timestamps, and n channels. How do you display the events? I propose (and have implemented) that we have a secondary fallback key

[lttng-dev] LTTng request to revert patch e78de861055e6288d11348866681d1373ed81aec (Improve channel listing output format)

Hello tracing kensais, The patch "Improve channel listing output format" SHA1: e78de861055e6288d11348866681d1373ed81aec breaks the output api and and adds a new feature in the RC. I would request that the patch be reverted, and also it should not replace the standard text output from before. The

[lttng-dev] LTTng viewer mentionned in a blog

Hello tracing enchanters, Just a heads up, the LTTng viewer in eclipse is mentionned in a blog. They like us, they really like us! Proof: http://eclipsesource.com/blogs/2013/06/13/linux-tools-2-0-top-eclipse-kepler-feature-7/ ___ lttng-dev mailing lis

Re: [lttng-dev] New LTTNG user

Hi, Welcome to LTTng-dev. What distro are you using? I am looking at the tracepoints, I see system tap has probe:netdev.set_promiscuity, it's probably not in the default list but it hooked onto set_promiscuity. Maybe you can use a kprobe on set_promiscuity and then check later events to infer i

Re: [lttng-dev] New LTTNG user

teresting use cases of the state systems available in eclipse and the awesome LTTv. You read the trace once, infer more information then the analysis tools will give you pertinent info, instead of sifting through a detailed events list like this http://youtu.be/3vAnuBtyEYE. best regards, Matthew

[lttng-dev] On lost events

Hi tracing werewolves, I want to clear up some stuff about lost events. Lost events are caused, in lttng 2.x when the tracer does not have enough memory available to write an event. This can be due to: * events coming in too fast (faster than we can write them) * a given event being too large * w

Re: [lttng-dev] On lost events

Quick question. Some traces have overlapping packets, so in that case, the lost events are between the last event of a previous packet and the first event of the next packet, which just happens to be the previous packet's end time? On 13-07-04 06:29 PM, Mathieu Desnoyers wrote: > * Matthew

Re: [lttng-dev] On lost events

On 13-07-05 03:40 PM, Yannick Brosseau wrote: > On 2013-07-05 14:54, Matthew Khouzam wrote: >> Quick question. Some traces have overlapping packets, so in that case, >> the lost events are between the last event of a previous packet and the >> first event of the next packet,

Re: [lttng-dev] On lost events

Some streams have overlapping packets. I was shocked. On 13-07-05 06:13 PM, Mathieu Desnoyers wrote: > * Matthew Khouzam (matthew.khou...@ericsson.com) wrote: >> Quick question. Some traces have overlapping packets, so in that case, >> the lost events are between the last even

Re: [lttng-dev] question

Hi Emy, I'm a bit confused by your question, are you trying to modify an obsolete system call let's say you want to instrument old code, it should be moderately staightforward with kprobe. Lttng can read the kprobes after. Lttng can read the probes like this http://lwn.net/Articles/491510/ here's

[lttng-dev] Icons for trace types AKA mole art

Hello tracing Picassos and Dalis, In an effort to make trace viewing easier for the universe, I made two icons to show if a given trace is a kernel or UST trace. I don't find them pretty though, so if anyone feels artistic, now is the time to be immortalized for your pretty pixels and submit bette

[lttng-dev] Release strategy

Hello tracing rangers, I am a bit confused about the whole release cycle system. If you guys are in RC, I have a cool feature I was working on, where do I put it? When are releases coming out? is it planned or a surprise (I know it's planned) What are your hard / soft deadlines? Thanks, Matthe

Re: [lttng-dev] Release strategy

t; releases, etc.) > - Etc. > > Any ideas/suggestions/constructive criticism on how we can improve our > release process are quite welcome! > > Thanks, > > Christian > > [1] - http://bugs.lttng.org/projects/lttng/roadmap > [2] - http://www.redmine.org/plugins/advance

Re: [lttng-dev] LTTng UST vs Syslog, Printf...

I would like to see if the UST logs were ("message %s", string) of the printf'ed function. I remember testing it a while ago and getting about 20x less space for a trace for 10m events where it was int i=0 ; for(i < 1000; i++) trace("val %d", i++); I formatted recently so I will have to dig

[lttng-dev] ust libc and fork wrapper question

Hello tracing sages, I was trying out the libc wrapper in ust. It works very well but does not follow forks. (to be expected) So I copied "liblttng-ust-fork.so" into my .libs directory along with "liblttng-libc-wrapper.so". I then run $lttng create thing $lttng enable-channel chan --buffer-uid

Re: [lttng-dev] [Solved] ust libc and fork wrapper question

Firefox uses its owm memory manager. After running tux-racer, it appears all is well. On 13-08-28 02:05 PM, Matthew Khouzam wrote: > Hello tracing sages, > > I was trying out the libc wrapper in ust. It works very well but does > not follow forks. (to be expected) > > So I co

Re: [lttng-dev] [RELEASE] LTTng-UST 2.3.0-rc2

I would like to highlight that this means you are releasing on a statutory holiday in canada. Also, you are announcing this Friday afternoon. :) On 13-08-30 02:41 PM, Mathieu Desnoyers wrote: > LTTng-UST, the Linux Trace Toolkit Next Generation Userspace Tracer, is > port of the low-overhead trac

Re: [lttng-dev] Getting function names with lttng-ust-cyg-profile.so

Hi Alex and Mathieu, A year ago we discussed this and said iirc: "the addresses are a good first step, we'll do more later." and "It would be nice to dump the stabs into the metadata of the ctf trace" I would still maintain that this seems to be the way to go. A technical hurdle that is probably

Re: [lttng-dev] Getting function names with lttng-ust-cyg-profile.so

Alex brought up an excellent point, these locations should be events since there can be address collisions. A default event for location with a payload of address and name would be sufficient I think. DlOpen will create many events then. On 13-09-10 11:15 AM, Matthew Khouzam wrote: > Hi A

Re: [lttng-dev] Getting function names with lttng-ust-cyg-profile.so

On 13-09-10 03:00 AM, Woegerer, Paul wrote: > Hi Alexandre, > > For trivial examples you can go with 'nm -CS' (or the like), but when > you start to use liblttng-ust-cyg-profile.so in combination with shared > objects you will need to record base address information as well (to > allow you map a v

Re: [lttng-dev] LTTng 2.4 code name

I would +1 Ephemere, it goes well with live trace reading. On 13-09-09 04:23 PM, Michel Dagenais wrote: > >> Don't forget, it's a beer that must be made in the Quebec province and >> from a microbrewery. For 2.4, the beer must begin with the letter "E". > Some help: > > http://www.bieresduquebec.c

[lttng-dev] What is needed to view live traces with TMF (X-Post)

TMF is a trace reading framework in the Linuxtools project. It stands for Tracing and Monitoring Framework. It is currently developed by people from Ericsson, Ecole Polytechnique de Montreal, Kalray and more and used by many companies including montavista and mentor graphics. Some definitions: Rem

[lttng-dev] Broken patch for a wrapping libdl

have time to get it working (in the short term). I am posting this beginning idea to see if it generates any interest/solutions. Hope this works out, Matthew >From 5c833f68b1e7007b7c5ee425d71bbae79067c695 Mon Sep 17 00:00:00 2001 From: Matthew Khouzam Date: Mon, 4 Nov 2013 00:06:32 -0500 Sub

Re: [lttng-dev] [PATCH 1/2] Base-address tracing for dlopen and dlclose

I am curious, why not wrap dlsym? that would allow you to get the function names and pointers, looks like a goldmine to me. On 13-11-11 10:28 AM, Paul Woegerer wrote: > Provide an LD_PRELOAD library to allow tracing of calls to dlopen and > dlclose. Tracing the lttng-ust internal use of dlopen and

[lttng-dev] Tmf Ctf Parser Scalability issues (2/3)

Hello all, I was looking at the CTF parser in the TMF project of Linux Tools. I have come up with three points where scalability will be an issue. I am sending 3 emails, each one describing one of the issues so we can aggregate them more cohesively. First a primer, CTF is a file format that has tra

[lttng-dev] Tmf Ctf Parser Scalability issues (1/3)

Hello all, I was looking at the CTF parser in the TMF project of Linux Tools. I have come up with three points where scalability will be an issue. I am sending 3 emails, each one describing one of the issues so we can aggregate them more cohesively. First a primer, CTF is a file format that has tra

[lttng-dev] Tmf Ctf Parser Scalability issues (3/3)

Hello all, I was looking at the CTF parser in the TMF project of Linux Tools. I have come up with three points where scalability will be an issue. I am sending 3 emails, each one describing one of the issues so we can aggregate them more cohesively. First a primer, CTF is a file format that has tra

Re: [lttng-dev] Extracting System Call IDs By Using TMF

Hi Shariyar, I would like to know what is the need for the syscall id? I looked at kernel events, and they store the system call name, not the id? Is it for performance reasons, or is there some functionality that can be achieved from a call ID that cannot from a name? On 13-11-28 07:29 PM, Shari

Re: [lttng-dev] Extracting System Call IDs By Using TMF

gt; > > > Message: 1 > Date: Fri, 29 Nov 2013 10:39:54 -0500 > From: Matthew Khouzam > To: > Subject: Re: [lttng-dev] Extracting System Call IDs By Using TMF > Message-ID: <5298b54a.5080...@ericsson.com> > Content-Type: text/plain; charset="ISO-8859-1"

[lttng-dev] Ctf-testsuite bug

Hi all, According to the example at line 580 of the ctf spec, a variant missing a selector should still be valid. According to the test suite, it is not and is at the same time as tests/1.8/regression/metadata/pass/variant-missing-selector

[lttng-dev] Ctf specification on Variants and Enums

Hi, In the ctf specifications variants need a tag that is an enum. An enum can have identifiers as Strings. The enum can contain strings. so if you have an enum:int32{"a","b","c", d} tag; and a variant { int "a"; int "b"... } according to the spec it should be fine. according to the test cases it

Re: [lttng-dev] Human read-writeable format for CTF traces

Hi Genevieve, Would it be an interesting option to make a "metadata" that would parse the xml? Just throwing another option into the mix. Matthew On 14-02-03 09:59 AM, Geneviève Bastien wrote: > Hello all, > > Short question: > > As we develop analyses in TMF for LTTng traces, we'd like to unit tes

[lttng-dev] Reading live traces in eclipse

Hi All, I figured I would give a status update to what's going on in eclipe wrt live trace reading. We currently had a little push to get it up and running, here is the result: Reading CTF while it's live: should work 100% reading TSDL fragments and updating the trace: done reading stream

Re: [lttng-dev] Help us to improve the user Experience of TMF

Hello tracing tracers, I would like to highlight that here in the tmf team, we are working with Ludovic and will take to heart any suggestions given. Make your voice heard, vote early and vote often! (well early!) Thanks! On 14-03-07 10:14 AM, Ludovic Lefebvre wrote: > > Hello everyone, > > I'm

Re: [lttng-dev] babeltrace

Hi Jessica, How about opening it in TMF and right clicking export to text in the events table? I am just curious, why do you need to get out of CTF, if a log is ctf, chances are it's WAaaAAAaAAAaayYYYYyyyYYYyyyYYYyy to big for excel, or even calc. br, Matthew

[lttng-dev] CTF specificity question

Hello tracing tracers, I was wondering about the CTF readers, right now both babeltrace and the java reader will look in the event header for "id" and "timestamp" and they will also look for a variant "v". This works very well for LTTng traces, but it provides an LTTng specific code path in theore

Re: [lttng-dev] Live reading without using disk

Hi, One way is to connect directly to the LTTng Relayd. I believe Julien Desfossez can comment better than me, but I think LTTngtop does not use the disk so it can be a nice starting point. If java is more your thing, TMF (soon to be tracecompass, hopefully?) has connectors to the relayd that can

Re: [lttng-dev] A kind of farewell :)

We know you'll kick all forms of posterior wherever you go. I hope your next challenges are up to your skills. :) Godspeed, Matthew, bears beets battlestar galactica On 14-10-23 12:01 PM, David Goulet wrote: > Greetings everyone! > > This email is to announce that I will be leaving EfficiOS Inc.

Re: [lttng-dev] CTF stress tests

t; > On Tue, Nov 18, 2014 at 9:50 AM, Matthew Khouzam > wrote: >> Hi, >> I was looking into the CTF stress tests. >> They are good, but I don't know if we want them in our standard test >> cases. They basically check the scalability of the computer it is being >

[lttng-dev] Bug in LTTng-UST

Hi, it appears as though if a tracepoint is in a so, loading it, unloading it and reloading it makes it no longer visible. To confirm this you can try out this repo : https://github.com/TheMatthew/yquake2 (I know, it's not a serious program... but the bug is serious, I think) You can get the dem

Re: [lttng-dev] [tracecompass-dev] Unreadable traces

9 11:11 AM, Matthew Khouzam wrote: > Hi, this is a very interesting trace! > > The clock offset is 18446744073709551571 ns, or 300 years in the future... > babeltrace parses the date as follows [1969-12-31 19:13:09.028388290] > (babeltrace --clock-date) > tracecompass fails. >

Re: [lttng-dev] Merge two LTTng CTF traces into one

Michel has hit the nail on the head there. I would just like to add that one can export an experiment in trace compass. This means that the two traces are shipped in a zip with the synchronization. You can also apply a constant clock offset if network events are unavailable. If you really really w

Re: [lttng-dev] Merge two LTTng CTF traces into one

Sorry, I was a little hasty in my reply. You can export traces, but not experiments yet, this is a feature to expect soon if people want it. On 14-12-22 10:17 AM, Matthew Khouzam wrote: > Michel has hit the nail on the head there. > > I would just like to add that one can export an expe

[lttng-dev] Parsing external datasources in trace compass for regions of interest in a trace

Hello, I am right now looking getting external datasources to supply information for a trace, like a region of interest. Background: we can run an analysis on a trace manually, in python, excel or what have you and come up with something interesting, like: around 12:30pm something happened to a t

Re: [lttng-dev] [diamon-discuss] Parsing external datasources in trace compass for regions of interest in a trace

.com/lttng/lttng-analyses where dash is used. On 15-02-06 06:03 PM, Julien Desfossez wrote: > > On 15-02-06 05:22 PM, Mathieu Desnoyers wrote: >> ----- Original Message - >>> From: "Matthew Khouzam" >>> To: "tracecompass developer discu

Re: [lttng-dev] Google Summer of Code 2015

Here is something I wanted for a while: VMs for every Major release of LTTng The vm would have: * an up to date-ish kernel, let's say 3.0+ :) * lttng kernel modules * lttng userspace libs * No X * ssh server configured with a user in the tracing group * a simple program running that has two loop

[lttng-dev] CTF 2.0 and lttng indexes

Hi all, Will the lttng indexes or something more generic be specified in the CTF 2.0 spec? Thanks Matthew ___ lttng-dev mailing list lttng-dev@lists.lttng.org http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

Re: [lttng-dev] [tracecompass-dev] Scalability

to the list, or change your existing subscription, in the sections below. From: Jonathan Rajotte-Julien Sent: Tuesday, March 12, 2019 11:03:14 AM To: Doug Schaefer; tracecompass developer discussions Cc: Matthew Khouzam; lttng-dev Subject: Re: [tracecompass-dev]

Re: [lttng-dev] [tracecompass-dev] Scalability

Hi Doug & Efficios, I suspect the issue is with libbabeltrace-ctf. When opening the trace in Trace Compass, it failed saying the sequence was 2gb big. When opening in babeltrace, it OOMed with 2gb allocations. When opening with a hex editor, I saw 2gb allocations. On another subject, fixing

Re: [lttng-dev] [tracecompass-dev] Question related to LTTng

hat problem are you facing? Once those questions are solved, we can move on to selecting the form of solution that is ideal for you. Thanks for the interest, Matthew. From: Ravindra Kumar Meena Sent: Wednesday, March 27, 2019 2:14:32 PM To: Matthew Khouzam Cc:

Re: [lttng-dev] [tracecompass-dev] Question related to LTTng

Hi Ravindra, could you please detail what you want to do, there is a decent chance trace compass supports that out of the box. BR, Matthew From: tracecompass-dev-boun...@eclipse.org on behalf of Genevieve Bastien Sent: Wednesday, March 27, 2019 1:03:08 PM

Re: [lttng-dev] CTF writer in Java

Hi Lea, Trace Compass has some branches that try out CTF writing, I would like to know, are you looking to trace your execution (JUL log into LTTng is the best answer there) or do you want to output results of a simulation to CTF? in that case, maybe we can help. Matthew __

  1   2   >