Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Kernel Tracing With LTTng On Ubuntu Maverick

kdawson posted more than 3 years ago | from the truth-of-kernel dept.

Programming 88

francis-giraldeau writes "Linux Tracing Toolkit (LTTng) provides high-performance kernel tracing for Linux. This is the killer app for system level debugging and performance tuning. It's now easier than ever to install, with packages released for Ubuntu Maverick. The short introduction to kernel tracing shows how to interpret a simple kernel trace and relate it to strace. I would like to ask Slashdot readers what they would expect as features for a kernel tracing analysis tool, because I'm starting my PhD on this topic and looking for ideas. Also, I wonder why LTTng is not mainline yet. Will Linus Torvalds see the light in 2011?"

Sorry! There are no comments related to the filter you selected.

BUSTED !! (-1, Troll)

Anonymous Coward | more than 3 years ago | (#34215574)

damn ya buyz dunno how to keep this shit runnin?

Crowd-sourcing a degree... (2, Funny)

PsiCTO (442262) | more than 3 years ago | (#34215596)

Seriously? You're asking for a crowd-sourced original contribution to (I assume) Computer Science for your PhD? Are you going to defend "our" dissertation live on /. ?

;-)

Re:Crowd-sourcing a degree... (1, Interesting)

Anonymous Coward | more than 3 years ago | (#34215610)

I look forward to being the opponent.

Re:Crowd-sourcing a degree... (3, Insightful)

maxwell demon (590494) | more than 3 years ago | (#34215654)

Well, I guess his PhD would not be about imagining those features, but about implementing them. He asked for ideas what to implement, not for ideas how to implement it.

Re:Crowd-sourcing a degree... (1)

PsiCTO (442262) | more than 3 years ago | (#34224132)

Ummm, a PhD is about original contribution. Implementing a "feature" that someone else imagines is about algorithm implementation. An original contribution is about recognizing a hard problem and at least coming up with a novel solution. Ideally, the candidate also provides a thorough description of the problem in the language of her/his discipline. For the most part, if someone can describe a "feature" they'd like to see for LTtng, then they know the problem they want to solve. Defining the problem is often most of the work; once you know what you want to do, the rest is easy. Of course, there are problems that are known that don't have solutions, or at least not optimal solutions.

I'd say the poster is better off (and hopefully already doing this) working to become an expert (aka PhD) on the topic, say, by joining the ACM (maybe IEEE, but since they share databases in this area maybe not), and entering the terms "linux unix trace" into the ACM's digital library. I did that and got all kinds of interesting papers and potential topics, none of which are mentioned in this thread.

If soliciting thoughts from /. was enough to get one started on a PhD, then perhaps we ought have a /. category of Ph/.D. ;-)

Good luck!

Re:Crowd-sourcing a degree... (5, Insightful)

francis-giraldeau (1939602) | more than 3 years ago | (#34215678)

The reason is that I would like to make my research useful for tracing users, and I think the best way to do it is to ask people what they really need. I will give credits to those how helped my, why not? ;-)

Re:Crowd-sourcing a degree... (-1, Troll)

Anonymous Coward | more than 3 years ago | (#34216038)

Have you considered cunnilingus? Maybe a little of that on Linus's vulvae. If he doesn't see the light, then it might be time to pull out the Chewbacca dildos.

Re:Crowd-sourcing a degree... (-1, Flamebait)

Anonymous Coward | more than 3 years ago | (#34216136)

look, linus IS a cunt, but he doesn't HAVE a cunt. It gets confusing because he has an asshole and he is an asshole.

Maybe this will help:

class linux : public asshole, public cunt {
private:
asshole *m_asshole;
};

Re:Crowd-sourcing a degree... (1)

mevets (322601) | more than 3 years ago | (#34218188)

What people really need is dtrace. its the gold standard in this arena. Experiment with each of them for a day. Lttng is ok, but like the apostles, a bit thick and ordinary.

Get Ubuntu Zesty Zebra (-1, Troll)

Anonymous Coward | more than 3 years ago | (#34215602)

AKA Windows, with 90% market share for a reason, unlike Lolnux which has less market share than iOS.

Troll Truths, Flamebait Facts Offtopic Outings, Redundant Realities.

It's OK, Billy... (0)

Anonymous Coward | more than 3 years ago | (#34215704)

...now be a good boy and have your meds. Otherwise I'll have to call the bad big doctor.

(sorry: sometimes I can't help but feeding the trolls :)

Goal of the PhD work? (3, Insightful)

Anonymous Coward | more than 3 years ago | (#34215650)

What is the goal of your work? Do you want to compare kernel tracing solutions and identify critical features in the process of coming up with a reasonable taxonomy? Do you want to implement something? Do you have a specific application for kernel tracing (e.g. informing performance tuning measures in enterprise environments which would probably be of interest to businesses)? Just throwing together a list of desired features is not going to be of interest to anyone, I guess. You have to come up with a motivation for each of the features, argue why this feature is necessary for the application at hand or for any application of kernel tracing in general, cite literature that gives evidence for your assumptions and conclusions. Maybe if you told the people what kind of work you're interested in and what the interest of your advisor(s) is, in which reasearch context (department, university) you are working, they could make sensible suggestions as to which features might be interesting to you.

Re:Goal of the PhD work? (4, Informative)

francis-giraldeau (1939602) | more than 3 years ago | (#34215794)

Kernel tracing instrumentation is ready, now we need decent analysis tools. The problem is that there is so much data, that it's hard to interpret them. For the project, I have to come up with something that is new and better that what is already known. For example, we could get a better analysis than bootchart, or auto detect bottlenecks in a system (disk, CPU, memory, network, etc...). There are some work done to integrate userspace and kernel space tracing, virtual machine and host traces, dynamic and static trace points. For a distro, they could record a trace in background and send this information allong with the core dump when an application crash occur. That's all ideas!

Re:Goal of the PhD work? (2, Interesting)

hAckz0r (989977) | more than 3 years ago | (#34215874)

Here is another idea for you. How about hardware assisted "dynamic" (aka dynamically hooked) tracepoints via a custom Xen-like bare metal hypervisor? The OS and therefore its contained malware would know nothing of the inspection process, and best of all it could be OS independent if done at the hardware level. The control/diagnostics software could be running in a VM right next to the OS under test. Boot the hypervisor from CD and then load the original machines OS. Stealth rootkits would be a thing of the past. Simply boot the monitor before loading the OS under test and have a blast uncovering all kinds of malware in any OS of your choice.

Re:Goal of the PhD work? (1)

francis-giraldeau (1939602) | more than 3 years ago | (#34215904)

You're right, we can analyse abnormal situation with tracing. For example, if you have a trace of a system with correct behavior and one with a malware, it could be possible to do a "trace diff" and see what's different. As you may expect, this is not trivial diff!

Re:Goal of the PhD work? (2, Informative)

Lally Singh (3427) | more than 3 years ago | (#34217190)

Hello, I couldn't find another way to contact you, so here we are.

I'm finishing up a PhD in scalability & performance analysis, and have done a lot of work in instrumentation. A userland instrumentation tool is part of my final research. Instrumentation is in a terrible, terrible state -- save a few points of light -- and I'm happy to see someone else in this area!!

So, as you're starting out, some tips:

1) If you haven't already done so, investigate dtrace. While available on Mac OS & FreeBSD, it's worth picking up a virtual machine image of opensolaris & playing with it there.

2) Pick up a copy of: R. Jain, "The Art of Computer Systems Performance Analysis: Techniques for Experimental Design, Measurement, Simulation, and Modeling," Wiley- Interscience, New York, NY, April 1991, ISBN:0471503361. It's my new Bible.

Good luck, and hit me up if you'd like to chat. For my email address, I'm [my first name].[my last name]@gmail.com

Cheers,
-ls

Re:Goal of the PhD work? (1)

francis-giraldeau (1939602) | more than 3 years ago | (#34218546)

Great, thanks for this tip!

Re:Goal of the PhD work? (0)

Anonymous Coward | more than 3 years ago | (#34221956)

For desktop distros this would be invaluable IMO. Linux desktop suffers frequent slow downs and issues with UI responsiveness, which is next to impossible to reproduce. Info from the trace in the background could help frustrated users to file bugs, which would have a chance to be addressed by distro maintainers which are equally frustrated by rumors of their distro being sluggish which nobody can reproduce on a request.

ps. Lately kernel devs acknowledged these issues and are trying to fix them in latest releases and future bugs filled using these traces could be used to give them a feedback whether they were successful or not.

Re:Goal of the PhD work? (1)

wealthychef (584778) | more than 3 years ago | (#34215816)

People, just answer his effing question. He is just looking for ideas about what would be useful, not ideas about how to write his PhD thesis.

Yes (0, Funny)

Anonymous Coward | more than 3 years ago | (#34215674)

Yes, Linus will see the light in 2011. It will probably be in the form of light from a CFL, but it could be the light from sunrise. Heck, unlikely as it may be, it could even be when Melinda Gordon tells him to "go into the light". Anyway, he'll definitely see the light one way or another.

Re:Yes (1)

jimmydevice (699057) | more than 3 years ago | (#34215778)

I vigorously protest your insinuation that people living in Portland Or. are nerds, crunchy hippies and Indian engineers.

Re:Yes (1)

Joce640k (829181) | more than 3 years ago | (#34215812)

Linus doesn't see lights, he IS the light.

Re:Yes (1)

maxwell demon (590494) | more than 3 years ago | (#34217304)

Linus doesn't see lights, he IS the light.

I see. On the first day, God said: Let there be Linus ...

Re:Yes (1)

Shikaku (1129753) | more than 3 years ago | (#34220236)

On the first day, God said: make; make install

That it is safe to use in a production environment (5, Insightful)

Dug (9395) | more than 3 years ago | (#34215686)

Not that much point having a tracing tool if an inexperienced admin cannot safely use it on a live system which has a problem.

A problem already solved with DTrace on Solaris http://docs.sun.com/app/docs/doc/817-6223 [dpp]

Re:That it is safe to use in a production environm (2, Informative)

Dug (9395) | more than 3 years ago | (#34215730)

Another go this time with a working link

http://docs.sun.com/app/docs/doc/817-6223 [sun.com]

Re:That it is safe to use in a production environm (1)

h4rr4r (612664) | more than 3 years ago | (#34215754)

Not much point in having a tracing tool if it is owned by Oracle :)

I really liked solaris, I have an Ultra 5 as upgraded as it can be running an old version of solaris 10. Oracle has since decided I don't even deserve to be able to download bios updates, nevermind the OS. Stuff like that makes me want to avoid anything that says SUN on it.

Re:That it is safe to use in a production environm (1)

trasz (758585) | more than 3 years ago | (#34229346)

DTrace is Open Source, Free Software (FSF certified), thus, the fact that it's owned by Oracle doesn't really matter much. You don't need Solaris to use it; DTrace is fully functional in MacOS X and FreeBSD (in the latter, userland dtracing is available from 8.2).

Re:That it is safe to use in a production environm (2, Interesting)

francis-giraldeau (1939602) | more than 3 years ago | (#34215846)

With DTrace, you have to know what you are looking for in advance, while LTTng can trace in background in flight recording mode and record everything that is going on. Then, afterward you can have all the information you need, and this is invaluable when you have a hard to reproduce bug!

Re:That it is safe to use in a production environm (1)

trasz (758585) | more than 3 years ago | (#34229354)

You can do the same with DTrace. Read about DTrace feature called "speculation".

Re:That it is safe to use in a production environm (1)

Zero__Kelvin (151819) | more than 3 years ago | (#34215898)

"Not that much point having a tracing tool if an inexperienced admin cannot safely use it on a live system which has a problem. "

Right. Because everyone knows the best place to develop, debug, and profile code is on a production machine, and the person doing the development should be a system administrator, preferably with minimal experience.

Re:That it is safe to use in a production environm (5, Interesting)

Dug (9395) | more than 3 years ago | (#34216050)

"Not that much point having a tracing tool if an inexperienced admin cannot safely use it on a live system which has a problem. "

Right. Because everyone knows the best place to develop, debug, and profile code is on a production machine, and the person doing the development should be a system administrator, preferably with minimal experience.

I would say many people do know that the best place to understand the performance of a system in production is in production. If the vendors support techs can give an admin commands to run and know that a typo here or there will not result in a panic then that is a very useful feature.

Re:That it is safe to use in a production environm (1)

Zero__Kelvin (151819) | more than 3 years ago | (#34216604)

You are confusing profiling with tracing. [linux-mag.com]

Re:That it is safe to use in a production environm (1)

Junta (36770) | more than 3 years ago | (#34216902)

The profiling/tracing issue is moot. Either way the ability to have robust tracing/profiling/debugging tools that can start from the beginning or attach to a process in progress and safely report as much as possible is great for production environments.

The quickest way to characterize any problem is with low level trace information. Trying to think through all the possible differences between a test and production environment *usually* can produce results eventually, but stack traces, syscalls, and more show the failure so precisely that a reproduction procedure is often trivial.

Re:That it is safe to use in a production environm (1)

Dug (9395) | more than 3 years ago | (#34220678)

No confusion here. DTrace is useful for both profiling and tracing. More details here [lwn.net]

Re:That it is safe to use in a production environm (1)

Zero__Kelvin (151819) | more than 3 years ago | (#34222550)

"No confusion here. DTrace is useful for both profiling and tracing. More details here"

As is LTT-ng, so I guess your "point" is pointless. Furthermore, more details can easily be found on the lttng website, including a comparison to DTrace, and it doesn't do Linux, making it completely useless as a competitor to LTT-ng. Also, it was your ridiculous claim that such a tool is of no value unless it can be used by an inexperienced system administrator that I was rebutting.

Re:That it is safe to use in a production environm (1)

trasz (758585) | more than 3 years ago | (#34229362)

You're missing the fact that DTrace is safe to use, so it's impossible to crash the system just by tracing it.

Re:That it is safe to use in a production environm (1)

Zero__Kelvin (151819) | more than 3 years ago | (#34229434)

I certainly agree that you will never crash a Linux system using DTrace.

Re:That it is safe to use in a production environm (0)

Anonymous Coward | more than 3 years ago | (#34216610)

"Not that much point having a tracing tool if an inexperienced admin cannot safely use it on a live system which has a problem. "

Right. Because everyone knows the best place to develop, debug, and profile code is on a production machine, and the person doing the development should be a system administrator, preferably with minimal experience.

Shit happens; problems arise. I've had many cases where production was slow, and the problem could not be re-created in QA, DEV, or STG. Or environments where there is no "non-PROD" environment, like my Perforce servers. How many places have a STG or "DEV" server for their central source code system? What if commits were working fine, but all of a sudden they're slow and "nothing was changed".

DTrace is also useful for gathering general statistics on a system with minimal impact. It's accessible via Perl in Solaris as well, so you could write wrappers and such. You could have a version of top(1) which, when run, isn't the top process.

Often the non-PROD environments aren't as large as the PROD ones, so while the code works, it isn't to the scale that the developer thought it would be once it really gets hit hard.

In general, why would you purposely have less visibility on your system/s (regardless of the environment) rather than more?

Re:That it is safe to use in a production environm (1)

Zero__Kelvin (151819) | more than 3 years ago | (#34216696)

OK. Now just tell me that you wouldn't chase down the problem, but you would have an inexperienced system administrator do it. Then tell me you would never use the tool in any other way, and I'll concede that the OP's point that the tool was useless was spot on.

Re:That it is safe to use in a production environm (1)

trasz (758585) | more than 3 years ago | (#34229376)

The best place to investigate a problem that manifests itself on a production machine and cannot be easily reproduced on a development environment may be that machine - especially when doing it is safe. With DTrace, it is. With e.g. SystemTap - it's not.

Re:That it is safe to use in a production environm (1)

Zero__Kelvin (151819) | more than 3 years ago | (#34229430)

Excellent point in a thread about LTT

Ubuntu Only? (3, Insightful)

hduff (570443) | more than 3 years ago | (#34215700)

Why does the OP mention the Ubuntu package when the project releases a tarball?

There is no need to make news distro-centric when it does not need to be. The submitter should check to see what other binary packages are available or not mention them at all.

Re:Ubuntu Only? (0, Redundant)

h4rr4r (612664) | more than 3 years ago | (#34215726)

Why limit it to binaries?

Re:Ubuntu Only? (2, Insightful)

francis-giraldeau (1939602) | more than 3 years ago | (#34215854)

Binary packages are easy to install, that's it. I don't know of other LTTng integration inside a distro. If you prefer patching your own kernel and compiling tools from git repository, you're free to do it.

Re:Ubuntu Only? (3, Informative)

compudj (127499) | more than 3 years ago | (#34216030)

So far, LTTng has been mainly integrated in embedded distros: WindRiver Linux, Montavista Linux and STLinux currently ship with LTTng. The interesting news that is particular about Ubuntu here is that, by installing the LTTng packages from PPA, it is now possible to easily deploy the LTTng kernel and userspace tracers on a desktop-oriented distribution.

See the light? (4, Insightful)

codegen (103601) | more than 3 years ago | (#34215732)

Maybe I'm reading slashdot too early on a weekend morning, but I find the last statement of the summary particularly offensive. It seems like everyone who has some sort of kernel widget wants a PR campaign to get it included in the mainline. How about you finish your Ph. D. first and provide some convincing evidence as to why every single person running Linux has to have the tool? The trace tools are available as a package for anyone who wants them now. Why should the mainline be burdened with maintaining the package unless a significant number of users need it?

Re:See the light? (2, Insightful)

francis-giraldeau (1939602) | more than 3 years ago | (#34215836)

We are waiting for decent kernel tracing since a decade, while LTTng is readily available today. It's better than any other tools like perf, ftrace and dtrace. Microsoft Windows has the Event Tracing for Windows since 2003, and if Linux wants to be taken seriously, it has to be mainline and available without kernel patching. And, I think that users should not be experts to use that kind of tools.

Re:See the light? (3, Insightful)

Anonymous Coward | more than 3 years ago | (#34216036)

"if Linux wants to be taken seriously..."

Funny the one thing needed to be taken seriously is, by magic, the subject of your thesis.

Had you been working on, say, resizable ramdisks (I'm just making this up), then resizable ramdisks would have been the one thing needed in Linux for Linux to be taken seriously.

Ever considered humility?

Re:See the light? (1)

francis-giraldeau (1939602) | more than 3 years ago | (#34218570)

You're right, Linux is already considered seriously. It just can be even better. Happy to know that you too are contributing to this.

Re:See the light? (2, Insightful)

h4rr4r (612664) | more than 3 years ago | (#34216084)

Linux is taken quite seriously and to use this sort of tool and not be an expert is pointless.

Re:See the light? (1)

codepunk (167897) | more than 3 years ago | (#34216202)

Yea because Linux will never be taken seriously, give me a break. If I want to trace the performance or a particular chunk of code within linux I don't need a tool to do it I have the source code and the ability to modify it. If my boss comes to me ant tells me hey I need this to run faster than competitor xyz you had better believe I am going to make that happen with or without LLT. Sure it may make it easier to do so but if shooting for performance it would also be the first thing I disable.

Re:See the light? (2, Interesting)

Anonymous Coward | more than 3 years ago | (#34216484)

Actually ... While I was the maintainer, IBM's had a team of people working on LTT for a period of 3 years before pulling the plug on their involvement because they saw that all the money they were pouring in there wasn't leading to a mainlining.

Why were they interested in kernel tracing? Well ... When a customer of theirs has one of his 10,000 servers misbehaving in production, they can't afford telling him to just take it offline for diagnostics. They have to find (and fix) the problem in the field. There are very few tools that allow you to do that. Oh, and having the source code and being able to rebuild is just not an option in those cases. After I passed on maintainership, Google did some work on kernel tracing with the LTTng developers with goals very similar to IBM's: misbehaving machines in server farms should not need to taken offline for diagnostics.

As for shooting the performance, I suggest you read up on LTTng's literature. The current team has done a stupendous job at deconstructing that myth.

You're right, Linux is being taken very seriously. Hence the need for these kinds of tools.

Thanks,

Karim Yaghmour

Re:See the light? (1)

codepunk (167897) | more than 3 years ago | (#34216238)

Oh and another thing what the hell good is event tracing in windows anyhow? It may give me visibility to a issue but I have no way to solve it. Not like I can crack open the windows source and fix it.

Re:See the light? (1)

francis-giraldeau (1939602) | more than 3 years ago | (#34218592)

Event Tracing for Windows is great, and they have done a good job on this.

Re:See the light? (0)

Anonymous Coward | more than 3 years ago | (#34216308)

I see you are really ignorant on what's in the linux kernel. There is already `perf' and `ftrace'. Want' a third one ?

Re:See the light? (0)

Anonymous Coward | more than 3 years ago | (#34216526)

"if Linux wants to be taken seriously"

Plus "Will Linus see the light?" Interesting attitude. Not fond of it.

Re:See the light? (1, Insightful)

Anonymous Coward | more than 3 years ago | (#34216830)

and if Linux wants to be taken seriously

Linux isn't taken seriously at all! It only has close to 50% of the server market-share and a near monopoly on supercomputers. Look, when you have something workable we might talk, but until then you're just another PHD that has produced absolutely nothing of value.

academic hyperbole (1)

luis_a_espinal (1810296) | more than 3 years ago | (#34217650)

We are waiting for decent kernel tracing since a decade, while LTTng is readily available today. It's better than any other tools like perf, ftrace and dtrace. Microsoft Windows has the Event Tracing for Windows since 2003, and if Linux wants to be taken seriously, it has to be mainline and available without kernel patching. And, I think that users should not be experts to use that kind of tools.

You might be Ph.D student, but apparently you are disconnected from industrial reality. Linux not being taken seriously? Are you f* kidding me? Is that going to be part of your problem statement at the start of your dissertation?

Seriously. You take the cake in the academic hyperbole department.

Re:academic hyperbole (1)

francis-giraldeau (1939602) | more than 3 years ago | (#34218600)

You are right. Linux is taken seriously. And it just can be better with LTT.

Re:academic hyperbole (0)

Anonymous Coward | more than 3 years ago | (#34220152)

Pissy Pissy. Keep in mind that this is someone that is learning and developing as an individual. In a sense, they are trying to prove themselves to their world on their problem.

Re:See the light? (1, Interesting)

Anonymous Coward | more than 3 years ago | (#34216264)

Sorry, I'm not going to bother registering - I read /. quite steadily but don't usually ever feel the need to add more than what's already said. You can google me around, though, I'm easy to find.

FWIW, I introduced LTT in 1999 and lobbied kernel developers for inclusion for 6 years before giving maintainership to someone else. LTTng is in fact a complete rewrite of LTT and I've got little do with the project these days. I had little to do with its authoring and it likely has none of my code.

I do take issue, however, with your posturing. The fact that LTT was there before DTrace and still today Linux lacks equivalent functionality speaks volumes about some of the lesser known aspects of the kernel development process: namely that disruptive changes are insanely hard to mainline. It's one thing to ask for proof of the need. It's another to ignore the proof that's already out there and the project's history.

Francis doesn't need to finish his PhD to prove usefulness. Simply because, combined, there have already been half a dozen PhDs and Masters degrees done on Linux kernel tracing already.

So please, do *YOUR* research before asking others to do theirs.

Cheers,

Karim Yaghmour

Re:See the light? (1)

m50d (797211) | more than 3 years ago | (#34216534)

Because the mainline refuses, at length, to provide a stable API for the package to target. The kernel documentation basically says "if you think you want a stable API, you actually want to get your package into the mainline kernel (and if your licensing terms won't let you do this, your problem)." To which my response is: fine, but in that case it's your responsibility to accept any and all reasonably coded modules, even if only a tiny proportion of users will want to enable them.

see the light? (1)

fuliginous (1059354) | more than 3 years ago | (#34215768)

"see the light" - you are making the assumption of it being something that casts light; I suspect Linus Torvalds judges on presented evidence and so far apparently judges the argument hasn't been carried.

Re:see the light? (1)

francis-giraldeau (1939602) | more than 3 years ago | (#34215866)

Really, he doesn't care. I whish he would care about this core feature.

Re:see the light? (1)

h4rr4r (612664) | more than 3 years ago | (#34216092)

Or just maybe he disagrees with you. Maybe you can write your own OS, then you can make these decisions.

Re:see the light? (1)

volkerdi (9854) | more than 3 years ago | (#34216114)

Maybe he cares about kernel bloat.
Maybe he cares that a feature that few people would use might have side effects for all.

LD_AUDIT anyone?

Re:see the light? (0)

Anonymous Coward | more than 3 years ago | (#34217436)

Good point. Made me think "kernel" hmmm what might a kernel be. The tiny minimal essentials of certain services? So what place is there in the kernel for something like LTT?

Some googling (4, Insightful)

diegocg (1680514) | more than 3 years ago | (#34215824)

Also, I wonder why LTTng is not mainline yet

Well, a bit of searching [lkml.org] would have answered your question [lttng.org]

The LTTng maintainer has been working for months (years?) to get the kernel tracing into a decent shape. These days the Linux tracing support is wonderful, and not just for LTT - perf, ftrace and systemtap are awesome tools (and more powerful than LTTng in some ways). In fact perf can do all what the web page says and it seems to be more simple for my taste

Re:Some googling (1)

francis-giraldeau (1939602) | more than 3 years ago | (#34215878)

Trace points in the kernel are available, and this is great, but there are many more than that. You need a good ring buffer lock less to not impact performances and all the infrastructure for this. For example, you can't do flight recording with perf and it's impact performance is greater due to less sophisticated ring buffers.

Re:Some googling (2, Informative)

diegocg (1680514) | more than 3 years ago | (#34215928)

Well, ftrace has a lockless ring buffer. And eventually all the ring buffers are going to be unified [lwn.net] ...

Re:Some googling (1)

francis-giraldeau (1939602) | more than 3 years ago | (#34218610)

It's getting there, we can be happy of that!

test (-1)

Anonymous Coward | more than 3 years ago | (#34215840)

This is only a test. Please do not waste mod points sending this to oblivion. Thanks.

trace or analysis tool? (2, Insightful)

hraponssi (1939850) | more than 3 years ago | (#34215886)

are you talking about a trace or a data analysis tool? if you plan to use LTT to get a trace and then help the user analyse it, maybe you are more into analysis than tracing. then your question could be a bit misleading. Anyway, you would probably end up trying it all out, adding some features to make it all easier to trace as you try to use the existing stuff and analyse the results and so on as you progress. And if you are into trace data analysis (as opposed to tracing) then your domain of kernel trace data analysis is just one application of data analysis. there you need to look into data analysis methods, statistical methods, machine learning, etc. depending on what kind of analysis you like and need. it is somewhat different depending on your goals such as performance data, behaviour trace and analysis, etc.. for some more behaviour related stuff you can look into domains of program comprehension, behaviour analysis and modeling in general, software reverse engineering, specification mining, etc. anyway, at least i would be interested to see some results on this kind of stuff if you go with it and have some means to follow on it and provide feedback..mainstream or not most of this stuff never ends up anywhere or is available at all.

Re:trace or analysis tool? (1)

francis-giraldeau (1939602) | more than 3 years ago | (#34215932)

Yeah, data mining techniques may be relevant since the huge trace size that we can get. Trace reduction techniques, algorithms to index data. One of the thing that is particular to trace analysis is the temporal nature of events, that may lead to something...

Because (1)

cratermoon (765155) | more than 3 years ago | (#34216104)

As I recall, Linus has pretty strong opinions [linuxmafia.com] on why it's a Good Thing that kernel development isn't "so easy a caveman could do it".

Re:Because (1)

LingNoi (1066278) | more than 3 years ago | (#34216270)

That post is over ten years old.

Re:Because (1)

cratermoon (765155) | more than 3 years ago | (#34216290)

So am I. Your point, exactly?

Re:Because (1)

LingNoi (1066278) | more than 3 years ago | (#34216438)

It's irrelevant unless you can show that Linus still holds such an opinion today.

Re:Because (1)

cratermoon (765155) | more than 3 years ago | (#34216918)

There's nothing to indicate he's changed his views. Linus has otherwise remained largely silent on the issue since this comment in 2000. Normally that would indicate no change, so the onus is really on you to show any evidence that would indicate any new opinion. Yes, various debugging facilities have made it into the mainline kernel. Show me the evidence that Linus now likes debuggers, or more to the original point, that he now thinks kernel development should be easy.

Re:Because (1)

LingNoi (1066278) | more than 3 years ago | (#34220070)

What a stupid statement. Of course the onus isn't on me. Do I also need to come up with evidence to show Bill Gates doesn't think we still need more then 640KB of ram?

Newsflash for you, people change their opinions all the time.

Re:Because (2, Interesting)

larry bagina (561269) | more than 3 years ago | (#34217768)

He has expressed similar sentiments more recently as well (eg from 2007 on git's use of c vs c++)

C++ is a horrible language. It's made more horrible by the fact that a lot of substandard programmers use it, to the point where it's much much easier to generate total and utter crap with it. Quite frankly, even if the choice of C were to do *nothing* but keep the C++ programmers out, that in itself would be a huge reason to use C.

Some googling (-1, Offtopic)

emitsu (1939908) | more than 3 years ago | (#34216674)

wow...Linux tracing support is wonderful my blog: http://emitsu.ru/ [emitsu.ru]

Great! Maybe Linux can regain usefulness in media (1)

drwho (4190) | more than 3 years ago | (#34219592)

Yea, maybe it's just un Ubuntu problem, but on both my laptop (Inspiron 700m) and my desktop (with an Nvidia card) crash about 10% on the desktop and 40% on the laptop, when viewing videos and occasionally when listening to music. It's pretty sad, as I need this stuff for my classwork. I pulled an old celeron XP box out of the closet, tranplanted some ram into it, and everything works fine. Pretty sad.

A scripting/domain-specific language for LTTng ? (1)

sandGorgons (1528485) | more than 3 years ago | (#34219914)

One of the biggest selling points for DTrace is its scripting language. It is extremely powerful and you can find dtrace scripts shared by others that allow you to do very powerful system stats gathering (e.g. here [goo.gl] ) How about doing something similar for LTTng - you could even do something simple like Lua hooks for LTTng

Re:A scripting/domain-specific language for LTTng (0)

Anonymous Coward | more than 3 years ago | (#34223358)

See systemtap.

Zero conclusions (0)

Anonymous Coward | more than 3 years ago | (#34221254)

Slashdot: Put news in, take stupidity out. Zero actually useful comments.

Where does this leave mine?

The light? (1)

DaVince21 (1342819) | more than 3 years ago | (#34233878)

Will Linus Torvalds see the light in 2011?

Oh come on, Linus isn't inside coding all day, every day, you know.

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?