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!

Non-Deathmatch: Preempt v. Low-Latency Patch

timothy posted more than 12 years ago | from the floor-wax-and-dessert-topping dept.

Linux 178

LiquidPC writes: "In this whitepaper on Linux Scheduler Latency, Clark Williams of Red Hat compares the performance of two popular ways to improve kernel Linux preemption latency -- the preemption patch pioneered by MontaVista and the low-latency patch pioneered by Ingo Molnar -- and discovers that the best approach might be a combination of both."

cancel ×

178 comments

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

FP! (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205540)

first post bitches!

Ximian (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205544)

Interesting. Should be related to the article about GNOME earlier.

co-op! (2, Funny)

FigBugDeux (257259) | more than 12 years ago | (#3205546)

whats wrong with cooperative multitasking?

Re:co-op! (0)

Anonymous Coward | more than 12 years ago | (#3205561)

You can't trust other software to function correctly.

Re:co-op! (1)

clone304 (522767) | more than 12 years ago | (#3205598)


Uhh, that was either a joke or a troll. Either way you got caught!

.

Re:co-op! (0)

Anonymous Coward | more than 12 years ago | (#3205614)

People really do have low expectations of others these days... *sigh*

Re:co-op! (0)

Anonymous Coward | more than 12 years ago | (#3205685)

hey - millions of macintosh users can't be wrong!

Re:co-op! (0)

Anonymous Coward | more than 12 years ago | (#3205997)

hey - millions of macintosh users can't be straight!

Re:co-op! (0)

Anonymous Coward | more than 12 years ago | (#3206089)

Hey.. Mac OS ain't Co-Op anymore..
it was up until Mac OS 8.. but 9 and X are preemptive.

Re:co-op! (0)

Anonymous Coward | more than 12 years ago | (#3205690)

The same thing that is wrong if Co-op food.
Its full of lentils.

Re:co-op! (2, Insightful)

lightfoot jim (441918) | more than 12 years ago | (#3205753)

I hope you're kidding! In a co-op system, each program monopolizes all the resources of the machine until it voluntarily gives up control. If the instruction to return control to other processes is preceded by instructions which could result in an infinite loop or wait state, the machine is then inaccessible. Situations like this exist in preemptive systems as minor annoyances. For example, if you ssh from and openBSD machine to a sshd server and then pull the network cable from the sshd server, the tty on the openBSD machine will stop responding to input. Since the BSD kernel is preemptive, you can switch tty's and just kill the process of the frozen tty. In a coop setup, the key sequence to switch virtual consoles would be ignored and you'd have to reboot the machine.

Re:co-op! (-1)

Genghis Troll (158585) | more than 12 years ago | (#3205787)

It'd be a step up from linux, anyway!

I preempted the current Linux kernel (-1, Flamebait)

Anonymous Coward | more than 12 years ago | (#3205556)

on my home system.

It runs better now that I've installed the NT kernel.

Re:I preempted the current Linux kernel (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205581)

It is now official - HighTImes has confirmed: *BSD is for stoners.

Yet another crippling bombshell hit the beleaguered *BSD community when recently HighTImes confirmed that *BSD accounts for less than a fraction of 100 percent of all servers. Coming on the heels of the latest MadMagazine survey which plainly states that *BSD has lost more market share, this news serves to reinforce what we've known all along. *BSD is collapsing in complete disarray, as further exemplified by failing dead last [microsoft.com] in the recent Sys Admin comprehensive networking test.

You needn't be Wavy Gravy to predict *BSD's future. The hand writing is on the wall: *BSD faces a bleak future. In fact there won't be any future at all for *BSD because *BSD is dying. Things are looking very bad for *BSD. As many of us are already aware, *BSD continues to lose market share. Red ink flows like a river of blood. FreeBSD is the most endangered of them all, having lost 3% of its core developers.

Let's keep to the facts and look at the numbers.

OpenBSD leader Theo states that there are 7000 users of OpenBSD. How many users of NetBSD are there? Let's see. The number of OpenBSD versus NetBSD posts on Usenet is roughly in ratio of 5 to 1. Therefore there are about 7000/5 = 1400 NetBSD users. BSD/OS posts on Usenet are about half of the volume of NetBSD posts. Therefore there are about 700 users of BSD/OS. A recent article put FreeBSD at about 80 percent of the *BSD market. Therefore there are (7000+1400+700)*4 = 36400 FreeBSD users. This is consistent with the number of FreeBSD Usenet posts.

Due to the troubles of Walnut Creek, abysmal sales and so on, FreeBSD went out of business and was taken over by BSDI who sell another troubled OS. Now BSDI is also dead, its corpse turned over to yet another charnel house.

All major surveys show that *BSD has steadily declined in market share. *BSD is very sick and its long term survival prospects are very dim. If *BSD is to survive at all it will be among OS hobbyist dabblers. *BSD continues to decay. Nothing short of a miracle could save it at this point in time. For all practical purposes, *BSD is for stoners.

Fact: Slashdot is an abomination...

Re:I preempted the current Linux kernel (0)

clone304 (522767) | more than 12 years ago | (#3205616)


Lol!! Mod parent up for exhibiting L337 trolling skills!

.

I did better (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205681)

I got rid of toy OS's and machines like windows and intel and linux and got a SUN.

How yo like that, beyach

Re:I did better (0)

Anonymous Coward | more than 12 years ago | (#3205693)

Stupid, we all get the Sun. Those who don't leave the house notwithstanding.

Re:I did better (0, Flamebait)

clone304 (522767) | more than 12 years ago | (#3205710)


You're an idiot.

Re:I did better (3, Funny)

(outer-limits) (309835) | more than 12 years ago | (#3205804)

The only real OS is MVS, aks OS/390 aka Z/OS. It is truly an operating system. It makes no concessions for mere mortals, it is made to run a machine. It hails from 1966. All bow down before it.

The Linux kernel preemption project (5, Informative)

BrianGa (536442) | more than 12 years ago | (#3205568)

Check out this [gatech.edu]
comprehensive guide to Linux Latency.

Re:The Linux kernel preemption project (-1, Offtopic)

Fagtastic (566399) | more than 12 years ago | (#3205588)

karma whore!

Re:The Linux kernel preemption project (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205599)

negative karma whore!

Re:The Linux kernel preemption project (1, Interesting)

Anonymous Coward | more than 12 years ago | (#3206126)

Check out this story [monolinux.com] . It was posted several hours before Slashdot got around to it.

Pretty shameful for /.

The Right Reasons (0)

Daveman692 (558544) | more than 12 years ago | (#3205577)

This is why I love open source software. Linux will push ahead as Windows is left in the dust.

Re:The Right Reasons (-1, Flamebait)

Anonymous Coward | more than 12 years ago | (#3205583)

This is why I love open source software. Linux will push ahead as Windows is left in the dust.

  • Yeah, because we've really seen that thus far....

Re:The Right Reasons (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205632)

I want to say I'm completely disgusted by all the wankers who take slashdot seriously. Did you know that neither cmdrTaco nor anyone else on the slashdot moderation crew knows calculus? That fact alone should deter anyone from slashdot. Linux is a pile of shit anyway. Would you even consider paying for linux if it wasn't free? Of course not! If you think linux is so special, I have swamp land in Florida for sale. Email Me:-)

Linux_is_for_losers@hotmail.com

Re:The Right Reasons (0)

Anonymous Coward | more than 12 years ago | (#3205662)

yeah , but its a better piece of shit than windows.
if we wanted real OS's on real machines we would be using SUN or an IBM OS/390

Re:The Right Reasons (-1)

ArchieBunker (132337) | more than 12 years ago | (#3205707)

95% of desktop users would beg to differ with you... Hell even 70% of the fucktards here admit to using IE for browsing. Do those Z series mainframes have a graphics option? I don't want to be stuck in 3270 land forever.

Re:The Right Reasons (0, Offtopic)

clone304 (522767) | more than 12 years ago | (#3205728)


Fsck IE! I'm using Lynx count me among the 5% and 30% respectively.
Of course, I do feel kindof naked without all the pretty pictures.. Wait, what are we arguing about again? Ohh, you said something irrelevant and stupid, so I laughed at you. Then I posted this stupid message. Laugh at me.

.

Re:The Right Reasons (0, Flamebait)

clone304 (522767) | more than 12 years ago | (#3205742)


Ohh yeah.. I forgot to tell you. Fsck IE. Any fucktard that uses slow ass IE when Opera is available is a fucking moron.

.
And it doesn't surprise me at all that 95% of desktop users are morons. Hell, you probably use IE, and you chose Archie Bunker as a nick. You MUST be a moron. Thanks for speaking up for your species. Now please sit down and let the IQ-enabled people talk.

Re:The Right Reasons (0, Offtopic)

clone304 (522767) | more than 12 years ago | (#3205668)


Swamp land in Florida!!?!? Like the kind I could build a really successful theme park on? Cool!!

Wait.. Must stop feeding trolls... ahhh!! Can't help myself.. ARGGGHHH!!! AIYEEEE!!

.

Re:The Right Reasons (0)

Anonymous Coward | more than 12 years ago | (#3205798)

Yes, thats why Windows is used on most desktops .

Thats why Windows has a db behind its fs, and alows sql statments to be proformed on it.

While linux has a handfull of users and a default fs that much its like looking at FAT.

Trolling fun... Go on, Mod down! MOD DOWN!

Re:The Right Reasons (2)

WNight (23683) | more than 12 years ago | (#3205875)

If you want a database for a file system they've been out for decades.

However, it's debatable if this is a good thing. Providing transactions is good, but why write an SQL interface for the whole file system? And then have to massage files into specific data types, or a whole bunch of BLOBs which you can't perform any complex operations on...

If MS does come out with this filesystem, they won't make it a default. It'd rock for some uses and suck for others. Much like NTFS.

Re:The Right Reasons (0)

Anonymous Coward | more than 12 years ago | (#3206205)

Thats why Windows has a db behind its fs, and alows sql statments to be proformed on it.

That's a really stupid idea. The OS design community abandoned that idea 20 years ago. As usual, Microsoft is decades behind the state of the art. But, hey, what can you expect from a company whose technological expertise is based mostly on college dropouts and interns?

What Linux does have is a number of powerful, efficient, and innovative file systems.

Re:The Right Reasons (0)

markyd (517099) | more than 12 years ago | (#3206259)

Thats why Windows NT has had a pre-emptable kernel since version 3.1? Linux is finally catching up with Windows NT in these areas, we've had a journaling filesystem (NTFS) for years as well.

preempted (-1, Troll)

Fagtastic (566399) | more than 12 years ago | (#3205578)

kinda like that time my cock got preempted by a huge load of shit in linus's ass

Wow!, what an insightful conclusion. (0)

Ignominious Cow Herd (540061) | more than 12 years ago | (#3205590)

Hmmm, let's see here. The Low Latency patch makes the slowest parts of the kernel faster or breaks them into smaller pieces. The Preempt Patch allows the kernel to be interrupted in lots of places. Exactly how could combining these NOT be a good idea?

Re:Wow!, what an insightful conclusion. (0, Offtopic)

clone304 (522767) | more than 12 years ago | (#3205643)


Yeah.. When I read this on OSNews, I couldn't help thinking this was really old news.
Now to see it on /. after OSNews, I'm starting to think there's a stupidity virus travelling amongst the IT groupie crowd..

.

Re:Wow!, what an insightful conclusion. (2, Informative)

cs668 (89484) | more than 12 years ago | (#3205683)

I agree with you except there is a critical difference between assuming that this would be the case and demonstrating that the assumptions are true.

Clark Williams did a lot of work to prove that the assumptions you would have when looking at combining the two patches hold.

Re:Wow!, what an insightful conclusion. (0, Offtopic)

1155 (538047) | more than 12 years ago | (#3205688)

IT groupies?

Do they like.. travel around with you, worshiping the ground you walk on, and fetch your beer while you hack the kernel?

Re:Wow!, what an insightful conclusion. (0, Flamebait)

clone304 (522767) | more than 12 years ago | (#3205749)


Know they're usually 13 yr old trolls who end up breaking their necks trying to suck their own dicks.. But, you can believe what you want..

Re:Wow!, what an insightful conclusion. (0)

Anonymous Coward | more than 12 years ago | (#3205766)

If you're going to try to comment on one's stupidity, at least try to spell properly!

Re:Wow!, what an insightful conclusion. (0, Troll)

clone304 (522767) | more than 12 years ago | (#3205769)


Know shit!! Wat teh fack waz I thincking? You got me.

.

No way can you reach it.. (0)

Anonymous Coward | more than 12 years ago | (#3205925)

>"13 yr old trolls who end up breaking their >necks trying to suck their own dicks"

"Do people actually try that??"
"Yup, some do"
"I tried once, I couldnt reach it."
"Sick bastard."

http://Slashwife.com (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205622)

SLASHWIFE
News for negroes. Stuff for darkies.

WATERMELON UNDER THE GPL

POPEYES SELLING CUSTOMER DATA

BREW YOUR OWN 40s OR STEAL THEM

PORCH SITTING PIONEER DEAD AT 54

BOOK REVIEW: READ? FUCK DAT!

ASK SLASHWIFE: GOOD CRACK FOR CHEAP?

YOUR RIGHTS ON WELFARE: BUYING SMOKES WITH FOODSTAMPS

DNA TESTING NARROWS FATHER TO 1 IN 150

Mirror (-1)

Alan_Thicke (553655) | more than 12 years ago | (#3205629)

Here is a mirror. [yimg.com]

Re:Mirror (0)

Anonymous Coward | more than 12 years ago | (#3205795)

And some wonder why you post at -1. Oh well, at least it's an improvement on the "Alan Thicke is dead" posting.

Linux For The Masses (-1, Troll)

Haxx (314221) | more than 12 years ago | (#3205638)



All this attention to Linux problems and security flaws is good. It means we are spreading. Welcome to the mainstream.

-We are putting probes into Mars orbit and analyzing light from galaxies that are 15.5 billion light years away yet we are at war with people that wipe thier ass with thier hands and don't allow woman to show thier faces in public.

Sad day ... Stephen King dead at 54 (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205641)

I just heard some sad news on talk radio - Horror/Sci Fi writer Stephen King was found dead in his Maine home this morning. There weren't any more details. I'm sure everyone in the Slashdot community will miss him - even if you didn't enjoy his work, there's no denying his contributions to popular culture. Truly an American icon.

Re:Sad day ... Stephen King dead at 54 (-1, Troll)

clone304 (522767) | more than 12 years ago | (#3205692)


Woohoo!! Finally! I'm so glad that asshole is finally dead. Slashdot Unite!! We will celebrate with a joint DDOSing of his publishers webservers!! Death to the King!!

I don't understand (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205645)

OK, I am kinda scared to ask this.... What is Linux? I don't get it. I heard about Linux and how it is elite and all that, but I don't get it.

There's some sort of stuff called "Mandrake Linux" and "Red Hat Linux" and some others, but there are also "kernals" and "unstable releases" and things I don't understand. Can someone explain this to me?

I tried to get Linux for my Windows 200, Service Pack 2 machine but it didn't work. I went www.redhat.com (because I hear red hat a lot) and downloaded something called "Red Hat Linux 7.2". There were other things called "Samba" and "RPM" and some other stuff on their download page, but I didn't know what that was.

So I downloaded something from their webpage, but I couldn't install it. I tried opening it with winzip, but that didn't work. I also tried ending it with ".exe" since it didn't have a dot-something, but htat didn't work either.

I don't understand, every time I hear people talk about Linux, they say how fun it is. This isn't fun at all - it makes me feel dumb for not understanding.

Re:I don't understand (-1, Flamebait)

Anonymous Coward | more than 12 years ago | (#3205675)

HAHA, you are stupid n00b. You don't know shit about Linux or computers, you dumb fuck. You think people here will help you? Bullshit, look you already got a (-1). I'm posting anonymously because I will get one too and I don't want to hurt my karma, but you goddamn you are dumb and you will never get Linux. Hahahahahhaha, you suck.

Re:I don't understand (-1, Troll)

Anonymous Coward | more than 12 years ago | (#3205746)

Yeah man!!!! Linux rulez and ppl who uze windoze r dumb! LOLOLOL!!!!!

MOD PARENT UP (-1, Offtopic)

Anonymous Coward | more than 12 years ago | (#3205712)

You know, the hypocracy of this site astounds me. You slam Microsoft for not treating their users right, and then you mod this poor SOB down. This is a prime chance to introduce someone to the joy of Linux, and he gets flamed and modded down.

Next week, you will be bitching about how Linux doesn't have enough popular support. Start here, damn it!! Make a difference starting with this one person!

Re:I don't understand (0)

Anonymous Coward | more than 12 years ago | (#3206343)

I see you've been studying the Trolling FAQ and you seem to have put many of the key principals into effect in your post. It's a little too obvious though.

Never mind, keep practicing.

More serious player in the shadows (3, Interesting)

CathedralRulz (566696) | more than 12 years ago | (#3205646)

The article doesn't mention this, but something some folks aren't aware of is that MontaVista is a serious linux partner with IBM. If the technologies described in the white paper can be merged, then the real effect can have a more significant impact in the embedded application/ PowerPC products from the World's 9th largest corporation.

Re:More serious player in the shadows-IBM (0)

Anonymous Coward | more than 12 years ago | (#3205966)

So in an indirect fashion, we have IBM to thank.

Pioneers will always... (0)

Anonymous Coward | more than 12 years ago | (#3205677)

ask...is it or is it not, a good read?

# [slashdot.org]

Oooooh (4, Funny)

NiftyNews (537829) | more than 12 years ago | (#3205678)

Don't miss the thrilling link to the debate on whether it is PreemptAble or PreemptIble... [linuxdevices.com]

Re:Oooooh (1)

clone304 (522767) | more than 12 years ago | (#3205762)


Nah. I'll just wait for Taco to post the definitive answer once the debate is over.. ;)

.

SMP Scheduling Latency... (0, Offtopic)

MonkeyBot (545313) | more than 12 years ago | (#3205680)

...definitely sucks ass in Linux. I don't know if it is better or worse in other OSes, but we have several dual processor machines and two quad Xeon machines at work running Linux, and it seems as though if the first processor is running at max capacity, then you're just kind of shit outta luck if you try to start a process--it sometimes takes as 5-10 seconds before it'll even notice you've tried to run something and assign it to another processor that isn't doing anything.
Does anyone else have this problem? I'm hoping it's just some kind of dumbass configuration mistake on my part and not just the way it is...

Re:SMP Scheduling Latency... (0)

Anonymous Coward | more than 12 years ago | (#3205739)

I believe that is being adressed in with the O(1) scheduler. Hopefully it will be in 2.6, does anyone know for sure?

O(1) already integrated... (2, Informative)

Anonymous Coward | more than 12 years ago | (#3205750)


Ingo Molnar's O(1) scheduler was integrated
into the development tree back around Linux 2.5.4
So it's already in there.

Preemption was integrated about the same time.

Re:SMP Scheduling Latency... (1)

CelestialWizard (13685) | more than 12 years ago | (#3205773)

It has been in the -ac (and subsequently in -mjc) tree for quite a while.

2.5 merged it around 2.5.4 (IIRC)

Works very well here (1)

robinjo (15698) | more than 12 years ago | (#3205949)

I've found SMP on Linux to work very well. There are definitely no delays on my dual cpu workstation and dual cpu server. I've even done some extensive bvenchmarks which show that Linux scales very well. At least compared to NT4/W2K where the results were dreadful.

I'm sorry that I can't offer any help but I hope you find where the problem is.

Non-deathmatch, eh? (0, Troll)

zapfie (560589) | more than 12 years ago | (#3205689)

Non-deathmatch, eh? I think I smell a Preempt vs. Low-Latency Quake 3 mod in the works..

Re:Non-deathmatch, eh? (1)

sloanster (213766) | more than 12 years ago | (#3206054)

Sheesh, I've been running low latency and/or preemptive patches for over a year. It's the only way to quake - that, and a good 3d video card.
(used to run voodoo3, now I run nvidia geforce)

Re:Non-deathmatch, eh? (0)

Anonymous Coward | more than 12 years ago | (#3206166)

How is this a troll? What is this trolling?

One good way to reduce kernel latency.. (5, Interesting)

thesupraman (179040) | more than 12 years ago | (#3205760)


One thing not mentioned so far is that one of THE largest scheduler latency problems comes from the driver for a PS/2 mouse, a very common item to be found plugged into servers which have no need for it. By removing the PS/2 mouse (and driver..) a significant latency improvement can be gained!

It's a pity that most USB mice don't seem to provide quite the quality of use as the PS/2 items (although this is probably also a driver issue)

Loy latency can be an advantage, but it is important that the cost of the lower latency is not an increase in total load, as in reality the lower latency does not provide a large gain in performance for most desktop or server roles, but rather is a measure more often used in real time systems, which it can make the difference between a system working or not.

An example of this is in an ignition ECU for a V12 engine at 6000 RPM, a (pair of) plug is firing every 1/600th of a sceond (1.66ms), but the accuracy of the firing even must be in the order of 10us, which is not yet reachable be any 'standard' unix kernel, but quite easy to get on a much simpler ECU (I use an SH-2 at 24 MHz) than you would notmally find using a true real-time kernel.

With some developments is may be possibly for a form of linux to reach this level, which would be fantastic, as a LOT of time is spent in embedded development providing 'operating system' level functionality around the actual application code, and with embedded processors getting faster, and memory getting cheap, embedding *nix has become much more of a possibility.

Re:One good way to reduce kernel latency.. (4, Funny)

clone304 (522767) | more than 12 years ago | (#3205799)


Honestly, I'm not trying to troll you here, but why would you WANT to run a *nix kernel on hardware that's responsible for engine timing? Especially when you apparently already have tech that works. Is the idea just to make vehicles that much harder for people to maintain? The day my mechanic keeps a sysadmin on duty so he patch my buggy Linux 4.5.3 ECU is the day I put a gun to my head and pull the trigger.

Aside from that. Way to inform the masses.

Re:One good way to reduce kernel latency.. (2, Funny)

Jah-Wren Ryel (80510) | more than 12 years ago | (#3206203)

Forget about mechanics. Just let us telnet into the engine and we can fix it ourselves.

Re:One good way to reduce kernel latency.. (0)

Anonymous Coward | more than 12 years ago | (#3205905)

It's a pity that most USB mice don't seem to provide quite the quality of use as the PS/2 items

I use the Microsoft Wheelmouse Optical. You can find it for about $20, and it works flawlessly under Linux. (gpm has no trouble sorting out which buttons are which, and the scroll wheel works.) Use You can use it as a PS/2 mouse with an included adapter, but I always use it as a USB mouse.

Get a KVM switch that will switch USB, and plug your mouse and keyboard into it. Then remove PS/2 support from your servers' Linux kernels, and you are good to go.

ps. Here is my /etc/gpm.conf file. This is from Debian.

device=/dev/usbmouse
responsiveness=
repeat_ty pe=raw
type=imps2
append=""
sample_rate=

Re:One good way to reduce kernel latency.. (0)

Anonymous Coward | more than 12 years ago | (#3205976)

It's a shame that FreeBSD implemented both of these latency-reducing setups, as well a 1-order scheduler AND a working VM systems years ago, huh? What a toy...

Re:One good way to reduce kernel latency.. (0)

Anonymous Coward | more than 12 years ago | (#3206058)

What are you talking about?

FreeBSD has terrible worst case latency, and is fully non-preemptable. Perhaps once SMPng is done, they can leverage that work to provide preemptability.

FreeBSD also certainly does not include a multiqueue O(1) scheduler; there is no reason to, since only one process can be in the kernel at a time, there is no reason to have per CPU run queues.

You idiot "advocates" who don't know what you're talking about, but are sure that everything that isn't your preferred system is crap do more harm to your preferred system than any competitor possibly could.

Time to ease up. (0)

Anonymous Coward | more than 12 years ago | (#3206286)

Look carefully at the history of the postings. I use to think that a lot of these bad postings were overzealous users, but I no longer think so. Slashdot use to be really interesting, but it is now controlled by trolls. But, I am really starting to believe that many of these trolls are paid ppl whose jobs are meant to destroy OSS esp. Linux. But other than that, thanks for the info. I get tired of all the FUD about who has what.
Personally, I only run and code on linux these days so I no longer am aware of who has what.

Re:One good way to reduce kernel latency.. (0)

Anonymous Coward | more than 12 years ago | (#3206410)

Its hard to put such a stringent lower bound on interrupt servicing with TLB's, many layers of caching, asynchronous DRAM etc etc. Hard to construct model's and proof's for, unlike with simple controllers where you can pretty much run through all the possible scenarios in your head.

You wouldnt want to use a PC for something like that, wether linux was pre-emptable or not.

Interrupt priorities (4, Funny)

ChaosDiscord (4913) | more than 12 years ago | (#3205761)

Most RTOS's prioritize device interrupts so that important interrupts ("shut down the reactor NOW!") are serviced first and lower priority interrupts ("time to make the doughnuts") are serviced later.


Clearly, most RTOS designers have their priorities backwards.



Mmmm, donuts.

liquidpc (0, Flamebait)

blendin (468979) | more than 12 years ago | (#3205807)

you are a fuckin cheat...you will never defeat me

What's up with the degrading performance? (2, Informative)

A Commentor (459578) | more than 12 years ago | (#3205811)

The low-latency patches had a maximum recorded latency of 1.3 milliseconds, while the preemption patches had a maximum latency of 45.2ms.

A 2.4.17 kernel patched with a combination of both preemption and low-latency patches yielded a maximum scheduler latency of 1.2 milliseconds, a slight improvement over the low-latency kernel. However, running the low-latency patched kernel for greater than twelve hours showed that there are still problem cases lurking, with a maximum latency value of 215.2ms recorded. Running the combined patch kernel for more than twelve hours showed a maximum latency value of 1.5ms.

So after only 12, the low-latency patch degraded by an ungodly amount (1.3 -> 215.2 ms)!! and even the combined patch had a 25% degraded performance(1.2 -> 1.5 ms)!

Embedded systems must have a very high uptime, it's not acceptable to reboot the machine every day to maintain performance. Many embedded systems require a downtime of less than 5 minutes per year. That doesn't give you much time to reboot the machine just for performance issues.

Re:What's up with the degrading performance? (3, Insightful)

dmiller (581) | more than 12 years ago | (#3205843)

It has got nothing to do with how often the machine is rebooted and everything to do with the frequency of latency increasing events.

The event which caused the 215ms even probably only happens once or twice per day. Perhaps it was some weird code path that the LL patch didn't touch, or some unlikely combination of events occuring "simultaneously"

Re:What's up with the degrading performance? (4, Interesting)

tempest303 (259600) | more than 12 years ago | (#3205849)

What's up with the degrading performance?

It's called a bug - they'll figure it out. ;)

it's not acceptable to reboot the machine every day to maintain performance

Hey, it worked for NT4 admins, why not for embedded developers? *rimshot*

Sorry. But seriously, anyone looking for hardcore low latency in Linux right now for systems that need that buzzword compliant "five 9's" should probably wait on using Linux until it's READY. Make no mistake, with as much interest and developer hours as are going into it, Linux WILL make it into this market, and it will succeed; it's merely a matter of time. (and hell, at this rate, it may not be long...)

Re:What's up with the degrading performance? (4, Informative)

Fluffy the Cat (29157) | more than 12 years ago | (#3205857)

So after only 12, the low-latency patch degraded by an ungodly amount (1.3 -> 215.2 ms)!!

You're misinterpreting the figures. After a short benchmarking, the worst figure recorded was 1.3ms. After the machine had been left up for 12 hours (thereby allowing there to be much more time for something odd to crop up), the worst figure recorded was 215.2ms. That doesn't mean that the performance had degraded - it means that over the course of those 12 hours, something happened that caused latency to peak at 215.2ms. It might be something that happens once every 12 hours, for instance.

Re:What's up with the degrading performance? (1)

rabidcow (209019) | more than 12 years ago | (#3205881)

It might even be something that happens just as often on with the combination patch as with the low-latency patch, except the combo got lucky.

Re:What's up with the degrading performance? (4, Informative)

SurfsUp (11523) | more than 12 years ago | (#3206113)

It might even be something that happens just as often on with the combination patch as with the low-latency patch, except the combo got lucky.

If you'd actually read the article you'd know that this can't happen with the preempt patch + low-latency, not unless a spinlock gets jammed, then you have much worse problems. The preempt patch takes care of scheduling events that occur during normal kernel execution (and it does this much more reliably than the low latency patch) but since preemption isn't allowed while spinlocks are held, it can't do anything about latency due to spinlocks. This explains the apparently worse performance of the preempt patch - you're just seeing the spinlock latency there.

The low latency patch breaks up the spinlocks with explicit scheduling points, which is pretty much the only approach possible without really major kernel surgery. That's why the combination works so well. In fact, the parts of the low latency patch that aren't connected with breaking up spinlocks aren't doing anything useful and should be omitted. The worst-case performance won't change.

Re:What's up with the degrading performance? (5, Insightful)

steveha (103154) | more than 12 years ago | (#3205878)

Well, to be precise, the worst-case value "degraded". And I'm not sure "degraded" is the correct word. With a huge torture load put on the kernel, during a 15-hour interval, at least once the latency value was 215.2 msec. This could mean that there is a possible latency condition that happens under torture load approximately once every 15 hours. It could also mean that after 15 hours, your chance goes up so it could happen much more often than that, but we don't know that. It could even mean that there is a possible 215 msec latency condition that happens under torture load approximately once every 30 hours, and it happened to occur during the first 15 hours.

Embedded systems must have a very high uptime, it's not acceptable to reboot the machine every day to maintain performance.

True that. Which is why the author of that article made the point that combining the two patches is the best way to go, since he ran the torture test for 15 hours and it didn't go over 1.5 msec even once.

Note that for many purposes, a worst-case latency of 1.5 msec is ample. I don't think there is any version of Windows that goes that low; I don't even think BeOS (legendary for low latency) goes that low. As the author noted, if you are driving a chemical processing factory or something like that, you need hard realtime and you should use something other than Linux kernel 2.4.x!

steveha

Measuring latencies? (3, Interesting)

acordes (69618) | more than 12 years ago | (#3205894)

Here's a question. How do you go about doing fine grained measurements of these latencies? Every time I've tried doing timings with Linux I've had problems being able to get accurate, fine grained results.

Re:Measuring latencies? (-1, Flamebait)

Anonymous Coward | more than 12 years ago | (#3205936)

Here's a question. How do you go about doing fine grained measurements of these latencies? Every time I've tried doing timings with Linux I've had problems being able to get accurate, fine grained results.

That's an excellent question. I'm glad you asked. The way you measure low latencies is to stick a thermometer up your ass and measure the amount of time (the latency) it takes for you to jump.

It's all for not as long as HZ=100 on x86 (0)

Anonymous Coward | more than 12 years ago | (#3205903)

We live in an age of multi GHz machines, yet we only have a timer that can go off 100 times per second on IA32. When can we join the real world (Alpha, Sparc) and get a 1024 HZ timer -- or even better -- get rid of HZ altogether. Realtime MIDI on Linux sucks unless it is the only process monopolising 100% of the CPU.

Nice Try Montavista (0)

Anonymous Coward | more than 12 years ago | (#3205920)

Unfortunately, the preemptable kernel was not pioneered by Montavista. The real developers were by the Real-Time Linux [rtlinux.com] project. I believe there are several long-standing law-suits between the two. All of which the RTL people have won.

You can make a difference in the Linux kernel! (0)

Anonymous Coward | more than 12 years ago | (#3205928)

You can start by submitting patches to the Linux Kernel Mailing List of code indented to 5 spaces, as is the standard in south america. The kernel maintainers always appreciate well indented code, and Linus usually integrates such patches almost immediately. But I must warn you - patches involving fewer than 50 files are rarely accepted by Linus. He wants to see a lot of effort on your part and weed out those "code indenter wannabees". Fortune 500 companies are always on the look out for industrious, self-starting code (re)indenters and are willing to pay big $$$. Get indenting today!

Custom Redhat RPM's (2)

nuintari (47926) | more than 12 years ago | (#3205999)

Anyone know if Redhat is planning on offering lower latency kernel RPM's for those of us who are loath to patch and recompile a kernel JUST to try something new out to see if we like it. Its kind of nice if I can drop in a quick RPM, decide weather I like it, and THEN compile a trimmed kernel properly if need be.

I'm just lazy. :-)

la... (2, Funny)

nickynicky9doors (550370) | more than 12 years ago | (#3206057)

...tency

A good paper, but... (2)

software_non_olet (567318) | more than 12 years ago | (#3206142)

I'm missing on Clark Williams' paper how the patches influenced the OS overhead.

NGPT in 2.4.19-pre3. (1)

tyrr (306852) | more than 12 years ago | (#3206180)

On the related note Next Generation POSIX Threads (NGPT) [ibm.com] made it into the official kernel (2.4.19-pre3). Kudos to the team.


So many very good things happen to Linux kernel! I am impressed.

QNX vs. Linux (1, Informative)

Anonymous Coward | more than 12 years ago | (#3206208)

I would like to see similar response graphs for QNX or other RTOS's for comparisons sake.

Anyway IMHO to make a real assesment for any 'hard' realtime tasks is much too much effort for most of the readers here. =)

But here are more white papers than you can shake a stick at....

http://www.ece.umd.edu/serts/bib/index.shtml

Re:QNX vs. Linux (2, Interesting)

Anonymous Coward | more than 12 years ago | (#3206262)

"Interrupt and process latency

All times given below are in microseconds (sec).

Processor_______Context____Interrupt Latency
Pentium/133_____1.95_______4.3
Pentium/1 00_____2.6________4.4
486DX4__________6.75_______ 7
386/33__________22.6_______15

With nested interrupts, these interrupt latencies represent the worst-case latency for the highest priority interrupt. Interrupt priority is user-definable and the interrupt latency for lower-priority interrupt sources is defined by the user's application-specific interrupt handlers. "

Interrupt latency in Windows CE 3.0 (2, Informative)

jawahar (541989) | more than 12 years ago | (#3206269)

Well, Windows CE 3.0 provides 50 ms latency response time running on a 166 MHz Pentium.

There was a fight? (2)

Anders (395) | more than 12 years ago | (#3206316)

I never realized there was competition between the two. I did hear the low-latency crowd claim that it was lower risk due to its less invasive nature. However, that hardly says anything about the performance of either approach - or that they should be mutually exclusive.

Two wrongs doesn't make a right, and vice-versa (but two Wrights make an airplane).

Low-latency and Preemptible on my Notebook (3, Funny)

hazzzard (530181) | more than 12 years ago | (#3206336)

I am using a low-latency kernel on my notebook at the moment and I can report the following behavior:
  • In X (KDE), I can move windows around, load programs, webpages etc. without my MP3-player ever beginning to skip.
  • When doing massive file IO, the MP3-player begins to skip. tar cvzf file.tar.gz bla/ is still ok, but cp -R bla1 bla2 causes massive skipping.
  • When I use the notebook as a samba server, things get worse. Still, massive skpping. Additionally, the samba becomes dog-slow and even the mouse falls asleep.
  • Often times, after such phases of heavy load, the skipping and sound-distortion remains! So I have to reboot the machine from time to time to enjoy music again. Closing the player and opening it again is not enough. Somehow, under heavy load things get messed up enough to make a recovery impossible.
I did use the preemptive patch before, but performance under heavy load was even worse and the similar problems with rebooting occurred. I was using kernel 2.4.12 for preemptive and I am using kernel 2.4.17 currently. The machine is a Celeron 466 with 128 megs of ram. Still, the low-latency patch makes sense for machines that are primary for playing MP3s and reading emails (that's what my notebook is), but not for desktops with a wider variety of usage patterns. It's just not ready for primetime yet, but it's promising and fun!

Weighted mid points (1)

AntDaniel (553730) | more than 12 years ago | (#3206385)

Why the suprise? So many time I find that the best solution to a problem is a compromise between two or more extreme solutions.
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?