Comments on: Linux is Bloated, Just Not How You Think http://www.linux-mag.com/id/7536/ Open Source, Open Standards Sat, 05 Oct 2013 13:48:18 +0000 hourly 1 http://wordpress.org/?v=3.1 By: Elwood Davi http://www.linux-mag.com/id/7536/#comment-990659 Elwood Davi Sun, 09 Jun 2013 15:06:59 +0000 http://www.linux-mag.com/id/7536/#comment-990659 Pains that are due to bloating will feel sharp and cause the stomach to cramp. These pains may occur anywhere in the body and can change locations quickly....... See all of the most recently released posting on our new website <http://www.healthmedicinebook.com/ Pains that are due to bloating will feel sharp and cause the stomach to cramp. These pains may occur anywhere in the body and can change locations quickly…….

See all of the most recently released posting on our new website
<http://www.healthmedicinebook.com/

]]> By: eavedrop44 http://www.linux-mag.com/id/7536/#comment-628735 eavedrop44 Thu, 06 Dec 2012 18:40:05 +0000 http://www.linux-mag.com/id/7536/#comment-628735 The second will have to block not only the running task, but those scheduled for it until the block is freed by the first core.<a href="http://www.sport-only.com" rel="nofollow">Sport Only</a> <a href="http://www.sportrecycle.com" rel="nofollow">Sport Recycle</a> <a href="http://www.sportscarexpo.com" rel="nofollow">Sport Car expo</a> The second will have to block not only the running task, but those scheduled for it until the block is freed by the first core.Sport Only
Sport Recycle
Sport Car expo

]]> By: eavedrop44 http://www.linux-mag.com/id/7536/#comment-519335 eavedrop44 Tue, 30 Oct 2012 05:39:02 +0000 http://www.linux-mag.com/id/7536/#comment-519335 This is often a cause of problems in the database world where multiple clients may be executing queries on desparent data.<a href="http://www.painting-inn.com" rel="nofollow">painting</a> This is often a cause of problems in the database world where multiple clients may be executing queries on desparent data.painting

]]> By: Ryan http://www.linux-mag.com/id/7536/#comment-25435 Ryan Mon, 07 Nov 2011 05:51:44 +0000 http://www.linux-mag.com/id/7536/#comment-25435 Clearly it's just fine for embedded stuff, since it's completey dominating that market. My router runs Linux, my NAS runs Linux, my phone runs Linux. Not because I cared what they were running when I bought them - just out of chance and because it's *that* ubiquitous. Microkernels are awful. Anyone who has ever tried to implement one has eventually cried foul and run home vowing never to try again. Linus is right on the matter. Anyone who disagrees needs an education in reality. Clearly it’s just fine for embedded stuff, since it’s completey dominating that market. My router runs Linux, my NAS runs Linux, my phone runs Linux. Not because I cared what they were running when I bought them – just out of chance and because it’s *that* ubiquitous.

Microkernels are awful. Anyone who has ever tried to implement one has eventually cried foul and run home vowing never to try again. Linus is right on the matter. Anyone who disagrees needs an education in reality.

]]> By: Ryan http://www.linux-mag.com/id/7536/#comment-25425 Ryan Mon, 07 Nov 2011 05:42:29 +0000 http://www.linux-mag.com/id/7536/#comment-25425 Linus makes a comment about Linux having a bit of bloat and out come the peanut gallery suggesting how to fix it. Hurd!? Oh sure, let's fix a small problem (minor bloat) with a fucking huge problem (developers who can't make a single release in 2 decades). Go eat your peanuts you fucking choad. Linus makes a comment about Linux having a bit of bloat and out come the peanut gallery suggesting how to fix it.

Hurd!? Oh sure, let’s fix a small problem (minor bloat) with a fucking huge problem (developers who can’t make a single release in 2 decades).

Go eat your peanuts you fucking choad.

]]> By: Ryan http://www.linux-mag.com/id/7536/#comment-25417 Ryan Mon, 07 Nov 2011 05:38:54 +0000 http://www.linux-mag.com/id/7536/#comment-25417 Shut up you fucking troll. Shut up you fucking troll.

]]> By: bendib http://www.linux-mag.com/id/7536/#comment-7048 bendib Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7536/#comment-7048 <p>5MB is bloated? This really should not even be talked about at this point, linux is much bigger, but bloated? I respect Linus, nearly worship the guy, in fact, but it is not bloated. This is goofy. Bigger, not bloated. Linus needs to see that his OS has evolved into a respectable and featureful desktop OS, and quit trying to make it into this little squeaker for routers and sony TVs. </p> 5MB is bloated? This really should not even be talked about at this point, linux is much bigger, but bloated? I respect Linus, nearly worship the guy, in fact, but it is not bloated. This is goofy. Bigger, not bloated. Linus needs to see that his OS has evolved into a respectable and featureful desktop OS, and quit trying to make it into this little squeaker for routers and sony TVs.

]]> By: corktowner http://www.linux-mag.com/id/7536/#comment-7049 corktowner Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7536/#comment-7049 <p>Bloated compared to it\'s roots, certainly. But it is still a lean and mean OS which runs circles around the competition. The ability to support older, resource limited systems as well as the ability to run supercomputers is the ultimate compliment for Linus and the many developers. I am thoroughly impressed and in awe of the power and abilities of Linux. Congratulations for the great work to all involved, you are the best of the best. You make computers fun and productive. I thank you all. </p> Bloated compared to it\’s roots, certainly. But it is still a lean and mean OS which runs circles around the competition. The ability to support older, resource limited systems as well as the ability to run supercomputers is the ultimate compliment for Linus and the many developers. I am thoroughly impressed and in awe of the power and abilities of Linux. Congratulations for the great work to all involved, you are the best of the best. You make computers fun and productive. I thank you all.

]]> By: rgmenke http://www.linux-mag.com/id/7536/#comment-7050 rgmenke Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7536/#comment-7050 <p>I know this is a devisive issue but maybe he ought to re-think his stance on microkernals? </p> I know this is a devisive issue but maybe he ought to re-think his stance on microkernals?

]]> By: jgabler http://www.linux-mag.com/id/7536/#comment-7051 jgabler Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7536/#comment-7051 <p>... or his stance on the definition and use of the word \"bloated\". With only 3% of merges on his plate, maybe Linus is getting bored? </p> … or his stance on the definition and use of the word \”bloated\”. With only 3% of merges on his plate, maybe Linus is getting bored?

]]> By: rhkramer http://www.linux-mag.com/id/7536/#comment-7052 rhkramer Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7536/#comment-7052 <p>Two comments:</p> <p> * If Linus says the kernel is bloated, I believe him--he\'s in a far better position to know and judge than I am. And, it wouldn\'t surprise me to find that it\'s bloated in various ways--maybe leftover cruft, maybe too much stuff in the \"base\" kernel that could be in modules, or ???</p> <p> * I\'d really like to find a Linux kernel and small something built around it that is not too much greater than, for example, MsDos 6.2 or similar (being the last number I remember using). I have tried things like Freesco, but I don\'t like that approach (iiuc, the typical Linux approach) for a number of reasons.<br /> * I\'d like a kernel that can fit on one say 1.44 MB floppy along with a few programs so I can boot from a floppy and run those programs<br /> * without the contortions that, for example, Freesco goes through--if I can remember correctly, they store a compressed image of a fairly small Linux system on the floppy, but when uncompressed it uses a great deal of RAM (again, from my imperfect memory), on the order of 25 MB or more. </p> <p>Dos didn\'t work that way, and I\'d prefer something that worked more like the old dos approach. </p> Two comments:

* If Linus says the kernel is bloated, I believe him–he\’s in a far better position to know and judge than I am. And, it wouldn\’t surprise me to find that it\’s bloated in various ways–maybe leftover cruft, maybe too much stuff in the \”base\” kernel that could be in modules, or ???

* I\’d really like to find a Linux kernel and small something built around it that is not too much greater than, for example, MsDos 6.2 or similar (being the last number I remember using). I have tried things like Freesco, but I don\’t like that approach (iiuc, the typical Linux approach) for a number of reasons.
* I\’d like a kernel that can fit on one say 1.44 MB floppy along with a few programs so I can boot from a floppy and run those programs
* without the contortions that, for example, Freesco goes through–if I can remember correctly, they store a compressed image of a fairly small Linux system on the floppy, but when uncompressed it uses a great deal of RAM (again, from my imperfect memory), on the order of 25 MB or more.

Dos didn\’t work that way, and I\’d prefer something that worked more like the old dos approach.

]]> By: blouis79 http://www.linux-mag.com/id/7536/#comment-7053 blouis79 Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7536/#comment-7053 <p>Something perhaps less bloated and huge could be a GNU/Hurd-type kernel. But 19 years of development has yet to result in something fast and stable. So a \"bloated\" kernel that works may be good enough for now. </p> Something perhaps less bloated and huge could be a GNU/Hurd-type kernel. But 19 years of development has yet to result in something fast and stable. So a \”bloated\” kernel that works may be good enough for now.

]]> By: ctryon http://www.linux-mag.com/id/7536/#comment-7054 ctryon Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7536/#comment-7054 <p>I think the worst \"bloat\" is in terms of the number of features which the kernel can <em>potentially</em> support. This can be a huge problem for developers -- as the complexity of the kernel increases, so do the number of unexpected consequences for any changes made. Testing gets harder, security gets more complicated, and bugs increase.</p> <p>On the good side though, the typical user probably only loads up (I\'m guessing) 5% to 10% of that kernel when they actually boot their machine. The rest is for drivers or architectures that they don\'t have and don\'t care about. The end result is that the kernel still runs relatively fast and bug-free.</p> <p>Of course, the rapid churn of development, and even the fundamental re-factoring that does take place create a huge headache for those companies trying to maintain links between their proprietary drivers and the rest of the kernel. (e.g., VMware, ATI, Nvidia, etc.) Without getting into the ethical flame war about proprietary code in the Linux kernel, it does create problems for vendors and users alike. </p> I think the worst \”bloat\” is in terms of the number of features which the kernel can potentially support. This can be a huge problem for developers — as the complexity of the kernel increases, so do the number of unexpected consequences for any changes made. Testing gets harder, security gets more complicated, and bugs increase.

On the good side though, the typical user probably only loads up (I\’m guessing) 5% to 10% of that kernel when they actually boot their machine. The rest is for drivers or architectures that they don\’t have and don\’t care about. The end result is that the kernel still runs relatively fast and bug-free.

Of course, the rapid churn of development, and even the fundamental re-factoring that does take place create a huge headache for those companies trying to maintain links between their proprietary drivers and the rest of the kernel. (e.g., VMware, ATI, Nvidia, etc.) Without getting into the ethical flame war about proprietary code in the Linux kernel, it does create problems for vendors and users alike.

]]> By: noahspurrier http://www.linux-mag.com/id/7536/#comment-7055 noahspurrier Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7536/#comment-7055 <p>This is a misleading way to look at it. Most of the kernel source is drivers, modules, and architecture specific code. Only 5% of the kernel source is the actual core kernel. In other words, 90% of the kernel source is stuff that you will never use in your particular binary release. Most of it never even gets built.</p> <p> http://cycle-gap.blogspot.com/2008/07/linux-kernel-development-stats-from.html</p> <p>--<br /> Noah </p> This is a misleading way to look at it. Most of the kernel source is drivers, modules, and architecture specific code. Only 5% of the kernel source is the actual core kernel. In other words, 90% of the kernel source is stuff that you will never use in your particular binary release. Most of it never even gets built.

http://cycle-gap.blogspot.com/2008/07/linux-kernel-development-stats-from.html

–
Noah

]]> By: beniji http://www.linux-mag.com/id/7536/#comment-7056 beniji Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7536/#comment-7056 <p>Bloat is all relative. Anyone tried compiling Vista recently? I\'m guessing it takes a while longer than linux 2.6.</p> <p>The monolithic architecture makes writing device drive code simpler but clearly not great for non-embedded stuff. </p> Bloat is all relative. Anyone tried compiling Vista recently? I\’m guessing it takes a while longer than linux 2.6.

The monolithic architecture makes writing device drive code simpler but clearly not great for non-embedded stuff.

]]>