...the 6501 chip in the C64 limited you to 127 byte either direction relative branching. Anything else had to be handled by an explicit jmp, which meant you couldn't dynamically load your program into other areas of memory (unless you had isolated branch instructions within 127 bytes of every call and didn't use a single jmp - *nearly* impossible) - which kind of blows serious multitasking!
You are talking about relocation, which has nothing to do with multitasking
Anyway, even code with absolute branches can be loaded wherever you want in memory, it just means the loader/linker has to be able to do the necessary fixups.
Let me get this straight... you want MS to stop giving away bundled software for free so competitors (many of whom are free..i.e.Open Office) can compete? Where's the logic in that one?
Not just any software - it's media-playing software that's specifically at issue. Obviously, if anybody can gain control of the media-playing sofware, they gain control of the content as well. So this goes far beyond generic software bundling.
Equally obviously, Microsoft is intent on gaining control of the PC media channels, and to all appearances, will succeed if no regulatory body steps in.
No, Contiki does cooperative multitasking. The reason for not supporting pre-emptive multitasking is that it would unnecessarily increase the complexity not only of the operating system, but also of the applications that would run under it. Pre-emptive multitasking is primarily useful in general purpose multiuser operating systems such as *nix, or in real-time systems where response time is critial. Contiki does not fit in either of those categories.
Hah, apparently this didn't stop the Geoworks people from pulling off a fully preemptive OS on the lowly 8088.
"SuSE/KDE came damned close to meeting or beating Windows XP. I suspect that "anytime soon" they WILL meet or beat WinXP."
Yeah because MS never improves anything by the next version. *eyeroll*
But they don't improve fast enough, compared to the massively parallel development of OSS. And even if they manage to break that logjam, it would inevitably be at the expense of security, which is already slowly killing them.
all signs are point to Doom3 being a lot longer than a few months away
According to my fuzzy recollection, JC was saying that Microsoft is paying them large amounts of money to hold off releasing until the XBox version is ready. That must be some large pile of money.
I think he also said that id would use the time profitably, to make Doom 3 even better. It's apparently been release-ready for quite some time.
The difference between us and an LSF is that we perform well BEFORE you run the repacker, and we merely perform even better after you run it. LSF's required that you run the repacker to get good read performance, we don't. V4 kicks V3's butt without the repacker by a lot (due to dancing trees, allocate on flush, extents, and ending the use of BLOBS, among other things). With the repacker, it will just kick it harder.
Hi Hans! Congratulations on your release and your latest achievements. You really are leading the field in filesystem innovation.
Root is a prime candidate for a small (100MB should do it) ext2 system mounted "sync". You don't need decent write performance on root; in fact, many sysadmins make it read-only. Journalling is pointless if you're only writing to the filesystem once every 6 months to add a new user.
On the contrary, that's exactly the case where you should always journal, and with full data journalling. You don't care about write performance, since you hardly ever do it, but you do care at lot about keeping your root filesystem consistent.
I was there, two critical notes: - Too commercial (it needs commercial stuff, but it needed more non-commercial stuff too).
I was there too, and I disagree. It had exactly the right balance. I was very pleased to see that all the bubble companies are gone, replaced by businesses with serious business plans, and for the most part, also committed to supporting Linux in general.
- 80 to 90 percent of the speaches in German.
Mine wasn't, except for "Weiviel leute hier konnen mien Deutsch verstehen?".
The reason just windows is because that as much as we hate it, we are in the minority of computer uses, they are not going to Bata test a new technology on a system that only a maximum of 5% of computer users will have (and yes I am being overly optimistic here) if this works for them the next platform will be Mac.
Oh good, the test is to see how well it works to discriminate against a minority in a democratic system.
I've got an idea, how about limiting the initial test to whites, as they are the majority. If it works for them, try adding blacks the next time, and if everything's still ok, then maybe hispanics after that. After all, democratic systems are based on majorities, aren't they?
yes, I know I can run KDE (in fact I use the CD player kscd from that environment) but having heard how RH mucked around with KDE to 'unify' it with gnome, I'm leery of that
Have no fear:
xinit/usr/bin/kde3
(you know what to do if you have kde2)
Or if you want to run both at the same time, go to a text console and:
xinit/user/bin/kde3 --:1
I feel your pain re Gnome. At least the desktop itself and panel don't seem to crash any more. It's getting more useable too, but you need time lapse photography to notice the changes. I'll even go out on a limb and suggest that the closer Gnome moves in the KDE's direction, the more usable it gets, and frankly, I don't have a problem with that.
By the way, Gnome2 boots about 30% faster than KDE3. They're both patheticly slow in that department, though:-)
1) Had most of the screws fall out 2) An HD make that "I'm about to die." squealing sound.
Oops, right, I forgot about that - I had to replace the hard disk too, as it was making that kamikaze sound.
So:
1) Replaced the keyboard after 1 1/2 years ($175)
2) Replaced the hard disk after 1 1/2 years ($180, did it myself)
3) Replaced the battery after 1 3/4 years ($250)
Total $605. Sony tax.
The original cost $1450, then there was an extended warranty ($200) which Best Buy never made good on. So, actually $805 went down the drain with that machine. It's been pretty glitchy too, in a number of ways. The latest is, it sometimes thinks the mouse is a keyboard after coming out of suspend. As you might imagine, that's quite annoying.
If your warranty is up, and something breaks, you're really out of luck. This is why I suggest that laptop owners buy an extended warranty of some sort for their machines.
I did. I bought the 5 year extended warranty from Best Buy, along with my VAIO F540. A little over a year the keyboard broke, and following instructions from Best Buy, I had it replaced. Then I started getting the runaround between Best Buy and the company they farm the warranty stuff out to. More than a year later, I'm still trying to collect on the warrany. I'm out of pocket $175.
1. Don't buy anything from Best Buy, they don't care about you, and they do not make good on their warranties (at least mine).
2. Buy a Samsung. The are cheaper than VAIOs and, I've heard, better constructed.
Both are known to be defective on VAIOs. My battery lasted just over a year (and they are *expensive*) and the keyboard just a little longer. A quick web search shows I'm just one of many.
Also, I'm still waiting to get my warrantee refund on the keyboard, I replaced it at my expense more than a year ago.
Okay, but without drivers for that wireless hardware (which is closed and proprietary), you've got yourself a nice little doorstop to hack at your own will...
The drivers probably take the form of kernel modules. (The only other alternatives are user space programs - very unlikely - or code linked into the kernel, in which case it has to be GPLed.) So we can continue using those modules, as long as we stick with the 2.4.5 kernel version they're using (sick, I know). We could possibly even write shims to forward-port the binary modules to later kernel versions, if there turned out to be enough interest.
So we're not stuck for drivers, sorry for assuming that everybody would see that right away.
This type of license strong-arming is the reason embeded market should choose BSD.
Sure, Linksys would have used BSD if BSD worked as well for them as Linux does. But Linux is better, supports more hardware, and supplies more of the features they need. That's because more programmers work on Linux, and that in turn is because of the licence - we know that our contributions to the public are going to *stay* public.
And by the way, I like BSD. I think there's some stuff in BSD that's better than Linux. I'm no fanatic, I'm just explaining the phenomenon.
Anyway, what's the big deal for Linksys if they accidently ended up letting everybody reprogram their hardware? It just means they sell more of it and make more money. That's gotta hurt, huh?
Good first step, but that's far from the only Linksys device running Linux. Where are the sources for the other devices?
Of course you're correct about that. But what's wrong with being nice to them just now. Let's watch them be shocked when sales of the unit tick upward, then let's ask for more, ok?:-)
Of course, the above depends on being able to reflash the thing successfully, and as far as I can see, that's going to be a whole lot easier than the XBox was, plus more useful, including being useful to Linksys. Unlike Microsoft, they make money every time they sell one.
It's nice that you can see the GPL bits running on their box (Hey running a 2.4.5 kernel) - but it doesn't give you drivers, or scripts, etc.
The original flash image was decoded (by Andrew Miklas) as a cramfs filesystem. We have *all* the components available in binary form. Unless there is crytographic checking in the bootloader (i.e. a signed flash image) we're all set to go make our own images complete with Linksys's proprietary binaries and our hacked/improved GPL binaries.
To my knowledge, nobody has done this yet. I hope that doesn't last long. These units will make lovely general-purpose embedded machines if we can put our own code in them. We'd have to rely on Linksys binaries for some of the hardware, but personally I have no ideological problem with that. What I want is to be able to fix some bugs of the bugs and interface stupidities in the darn thing, and to add some of my own functionality, such as being able to ssh into it. Of course, I'd like it even more if Linksys released the full hardware specs, but hey. It's a start.
Now, I see the kind of hacking I described above, and which I fully intend to get involved in, as nothing but good for Linksys. If it turns out we can reflash the unit as it appears we can, I for one will be in the market for a few more of these.
...it is *necessary* in order for open-source projects to reach the mass-market (unless they sell themselves out) that any spokespeople which they have (and Linus is certainly this) pander to the needs and perceptions of the mass majority. And Linus, in this case, did *not* do this.
Why, you should email him and give him a piece of your mind then. For my part, I felt the article his responses were artfully crafted for exactly the target audience he wished to address, and just so you know, that is not the general public. It is people like you and me, technophiles who hang out out the net and help define the future directions of technology, and hence, society. To put it less delicately, he's aiming at people who have a clue.
If your wife needs someone to speak more directly to her level of knowledge, point her at Mad Dog Hall. Now there is a man who speaks beautifully to the masses and enjoys doing it.
Linux was influenced heavily by Minix, yes. But Linus never had access to Minix source code when he began writing what would end up being called Linux, so your conspiracy theory of Linus borrowing code from Minix is non-viable.
Well, considering that the Minix code was included with a book written by Tannenbaum, I think you're a little wide of the mark there. What Linus wasn't allowed to do, or anyone else for that matter, was contribute code back to Tannebaum's OS.
Anyway, it's well-known Linux is a from-scratch implementation, which apparently started as a terminal emulator (and that possibly explains why the console handling code still sucks so much, but there you go). Andy Tannenbaum has had plenty of time to complain about any plagiarism, if Linus indulged in it, and trust me, he would have.
I don't think it's entirely because Linus doesn't speak english natively, either, to fend off potential criticism: I speak three languages, and I know *plenty* of efl (english as a foreign language) speakers who speak english which isn't as good as Linus's (which is excellent) who can explain technical issues with more clarity.
Having heard him speak often enough, I'm reasonably sure Linus is a more accomplished English speaker than you are. But more to the point, I question why you even felt compelled to raise the issue.
I'm very sorry that your wife needs to have forking explained to her. Ah, take that how you will.
I just read the FAQ (Mostly for developers). One part caught my attention:
Q: How does the module remove code work?
A: It stops the machine by scheduling threads for every other CPU,
then they all disable interrupts. At this stage we know that noone
is in try_module_get(), so we can reliably read the counter. If
zero, or the rmmod user specified --wait, we set the live flag to
false. After this, the reference count should not increase, and
each module_put() will wake us up, so we can check the counter
again.
To me that sounds like a very expensive system call.
It is, but you don't execute it hundreds of times per second.
Slashdot Mirror
...the 6501 chip in the C64 limited you to 127 byte either direction relative branching. Anything else had to be handled by an explicit jmp, which meant you couldn't dynamically load your program into other areas of memory (unless you had isolated branch instructions within 127 bytes of every call and didn't use a single jmp - *nearly* impossible) - which kind of blows serious multitasking!
You are talking about relocation, which has nothing to do with multitasking
Anyway, even code with absolute branches can be loaded wherever you want in memory, it just means the loader/linker has to be able to do the necessary fixups.
Let me get this straight... you want MS to stop giving away bundled software for free so competitors (many of whom are free..i.e.Open Office) can compete? Where's the logic in that one?
Not just any software - it's media-playing software that's specifically at issue. Obviously, if anybody can gain control of the media-playing sofware, they gain control of the content as well. So this goes far beyond generic software bundling.
Equally obviously, Microsoft is intent on gaining control of the PC media channels, and to all appearances, will succeed if no regulatory body steps in.
Hah, apparently this didn't stop the Geoworks people from pulling off a fully preemptive OS on the lowly 8088.
"SuSE/KDE came damned close to meeting or beating Windows XP. I suspect that "anytime soon" they WILL meet or beat WinXP."
Yeah because MS never improves anything by the next version. *eyeroll*
But they don't improve fast enough, compared to the massively parallel development of OSS. And even if they manage to break that logjam, it would inevitably be at the expense of security, which is already slowly killing them.
So say someone was gonna goto next years event, does anyone who was there this year have any good tips? What to expect, where to stay, drink, etc...
There is no such thing as a bad place to drink in Germany.
all signs are point to Doom3 being a lot longer than a few months away
According to my fuzzy recollection, JC was saying that Microsoft is paying them large amounts of money to hold off releasing until the XBox version is ready. That must be some large pile of money.
I think he also said that id would use the time profitably, to make Doom 3 even better. It's apparently been release-ready for quite some time.
The difference between us and an LSF is that we perform well BEFORE you run the repacker, and we merely perform even better after you run it. LSF's required that you run the repacker to get good read performance, we don't. V4 kicks V3's butt without the repacker by a lot (due to dancing trees, allocate on flush, extents, and ending the use of BLOBS, among other things). With the repacker, it will just kick it harder.
Hi Hans! Congratulations on your release and your latest achievements. You really are leading the field in filesystem innovation.
Root is a prime candidate for a small (100MB should do it) ext2 system mounted "sync". You don't need decent write performance on root; in fact, many sysadmins make it read-only. Journalling is pointless if you're only writing to the filesystem once every 6 months to add a new user.
On the contrary, that's exactly the case where you should always journal, and with full data journalling. You don't care about write performance, since you hardly ever do it, but you do care at lot about keeping your root filesystem consistent.
do you think that really deserved a Troll?
No. But of course I did not do the modding.
The takeaway from that one is: if you claim someone did say a certain thing, you'd better provide a link.
AFAIK Linus has stated that most if not all of the code that SCO is B1tchin' about has be[en] rewritten in 2.6.
Linus has stated no such thing.
Linus has stated that the SCO suit is not about IP, either copyrights or patents, on the contrary, it is over a contract with IBM.
I was there, two critical notes:
- Too commercial (it needs commercial stuff, but it needed more non-commercial stuff too).
I was there too, and I disagree. It had exactly the right balance. I was very pleased to see that all the bubble companies are gone, replaced by businesses with serious business plans, and for the most part, also committed to supporting Linux in general.
- 80 to 90 percent of the speaches in German.
Mine wasn't, except for "Weiviel leute hier konnen mien Deutsch verstehen?".
"Security through obsurity is worthless"
Everyone says this, but not one understands what it means.
What it means is that obscurity is not sufficient for security. It does not mean that obscurity is not helpful as part of an overall security system.
On the contrary, it means exactly that. A false sense of security is the worst possible state of affairs for what you call "overall security".
The reason just windows is because that as much as we hate it, we are in the minority of computer uses, they are not going to Bata test a new technology on a system that only a maximum of 5% of computer users will have (and yes I am being overly optimistic here) if this works for them the next platform will be Mac.
Oh good, the test is to see how well it works to discriminate against a minority in a democratic system.
I've got an idea, how about limiting the initial test to whites, as they are the majority. If it works for them, try adding blacks the next time, and if everything's still ok, then maybe hispanics after that. After all, democratic systems are based on majorities, aren't they?
yes, I know I can run KDE (in fact I use the CD player kscd from that environment) but having heard how RH mucked around with KDE to 'unify' it with gnome, I'm leery of that
/usr/bin/kde3
/user/bin/kde3 -- :1
:-)
Have no fear:
xinit
(you know what to do if you have kde2)
Or if you want to run both at the same time, go to a text console and:
xinit
I feel your pain re Gnome. At least the desktop itself and panel don't seem to crash any more. It's getting more useable too, but you need time lapse photography to notice the changes. I'll even go out on a limb and suggest that the closer Gnome moves in the KDE's direction, the more usable it gets, and frankly, I don't have a problem with that.
By the way, Gnome2 boots about 30% faster than KDE3. They're both patheticly slow in that department, though
BTW, I am a former VAIO owner who:
1) Had most of the screws fall out
2) An HD make that "I'm about to die." squealing sound.
Oops, right, I forgot about that - I had to replace the hard disk too, as it was making that kamikaze sound.
So:
1) Replaced the keyboard after 1 1/2 years ($175)
2) Replaced the hard disk after 1 1/2 years ($180, did it myself)
3) Replaced the battery after 1 3/4 years ($250)
Total $605. Sony tax.
The original cost $1450, then there was an extended warranty ($200) which Best Buy never made good on. So, actually $805 went down the drain with that machine. It's been pretty glitchy too, in a number of ways. The latest is, it sometimes thinks the mouse is a keyboard after coming out of suspend. As you might imagine, that's quite annoying.
If your warranty is up, and something breaks, you're really out of luck. This is why I suggest that laptop owners buy an extended warranty of some sort for their machines.
I did. I bought the 5 year extended warranty from Best Buy, along with my VAIO F540. A little over a year the keyboard broke, and following instructions from Best Buy, I had it replaced. Then I started getting the runaround between Best Buy and the company they farm the warranty stuff out to. More than a year later, I'm still trying to collect on the warrany. I'm out of pocket $175.
1. Don't buy anything from Best Buy, they don't care about you, and they do not make good on their warranties (at least mine).
2. Buy a Samsung. The are cheaper than VAIOs and, I've heard, better constructed.
Both are known to be defective on VAIOs. My battery lasted just over a year (and they are *expensive*) and the keyboard just a little longer. A quick web search shows I'm just one of many.
Also, I'm still waiting to get my warrantee refund on the keyboard, I replaced it at my expense more than a year ago.
Do yourself a favor and buy a Samsung.
Okay, but without drivers for that wireless hardware (which is closed and proprietary), you've got yourself a nice little doorstop to hack at your own will...
The drivers probably take the form of kernel modules. (The only other alternatives are user space programs - very unlikely - or code linked into the kernel, in which case it has to be GPLed.) So we can continue using those modules, as long as we stick with the 2.4.5 kernel version they're using (sick, I know). We could possibly even write shims to forward-port the binary modules to later kernel versions, if there turned out to be enough interest.
So we're not stuck for drivers, sorry for assuming that everybody would see that right away.
This type of license strong-arming is the reason embeded market should choose BSD.
Sure, Linksys would have used BSD if BSD worked as well for them as Linux does. But Linux is better, supports more hardware, and supplies more of the features they need. That's because more programmers work on Linux, and that in turn is because of the licence - we know that our contributions to the public are going to *stay* public.
And by the way, I like BSD. I think there's some stuff in BSD that's better than Linux. I'm no fanatic, I'm just explaining the phenomenon.
Anyway, what's the big deal for Linksys if they accidently ended up letting everybody reprogram their hardware? It just means they sell more of it and make more money. That's gotta hurt, huh?
Good first step, but that's far from the only Linksys device running Linux. Where are the sources for the other devices?
:-)
Of course you're correct about that. But what's wrong with being nice to them just now. Let's watch them be shocked when sales of the unit tick upward, then let's ask for more, ok?
Of course, the above depends on being able to reflash the thing successfully, and as far as I can see, that's going to be a whole lot easier than the XBox was, plus more useful, including being useful to Linksys. Unlike Microsoft, they make money every time they sell one.
It's nice that you can see the GPL bits running on their box (Hey running a 2.4.5 kernel) - but it doesn't give you drivers, or scripts, etc.
The original flash image was decoded (by Andrew Miklas) as a cramfs filesystem. We have *all* the components available in binary form. Unless there is crytographic checking in the bootloader (i.e. a signed flash image) we're all set to go make our own images complete with Linksys's proprietary binaries and our hacked/improved GPL binaries.
To my knowledge, nobody has done this yet. I hope that doesn't last long. These units will make lovely general-purpose embedded machines if we can put our own code in them. We'd have to rely on Linksys binaries for some of the hardware, but personally I have no ideological problem with that. What I want is to be able to fix some bugs of the bugs and interface stupidities in the darn thing, and to add some of my own functionality, such as being able to ssh into it. Of course, I'd like it even more if Linksys released the full hardware specs, but hey. It's a start.
Now, I see the kind of hacking I described above, and which I fully intend to get involved in, as nothing but good for Linksys. If it turns out we can reflash the unit as it appears we can, I for one will be in the market for a few more of these.
...it is *necessary* in order for open-source projects to reach the mass-market (unless they sell themselves out) that any spokespeople which they have (and Linus is certainly this) pander to the needs and perceptions of the mass majority. And Linus, in this case, did *not* do this.
Why, you should email him and give him a piece of your mind then. For my part, I felt the article his responses were artfully crafted for exactly the target audience he wished to address, and just so you know, that is not the general public. It is people like you and me, technophiles who hang out out the net and help define the future directions of technology, and hence, society. To put it less delicately, he's aiming at people who have a clue.
If your wife needs someone to speak more directly to her level of knowledge, point her at Mad Dog Hall. Now there is a man who speaks beautifully to the masses and enjoys doing it.
Linux was influenced heavily by Minix, yes. But Linus never had access to Minix source code when he began writing what would end up being called Linux, so your conspiracy theory of Linus borrowing code from Minix is non-viable.
Well, considering that the Minix code was included with a book written by Tannenbaum, I think you're a little wide of the mark there. What Linus wasn't allowed to do, or anyone else for that matter, was contribute code back to Tannebaum's OS.
Anyway, it's well-known Linux is a from-scratch implementation, which apparently started as a terminal emulator (and that possibly explains why the console handling code still sucks so much, but there you go). Andy Tannenbaum has had plenty of time to complain about any plagiarism, if Linus indulged in it, and trust me, he would have.
I don't think it's entirely because Linus doesn't speak english natively, either, to fend off potential criticism: I speak three languages, and I know *plenty* of efl (english as a foreign language) speakers who speak english which isn't as good as Linus's (which is excellent) who can explain technical issues with more clarity.
Having heard him speak often enough, I'm reasonably sure Linus is a more accomplished English speaker than you are. But more to the point, I question why you even felt compelled to raise the issue.
I'm very sorry that your wife needs to have forking explained to her. Ah, take that how you will.
I just read the FAQ (Mostly for developers). One part caught my attention:
Q: How does the module remove code work?
A: It stops the machine by scheduling threads for every other CPU,
then they all disable interrupts. At this stage we know that noone
is in try_module_get(), so we can reliably read the counter. If
zero, or the rmmod user specified --wait, we set the live flag to
false. After this, the reference count should not increase, and
each module_put() will wake us up, so we can check the counter
again.
To me that sounds like a very expensive system call.
It is, but you don't execute it hundreds of times per second.