Seems to me that would be a good thing, as we could use each and see which one actually performs better. They actually discussed this on the Linux Kernel Mailing List and decided bast on the experience of having plugable IO Schedulers that all you do is you end up with several different specialized schedulers that excel at their own niche thing, and no end users have a CLUE which they really need for their task.
They want to go with a single scheduler in mainline to that one can perform the best in pretty much all cases, instead of creating a new one for each obscure niche and no one really knowing whats best for their usage.
Funny you call KHTML broken, when KHTML is far more standards compliant than Firefox. Firefox being a descendent of Netscape simply inherited the code that many of the older sites developed themselves to be broken for, as well as having a better 'lets clone IE mode'.
Yes you can. You get the 32bit version of nspluginwrapper and replace the 64bit one in your install. Since nspluginwrapper is run as a separate process and is simply embedded in the Konqueror window it means you shouldn't have any problems watching flash with the 32bit plugin.
AMD/ATI NEVER said they were going to open source their drivers. They simply said some fluff like they're going to work closer with the community and some such, some blogger took that as 'OPEN SOURCING DRIVERS!'. It is possible ATI will... but fglrx drivers suck, I don't think the OSS community wants the source and would much prefer the driver specs.
I think what he means is atime will only be updated if the the old if the files been modified since the last atime (well, probably at the same time as the last mtime, that way you can just tell if its been read since being written).
On Konqueror 3.5.7 (from openSuSE 10.2) with Java installed and enabled it worked perfectly. As in it created a full screen java window that I couldn't close any way other than to just keep clicking through the messages till it closed itself. alt+f3 (to bring up the application menu), alt+f4 (wanted to close the konqueror window, didn't work cause I had tabs open), alt+tab (it would flash the alt+tab frame but no window would rise above it), I even tried to open yakuake but that was just opened behind it.
I'm sure I could have just done the old ctrl+alt+f1, login, killall -s 9 java, ctrl+alt+f7 or similar to fix it anyways, but still quite annoying. After seeing that I simply went into Settings and disabled java globally, not like anyone actually uses java applets anyways.
Actually the quote from the in the summary from Linus is part of a larger email where hes dismissing the idea of using the plugscheduler (I can't seem to recall the exact name) that would make the CPU scheduler plugin based like the IO scheduler currently is. His reasoning against it were largely BECAUSE what they learned from having the IO scheduler plugin based and its something Linus as well as the subsystem maintainers DON'T want to repeat.
Anyways, you can still just apply Con's patch to the kernel to use his scheduler instead of the old scheduler (and if he keeps maintaining it, you'll be able to use SD instead of CFS). Don't forget that we haven't even had a kernel released using CFS!
KQemu hasn't been GPL for very long, so it hasn't so there hasn't been very many people looking at it for very long. Also the developer of KQemu probably hasn't even ASKED it to be merged (or maybe he has, I don't read the LKML). It's also possible he submitted it to be merged and theres a few things they want him to work out first (Xen has been trying to be merged for a LONG time now).
Actually there is a 'memoryTOP' coming soon (well partly here already!). Right now vmres and vmsize are INCREDIBLY misleading numbers (in some ways they're almost worse than no numbers).
on lwn.net there was an article about a kernel patch that would give more detailed information to user space about applications memory usage (i.e. like what pages are shared, and shared between how many applications, shared where only 1 application uses it, or shared where every single application on the system uses it makes a MASSIVE difference in the actual memory consumption).
Also don't even start looking at the numbers for Xorg, they are down right horrible lies. The reason is because, 1.) Xorg maps your video memory into itself, which artificially raises its memory consumption, and 2.) Applications can store lots of stuff (i.e. pixmaps) inside of Xorg's memory space (Firefox does this extensively).
And I say you're lucky. On the X300 in my laptop I can't even log into SL (it seems to run at around 1 frame a minute on just the login screen while spamming that it failed to allocate memory, a similar error message to the one with regards to/dev/shm not being mounted, but completely unrelated to (/dev/shm is mounted perfectly fine). Because of my horrible experience with ATI cards, and pretty good experience with NVIDIA cards I've already been responsible for multiple people going with nVidia over ATI, and if I buy a new laptop theres no chance in hell it will have an ATI card in it unless they DRASTICALLY improve the support for the x300 in my laptop before then (and even then I probably wouldn't touch it).
The sad thing is, their windows drivers aren't much better...
A very, very, very long ways away from being merged into mainline. About a year ago I tried running Reiser4 as my root filesystem on my laptop for a good while. I had quite a few issues because of it as well, also the performance was pretty crappy. Now I've switched my laptop over to version 3, which I had been running on my desktop for a very long time and have had a much more pleasant experience.
Reiser4 has a very interesting design (I find the plugin approach interesting), but in practice the implementation just wasn't there yet from my usage of it.
Use software suspend instead of shutting off the machine. It boots up much faster as well as shuts down possibly a little faster.
My laptop (SuSE 10.2... possibly going to switch to Kubuntu Fiesty some time in the future) has an uptime of 44 days, though every night I suspend it to the disk (and several times during the day as well sometimes). On boot I generally run the command "swapoff -a && swapon -a; echo done swap; sync; echo done sync" just to bring anything that was swapped out back into main memory (the sync isn't really needed), it seems if you have too little swap space to fit all the memory into it part will be swapped out of the memory and won't be automatically loaded on resume.
All in all, it is FAR faster to suspend (both suspend2disk and suspend2ram are faster, though on my system suspend2disk is the one that was the most reliable, although I should test suspend2ram since I'm using a much more current kernel now).
There are currently 36,511 users currently logged into SecondLife right now. Somewhat hard to call that 'no one'. Back on December 29th they first broke having 20,000 users on at the same time. The population is growing quite quickly (much quicker than the Lindens can handle, though they have a few projects that should help greatly, such as switching to using MONO as the backend for their scripting system instead of their own hand rolled one, in testing they've had a 200x improvement in performance over theirs).
LL is working to move SL to standard protocols (a lot of which will have to be newly created, since in a lot of ways theres nothing equivalent that they could use).
Slashdot Mirror
Since when did the BSDs make their own desktop environments?
They want to go with a single scheduler in mainline to that one can perform the best in pretty much all cases, instead of creating a new one for each obscure niche and no one really knowing whats best for their usage.
The Source engine includes a heavily modified version of the Havok engine.
Funny you call KHTML broken, when KHTML is far more standards compliant than Firefox. Firefox being a descendent of Netscape simply inherited the code that many of the older sites developed themselves to be broken for, as well as having a better 'lets clone IE mode'.
Yes you can. You get the 32bit version of nspluginwrapper and replace the 64bit one in your install. Since nspluginwrapper is run as a separate process and is simply embedded in the Konqueror window it means you shouldn't have any problems watching flash with the 32bit plugin.
AMD/ATI NEVER said they were going to open source their drivers. They simply said some fluff like they're going to work closer with the community and some such, some blogger took that as 'OPEN SOURCING DRIVERS!'. It is possible ATI will... but fglrx drivers suck, I don't think the OSS community wants the source and would much prefer the driver specs.
I think what he means is atime will only be updated if the the old if the files been modified since the last atime (well, probably at the same time as the last mtime, that way you can just tell if its been read since being written).
On Konqueror 3.5.7 (from openSuSE 10.2) with Java installed and enabled it worked perfectly. As in it created a full screen java window that I couldn't close any way other than to just keep clicking through the messages till it closed itself. alt+f3 (to bring up the application menu), alt+f4 (wanted to close the konqueror window, didn't work cause I had tabs open), alt+tab (it would flash the alt+tab frame but no window would rise above it), I even tried to open yakuake but that was just opened behind it.
I'm sure I could have just done the old ctrl+alt+f1, login, killall -s 9 java, ctrl+alt+f7 or similar to fix it anyways, but still quite annoying. After seeing that I simply went into Settings and disabled java globally, not like anyone actually uses java applets anyways.
Actually the quote from the in the summary from Linus is part of a larger email where hes dismissing the idea of using the plugscheduler (I can't seem to recall the exact name) that would make the CPU scheduler plugin based like the IO scheduler currently is. His reasoning against it were largely BECAUSE what they learned from having the IO scheduler plugin based and its something Linus as well as the subsystem maintainers DON'T want to repeat.
Anyways, you can still just apply Con's patch to the kernel to use his scheduler instead of the old scheduler (and if he keeps maintaining it, you'll be able to use SD instead of CFS). Don't forget that we haven't even had a kernel released using CFS!
KQemu hasn't been GPL for very long, so it hasn't so there hasn't been very many people looking at it for very long. Also the developer of KQemu probably hasn't even ASKED it to be merged (or maybe he has, I don't read the LKML). It's also possible he submitted it to be merged and theres a few things they want him to work out first (Xen has been trying to be merged for a LONG time now).
It'll only increase the kernel foot print IF you compile them into the kernel, which they won't be enabled by default.
*cough* You didn't read Asimov's books, did you?
the movie wasn't a telling of the story I, Robot...
Actually its no longer the OSTG. Apparently they renamed themselves yet again and they're now SourceForge Inc.
SQLite is meant for smaller DBs. When you have >60,000 tracks you're going to be pushing the limits of 'small'.
Actually there is a 'memoryTOP' coming soon (well partly here already!). Right now vmres and vmsize are INCREDIBLY misleading numbers (in some ways they're almost worse than no numbers).
on lwn.net there was an article about a kernel patch that would give more detailed information to user space about applications memory usage (i.e. like what pages are shared, and shared between how many applications, shared where only 1 application uses it, or shared where every single application on the system uses it makes a MASSIVE difference in the actual memory consumption).
Also don't even start looking at the numbers for Xorg, they are down right horrible lies. The reason is because, 1.) Xorg maps your video memory into itself, which artificially raises its memory consumption, and 2.) Applications can store lots of stuff (i.e. pixmaps) inside of Xorg's memory space (Firefox does this extensively).
Be raped?
And I say you're lucky. On the X300 in my laptop I can't even log into SL (it seems to run at around 1 frame a minute on just the login screen while spamming that it failed to allocate memory, a similar error message to the one with regards to /dev/shm not being mounted, but completely unrelated to (/dev/shm is mounted perfectly fine). Because of my horrible experience with ATI cards, and pretty good experience with NVIDIA cards I've already been responsible for multiple people going with nVidia over ATI, and if I buy a new laptop theres no chance in hell it will have an ATI card in it unless they DRASTICALLY improve the support for the x300 in my laptop before then (and even then I probably wouldn't touch it).
The sad thing is, their windows drivers aren't much better...
Better to know the bias than have no clue what their bias is. Everyone is biased, might as well know up front what theirs is.
Let me guess, you have an ATI video card?
A very, very, very long ways away from being merged into mainline. About a year ago I tried running Reiser4 as my root filesystem on my laptop for a good while. I had quite a few issues because of it as well, also the performance was pretty crappy. Now I've switched my laptop over to version 3, which I had been running on my desktop for a very long time and have had a much more pleasant experience.
Reiser4 has a very interesting design (I find the plugin approach interesting), but in practice the implementation just wasn't there yet from my usage of it.
Use software suspend instead of shutting off the machine. It boots up much faster as well as shuts down possibly a little faster.
My laptop (SuSE 10.2... possibly going to switch to Kubuntu Fiesty some time in the future) has an uptime of 44 days, though every night I suspend it to the disk (and several times during the day as well sometimes). On boot I generally run the command "swapoff -a && swapon -a; echo done swap; sync; echo done sync" just to bring anything that was swapped out back into main memory (the sync isn't really needed), it seems if you have too little swap space to fit all the memory into it part will be swapped out of the memory and won't be automatically loaded on resume.
All in all, it is FAR faster to suspend (both suspend2disk and suspend2ram are faster, though on my system suspend2disk is the one that was the most reliable, although I should test suspend2ram since I'm using a much more current kernel now).
There are currently 36,511 users currently logged into SecondLife right now. Somewhat hard to call that 'no one'. Back on December 29th they first broke having 20,000 users on at the same time. The population is growing quite quickly (much quicker than the Lindens can handle, though they have a few projects that should help greatly, such as switching to using MONO as the backend for their scripting system instead of their own hand rolled one, in testing they've had a 200x improvement in performance over theirs).
LL is working to move SL to standard protocols (a lot of which will have to be newly created, since in a lot of ways theres nothing equivalent that they could use).
Clones AREN'T exact copies. At least with our current technology. Clones tend to die a lot quicker than the real things and develop more diseases.
Block the sub{directory,domain} that includes the ads. I doubt they'd do a better job for the doubleclick ads than they do for their own.