The BIOS is allowed to patch the cpu microcode. applications are not. Allowing an application to patch the cpu microcode would be a security flaw. icc does not emit microcode. Maybe you mean to say it can do microcode-aware instruction scheduling.
Wow, I read slashdot from the beginning. I remember the days before user IDs when "anonymous Coward" was an actual person, not a meme, and BoReD At WoRK used to post comments. When a reverse DNS lookup on slashdot lead you to triton.imagegroup.com....
I resisted the user ID stuff when the site's geek quotient started to go down, and newbies diluted things, but finally signed up when I just HAD to comment on some story or other. The site never lost it's geek cred even when 6 digit UIDs started to be the norm, so hopefully it will continue to be relevant and true in spirit to the old site. But even so it won't be the same without you!
I've been on Slashdot for what 15ish years now... before there were user IDs.. anyway
I've always hated the supposedly free market capitalism anti-government crap espoused by the supposedly enlightened intellectual libertarians who hand around the "open source" movement. But this is the worst I have ever seen, and it proves they are not intellectuals, they are rabid dogmatists.
The fire department DOES need to get paid for fighting fires, and when a non-insured home is on fire the "putting out an actual fire" service fee will need to be much higher than the "putting out a hypothetical fire" insurance fee. But the service fee can be paid OUT OF POCKET, or by putting a LIEN on the house. The cost to put the fire out was WAY less than the value of the home, and refusing to put it out ends up being a net destruction of wealth (not to mention the lives of the pets inside).
At this point the worst libertards say "well they didn't pay the insurance, so tough", and it completely escapes their consideration that MAYBE just MAYBE the fire department could have charged them $5000 or $10000 to put out the fire. An intellectual would see that alternate payment system as a good and just possibility, but a dogmatist doesn't need to think that hard because they already have all the answers.
Any economic system that fails to protect wealth and watches it burn away will lead to a society with LESS WEALTH. The fire department may have saved $1000 by not putting out the fire, but the society lost 200 times that when the home and its contents were consumed. If you don't have enough brains to improve upon that result, even within the libertarian worldview then you are an idiot, not a legitimate intellectual that people should bother listening to.
um yeah except the WHITE firefighters were the ones suing, after the CITY decided to throw out the results of the exam, and IIRC she was following precedent that the supreme court later reversed itself on........ so how did the supreme court reversing its own precedent become her fault? Oh yeah race-baiting fox news.
There may be additional hacks you need. Specifically comment out the lines in soundon or whatever they called it that uninstalls the alsa drivers, or you'll have to re-install the kernel after it deletes them. And be aware you may have to strangle udev to get it to STOP LOADING THE GODDAMN ALSA DRIVERS FIRST.
Actually I have oss4 for hdaudio, and alsa for usb audio working together on debian. It is was a royal bitch tho. I run this script instead of oss soundon/soundoff
#!/bin/bash set -x
if [ ! -d/lib/modules/$(uname -r)/oss ] then
soundoff
soundon
mv/lib/modules/$(uname -r)/kernel/oss/lib/modules/$(uname -r)/oss
depmod -a fi
What corporation do you own 10% of that does not confer limited liability to you via the concept of corporate personhood? In order to sue you personally they'd have to "pierce the corporate veil" which doesn't happen very often. I think you're BSing here.
Um, dude. That statement by itself isn't racist. The policy that the statement is meant to defend by camouflaging it, RACIAL PROFILING, is racist. And the statement is wrong. Studies (that I do not have links for, sorry) have shown that after dark when the race of a person driving a car is harder to see the racial disparity in traffic stops decreases.
Of COURSE they don't know how much a voluntary insurance plan is going to cost, since they can't FORCE you to sign up for it! Blue Cross doesn't know how much their plan costs in advance either.
constant fscking is the bug, not the intended behaviour, dude. If I had KNOWN it would fsck every reboot I would have done that. Now that people have heard my experience maybe they will choose that option.
Hey genius! It's called SARCASM! There is no need to insult me for sharing my experiences with ext4. I am not some idiot who thinks RAID is a backup either.
I am NOT happy with ext4, but in case you needed to know more about my system, my ext4 data is a MIRROR of the ext3 data, and I have a LTO-2 ultirum drive to backup my filesystem regularly. I have off-site backups in fireproof waterproof safes.
I also use e2image to get the metadata off the drive in case I need to reconstruct the contents.
Well, the fsck times are really fast compared to ext3, and thank god, because EVERY time I reboot it requires an fsck, complaining about group descriptor checksums. Even if I unmount my ext4 filesystem and remount it without rebooting it gets all fscked up. I have a 3TB ext4 fs on LVM on RAID, that was NOT converted from ext3, but built on brand new drives. My similar ext3 filesystem has had so such problems.
ext4 takes about 7 minutes to fsck, ext3 took hours. I hope they fix this soon.
the 1 error per 512 bits is not an EXPECTED error rate. It's a CORRECTABLE error rate. The actual error rate is going to be far below that, but due to bursts will sometimes be above it, which is why the extra block length is good.
OK, yes you COULD move the parity dta around but you'd get shitty performance. Hard drives are made so that each sector is independent of another. That makes each sector a seperate codeword on disk. What you are proposing is to introduce dependency between sectors, and that would mean having to read adjacent sectors in order to write a single sector, which means goin through 2 revolution of the disk instead of one.
I guess I should pre-emptively point out that for a hard drive you want to be able to modify each sector atomically, which means that a single sector corresponds to a single codeword, and increasing areal density means you need longer codewords to maintain error correction. So either you decrease the rate of the code, and use extra redundancy, which lower capacity and defeats the purpose of increasing areal density, or you us longer codewords at the same rate, which means using longer sectors.
the rate of a code measure how much redundundacy it has, correct. But why do you think block length doesn't matter? Just because you have high redundancy doesn't mean your errors are going to magically be recoverable. To actually recover the data you need enough distance between valid codewords so that when a codeword is perturbed by errors you can still see which valid codeword it is closest to. With short block lengths you get small decoding distances, and low error correcting power. If you learn information theory a bit better you'll see Claude Shannon's channel "capacity" theory assumes infinite block length, and it does that for a REASON.
Yep threads are faster, but what about stability and security? If there's a buffer overflow attack against apache with pre-forking can the attacker see everbody else's credit card numbers at the end of the SSL stream where it's been decrypted? Nope. But they could see them if you're running the multi threaded server, since there is no protection. And if the multithreaded server has a single thread die, they all die.
Slashdot Mirror
The BIOS is allowed to patch the cpu microcode. applications are not. Allowing an application to patch the cpu microcode would be a security flaw. icc does not emit microcode. Maybe you mean to say it can do microcode-aware instruction scheduling.
Wow, I read slashdot from the beginning. I remember the days before user IDs when "anonymous Coward" was an actual person, not a meme, and BoReD At WoRK used to post comments. When a reverse DNS lookup on slashdot lead you to triton.imagegroup.com....
I resisted the user ID stuff when the site's geek quotient started to go down, and newbies diluted things, but finally signed up when I just HAD to comment on some story or other. The site never lost it's geek cred even when 6 digit UIDs started to be the norm, so hopefully it will continue to be relevant and true in spirit to the old site. But even so it won't be the same without you!
Thanks Rob, and best of luck!
I've been on Slashdot for what 15ish years now... before there were user IDs.. anyway
I've always hated the supposedly free market capitalism anti-government crap espoused by the supposedly enlightened intellectual libertarians who hand around the "open source" movement. But this is the worst I have ever seen, and it proves they are not intellectuals, they are rabid dogmatists.
The fire department DOES need to get paid for fighting fires, and when a non-insured home is on fire the "putting out an actual fire" service fee will need to be much higher than the "putting out a hypothetical fire" insurance fee. But the service fee can be paid OUT OF POCKET, or by putting a LIEN on the house. The cost to put the fire out was WAY less than the value of the home, and refusing to put it out ends up being a net destruction of wealth (not to mention the lives of the pets inside).
At this point the worst libertards say "well they didn't pay the insurance, so tough", and it completely escapes their consideration that MAYBE just MAYBE the fire department could have charged them $5000 or $10000 to put out the fire. An intellectual would see that alternate payment system as a good and just possibility, but a dogmatist doesn't need to think that hard because they already have all the answers.
Any economic system that fails to protect wealth and watches it burn away will lead to a society with LESS WEALTH. The fire department may have saved $1000 by not putting out the fire, but the society lost 200 times that when the home and its contents were consumed. If you don't have enough brains to improve upon that result, even within the libertarian worldview then you are an idiot, not a legitimate intellectual that people should bother listening to.
um yeah except the WHITE firefighters were the ones suing, after the CITY decided to throw out the results of the exam, and IIRC she was following precedent that the supreme court later reversed itself on........ so how did the supreme court reversing its own precedent become her fault? Oh yeah race-baiting fox news.
care to do a calculation of the LD50 of tritum in water?
For the love of god, tritium decays by beta particle emission. Why the boy-who-cried-wolf nuclear panic over a beta emitter?
There may be additional hacks you need. Specifically comment out the lines in soundon or whatever they called it that uninstalls the alsa drivers, or you'll have to re-install the kernel after it deletes them. And be aware you may have to strangle udev to get it to STOP LOADING THE GODDAMN ALSA DRIVERS FIRST.
Whoops, meant to say "instead of /etc/init.d/oss" during init.
Actually I have oss4 for hdaudio, and alsa for usb audio working together on debian. It is was a royal bitch tho. I run this script instead of oss soundon/soundoff
#!/bin/bash
set -x
if [ ! -d /lib/modules/$(uname -r)/oss ] /lib/modules/$(uname -r)/kernel/oss /lib/modules/$(uname -r)/oss
then
soundoff
soundon
mv
depmod -a
fi
rmmod oss_hdaudio /usr/sbin/ossdetect -d && /usr/lib/oss/etc/legacy_devices && /usr/sbin/ossdevlinks -v /usr/sbin/savemixer -L -v
rmmod osscore
modprobe osscore vmix_loopdevs=1 &&
sleep 1 &&
modprobe oss_hdaudio &&
sleep 1 &&
sleep 1 &&
sh
sleep 1 &&
sleep 1 &&
sleep 1 &&
modprobe snd_usb_audio
What corporation do you own 10% of that does not confer limited liability to you via the concept of corporate personhood? In order to sue you personally they'd have to "pierce the corporate veil" which doesn't happen very often. I think you're BSing here.
Um, dude. That statement by itself isn't racist. The policy that the statement is meant to defend by camouflaging it, RACIAL PROFILING, is racist. And the statement is wrong. Studies (that I do not have links for, sorry) have shown that after dark when the race of a person driving a car is harder to see the racial disparity in traffic stops decreases.
Of COURSE they don't know how much a voluntary insurance plan is going to cost, since they can't FORCE you to sign up for it! Blue Cross doesn't know how much their plan costs in advance either.
Oh jeez this sounds like tin-foil hat wearing conspiracy theories. Require. Force. Government Takeover. Give me a break!
constant fscking is the bug, not the intended behaviour, dude. If I had KNOWN it would fsck every reboot I would have done that. Now that people have heard my experience maybe they will choose that option.
I did, and the the cause of the fscks was...(drumroll)..ext4.
oh cool, thanks
Hey genius! It's called SARCASM! There is no need to insult me for sharing my experiences with ext4. I am not some idiot who thinks RAID is a backup either.
I am NOT happy with ext4, but in case you needed to know more about my system, my ext4 data is a MIRROR of the ext3 data, and I have a LTO-2 ultirum drive to backup my filesystem regularly. I have off-site backups in fireproof waterproof safes.
I also use e2image to get the metadata off the drive in case I need to reconstruct the contents.
Well, the fsck times are really fast compared to ext3, and thank god, because EVERY time I reboot it requires an fsck, complaining about group descriptor checksums. Even if I unmount my ext4 filesystem and remount it without rebooting it gets all fscked up. I have a 3TB ext4 fs on LVM on RAID, that was NOT converted from ext3, but built on brand new drives. My similar ext3 filesystem has had so such problems.
ext4 takes about 7 minutes to fsck, ext3 took hours. I hope they fix this soon.
neo geo games were maximum 330 mega BITS, not megabytes.
the 1 error per 512 bits is not an EXPECTED error rate. It's a CORRECTABLE error rate. The actual error rate is going to be far below that, but due to bursts will sometimes be above it, which is why the extra block length is good.
OK, yes you COULD move the parity dta around but you'd get shitty performance. Hard drives are made so that each sector is independent of another. That makes each sector a seperate codeword on disk. What you are proposing is to introduce dependency between sectors, and that would mean having to read adjacent sectors in order to write a single sector, which means goin through 2 revolution of the disk instead of one.
I guess I should pre-emptively point out that for a hard drive you want to be able to modify each sector atomically, which means that a single sector corresponds to a single codeword, and increasing areal density means you need longer codewords to maintain error correction. So either you decrease the rate of the code, and use extra redundancy, which lower capacity and defeats the purpose of increasing areal density, or you us longer codewords at the same rate, which means using longer sectors.
the rate of a code measure how much redundundacy it has, correct. But why do you think block length doesn't matter? Just because you have high redundancy doesn't mean your errors are going to magically be recoverable. To actually recover the data you need enough distance between valid codewords so that when a codeword is perturbed by errors you can still see which valid codeword it is closest to. With short block lengths you get small decoding distances, and low error correcting power. If you learn information theory a bit better you'll see Claude Shannon's channel "capacity" theory assumes infinite block length, and it does that for a REASON.
Yep threads are faster, but what about stability and security? If there's a buffer overflow attack against apache with pre-forking can the attacker see everbody else's credit card numbers at the end of the SSL stream where it's been decrypted? Nope. But they could see them if you're running the multi threaded server, since there is no protection. And if the multithreaded server has a single thread die, they all die.
ummm, because in dual core AMD chips the two cores talk over the SRQ, not over hypertransport links? How dumb do you think AMD is?