Netscape 4.x was ASS. It was okay, but unacceptably buggy and slow. What makes you think 5 was going to be any improvement? I'm assuming that they were still basing it on the original codebase (from version 1). They needed that new layout engine BADLY. Gecko ran circles around 4.x in rendering, none of that O(N^2) wait blowup if you picked the wrong nesting of tags.
Everyone would have hated 5 just as much as 4... they would still switch to IE.
It's sad, but I'm glad Netscape didn't try for a 5 before they switched codebases. I was like: stick a fork in it, it's done.
The use of ',' and ';'. Generally, you use ',' to indicate a list. Therefore it makes perfect sense it is used to create parameter lists, and when used as an operator, "returns" the expression to the right. The semicolon is often used at the end of a clause or list, therefore it does not defer to the right and thus is a suitable indicator for a logical break. I would think language programmers at some point flirted with the idea of using the period for an end-of-statement marker, but perhaps because it is also used as a decimal point in the ASCII character set, they might have been worried about determining whether a trailing decimal point after a numeric expression indicated end-of-statement or a decimal point missing the rest of it's significand.
Don't you have network jacks every few feet along the walls? Or in the floors in access panels?
That's how we do it here. We can move furniture, racks or equipment around, just replug the stuff in to whatever's closest. It's all on the same switch, so it's no big deal.
I just have my fingers crossed that my girlfriend gets her blog back," said software programmer Tom Gortell. "She feels like someone just sucked out her brains. I don't get it, it's just an online journal, right? But she feels like her entire life has been stolen."
And a prediction:
Researchers at University of Washington determine weblogs suck out bloggers souls
Posted by timothy on Wednesday June 16, @12:34AM from the just-like-cameras dept.
and moreso in rackmount external arrays with embedded controllers... whether fibre channel or scsi. Generally speaking you use RAID 5 because it gives you more flexibility and the unit manages all the details. Rebuiliding parity is transparent and painless.
Also, let me say that I would be uncomfortable with more than 6 drives in a RAID5, the chance of multiple failures increases, along with the size of the volume. But usually you have something that is dual channel, so you work with sets of 12. And right now 12 drives is the max you can fit in 2U that I've seen, so that's that right there. Thus I tend to think in groups of 6 when dealing with RAID 5 for those reasons...:P but you're right, it scales a bit better when you have more than one. Especially since 3Ware cards are not bad. An 8506-12 is like $700-$750. You trade two or three hard drives for the ability to have two RAID-5 volumes... not a bad deal.
Also, removing the extra wasted disks doesn't get you more storage, that's true. I was thinking along the lines that the storage capacity defines the RAID, not vice-versa. So if the RAID-1 configuration meets spec, then configuring it RAID-5 could reduce disk count. On the other hand, if you fix the configuration of the disks, then going RAID-5 makes sense if you can use the extra storage... (if the point is to maximize the storage, because you don't know the requirements yet).
And RAID-1 doesn't necessarily give you a performance advantage over RAID-5 either. Having N+M spindles versus N spindles makes little difference when N>3 (unless you have a massively parallel application, or really predicatable access... and a huge channel to the controller).
Anime, like any other consumable media, comes in three flavors:
1) Absolute shit served up with a marketing blitz 2) A valiant effort that self-destructs mid run (possibly because they ran out of money after the marketing blitz) 3) The overlooked gems.
I'm sure it'd be easy to make a similar list for any number of product areas:
Popular music, Reality TV, Feature Films, Websites, Taco Bell franchises, celebrity-sponsored hair care product lines, etc.
In this case you're spreading out the technology with new devices. All your sensors and coffee makers and whatnot still do only one thing (which is good, you have a clear division of responsibility... reduces complexity).
The computer controlling everything is still doing exactly what it is designed to do, be a generic computing platform that occaisionally interfaces with those annoying humans.
You have 6 bays in which to insert a (cold/hot) swap disk. Would you rather have 6*250GB/2 = 750GB of space, or (6-5)*250GB = 1250GB of space?
Keep in mind that in either case, if you lose a disk (doesn't matter which one), you're probably bringing the machine offline unless you're using hotswap (which you say is superflouous).
Get a decent RAID-5 hardware controller. Seriously.
Less wasted disks = less noise, less power, less heat, more room in the rack, and more storage.
I recommend an nStor 47*0s with 12 or so Seagate 200GB S-ATA drives (you want the ones that support command queuing... but 300GB Maxtors aren't bad either).
Get an Ultra-160/320 dual-port SCSI card, and hook that bad boy up.
Instant RAID-5 multi-terabyte storage that will utterly fly. About 8-9K, depending on where you get the constituent parts.
But there is no denser or faster configuration in existance for local storage (besides solid state)
The issue isn't that the context is gone... the issue is that the kernel is executing a non-waiting FPU instruction i.e. "fwait" on returning from the a context that flushes a user thread (i.e. return from signal handler, syscall after execve). Triggers the FPE, except the kernel isn't set up to handle FPEs properly from kernel space in this case. The problem is that the TS flag is set because it's switching tasks, so it receives a different exception, trap 7 (device_not_available). The purpose of that exception is to signal the kernel that a newly created process wants the FPU. So it attempts to set up the FPU... which ends up calling __clear_fpu again... heh... and the original exception isn't cleared yet... whoops.
What's really weird is I found this document, which details the potential problems of trying to use the FPU in a interrupt handler in the Linux kernel.
They brought up the potential of triggering this EXACT PROBLEM... quote "endless trap 7 activation"... only in this case they're talking about writing an interrupt routine, not returning from a signal handler. Still, they already discovered this misbehavior...
Well, you can't really call it that, though. It's was sort of by design (to make task switching faster). But the thing is you have to be ABSOLUTELY SURE that you never raise an FPE when TS is set, and you're NOT a user thread. That's what gets you burned here.
After poking around in the LKML, I've mostly figured it out. The kernel wasn't handling floating point exceptions correctly in the signal handler. The problem is that if the exception is triggered by the LAST instruction in the handler, the exception is attempted to be delivered to a signal context which no longer exists. The same thing was happening with execve... if you triggered it right before the execve syscall, the application context would be destroyed, and the pending exception would be pointing to a non-existant instruction. The exception handler would jump off into space trying to deliver SIGFPE...
So they changed __clear_fpu (which is called when doing a initial switch back to user space [I think]) to clear any pending FPU exceptions, because there was no way they could be handled anyway.
Missing an FPU exception doesn't sound so bad. I think someone was posting a better solution, which would attempt to handle it the right way... (I didn't really follow the more extensive patch, anyone care to explain?)
The article says it affects x86 (and x86-64) only.
So itanium, ppc, etc. are safe. But my other questions still remain.
Note that the person who reported the bug thought they were triggering a gcc bug. As it turns out, he munged his FPU assembly instructions. The GCC people rightly told him to contact the lkml... it's definitely an exception handling issue.
... that if you trigger a floating point exception inside a signal handler (specifically SIGALRM), the kernel doesn't handle it correctly, hanging the system. It appears to affect both SMP and UP kernels.
Some questions I have to those who may have been following this:
Does the crash occur without the syscalls in the signal handler/main process? Does the crash occur on SMP machines? Does the crash occur with other signals (PIPE, USR1, etc.) Does the crash occur on ppc, sparc, etc?
The textures, lighting effects, reflection maps, etc. are all pre-created (did you ever wonder why 3d Mark is a 250+ MB download?).
That's not nearly the same kind of thing.
POV-Ray generates the shadows, lighting, and often the textures, right on the spot. So a very short input file can generate a realistic 3d scene. And it does it using raytracing, so you get _real_ reflections, bump maps, and shadows. None of this is being done with GPUs, they still use precalulated texture maps and a scanline renderer (which is fine for interactive presentation and games... but again, all the hard work is done up front by the creator of the game/application/etc.)
Slashdot Mirror
Jesus, you people sound like whiney punks on the playground, counting and listing all your friends.
You're clearly better than us, what with your preachy signature and your goldplated nickname.
I've got vomity butterflies from laughing so hard.
Because in *nix, most every editor out there has a way to use ispell or aspell to check your document.
You can also run ispell/aspell on text document outside the editor... it has interactive search/replace.
Netscape 4.x was ASS. It was okay, but unacceptably buggy and slow. What makes you think 5 was going to be any improvement? I'm assuming that they were still basing it on the original codebase (from version 1). They needed that new layout engine BADLY.
Gecko ran circles around 4.x in rendering, none of that O(N^2) wait blowup if you picked the wrong nesting of tags.
Everyone would have hated 5 just as much as 4... they would still switch to IE.
It's sad, but I'm glad Netscape didn't try for a 5 before they switched codebases. I was like: stick a fork in it, it's done.
Cake is funny. Muffin? Not funny.
The use of ',' and ';'. Generally, you use ',' to indicate a list. Therefore it makes perfect sense it is used to create parameter lists, and when used as an operator, "returns" the expression to the right.
The semicolon is often used at the end of a clause or list, therefore it does not defer to the right and thus is a suitable indicator for a logical break.
I would think language programmers at some point flirted with the idea of using the period for an end-of-statement marker, but perhaps because it is also used as a decimal point in the ASCII character set, they might have been worried about determining whether a trailing decimal point after a numeric expression indicated end-of-statement or a decimal point missing the rest of it's significand.
It supports WEP (coming soon, WPA), and monitor mode. Wheeeee!!!
Don't you have network jacks every few feet along the walls?
Or in the floors in access panels?
That's how we do it here. We can move furniture, racks or equipment around, just replug the stuff in to whatever's closest. It's all on the same switch, so it's no big deal.
Just a little FYI... also, so is ECMA-script, and the CIL, IIRC.
I just have my fingers crossed that my girlfriend gets her blog back," said software programmer Tom Gortell. "She feels like someone just sucked out her brains. I don't get it, it's just an online journal, right? But she feels like her entire life has been stolen."
And a prediction:
Researchers at University of Washington determine weblogs suck out bloggers souls
Posted by timothy on Wednesday June 16, @12:34AM
from the just-like-cameras dept.
and moreso in rackmount external arrays with embedded controllers... whether fibre channel or scsi. Generally speaking you use RAID 5 because it gives you more flexibility and the unit manages all the details. Rebuiliding parity is transparent and painless.
:P but you're right, it scales a bit better when you have more than one.
Also, let me say that I would be uncomfortable with more than 6 drives in a RAID5, the chance of multiple failures increases, along with the size of the volume. But usually you have something that is dual channel, so you work with sets of 12. And right now 12 drives is the max you can fit in 2U that I've seen, so that's that right there.
Thus I tend to think in groups of 6 when dealing with RAID 5 for those reasons...
Especially since 3Ware cards are not bad. An 8506-12 is like $700-$750. You trade two or three hard drives for the ability to have two RAID-5 volumes... not a bad deal.
Also, removing the extra wasted disks doesn't get you more storage, that's true. I was thinking along the lines that the storage capacity defines the RAID, not vice-versa. So if the RAID-1 configuration meets spec, then configuring it RAID-5 could reduce disk count.
On the other hand, if you fix the configuration of the disks, then going RAID-5 makes sense if you can use the extra storage... (if the point is to maximize the storage, because you don't know the requirements yet).
And RAID-1 doesn't necessarily give you a performance advantage over RAID-5 either. Having N+M spindles versus N spindles makes little difference when N>3 (unless you have a massively parallel application, or really predicatable access... and a huge channel to the controller).
Anime, like any other consumable media, comes in three flavors:
1) Absolute shit served up with a marketing blitz
2) A valiant effort that self-destructs mid run (possibly because they ran out of money after the marketing blitz)
3) The overlooked gems.
I'm sure it'd be easy to make a similar list for any number of product areas:
Popular music, Reality TV, Feature Films, Websites, Taco Bell franchises, celebrity-sponsored hair care product lines, etc.
In this case you're spreading out the technology with new devices. All your sensors and coffee makers and whatnot still do only one thing (which is good, you have a clear division of responsibility... reduces complexity).
The computer controlling everything is still doing exactly what it is designed to do, be a generic computing platform that occaisionally interfaces with those annoying humans.
...the 51st state is because then we couldn't continue importing cheap marijuana from British Columbia... US drug laws are a bitch.
Also, I think we'd be hard pressed to allow a Socialist party on the ballot.
Three parties? That'd just confuse people.
just download it to your harddrive, recode it to MPEG4 (or MPEG2) and burn it on a DVD.
Get one of those set top player deals from the Asian market that can play MPEG4 and read DVDs. No problem!
Let me put it to you simply:
You have 6 bays in which to insert a (cold/hot) swap disk. Would you rather have 6*250GB/2 = 750GB of space, or (6-5)*250GB = 1250GB of space?
Keep in mind that in either case, if you lose a disk (doesn't matter which one), you're probably bringing the machine offline unless you're using hotswap (which you say is superflouous).
Get a decent RAID-5 hardware controller. Seriously.
Less wasted disks = less noise, less power, less heat, more room in the rack, and more storage.
I recommend an nStor 47*0s with 12 or so Seagate 200GB S-ATA drives (you want the ones that support command queuing... but 300GB Maxtors aren't bad either).
Get an Ultra-160/320 dual-port SCSI card, and hook that bad boy up.
Instant RAID-5 multi-terabyte storage that will utterly fly. About 8-9K, depending on where you get the constituent parts.
But there is no denser or faster configuration in existance for local storage (besides solid state)
Don't listen to those other ones... they're crazy.
God I wish I could edit posts.
The issue isn't that the context is gone... the issue is that the kernel is executing a non-waiting FPU instruction i.e. "fwait" on returning from the a context that flushes a user thread (i.e. return from signal handler, syscall after execve). Triggers the FPE, except the kernel isn't set up to handle FPEs properly from kernel space in this case. The problem is that the TS flag is set because it's switching tasks, so it receives a different exception, trap 7 (device_not_available). The purpose of that exception is to signal the kernel that a newly created process wants the FPU. So it attempts to set up the FPU... which ends up calling __clear_fpu again... heh... and the original exception isn't cleared yet... whoops.
What's really weird is I found this document, which details the potential problems of trying to use the FPU in a interrupt handler in the Linux kernel.
They brought up the potential of triggering this EXACT PROBLEM... quote "endless trap 7 activation"... only in this case they're talking about writing an interrupt routine, not returning from a signal handler. Still, they already discovered this misbehavior...
Well, you can't really call it that, though. It's was sort of by design (to make task switching faster). But the thing is you have to be ABSOLUTELY SURE that you never raise an FPE when TS is set, and you're NOT a user thread. That's what gets you burned here.
The article doesn't attempt to explain anything.
(Someone please correct me if I have this wrong)
After poking around in the LKML, I've mostly figured it out.
The kernel wasn't handling floating point exceptions correctly in the signal handler. The problem is that if the exception is triggered by the LAST instruction in the handler, the exception is attempted to be delivered to a signal context which no longer exists. The same thing was happening with execve... if you triggered it right before the execve syscall, the application context would be destroyed, and the pending exception would be pointing to a non-existant instruction. The exception handler would jump off into space trying to deliver SIGFPE...
So they changed __clear_fpu (which is called when doing a initial switch back to user space [I think]) to clear any pending FPU exceptions, because there was no way they could be handled anyway.
Missing an FPU exception doesn't sound so bad. I think someone was posting a better solution, which would attempt to handle it the right way... (I didn't really follow the more extensive patch, anyone care to explain?)
The article says it affects x86 (and x86-64) only.
So itanium, ppc, etc. are safe. But my other questions still remain.
Note that the person who reported the bug thought they were triggering a gcc bug. As it turns out, he munged his FPU assembly instructions.
The GCC people rightly told him to contact the lkml... it's definitely an exception handling issue.
... that if you trigger a floating point exception inside a signal handler (specifically SIGALRM), the kernel doesn't handle it correctly, hanging the system. It appears to affect both SMP and UP kernels.
Some questions I have to those who may have been following this:
Does the crash occur without the syscalls in the signal handler/main process?
Does the crash occur on SMP machines?
Does the crash occur with other signals (PIPE, USR1, etc.)
Does the crash occur on ppc, sparc, etc?
The textures, lighting effects, reflection maps, etc. are all pre-created (did you ever wonder why 3d Mark is a 250+ MB download?).
That's not nearly the same kind of thing.
POV-Ray generates the shadows, lighting, and often the textures, right on the spot. So a very short input file can generate a realistic 3d scene. And it does it using raytracing, so you get _real_ reflections, bump maps, and shadows. None of this is being done with GPUs, they still use precalulated texture maps and a scanline renderer (which is fine for interactive presentation and games... but again, all the hard work is done up front by the creator of the game/application/etc.)