Can you promise that no SIMD scatter can be performed with offsets that it shouldn't?
Yes I can. For instance in ATI r6xx it can only go to a surface defined by SX_MEMORY_EXPORT_BASE/SX_MEMORY_EXPORT_SIZE described on page 127 here http://www.x.org/docs/AMD/R6xx_3D_Registers.pdf In addition to that system memory is mapped to GPU via GPU VM page table, so only pages that were allocated by the process and that the kernel driver mapped into VM graphic context. See/usr/src/linux/drivers/gpu/drm/radeon and grep for VM_ So there are two layers of hardware enforced protection in addition to software command buffer parser that checks the addresses. Safe enough for you?
First, WebGL sends shader source code to the browser and the code is compiled and executed in OpenGL. This is no different from running any other OpenGL program on your machine. The remote attacker cannot make the GPU execute arbitrary hardware instructions, only whatever source he sends. The shaders pretty much execute in a sandbox (shader on GPU can only access buffers bound textures, vertex buffers, constant buffers, render targets etc etc). The access outside these buffers is not possible because the hardware enforces it (there is no way to even address outside texture or render target). It is little more complicated with compute shaders which have little more flexible addressing but they still cannot access anything outside global buffer (or OpenCL address space). It is like segment based protection in CPUs.
Latest GPUs have actual page table and VM, so on top of security protection from "segment" based addressing, there is also VM/page table based protection which only allows particular GPU context to access pages that have been allocated and mapped into it's VM.
The only real problem is a possibility of DOS attack caused by the fact that GPUs are not preemptable. Therefore if you send some complicated geometry or you write a shader that takes a very long time to execute (multiple nested loops+many pixels/vertices or compute threads) the draw can execute for a very long time. On Vista and later this will cause TDR and kill the trouble process. It happens all the time if you develop games or GPU compute apps. The only way to disable the watchdog is with a registry setting. On XP the watchdogs are implemented in the kernel part of graphic driver (ATI VPU Recover, and whatever nVidia has). This DOS is a little more problem in Linux since it doesn't have good watchdogs, the DOS should not crash X, but it will definitely lock the UI. Also, if you will bother to take look at the shader docs for AMD http://www.x.org/docs/AMD/r600isa.pdf you will see that the instruction set does not allow for truly infinite shaders, there are no arbitrary jumps, the loops cannot run forever (max loop count is 2^31), the flow control is only structured and easily verifiable. It is different for nVidia ISA which looks more like regular CPU and I think can do infinite loops.
On the upside Linux DRM drivers in kernel have pretty good command buffer parsers and validators, so it is hard for user-space driver to access memory that doesn't belong to it. On Vista and later the user-space driver doesn't even know GPU side addresses of its allocations and sends every render buffer with an allocation and patch list which is resolved and patched by VidMM and kernel mode driver, see D3DKMTRender function etc http://msdn.microsoft.com/en-us/library/ff547145%28v=vs.85%29.aspx
but then they wouldn't know where you are. I bet this "electronic device" will also report your location in realtime to the government for "bookkeeping".
Desire HD maybe, but Inspire 4G (same thing for ATT, little different radio) is locked shut. The process of unlocking it is long as well (downgrade to Desire HD ROM, and then exploting DHD radio).
I have bought collector's edition when it came out only because they promised a Linux port. So fuck you Sweeney. BTW, it sucked big time, UT3 is a lagfest.
Why does Goofle need two operating systems? I just don't see the appeal. Android makes some sense (save for Java, which fortunately came to bite Goofle in the ass, so maybe it will teach them something). It works, it runs on mobile devices and can scale to bigger machines and runs crap integrated with Goofle cloud. At least Goofle is not afraid to kill products (see Wave). So maybe they will get rid of it.
are texting like crazy and their thumbs are getting red and bleeding in the vain attempt to have 3x more chance to have sex. They all forget that 3 * 0 = 0.
return their 16M IPv4 addresses, just look at the map http://xkcd.com/195/ HP, DEC, Ford, Xerox, Bell Labs, Apple, MIT, USPS, DuPont, IBM, General Electric, Boeing, Prudential, Eli Lily, Halliburton. Why does plane, car, drug or chemical manufacturer or an insurance company need 16M publicly routable IP addresses? I guess HP has now all the DEC IPs, so they have 32M, WTF!
Just CmdrTaco trolling for ad impressions From TFA:
“slightly rewriting the Linux code so that each core kept a local count, which was only occasionally synchronized with those of the other cores, greatly improved the system’s overall performance.”
“The fact that that is the major scalability problem suggests that a lot of things already have been fixed. You could imagine much more important things to be problems, and they’re not. You’re down to simple reference counts.” Kaashoek said. “Our claim is not that our fixes are the ones that are going to make Linux more scalable,”
It is really hard to argue that selling additional copies of software will create more jobs. Maybe little with packing of software boxes and tech support. Otherwise all extra copies of software sold are pure profit. All it achieves is to transfer money from software users to software companies shaderholders.
That also holds your prescriptions etc. It is very easy for the computer to check all the interaction possibilities. Not so much for a human, no matter how well trained.
Central login by definition links your multiple accounts to a single identity. In most cases it is not a problem. But do you really want somebody to know you login with the same ID to you bank, health insurance and pr0n site? I don' think so. I'd prefer to have several identities on-line. One for secure stuff (bank, financial, medical info etc), one for shopping, one for unimportant stuff like forums, diggs, facespaces etc and one or many for things that I may not be so proud off like pr0n sites. The quality of the passwords I use on these tiers of logins should be appropriate for the importance of the account.
The problem is nobody owns it. This is what got GRUB developers in trouble. It is just there as an artifact of aligning first partition to full cylinder. Which is not requirement either, fdisk just did it so then everyone else followed. Since nobody owns it and it is not specified anywhere it has become free for all to mess with it. And hilarity ensued.
Tracks were on floppies on Amiga;-) PC has Sectors, Heads and Cylinders -> http://en.wikipedia.org/wiki/Cylinder-head-sector Or LBA (Logical Block Address) sector number. That's the way system worked for decades.
It is not in any "specs" Most partitioning programs including original DOS fdisk aligned the first partition the next full cylinder. There is actually no need for this because there is a starting sector field in MBR partition table so first partition can start right after the MBR and there wouldnt be space to put anything after MBR. See http://en.wikipedia.org/wiki/Master_boot_record If you really want to make your bootloader bigger than a single sector, create a small partition, put the rest of the bootloader code there. Make MBR load from the rest of your parition and then look at the real active partition and boot by default from there. Not that hard and completely safe.
While MBR has some function, the rest of sectors between MBR and the first partition was always a great area. Many MBR viruses put their stuff there. Many stupid programs use it to store DRM data, so they can check whether they were copied to other computer If GRUB is using this region too, it is equally stupid. There is no protocol for allocating this area and there is no guarantee that this data is not going to be overwritten by any other stupid program. So nothing to see here, move aling, it is just Core Wars between stupid programs. GRUB developers should have known better.
Does it have to refresh constantly and run at 60fps? The way I use debugger, most of the time it waits for input from me. My biggest gripe with gdb is how awkward to use it is and how crappy all visual overlays for gdb are, bot how fast it is.
It makes it much easier to notice how dumb people are without even leaving your home. So while, overall it may have positive influence, it also makes you lose faith in humanity.
Yeah, somehow I fail to be excited with it. But, if you look at all these Android tablet devices that begin to be available right now, they are Android 1.5 or 1.6. How pathetic.
Slashdot Mirror
Can you promise that no SIMD scatter can be performed with offsets that it shouldn't?
Yes I can. For instance in ATI r6xx it can only go to a surface defined by SX_MEMORY_EXPORT_BASE/SX_MEMORY_EXPORT_SIZE described on page 127 here http://www.x.org/docs/AMD/R6xx_3D_Registers.pdf /usr/src/linux/drivers/gpu/drm/radeon and grep for VM_
In addition to that system memory is mapped to GPU via GPU VM page table, so only pages that were allocated by the process and that the kernel driver mapped into VM graphic context. See
So there are two layers of hardware enforced protection in addition to software command buffer parser that checks the addresses. Safe enough for you?
First, WebGL sends shader source code to the browser and the code is compiled and executed in OpenGL. This is no different from running any other OpenGL program on your machine. The remote attacker cannot make the GPU execute arbitrary hardware instructions, only whatever source he sends.
The shaders pretty much execute in a sandbox (shader on GPU can only access buffers bound textures, vertex buffers, constant buffers, render targets etc etc). The access outside these buffers is not possible because the hardware enforces it (there is no way to even address outside texture or render target). It is little more complicated with compute shaders which have little more flexible addressing but they still cannot access anything outside global buffer (or OpenCL address space). It is like segment based protection in CPUs.
Latest GPUs have actual page table and VM, so on top of security protection from "segment" based addressing, there is also VM/page table based protection which only allows particular GPU context to access pages that have been allocated and mapped into it's VM.
The only real problem is a possibility of DOS attack caused by the fact that GPUs are not preemptable. Therefore if you send some complicated geometry or you write a shader that takes a very long time to execute (multiple nested loops+many pixels/vertices or compute threads) the draw can execute for a very long time. On Vista and later this will cause TDR and kill the trouble process. It happens all the time if you develop games or GPU compute apps. The only way to disable the watchdog is with a registry setting. On XP the watchdogs are implemented in the kernel part of graphic driver (ATI VPU Recover, and whatever nVidia has).
This DOS is a little more problem in Linux since it doesn't have good watchdogs, the DOS should not crash X, but it will definitely lock the UI. Also, if you will bother to take look at the shader docs for AMD http://www.x.org/docs/AMD/r600isa.pdf you will see that the instruction set does not allow for truly infinite shaders, there are no arbitrary jumps, the loops cannot run forever (max loop count is 2^31), the flow control is only structured and easily verifiable. It is different for nVidia ISA which looks more like regular CPU and I think can do infinite loops.
On the upside Linux DRM drivers in kernel have pretty good command buffer parsers and validators, so it is hard for user-space driver to access memory that doesn't belong to it. On Vista and later the user-space driver doesn't even know GPU side addresses of its allocations and sends every render buffer with an allocation and patch list which is resolved and patched by VidMM and kernel mode driver, see D3DKMTRender function etc http://msdn.microsoft.com/en-us/library/ff547145%28v=vs.85%29.aspx
but then they wouldn't know where you are.
I bet this "electronic device" will also report your location in realtime to the government for "bookkeeping".
Desire HD maybe, but Inspire 4G (same thing for ATT, little different radio) is locked shut. The process of unlocking it is long as well (downgrade to Desire HD ROM, and then exploting DHD radio).
everyone selling the bulbs as heaters or for "industrial use".
I have bought collector's edition when it came out only because they promised a Linux port.
So fuck you Sweeney.
BTW, it sucked big time, UT3 is a lagfest.
Why does Goofle need two operating systems?
I just don't see the appeal.
Android makes some sense (save for Java, which fortunately came to bite Goofle in the ass, so maybe it will teach them something). It works, it runs on mobile devices and can scale to bigger machines and runs crap integrated with Goofle cloud.
At least Goofle is not afraid to kill products (see Wave). So maybe they will get rid of it.
I think cloud gaming sucks so if there will be only one company doing it for next 17 years, it's for the better.
are texting like crazy and their thumbs are getting red and bleeding in the vain attempt to have 3x more chance to have sex. They all forget that 3 * 0 = 0.
return their 16M IPv4 addresses, just look at the map
http://xkcd.com/195/
HP, DEC, Ford, Xerox, Bell Labs, Apple, MIT, USPS, DuPont, IBM, General Electric, Boeing, Prudential, Eli Lily, Halliburton.
Why does plane, car, drug or chemical manufacturer or an insurance company need 16M publicly routable IP addresses?
I guess HP has now all the DEC IPs, so they have 32M, WTF!
Just CmdrTaco trolling for ad impressions
From TFA:
“slightly rewriting the Linux code so that each core kept a local count, which was only occasionally synchronized with those of the other cores, greatly improved the system’s overall performance.”
“The fact that that is the major scalability problem suggests that a lot of things already have been fixed. You could imagine much more important things to be problems, and they’re not. You’re down to simple reference counts.” Kaashoek said. “Our claim is not that our fixes are the ones that are going to make Linux more scalable,”
It is really hard to argue that selling additional copies of software will create more jobs. Maybe little with packing of software boxes and tech support. Otherwise all extra copies of software sold are pure profit. All it achieves is to transfer money from software users to software companies shaderholders.
That also holds your prescriptions etc.
It is very easy for the computer to check all the interaction possibilities. Not so much for a human, no matter how well trained.
Central login by definition links your multiple accounts to a single identity. In most cases it is not a problem. But do you really want somebody to know you login with the same ID to you bank, health insurance and pr0n site? I don' think so. I'd prefer to have several identities on-line. One for secure stuff (bank, financial, medical info etc), one for shopping, one for unimportant stuff like forums, diggs, facespaces etc and one or many for things that I may not be so proud off like pr0n sites. The quality of the passwords I use on these tiers of logins should be appropriate for the importance of the account.
The problem is nobody owns it. This is what got GRUB developers in trouble. It is just there as an artifact of aligning first partition to full cylinder. Which is not requirement either, fdisk just did it so then everyone else followed.
Since nobody owns it and it is not specified anywhere it has become free for all to mess with it. And hilarity ensued.
Tracks were on floppies on Amiga ;-)
PC has Sectors, Heads and Cylinders -> http://en.wikipedia.org/wiki/Cylinder-head-sector
Or LBA (Logical Block Address) sector number.
That's the way system worked for decades.
It is not in any "specs"
Most partitioning programs including original DOS fdisk aligned the first partition the next full cylinder. There is actually no need for this because there is a starting sector field in MBR partition table so first partition can start right after the MBR and there wouldnt be space to put anything after MBR. See http://en.wikipedia.org/wiki/Master_boot_record
If you really want to make your bootloader bigger than a single sector, create a small partition, put the rest of the bootloader code there. Make MBR load from the rest of your parition and then look at the real active partition and boot by default from there. Not that hard and completely safe.
While MBR has some function, the rest of sectors between MBR and the first partition was always a great area.
Many MBR viruses put their stuff there. Many stupid programs use it to store DRM data, so they can check whether they were copied to other computer
If GRUB is using this region too, it is equally stupid. There is no protocol for allocating this area and there is no guarantee that this data is not going to be overwritten by any other stupid program.
So nothing to see here, move aling, it is just Core Wars between stupid programs.
GRUB developers should have known better.
Does it have to refresh constantly and run at 60fps?
The way I use debugger, most of the time it waits for input from me. My biggest gripe with gdb is how awkward to use it is and how crappy all visual overlays for gdb are, bot how fast it is.
to make him semi famous again. Maybe somebody will start making fun of him again.
Did he say how he wants to change the world?
If for better, I'd say he is not doing too well.
but he never cries.
a space horror movie. Or some under water slasher/ghost ship movie.
It just could not end well with a name like that.
It makes it much easier to notice how dumb people are without even leaving your home.
So while, overall it may have positive influence, it also makes you lose faith in humanity.
Yeah, somehow I fail to be excited with it.
But, if you look at all these Android tablet devices that begin to be available right now, they are Android 1.5 or 1.6.
How pathetic.