Interesting. Does it also supporting using the logic for system on a chip features, like implementing the memory controller? (Obviously in addition to the main draw of reconfigurable accelerators and co-processors.)
I ask because most of Xilinx's past offerings with hard CPU cores were aimed mostly at creating an System on a chip, while still potentially supporting reconfigurable accelerators if you could pull it off (partial reconfiguration is rather tricky[1]).
[1] Of course using partial reconfiguration as a way of live patching a system with negligible downtime can be even trickier than using it for replaceable accelerators that are inactive while reconfiguring, but people have still pulled off the former feat.
Many of us had thought about it, but it is very tricky to pull off. First standardizing the FPGA frabric is essential, or apps would need a different partial config bitstream for each CPU manufacturer. Further, standard FPGA frabric is not optimized for minimizing reconfiguration time.
In many setups, we are talking ten or hundreds of milliseconds. Not bad for starting an application, but unless apps claim exclusive us of the FPGA, then to quote Andy Dodd's post above: "context switching would be a real bitch".
Further, to get maximum possible performance likely requires exposing parts of the CPU to the fabric. A malicious application could potentially permanently damage the processor unless special step are taken. For example, the outputs of the CPU would need to be connected to FPGA pads that cannot be configured to be output pads, or a program could effectively short VDD and GND, literally causing the chip to fry itself.
Further, if great care is not taken, the FPGA configuration could deliberately time its outputs to change as the CPU clock ticks, attempting to introduce metastability to the processor. The processors are designed to be resistant to inputs randomly causing metastability of an input latch, but continuous deliberate attempts to introduce metastability are not usually factored into CPU design.
Most FPGAs in the field are coupled with EEPROMs that are triggered by the FPGA's own startup sequence, and emit a bitstream that is basically exactly the same as the data that would be sent if programing via JTAG. The FPGA start-up logic just transfers this data directly onto the interior scan chain that holds the configuration.
The chips invariably support other options too, such as the already mentioned JTAG based programming, and sometimes more complicated schemes.
There also exist a few that have integrated the EEPROM into the same package as the FPGA (probably not on the same die, but it could potentially be.
They do suck, but you get used to the ways they suck after a while. For example, I found it hard to move to a different synthesis tool from XST, because the other ones lacked some nice little convenience features of XST, despite the better RTL-level synthesis of the alternative tools.
I'm sure I'd be driven batty if I ever tried to use an Altera FPGA, simply because some of the features I'm used to will not be there, even though the features are emphatically non-essential. My mental model of an FPGA is also based heavily on the Virtex and Spartan designs, and it would take some work to adjust them.
Take something like a digital audio recorder as the core, and add a walkman cassett head, and peice them together with a few passive components, and you have a simple, cheap and effective device to skim credit cards.
Later you download the recorded audio (it is a Digital audio recorder) and run it through say a quick matlab script, and you decode the card data.
Hmm... Odd considering that from a group of graduation seniors form another good program (not quite at the level of say Cornell, but far, far better the average state school) had approximately $40k to $50k average. If the starting salaries are differing that much between schools, something has gone horribly wrong. Some amount differing should be expected, but a difference that large is downright disturbing.
Wikileaks has leaked plenty about other governments, as well as corporations local and foreign. Heck it used to be the case that the majority was non-US. I've not looked in a while, but that could still be the case.
The problem is that its the US leaks that get all the Attention in the US news (not to mention that much of what they have on other governments just is not as "juicy" as some of the US leaks).
If you have have a copy of the Taliban's membership roster, or North Koera's nuclear plans, or something similar, Wikileaks would be more than happy to publish that.
Please keep in mind that that is the average salary counting those students who got a M.Eng degree, which is a joke of a degree even at good schools like Cornell. That degree exists soley for the ability to claim to have a master's Degree.
Furthermore, I tend to doubt the numbers reported there are actually representative starting salaries, as the vast majority of companies I've seen do not have starting salaries that high. I'm wondering how many people responded to their survey giving their expected salary in a few years, or something to that effect. Either that or a few people got million dollar a year jobs through a "friend of the family".
It seems unlikely that a granularity of 1/80 of a mile is really needed. Are there frequently more than one railway location in a mile? I mean if not, why not just number them by mile? I mean the US does that for exists on major roadways. (I admit that it would be stupid for the UK to change now, except to metric.)
As for money, that is one area we will likely never see real metric conversion. After all names like "25 centiPound coin", or "deciDollar" sound terrible, not to mention the idea of spending several "kiloEuro" on something expensive.
Quite simply though, the better way is to simply replace ICANN with something international, and under no country's jurisdiction. Say a new UN treaty organization, or something similar.
An even better (although unfeasible) method would be to kill off the GTLDs, making US control of ICANN moot, since ICANN could not screw with any country's ccTLD, since that literally could be considered an act of war (overtly attempting to sabotage the infrastructure of another country), and without GTLDs only the ccTLDs matter.
It is rather interesting to look at how Google's results are structured. For example, the list of results is actually an HTML ordered list. Each search result is a list item. Prior to the ordered list, there is a hidden level 2 heading of "search results".
The only thing I would object to is the fact that additional OneBox-style items may appear at the bottom of the page, inside the ordered list that contains search results. That should be fixed, as those are not actually results. Fortunately the ones down there are usually suggesting different types of searches, like video or news searches, so most people will not mistake them for results.
That is definitely not even remotely standard for the US. It is a UK thing being a holdover from prior times since prior to 1985 the health care industry did use stones as part of measuring weight.
But the average American's reaction to hearing that somebody weighs 12 stones would be confusion. After all, to different stones (chunks of rock) don't weight the same.
Even well educated Americans would not need to pause for a moment upon hearing that. They would know of the unit, but may or may not know how large it is. If they do, they would still need to mentally convert it to pounds.
He is saying that he still expects Gmail, the Android Market, and the other Google Apps will continue to ship on the phones by default. I agree. All the market availability does is enable app updates more frequently than OS upgrades.
I will also dispute the Google apps being the core apps. The core apps are those like Contacts and Home, which would be present in an Android build you compiled from source.
Alternative root systems have never caught on, because nobody wants a domain that most people cannot resolve.
What you describe would still only have one root zone. By mirroring the zone files you end up with only one zone. All you are really proposing is splitting up policy making for gTLDs, meaning that some gTLDs will be Euro ones, subject to EUCANN policies, etc, and others will be US gTLDs, subject to ICANN policies.
The group feature is already there for those that want it, but it is not enabled by default in most distro kernels.
They create a hierarcy. The scheduler looks at schedulable items, not caring if they are threads or groups, and uses its normal algorithms (including nice levels) to decide which to run next. If it chooses a group, it repeats this process for the group, and of course, all time used by processes or subgroups count towards the group's running time. It repeats until it gets an actual thread, and runs that.
I'm not sure how one sets the nice level for a group though. I only learned about this feature yesterday when reading the LKML thread.
Heck most of the time you should be able to get away with an a pair of omnidirectional antennas, one in the Faraday cage (house) and one outside connected by RG6. A high gain directional should really only be needed if you also have pretty poor signal outside the house.
Hold on a second. Let's say that the repeater has a 3 dB gain (gain of 2) in both uplink and downlink. The phone will adjust its transmit gain to be just loud enough that the closest tower can hear, just like always, no? So the towers should be getting a signal no stronger than if the building simply were not there.
In all reality though, most people only need a a gain of 1 (0 dB). The real key is to get the signal through the wall without the associated losses. Heck, many people could get by with a pair of identical antennas connected by coax, despite that having a gain of significantly less than 1.
There cannot be multiple root zones. That just is not feasible. But the US only controls the gTLDs and one ccTLD. The other TLDs are owned by their respective countries. The US cannot have them removed from the root without causing a major incident, the result of which would include the Internet Society appointing some other organization as IANA (and thus killing ICANN and US control of the DNS).
Explanation: ICANN exists solely to be a policy creating shell around the IANA. Legally the Internet Society (ISOC) is the only company in the world with any reasonable claim to being able to appoint an IANA. The organizations the ISOC consists of (IETF, IESG, IAB, RFC-EDITOR) had agreed to allow ICANN to have the role of IANA, but they could revoke that agreement.
If an RFC is published naming a new international entity as the IANA there is little reason to suspect the companies running the root servers would not use the root zone published by the new IANA, as long as they got similar say in policy making as they had with ICANN.
Sure. Hopefully you know what a TTY is, but in case you don't it it a virtual or real terminal. When you open up an xterm you create one. If you don't have x-windows installed, you reach one, etc.
Well Linus had an idea about using a grouping functionality that was already in the Kernel to allow all the processes (technically actually all the kernel threads) running from one TTY to be grouped together for scheduling.
The result of that is that if you are running 99 processes in one xterm that could consume all of your CPU, and you open another xterm, one one just one process that wants 100% CPU, each xterm's processes gets 50% of the CPU, rather than one getting 99% and the other getting only 1%.
But lets say you only had that first xterm. Since each of those processes are not getting nearly the processor amount they desire, normally the scheduler sees them as nearly starved, and the next process that only wants 5% of CPU does not get much preferential treatment for giving up most of it's time. However, with the grouping, the scheduler can see that those 99 processes are related, and they are not really starved, since as a group they are getting 100%. So now when this other app that wants only 5% comes along, the scheduler might give it pretty much all of that 5% rather than the mere 1% it would have been getting before, and so that app (probably a web browser or something) remains nice and responsive.
That is not 100% accurate, since I've simplified some things a little, especially with regard to the working of the scheduler, but it should give you the idea. Eventually, more heuristics might be added, so that a GUI application that launches a bunch of threads and hogs the CPU might have all it's threads grouped, so they don't hurt responsiveness of interactive apps either.
I went to a a college much smaller than the one in question, and virtually all of my teachers were always in the room the whole test. There were a few exceptions though, mostly with the small senior classes that had 6-15 students. In that case the teachers realized that no student could get away with cheating unless everybody colluded, and nobody told him/her.
The prisoner's dilemma in exact reverse, but with more participants. Normally game theory would say that everybody would go for it, but once you add ethics you cannot expect 100% participation, so the best course of action is to not cheat.
That clause says the Constitution, federal laws, and treaties overrule state constitutions and law. to interpret that last "Constitution" to mean the Federal Constitution would require that Federal Laws override the Federal Constitution, which is absurd.
Slashdot Mirror
Interesting. Does it also supporting using the logic for system on a chip features, like implementing the memory controller? (Obviously in addition to the main draw of reconfigurable accelerators and co-processors.)
I ask because most of Xilinx's past offerings with hard CPU cores were aimed mostly at creating an System on a chip, while still potentially supporting reconfigurable accelerators if you could pull it off (partial reconfiguration is rather tricky[1]).
[1] Of course using partial reconfiguration as a way of live patching a system with negligible downtime can be even trickier than using it for replaceable accelerators that are inactive while reconfiguring, but people have still pulled off the former feat.
Many of us had thought about it, but it is very tricky to pull off. First standardizing the FPGA frabric is essential, or apps would need a different partial config bitstream for each CPU manufacturer. Further, standard FPGA frabric is not optimized for minimizing reconfiguration time.
In many setups, we are talking ten or hundreds of milliseconds. Not bad for starting an application, but unless apps claim exclusive us of the FPGA, then to quote Andy Dodd's post above: "context switching would be a real bitch".
Further, to get maximum possible performance likely requires exposing parts of the CPU to the fabric. A malicious application could potentially permanently damage the processor unless special step are taken. For example, the outputs of the CPU would need to be connected to FPGA pads that cannot be configured to be output pads, or a program could effectively short VDD and GND, literally causing the chip to fry itself.
Further, if great care is not taken, the FPGA configuration could deliberately time its outputs to change as the CPU clock ticks, attempting to introduce metastability to the processor. The processors are designed to be resistant to inputs randomly causing metastability of an input latch, but continuous deliberate attempts to introduce metastability are not usually factored into CPU design.
Most FPGAs in the field are coupled with EEPROMs that are triggered by the FPGA's own startup sequence, and emit a bitstream that is basically exactly the same as the data that would be sent if programing via JTAG. The FPGA start-up logic just transfers this data directly onto the interior scan chain that holds the configuration.
The chips invariably support other options too, such as the already mentioned JTAG based programming, and sometimes more complicated schemes.
There also exist a few that have integrated the EEPROM into the same package as the FPGA (probably not on the same die, but it could potentially be.
They do suck, but you get used to the ways they suck after a while. For example, I found it hard to move to a different synthesis tool from XST, because the other ones lacked some nice little convenience features of XST, despite the better RTL-level synthesis of the alternative tools.
I'm sure I'd be driven batty if I ever tried to use an Altera FPGA, simply because some of the features I'm used to will not be there, even though the features are emphatically non-essential.
My mental model of an FPGA is also based heavily on the Virtex and Spartan designs, and it would take some work to adjust them.
Take something like a digital audio recorder as the core, and add a walkman cassett head, and peice them together with a few passive components, and you have a simple, cheap and effective device to skim credit cards.
Later you download the recorded audio (it is a Digital audio recorder) and run it through say a quick matlab script, and you decode the card data.
DSL is a perfectly good name, DLSR - reflex=DSL.
One problem is that DSL already has two common tech expansions (Digital Subscriber Line, Domain Specific Language).
The other downside is that DSL does not imply interchangeable lenses and would technically be an accurate name for some smaller form cameras.
Hmm... Odd considering that from a group of graduation seniors form another good program (not quite at the level of say Cornell, but far, far better the average state school) had approximately $40k to $50k average. If the starting salaries are differing that much between schools, something has gone horribly wrong. Some amount differing should be expected, but a difference that large is downright disturbing.
Wikileaks has leaked plenty about other governments, as well as corporations local and foreign. Heck it used to be the case that the majority was non-US. I've not looked in a while, but that could still be the case.
The problem is that its the US leaks that get all the Attention in the US news (not to mention that much of what they have on other governments just is not as "juicy" as some of the US leaks).
If you have have a copy of the Taliban's membership roster, or North Koera's nuclear plans, or something similar, Wikileaks would be more than happy to publish that.
Please keep in mind that that is the average salary counting those students who got a M.Eng degree, which is a joke of a degree even at good schools like Cornell. That degree exists soley for the ability to claim to have a master's Degree.
Furthermore, I tend to doubt the numbers reported there are actually representative starting salaries, as the vast majority of companies I've seen do not have starting salaries that high. I'm wondering how many people responded to their survey giving their expected salary in a few years, or something to that effect. Either that or a few people got million dollar a year jobs through a "friend of the family".
It seems unlikely that a granularity of 1/80 of a mile is really needed. Are there frequently more than one railway location in a mile? I mean if not, why not just number them by mile? I mean the US does that for exists on major roadways. (I admit that it would be stupid for the UK to change now, except to metric.)
As for money, that is one area we will likely never see real metric conversion. After all names like "25 centiPound coin", or "deciDollar" sound terrible, not to mention the idea of spending several "kiloEuro" on something expensive.
Quite simply though, the better way is to simply replace ICANN with something international, and under no country's jurisdiction. Say a new UN treaty organization, or something similar.
An even better (although unfeasible) method would be to kill off the GTLDs, making US control of ICANN moot, since ICANN could not screw with any country's ccTLD, since that literally could be considered an act of war (overtly attempting to sabotage the infrastructure of another country), and without GTLDs only the ccTLDs matter.
It is correct that the first link goes to Google health.
What would be wrong to say is that the first result goes to Google Health.
The format of the main column of a Google web search results page is as follows:
The only problem is that Google does not mark the OneBox ( http://googlesystem.blogspot.com/2006/07/google-onebox-results.html ) with a shaded background or something.
It is rather interesting to look at how Google's results are structured. For example, the list of results is actually an HTML ordered list. Each search result is a list item. Prior to the ordered list, there is a hidden level 2 heading of "search results".
The only thing I would object to is the fact that additional OneBox-style items may appear at the bottom of the page, inside the ordered list that contains search results. That should be fixed, as those are not actually results. Fortunately the ones down there are usually suggesting different types of searches, like video or news searches, so most people will not mistake them for results.
That is definitely not even remotely standard for the US. It is a UK thing being a holdover from prior times since prior to 1985 the health care industry did use stones as part of measuring weight.
But the average American's reaction to hearing that somebody weighs 12 stones would be confusion. After all, to different stones (chunks of rock) don't weight the same.
Even well educated Americans would not need to pause for a moment upon hearing that. They would know of the unit, but may or may not know how large it is. If they do, they would still need to mentally convert it to pounds.
He is saying that he still expects Gmail, the Android Market, and the other Google Apps will continue to ship on the phones by default. I agree. All the market availability does is enable app updates more frequently than OS upgrades.
I will also dispute the Google apps being the core apps. The core apps are those like Contacts and Home, which would be present in an Android build you compiled from source.
How can he be detained in his absence? If he is not there you can't detain him. I'm obviously not understanding something here.
Alternative root systems have never caught on, because nobody wants a domain that most people cannot resolve.
What you describe would still only have one root zone. By mirroring the zone files you end up with only one zone. All you are really proposing is splitting up policy making for gTLDs, meaning that some gTLDs will be Euro ones, subject to EUCANN policies, etc, and others will be US gTLDs, subject to ICANN policies.
The group feature is already there for those that want it, but it is not enabled by default in most distro kernels.
They create a hierarcy.
The scheduler looks at schedulable items, not caring if they are threads or groups, and uses its normal algorithms (including nice levels) to decide which to run next. If it chooses a group, it repeats this process for the group, and of course, all time used by processes or subgroups count towards the group's running time. It repeats until it gets an actual thread, and runs that.
I'm not sure how one sets the nice level for a group though. I only learned about this feature yesterday when reading the LKML thread.
Why throw it away? surely once you come back you could still hook up a a set-top box to the tv via HDMI or component, no?
A bigger concern would be if the TV's PSU will support 120 V or not. If it cannot, that is a real deal breaker.
Heck most of the time you should be able to get away with an a pair of omnidirectional antennas, one in the Faraday cage (house) and one outside connected by RG6. A high gain directional should really only be needed if you also have pretty poor signal outside the house.
Hold on a second. Let's say that the repeater has a 3 dB gain (gain of 2) in both uplink and downlink. The phone will adjust its transmit gain to be just loud enough that the closest tower can hear, just like always, no? So the towers should be getting a signal no stronger than if the building simply were not there.
In all reality though, most people only need a a gain of 1 (0 dB). The real key is to get the signal through the wall without the associated losses. Heck, many people could get by with a pair of identical antennas connected by coax, despite that having a gain of significantly less than 1.
There cannot be multiple root zones. That just is not feasible. But the US only controls the gTLDs and one ccTLD. The other TLDs are owned by their respective countries. The US cannot have them removed from the root without causing a major incident, the result of which would include the Internet Society appointing some other organization as IANA (and thus killing ICANN and US control of the DNS).
Explanation: ICANN exists solely to be a policy creating shell around the IANA. Legally the Internet Society (ISOC) is the only company in the world with any reasonable claim to being able to appoint an IANA. The organizations the ISOC consists of (IETF, IESG, IAB, RFC-EDITOR) had agreed to allow ICANN to have the role of IANA, but they could revoke that agreement.
If an RFC is published naming a new international entity as the IANA there is little reason to suspect the companies running the root servers would not use the root zone published by the new IANA, as long as they got similar say in policy making as they had with ICANN.
Sure.
Hopefully you know what a TTY is, but in case you don't it it a virtual or real terminal. When you open up an xterm you create one. If you don't have x-windows installed, you reach one, etc.
Well Linus had an idea about using a grouping functionality that was already in the Kernel to allow all the processes (technically actually all the kernel threads) running from one TTY to be grouped together for scheduling.
The result of that is that if you are running 99 processes in one xterm that could consume all of your CPU, and you open another xterm, one one just one process that wants 100% CPU, each xterm's processes gets 50% of the CPU, rather than one getting 99% and the other getting only 1%.
But lets say you only had that first xterm. Since each of those processes are not getting nearly the processor amount they desire, normally the scheduler sees them as nearly starved, and the next process that only wants 5% of CPU does not get much preferential treatment for giving up most of it's time. However, with the grouping, the scheduler can see that those 99 processes are related, and they are not really starved, since as a group they are getting 100%. So now when this other app that wants only 5% comes along, the scheduler might give it pretty much all of that 5% rather than the mere 1% it would have been getting before, and so that app (probably a web browser or something) remains nice and responsive.
That is not 100% accurate, since I've simplified some things a little, especially with regard to the working of the scheduler, but it should give you the idea.
Eventually, more heuristics might be added, so that a GUI application that launches a bunch of threads and hogs the CPU might have all it's threads grouped, so they don't hurt responsiveness of interactive apps either.
I went to a a college much smaller than the one in question, and virtually all of my teachers were always in the room the whole test. There were a few exceptions though, mostly with the small senior classes that had 6-15 students. In that case the teachers realized that no student could get away with cheating unless everybody colluded, and nobody told him/her.
The prisoner's dilemma in exact reverse, but with more participants. Normally game theory would say that everybody would go for it, but once you add ethics you cannot expect 100% participation, so the best course of action is to not cheat.
That clause says the Constitution, federal laws, and treaties overrule state constitutions and law. to interpret that last "Constitution" to mean the Federal Constitution would require that Federal Laws override the Federal Constitution, which is absurd.
And yes, it helps them greatly to be able to blow up things.
Next government program: Free dynamite for the blind.