Well, now you see Linus's point. If the buffer is being sent asynchronously after the write() call, and the user program writes to the buffer before the ethernet chip picks up the buffer via dma, then the buffer must be COW so that the ethernet chip can send the appropriate data.
The real problem is that in a zero-copy world, write() returns before the data is sent, and in FreeBSD there is no way for the kernel to signal the user program that the write() is complete and it is safe to re-use the buffer
I think that most people that I have seen recently using their laptops at the airport or on the plane were trying to watch a movie or playing solitaire!
It is very unclear to me. If a driver loads a configuration data file at boot time, does that file have to be open source too? What if I make a driver that loads a user space program into the kernel? Does the user space program become GPL too?
Linus himself has repeatedly said that loading closed source kernel modules is not a violation of his license.
Also, how 'open' would an open source driver have to be if he said otherwise? Does the firmware file for an on-board Xilinx FPGA being embedded in the driver mean that the VHDL or Verilog source code needs to be included? Even when these things are not compilable by any gpl tools?
While that is true, it always bugs me that it seems like every distro I use (and I use a lot) 'adds so much functionality' to the/etc/init.d init scripts scheme to make it just different enough to be incompatible with everything else, and it makes it a pain to make my own packages for multiple distro's.
What were they thinking when they made that thing? They author must have been blind or something (haha,actually a blind person would have done much better).
Now that operatic "Glaucoma" song is stuck in my head.
I love the dichotomy of the neo-con. Why do you need privacy when you shouldn't be hiding anything unless you are doing something wrong. Oh, and the government needs to make these embarrassing documents about the Korean War secret... National Security, you know...
Yes, but are you saying that contact-less smart cards are NOT vulnerable to attacks where the smart card power draw is measured as the article discusses?
...people often observed eerie lights in the sky in the hours before a quake...Of course, nobody really knows why these electrical phenomenon occurs before a quake. But they still can be observed.
Obviously, the earthquakes are caused by aliens flying ufo's after they place the earthquake-causing plasma bombs beneath the crust...
Slashdot Mirror
Well, now you see Linus's point. If the buffer is being sent asynchronously after the write() call, and the user program writes to the buffer before the ethernet chip picks up the buffer via dma, then the buffer must be COW so that the ethernet chip can send the appropriate data.
The real problem is that in a zero-copy world, write() returns before the data is sent, and in FreeBSD there is no way for the kernel to signal the user program that the write() is complete and it is safe to re-use the buffer
--jeffk++
I think that most people that I have seen recently using their laptops at the airport or on the plane were trying to watch a movie or playing solitaire!
--jeffk++
No, no, they will make it so that the bathroom door doesn't open while the commercial is playing. Problem solved.
--jeffk++
Yes, of course, but TOPIC=="linux drivers" && LINUX=="GPL"
--jeffk++
It is very unclear to me. If a driver loads a configuration data file at boot time, does that file have to be open source too? What if I make a driver that loads a user space program into the kernel? Does the user space program become GPL too?
Jeff
I think Linus's comments on KernelTrap here are pretty clear. A binary module is not automatically a derived work. Jeff
Linus himself has repeatedly said that loading closed source kernel modules is not a violation of his license.
Also, how 'open' would an open source driver have to be if he said otherwise? Does the firmware file for an on-board Xilinx FPGA being embedded in the driver mean that the VHDL or Verilog source code needs to be included? Even when these things are not compilable by any gpl tools?
--jeffk++
While that is true, it always bugs me that it seems like every distro I use (and I use a lot) 'adds so much functionality' to the /etc/init.d init scripts scheme to make it just different enough to be incompatible with everything else, and it makes it a pain to make my own packages for multiple distro's.
--jeffk++
Oh... good point... ;-)
jeff
But you are talking about reality.
The question really is a LEGAL loophole one.
--jeffk++
Is it still technically 'wiretapping' if there is no wire, just fibre-optics? ;-)
--jeffk++
Weird, download via Safari (via different computers) gave me corrupted dmg file, even though the mimetime was right. Download via firefox was good!
jeff
( the URL i'm loading is http://download.parallels.com/beta/Parallels-2.1.1 658.24-Mac.dmg
--jeffk++
Betcha the 'catch line' for leopard will be something like "This leopard CAN change its spots"... :-)
--jeffk++
Except for Schrodinger's cat....
--jeffk++
I think you are confusing carbon with 'classic'.
--jeffk++
I clicked on your link and saw the second place winner, The Association of International Glaucoma Societies.
What were they thinking when they made that thing? They author must have been blind or something (haha,actually a blind person would have done much better).
Now that operatic "Glaucoma" song is stuck in my head.
--jeff++
--jeffk++
Unfortunately, fascsim is a word loosing meaning:
l ish/efasc
http://orwell.ru/library/articles/As_I_Please/eng
( George Orwell - What is Fascism?)
--jeffk++
You were misled. The jury is allowed to present any verdict it wants without needing an explanation. Ask the judge next time about jury nullification.
I love the dichotomy of the neo-con. Why do you need privacy when you shouldn't be hiding anything unless you are doing something wrong. Oh, and the government needs to make these embarrassing documents about the Korean War secret... National Security, you know...
--jeff++
I tried that once. It was enlightening to see a mysql dump refuse to load into a fresh mysql installation because:
1) The table had a column marked as 'unique'
and
2) The data in the column in that table was NOT unique.
yet the first installation of mysql allowed it! (version 4.1)
jeff
read google's legal pdf as linked to by TFA. The information is useless because of a number of reasons.
--jeff++
Yes, but are you saying that contact-less smart cards are NOT vulnerable to attacks where the smart card power draw is measured as the article discusses?
--jeffk++
Obviously, the earthquakes are caused by aliens flying ufo's after they place the earthquake-causing plasma bombs beneath the crust...
--jeffk++