The question is especially interesting in the long term. Will this be just a limited trick to please the developers that like usual console tools on Ubuntu. ? Or will this be someday included by default on Windows with enough support to allow to run Linux applications ?
For that reason I always wanted that the win32/win64 API will be a international standard managed by an official organization like ISO. That way Microsoft will lost evil control of an API that is actually a very effective tool to kick off any concurrent. This will allow Wine to implement a good implementation based on detailed documentation.
This is curious that no government on the planet require that win32/win64 are international standard, while there usually require standard certification on a tone of less critical parts of there infrastructure. OpenOffice did a genius master move by making ODF an international standard before Microsoft was able to do the same. If only Wine would get enough support to standardize win32/win64 before Microsoft. Seem to be an impossible dream actually.
It's doable, the Wine project prove that point. It's just way more complicated... In addition to what "KiloByte" have said, Microsoft have the Linux source code as implementation reference to look at, while the Wine project didn't have a chance to look at the Microsoft kernel and libraries.
Whenever it's for hardware or for software, the point to make a project open source is to create a community to support and improve the project, not to make money. Starting from that point I found cheap clone a rater big advantage, because this offload the production problem and make the project accessible to more and more peoples with almost no management. Instead of fighting against clones, it better to take advantage of them.
While you are right about the actual situation, I think that more complicated attacks will be the next mainstream just after the easy attacks will be patched. And I doubt that modifying a few files of a squashfs image using his block device will require multi-gigabyte images.
Because the squashfs image of OpenELEC is read only and mostly one of the know images already published, the operations you describes could be done on the hacker PC. It can then find a way to generate the hacked image from the original by analyse the difference between the two. This recipe can then be directly applied to the partition of the target device probably without too much processing and memory.
I agree that this is much more work, but maybe not so difficult for a motivated hacker.
What is the SSH login? Shortcut: #SSH Login Currently the login into OpenELEC has fixed settings.
Login: root
Password: openelec
How do I change the SSH password? Shortcut: #SSH Password change At the moment it's not possible to change the root password as it's held in a read-only filesystem. However, for the really security conscious advanced user, you can change the password if you build OpenELEC from source. Also you can consider logging in with ssh keys and disabling password logins.
It's possible to remount a file system with new permissions. See the details here: https://www.gnu.org/software/l... Or to mount it in a other folder with different permissions. Or to directly access the partition under/dev/sda1.
To make a read-only file system work as expected you have to use a hardware way to prevent writing to the memory. For example NOR SPI flash memory usually have a write protection pin. Of course that pin must be protected against unwanted operation that could drive it. In that case you can expect having a clean state after a boot. Secure boot is an other method to give the same kind of clean state after boot.
You still have issues that can live in the SDRAM as long as the device will run...
Maybe you should realize that every SDcard, eMMC, WiFi, or Bluetooth chip, contain a processor (generally 32 bits this days) to offload your main CPU. Seriously, if a main processor with a RTOS will be able to do all by itself, then all peripherals would be virtual with just AFE to convert to/from analog signals.
The reality is that toady even small controllers contain DMA engine, interrupt controller and peripheral state machines good enough to offload the core from most of the communication fast low level details. Look at the implementation details of a Ethernet MAC and PHY for example. Good luck to recreate something like that with just a CPU and a AFE...
Dedicated peripherals are nothing different. Depending of there complexity and performances, a extra CPU/DSP/FPGA could be a far better design than trying to do the same thing with a RTOS on a central processor and a AFE. Not counting that the vast majority of the industry is going more and more toward highly integrated peripherals chips that integrate there own processor, and all of that for a very low price due to the massive production scale.
Even if your smartphone PCB let you think that there is just a single processor chip with his main cores because of the marketing, in reality it probably have something like a half dozen of extra dedicated CPU or DSP running all the peripherals fast low level details, not counting all the DSP like cores of the GPU itself...
I do mostly Linux embedded software and I like to develop directly on the remote targets that don't have a screen. In this context Visual Code Studio is unusable as virtually any graphic IDE.
So your theory is that no useful software can exists if not done with Visual Studio ? Seriously, I doubt that your router or your TV will ever boot in that case...
And this is also very clear: sub foo {
$var = 22;
$anothervar = 23;
if ($var == $anothervar) {
return $var1;
} }
While wasting a line for each open brace is not so a problem for the functions, this make code really annoying to read if there is a lot of consecutive if-then blocks. K&R coding style is widely used for a reason.
"At lest one study" is nowhere near to be an argument. You can probably find "at least one study" that claim whatever you wants on a so width subject as the style used by individuals. I can either say that at least 3 well known projects (Linux, Git, PHP) uses 8 spaces tabs for indentation.
Now the cool stuff about tabs is that it allow to be very easily adjusted in a perfectly reliably way by the user editor without touching the file content. I will add that 8 spaces tabs make code with too many indentations look wrong faster, and I found this a good feature in my experience.
If you do a search with "Softbank Microsoft" you will quickly realize that Softbank is in a very close relation with Microsoft. I hope to be false, but I am afraid that the ability to install a free Linux on any ARM devices will be a thing of the past in less than 10 years.
Probably true from the technical point of view, but not completely certain as the deeper integration into the CPU chip of the features that was before into the north/south bridges or peripheral chips could be implemented in a way to no longer support some legacy compatibility features required to support old OS drivers.
Well, as long a I can run recent Linux on them, it's probably on good thing in the long term as the original PC architecture is really out of date. But new processors will not just run Windows, OSX or Linux. There is a good chunk of specialized OS (many for realtime embedded systems) that will need to run on them. I doubt that there all have the drivers ready for a such big change.
There are many others ARM based kits that brings comparable capabilities and also run Linux (with open source compilers) for a fraction of the Intel price.
For most embedded projects, especially those running Linux, the instruction set architecture weight as much as a photo in the design decision. I myself run many armhf and arm64 architecture based embedded boards with the exact same Debian distribution that I run on my amd64 architecture workstations, laptop and servers.
Probably not UNIX. Linux have now so much specific system calls, ioctl, filesystems in addition to Linux specific user space tools like udev, and systemd that now Linux define a new standard by itself. UNIX will gradually be just be a supported compatibility.
If what you say it true, why Microsoft is still trying so hard to lock PC manufactures to Windows ? UEFI secure product key is essentially a way to ensure world domination on the PC market.
For the 64 bit instruction set, AMD have take advantage of the Intel Itanium mistake and it was there smarter move ever as this enabled AMD to negociate the instruction set symetric cross licencing agreement with Intel.
AMD did make a lot of technical innovations on CPU before, like copper metal interconnect, silicon on insular, integrated memory controller, hypertransport, multiple cores, exclusive caches, etc..., and lately integrated graphic. At some point Intel haved nothing to compete but there extremly agressive marketing worked long enough to develop a completely new generation of Intel core processors to gently catch up and finally surpassed AMD product. Intel HKMG and smaller gate process finally destroyed AMD that at the same time almost stopped his progression.
The announce look like there used good design decisions. Time will tell.
Slashdot Mirror
The question is especially interesting in the long term.
Will this be just a limited trick to please the developers that like usual console tools on Ubuntu. ?
Or will this be someday included by default on Windows with enough support to allow to run Linux applications ?
For that reason I always wanted that the win32/win64 API will be a international standard managed by an official organization like ISO. That way Microsoft will lost evil control of an API that is actually a very effective tool to kick off any concurrent. This will allow Wine to implement a good implementation based on detailed documentation.
This is curious that no government on the planet require that win32/win64 are international standard, while there usually require standard certification on a tone of less critical parts of there infrastructure. OpenOffice did a genius master move by making ODF an international standard before Microsoft was able to do the same. If only Wine would get enough support to standardize win32/win64 before Microsoft. Seem to be an impossible dream actually.
It's doable, the Wine project prove that point. It's just way more complicated...
In addition to what "KiloByte" have said, Microsoft have the Linux source code as implementation reference to look at, while the Wine project didn't have a chance to look at the Microsoft kernel and libraries.
Whenever it's for hardware or for software, the point to make a project open source is to create a community to support and improve the project, not to make money. Starting from that point I found cheap clone a rater big advantage, because this offload the production problem and make the project accessible to more and more peoples with almost no management. Instead of fighting against clones, it better to take advantage of them.
While you are right about the actual situation, I think that more complicated attacks will be the next mainstream just after the easy attacks will be patched.
And I doubt that modifying a few files of a squashfs image using his block device will require multi-gigabyte images.
Because the squashfs image of OpenELEC is read only and mostly one of the know images already published, the operations you describes could be done on the hacker PC. It can then find a way to generate the hacked image from the original by analyse the difference between the two. This recipe can then be directly applied to the partition of the target device probably without too much processing and memory.
I agree that this is much more work, but maybe not so difficult for a motivated hacker.
Good idea, but what grant our that your firewall is not hacked ?
Actually it could be the case:
http://www.theregister.co.uk/2...
OpenELEC FAQ disagree:
http://wiki.openelec.tv/index....
What is the SSH login?
Shortcut: #SSH Login
Currently the login into OpenELEC has fixed settings.
Login: root
Password: openelec
How do I change the SSH password?
Shortcut: #SSH Password change
At the moment it's not possible to change the root password as it's held in a read-only filesystem. However, for the really security conscious advanced user, you can change the password if you build OpenELEC from source. Also you can consider logging in with ssh keys and disabling password logins.
It's possible to remount a file system with new permissions. See the details here: /dev/sda1.
https://www.gnu.org/software/l...
Or to mount it in a other folder with different permissions.
Or to directly access the partition under
To make a read-only file system work as expected you have to use a hardware way to prevent writing to the memory. For example NOR SPI flash memory usually have a write protection pin. Of course that pin must be protected against unwanted operation that could drive it. In that case you can expect having a clean state after a boot. Secure boot is an other method to give the same kind of clean state after boot.
You still have issues that can live in the SDRAM as long as the device will run...
Maybe you should realize that every SDcard, eMMC, WiFi, or Bluetooth chip, contain a processor (generally 32 bits this days) to offload your main CPU.
Seriously, if a main processor with a RTOS will be able to do all by itself, then all peripherals would be virtual with just AFE to convert to/from analog signals.
The reality is that toady even small controllers contain DMA engine, interrupt controller and peripheral state machines good enough to offload the core from most of the communication fast low level details. Look at the implementation details of a Ethernet MAC and PHY for example. Good luck to recreate something like that with just a CPU and a AFE...
Dedicated peripherals are nothing different. Depending of there complexity and performances, a extra CPU/DSP/FPGA could be a far better design than trying to do the same thing with a RTOS on a central processor and a AFE. Not counting that the vast majority of the industry is going more and more toward highly integrated peripherals chips that integrate there own processor, and all of that for a very low price due to the massive production scale.
Even if your smartphone PCB let you think that there is just a single processor chip with his main cores because of the marketing, in reality it probably have something like a half dozen of extra dedicated CPU or DSP running all the peripherals fast low level details, not counting all the DSP like cores of the GPU itself...
I do mostly Linux embedded software and I like to develop directly on the remote targets that don't have a screen. In this context Visual Code Studio is unusable as virtually any graphic IDE.
So your theory is that no useful software can exists if not done with Visual Studio ?
Seriously, I doubt that your router or your TV will ever boot in that case...
Mind you Visual Studio Code is a wonderful thing. It's the first software from MS I have been using on a regular basis for real work ever.
For you maybe. Totally unusable in my case.
And this is also very clear:
sub foo {
$var = 22;
$anothervar = 23;
if ($var == $anothervar) {
return $var1;
}
}
While wasting a line for each open brace is not so a problem for the functions, this make code really annoying to read if there is a lot of consecutive if-then blocks. K&R coding style is widely used for a reason.
"At lest one study" is nowhere near to be an argument. You can probably find "at least one study" that claim whatever you wants on a so width subject as the style used by individuals. I can either say that at least 3 well known projects (Linux, Git, PHP) uses 8 spaces tabs for indentation.
Now the cool stuff about tabs is that it allow to be very easily adjusted in a perfectly reliably way by the user editor without touching the file content.
I will add that 8 spaces tabs make code with too many indentations look wrong faster, and I found this a good feature in my experience.
If you do a search with "Softbank Microsoft" you will quickly realize that Softbank is in a very close relation with Microsoft.
I hope to be false, but I am afraid that the ability to install a free Linux on any ARM devices will be a thing of the past in less than 10 years.
Probably true from the technical point of view, but not completely certain as the deeper integration into the CPU chip of the features that was before into the north/south bridges or peripheral chips could be implemented in a way to no longer support some legacy compatibility features required to support old OS drivers.
Well, as long a I can run recent Linux on them, it's probably on good thing in the long term as the original PC architecture is really out of date. But new processors will not just run Windows, OSX or Linux. There is a good chunk of specialized OS (many for realtime embedded systems) that will need to run on them. I doubt that there all have the drivers ready for a such big change.
+1
That's the right way for sure.
Any editor should automatically use this rule by default.
I will add that with the today screen width, 8 spaces tab is really not an issue.
photo -> photon, sorry for the typo.
What two pieces of information ?
There are many others ARM based kits that brings comparable capabilities and also run Linux (with open source compilers) for a fraction of the Intel price.
For most embedded projects, especially those running Linux, the instruction set architecture weight as much as a photo in the design decision. I myself run many armhf and arm64 architecture based embedded boards with the exact same Debian distribution that I run on my amd64 architecture workstations, laptop and servers.
Probably not UNIX.
Linux have now so much specific system calls, ioctl, filesystems in addition to Linux specific user space tools like udev, and systemd that now Linux define a new standard by itself.
UNIX will gradually be just be a supported compatibility.
If what you say it true, why Microsoft is still trying so hard to lock PC manufactures to Windows ?
UEFI secure product key is essentially a way to ensure world domination on the PC market.
The new CPU is "Zen", so the new staff might be buddhism monks :-D
For the 64 bit instruction set, AMD have take advantage of the Intel Itanium mistake and it was there smarter move ever as this enabled AMD to negociate the instruction set symetric cross licencing agreement with Intel.
AMD did make a lot of technical innovations on CPU before, like copper metal interconnect, silicon on insular, integrated memory controller, hypertransport, multiple cores, exclusive caches, etc..., and lately integrated graphic. At some point Intel haved nothing to compete but there extremly agressive marketing worked long enough to develop a completely new generation of Intel core processors to gently catch up and finally surpassed AMD product. Intel HKMG and smaller gate process finally destroyed AMD that at the same time almost stopped his progression.
The announce look like there used good design decisions. Time will tell.