Tag Info

New answers tagged

1

The /dev/kmsg represeted by the kmsg_fops structure which has file_operations type that represents standard operations with a file: const struct file_operations kmsg_fops = { .open = devkmsg_open, .read = devkmsg_read, .write_iter = devkmsg_write, .llseek = devkmsg_llseek, .poll = devkmsg_poll, .release = devkmsg_release, }; You ...


0

Have a look at the *_label files to see what is being reported - here's my i5: $ grep "" /sys/devices/platform/coretemp.?/hwmon/hwmon?/temp?_label /sys/devices/platform/coretemp.0/hwmon/hwmon1/temp1_label:Physical id 0 /sys/devices/platform/coretemp.0/hwmon/hwmon1/temp2_label:Core 0 /sys/devices/platform/coretemp.0/hwmon/hwmon1/temp3_label:Core 1 (grep "" ...


0

The path /usr/src/linux is the standard path where kernel sources live in a linux system. Depending on your distribution you can install linux kernel sources with your package manager, or you go to http://www.kernel.org and download and extract one. Download a kernel with git One of the most powerful ways is to download a kernel with git as super-user ...


2

That's a kernel bug, Debian bug #789037 aka upstream bug #99161. It was introduced in a recent kernel update, and you probably saw it after the reboot (to install RAM) because you're now running that kernel. The fix is already available; you need to install it (and reboot). (BTW: I saw this on some of our servers after an unplanned reboot due to a circuit ...


1

If you've over-committed memory, a lot of tmpfs may be on disk. You may need to page stuff in to process the shutdown. mlock() is likely to force a lot of the other memory to disk. As you indicate you are diskless, you are likely reading over the network. Run sar gathering all stats while the server is shutting down. (sar may not be installed by ...


0

I'm trying to do the same thing you are: get a 3G/4G modem to work on the ARDrone2. I'm not as far along as you are though, I only just managed to get my hands on the toolchain, and my first few attempts at a hello world have crashed. I don't really have an answer, only an idea. I did a quick google on your log messages, and I found this: ...


1

The TASK_INTERRUPTIBLE state has to do with the manner in which a task is put to sleep when it invokes the scheduler: its name is short for "interruptible sleep". It does not control preemption (and in fact Linux had TASK_INTERRUPTIBLE long before it was made (optionally) preemptible). Preemption is not sleep; an executing task kicked off its CPU is ...


1

The GPL gives the distributor of the software three options: a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, b) Accompany it with a written offer, valid for at least three years, to give any ...


0

GPL does not require a developer to provide source code to all requesters. It's stated clearly in GPL FAQ The GPL does not require you to release your modified version, or any part of it. You are free to make modifications and use them privately, without ever releasing them. This applies to organizations (including companies), too; an organization can ...


1

It sounds like it. These days, the best place to ask seems to be the Software Freedom Conservancy. http://sfconservancy.org/linux-compliance/about.html compliance@sfconservancy.org


0

Since you didn't specify if actually corrupting the FS is acceptable you may want to look at this Q&A: http://serverfault.com/questions/40302/how-to-corrupt-a-file-system :) Note: you'd need to corrupt the root filesystem or a swap partition to cause the system to panic, a regular non-critical partition will be just switched to readonly mode but the ...


1

systemd grabs the serial console and squelches the kernel, so I only get the bootloader messages Does it? It doesn't. The loglevel=1 in your kernel command line is responsible for telling kernel to stop logging. Try removing that statement (or explicitly setting it to loglevel=7). To stop systemd from logging its own status messages, use ...


2

A signal is a classic kernel-to-process communication in some cases. Sure, any process with appropriate UID can use kill() to send a signal, but signals like SIGCHLD or SIGWINCH almost always come from the kernel. Specific to Linux, you can look at the signalfd() system call. The usual signal handling is via an "upcall", but signalfd() gives a process a ...


1

The Device Mapper driver is missing. Run make nconfig or make menuconfig: Device Drivers ---> [*] Multiple devices driver support (RAID and LVM) ---> <*> Device mapper support and recompile your kernel.


4

Here's the output from one of my bridging Linux machines: With brctl showstp, you can print the list of interfaces and their spanning-tree parameters involved with the bridging. There, in parenthesis, you'll find the interface index of the constituent interfaces - that's what is referenced in brctl showmacs. # brctl show bridge name bridge id ...


2

The reason you can't alter the RTO specifically is because it is not a static value. Instead (except for the initial SYN, naturally) it is based on the RTT (Round Trip Time) for each connection. Actually, it is based on a smoothed version of RTT and the RTT variance with some constants thrown into the mix. Hence, it is a dynamic, calculated value for each ...


0

It seems %gs is reserved for GCC'c stack protection feature on x86 Linux kernel with CONFIG_CC_STACKPROTECTOR enabled in order to set up stack canaries. You can see some explanation at arch/x86/include/asm/stackportector.h.


1

Let's assume that we have 3 GiB of virtual address space available and that a process' text, stack, heap, and prior memory mappings together occupy only a small amount of address space (much less than 1 GiB). Then if this process requests a mapping of size 2GB then there is no problem, there is plenty of space to accommodate that mapping. (Actually, there ...


1

Some extra debug information like getting the call stack and some other things like that that are needed by gdb for debugging will be enabled. This will have slight impact in the performance. But you will see this mainly by using the tools using which most of the code runs in the kernel space. e.g. check the speed of a file copy, creating multiple threads ...


0

You could isolate some cpu cores from kernel scheduling using isolcpus parameter. Add this parameter to your grub.conf and reboot to take effect. You can find more info and example at http://nairobi-embedded.org/cpu_affinity.html#the-isolcpus-kernel-parameter


0

Some devices lose firmware on suspend, but btusb does not notice that. For some devices it has been fixed in newer kernels. The fix is that the module is automatically bumped, like you do manually on resume. Anyway it is a kernel bug. You can test newer kernels or report it upstream. A workaround is to make a wake up script that will restart it on ...


2

The kernel memory management doesn't use the concept of "program" to organize the memory, but pages. The kernel decides based on a 'least recently used' (this may have changed, but the latest references I could find say so), in which when the kernel under memory pressure (by swappiness, it swaps out those pages which are rarely accessed to give space to more ...


2

You have to have a kernel source with kernel hacking enabled. Then switch on CONFIG_DEBUG_KMEMLEAK and compile your kernel. When you have that running: echo scan > /sys/kernel/debug/kmemleak to switch it on and cat /sys/kernel/debug/kmemleak for a report. Original proposal reviewed on lwn.net There is a kmemleak.txt in the kernel source code ...


1

If you run customized kernels for your embedded hw and have some hw register/bit available you may be able to customize the kernel crash code to set a flag in that hw location which you'd check after reboot. If not AFAIK you're only chance is to configure your kernel core dumping facility. Indeed, it's risky to write to a 'live' filesystem, but you can use ...


2

Is it possible to see Where did this rule come from? Not in the sense "where can I look up the source of this rule". There are several ways to investigate the issue: the most evident is to grep all startup scripts on your system to see which uses ip rule at all, and then start reading them. Or you could start your system in single-user mode, and start ...


3

Ancient History. the term image comes from an old Digital Equipment Corporation term for the output from the compiler-> linker. the file is an image created by interpreting the code and so on through the linker to make an executable "Image" of your design.


2

From what I read in LinuxFR article (french) This cache was suffering bad performance bottleneck relatively to today's latency needs and security problems where sometimes vulnerabilities allowed attackers to poison this cache. Also it's average hit rate was <10%. There is now a small per-entry cache, but only for additionnal informations (TCP stats, ...


5

In math the kernel is the inverse image of a subset of the image of a some map, were the subset is equal to the identity element in the codomain. I'm certain these names derive from mathematical concepts as they are related significantly in various fields in mathematics. Considering Unix was derived in an academic environment it may be possible that it's use ...


2

You want some processor affinity (or CPU affinity). The relevant syscall is sched_setaffinity(2), but you should use it thru pthread_set_affinity_np(3) if you want to code your benchmarks for that. The related command is taskset(1) and you might use it on the commands you want to benchmark (or on your shell). If possible, take care that the machine is ...


0

There's a key distinction to be made about where header files come from: <time.h> is provided by glibc (e.g. the glibc-headers package) <linux/time.h> is provided by the Linux kernel headers. Changing the kernel and its header packages will not affect <time.h>. Only changing glibc will do that. You should find that glibc's ...


15

It doen't mean an "image" is just a 1:1 copy of a disk. As the photo represents the reality exactly as it was when shooting, an image of an executable program (or kernel) represents the program in a state, where it can the loaded (or unpacked) in the systems memory exactly as it is and then given control to it. That program can then start running from that ...


28

The word image also has a definition "A file that contains all information needed to produce a live working copy."


56

The Unix boot process has (had) only limited capabilities of intelligently loading a program (relocating it, loading libraries etc). Therefore the initial program was an exact image, stored on disc, of what needed to be loaded into memory and "called" to get the kernel going. Only much later things like (de-)compression were added and although more powerful ...


1

You can use Ftrace. Ftrace is a tracing utility built directly into the Linux kernel.


0

As others have said it doesn't really matter what distro you use. Probably the best way to start would be to install whatever distro you are comfortable with in a VM. From there you can build the kernel version from kernel.org and use default config (though you may want to turn on some of the kernel debug options). A couple things to note, Make will be ...


0

If you're looking for beginner simplicity then kernel hacking would be rather tough going. But Arch Linux would be a great distro to use to do Kernel Hacking. I've never submitted a patch, but just paying around with the same book you're reading, I found Arch perfect for the job.


3

Yes, you are correct. In particular, this means that the child will inherit all variables from the parent process with the value they had at the moment of the fork. However, if at a later step one of the parent or the child modifies one of these variables, the modification will be local to this process: if the child modify a variable, the parent process ...



Top 50 recent answers are included