New answers tagged

0

After adding the ARCH=x86_64 flag to all the make commands the Linux kernel was successfully compiled.


1

You might get somewhere browsing the torvalds git tree, eg for the file time/hrtimer.c. Click on blame and for each line number you see the last patch applied. You can also browse the history for older patches.


3

If kernel was compiled with CONFIG_IKCONFIG_PROC, you will have the original .config in /proc/config.gz So in that case to see the full options: zcat /proc/config.gz Checking in my Armbian/Jessie with kernel 4.5.2 and BTRFS: $zgrep -i btrfs /proc/config.gz CONFIG_BTRFS_FS=y CONFIG_BTRFS_FS_POSIX_ACL=y # CONFIG_BTRFS_FS_CHECK_INTEGRITY is not set # ...


-1

Have a look at sysctl. It can be used to query and set kernel parameters. sudo sysctl -a will display all current values.


1

You can try using this command, which is distribution-independent: ldconfig -p This gives a full list of the libraries installed. If you pipe it to less, you will be able to scroll up or down: ldconfig -p | less


1

You can't (re)compile a kernel module by simply extracting the source and running make in its subdirectory. Each distribution has a preferred method - in your case, see section 8.10. Compiling a Kernel of the Debian handbook. Specific drivers have to be enabled via the kernel's build configuration system. You would need to enable MOUSE_PS2_SENTELIC. ...


0

Alex answers your question. Compare lspci 00:00.0 Host bridge: Intel Corporation Core Processor DRAM Controller (rev 02) 00:02.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 02) 00:16.0 Communication controller: Intel Corporation 5 Series/3400 Series Chipset HECI Controller (rev 06) 00:19.0 Ethernet ...


0

For CPU_Frequency: The timestamp format is "secs.usecs". You can find more about this on ftrace documentation: https://www.kernel.org/doc/Documentation/trace/ftrace.txt I think it shows the transition delay between frequencies. For CPU_Idle: '4294967295' means an exit from the current state. The documentation can be found here: ...


1

NetworkManager gained support for bonding recently. I doubt it expects you to load the module manually. There's a minority of tutorials that don't tell you to load the module, so I wonder about the ifenslave script... hah. From the Debian 8 package, file ifenslave.pre-up: add_master() { # Return if $BOND_MASTER is already a bonding interface. [ -f ...


0

godlygeek's answer is good for understanding how the system works but the subsequent question that inevitably follows is: How to determine if a process has gone away? The correct way to wait on a process in another process group or session is to use kill(). Obviously, that is an unintuitive answer. You can't use the wait family of functions because the ...


0

A comment of push_to_pool function from /drivers/char/random.c says: /* * Used as a workqueue function so that when the input pool is getting * full, we can "spill over" some entropy to the output pools. That * way the output pools can store some of the excess entropy instead * of letting it go to waste. */ If I understand correctly, the kernel ...


2

Unsure if you are still looking for a response but I was running into the same problem trying to upgrade to 4.5.3 kernel. I found my system had gcc-5 installed and switching to that resolved the first problem (-fstack-protector-strong). I then received the second set of errors you listed about ndiswrapper/1.59. I believe this is a bug in that software but ...


2

Reducing the priority of a process will only make it run for longer. So the only possible side effects would be: Its memory usage being assigned longer because it runs slower, and since it remains active, it will remain in active memory or cause paging (swap in and out). This is seldom a problem unless you're tight on available memory. If the process ...


0

You may have leftover metadata from earlier partitioning. Like a boot partition that is used without LVM but which has LVM metadata because an installer prepared it to be used via LVM. https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/452350 (German) https://debianforum.de/forum/viewtopic.php?f=33&t=154131 Yes this is not a full-blown solution, but ...


0

This is Oom(Out of memory) killer. When your system runs out of memory, the linux kernel kills processes to free memory. An heuristic determines which process is the best candidate to get memory freed without damaging the system (typically, root owned processes are not best candidates). More details here : How OOM killer decides which process to kill first? ...


2

It's talking about killing a child process of the process nominated to be sent a KILL signal.


0

When you add a network interface to a bridge (e.g. brctl addif br0 eth0), all received packets are diverted to the bridge. If you try to set an IP address on the interface, it will be accepted, but you won't be able to use it to receive any packets - this is a common pitfall. In the source code, the bridge sets an rx_handler on the interface, and consumes ...


2

The key is that tables are grouping things by design intention. All your rules intended for filtering are in this place, all your NAT rules over there. Chains are sequences of rules, and the default chains are traversed at specific points in the path of a packet. In theory, you could add a rule that does filtering to, say, the NAT table. But the front end ...


1

Here is the link for your binary blobs. Quote from the link: Binary firmware blobs from the Debian non-free archive are installed when no good Free Software alternative exists.


1

make install simply copies the kernel image to the /boot directory. make modules_install copies the modules to /lib/modules/kernel-version/. Most linux distributions these days boot using grub, so you need to run update-grub to notice the new kernel image in /boot, and add an entry to boot it to the grub configuration file so you get the option to boot ...


1

Is is hardly documented and largely depends on a platform. For x86, next available id is assigned to CPU in the function generic_processor_info() So, for x86, cpu ids are depending on order in which we would call that function. It is called when APIC (interrupt controller) is initialized, while APIC settings are taken from ACPI MADT table and the ACPI ...


1

I had to copy the modules from Qubes and regenerate the initramfs using the sudo dracut -f command.


0

No, it is not a strict subset. It is not even a subset. Here is a demonstration, on a desktop PC running a major GNU/Linux distribution without any customisations that should affect the result, that there is at least one datum present in sysfs that is not present in procfs: $ grep -ir `cat /sys/block/sda/device/model | cut -f1 -d' '` /sys 2>/dev/null ...


1

Sysfs is not at all a subset of procfs. They serve different purposes, see the answers to What is the difference between procfs and sysfs?.


0

Isn't /boot mounted readonly, by chance? If so, mount -o rw,remount /boot && yum update kernel && mount -o ro,remount /boot (note: I'm rarely seen at centos hosts being an ALT Linux developer so YMMV)


1

[Self answer] While this is less than satisfying, we essentially went with @sjsam's advice and built a list of kernel versions by looking at the default kernel versions that ship with RedHat Enterprise Linux. Looking at versions of RHEL that are still in support today (April 2016), this gives us the list: 2.6.18 2.6.32 3.10.0 4.X (just for good measure, ...



Top 50 recent answers are included