Linux kernel, Linux general

memleak in fs/char_dev.c : alloc_chrdev_region?

Submitted by Anonymous
on December 7, 2009 - 8:50am

snippet:

int alloc_chrdev_region(dev_t *dev, unsigned baseminor, unsigned count,
const char *name)
{
struct char_device_struct *cd;
cd = __register_chrdev_region(0, baseminor, count, name);
if (IS_ERR(cd))
return PTR_ERR(cd);
*dev = MKDEV(cd->major, cd->baseminor);
return 0;
}

Masive HTB file lots of filters low performance

Submitted by ionut
on December 4, 2009 - 11:25am
Linux

#!/bin/bash
#
# HTB Script v0.0.1
#

# Clase
#tc qdisc del dev eth0 root

# 1: root disc
tc qdisc add dev eth0 root handle 1:1 htb default 120 r2q 10
# 1:1 root class
tc class add dev eth0 parent 1: classid 1:0x1 htb rate 950Mbit prio 0

# 1:50 Metro class
tc class add dev eth0 parent 1:0x1 classid 1:0x50 htb rate 800Mbit prio 0

Masive HTB file lots of filters low performance

Submitted by ionut
on December 4, 2009 - 11:11am
Linux

#!/bin/bash
#
# HTB Script v0.0.1
#

# Clase
#tc qdisc del dev eth0 root

# 1: root disc
tc qdisc add dev eth0 root handle 1:1 htb default 120 r2q 10
# 1:1 root class
tc class add dev eth0 parent 1: classid 1:0x1 htb rate 950Mbit prio 0

# 1:50 Metro class
tc class add dev eth0 parent 1:0x1 classid 1:0x50 htb rate 800Mbit prio 0

Wuzzup!

Submitted by Anonymous
on December 1, 2009 - 9:18pm
Linux

Association inode - pathname requested by a process

Submitted by Anonymous
on November 20, 2009 - 5:21pm
Linux

Hi all

i want create a LSM function that checks the properties of an inode using as criteria for filtering the pathname requested by a process, not this with symlinks resolved. In what part of the fs code is this information available? There's an existent security hook suitable for this purpose?

Thanks for replies.

Kernel 2.6.27.7-9 for Suse 11.1 help required with CPU Clock freq/speed/ticks

Submitted by subhan222
on November 19, 2009 - 6:57am
Linux

Kernel 2.6.27.7-9 for Suse 11.1 help required with CPU Clock freq/speed/ticks

Problem Statement: High CPU clock speed/freq
Requirement: Obtain a stable CPU clock ticks/speed/freq on Suse 11.1 with kernel 2.6.27.7-9
Questions: Please refer to the bottom of this page

Artifacts in Use:
-----------------------------------
OS:					Suse 11.1

BNX2 trouble (2.6.31)?

Submitted by Anonymous
on November 16, 2009 - 4:29am
Linux

Today got server freezes. In log last message:
Nov 15 23:04:05 !!! kernel: [433562.050053] ------------[ cut here ]------------
Nov 15 23:04:05 !!! kernel: [433562.050070] WARNING: at /build/buildd/linux-2.6.31/net/sched/sch_generic.c:246 dev_watchdog+0x262/0x270()
Nov 15 23:04:05 !!! kernel: [433562.050073] Hardware name: PowerEdge 1950

USB2 Driver Crash after enumeration

Submitted by kerneltrap
on November 6, 2009 - 4:23pm
Linux

Hi,

I am trying to port Synopsys OTG USB driver 2.90. to Xtensa Linux 2.6.29.1.

I got Synopsys OTG USB driver which was for the IPMate platform.

It has the support for the LM_bus and PCI bus. I am not using either

LM bus or PCI for the USB but Xtensa is using Platform bus, So I tried to use the same.

Generic system call interceptor

Submitted by nsa310
on November 3, 2009 - 9:03am
Linux

Hi all !
I am working on writing a generic system call handler, which can intercept (almost) all linux system calls. I have already written a simple module which intercepts chdir() system call. But, the thing is it doesn't sounds like a good idea to write separate function for each system call. So, I am trying to write a generic function which can intercept all the system calls.

premmpt_disable on all cpus

Submitted by Anonymous
on November 3, 2009 - 2:35am
Linux

my requirement is as follows.

I want disable preemption on all cpus. and then call a function

with on_each_cpu(preempt_disable, NULL, 1, 1); call I dont think it will work.

is there a way to achieve it ?

NFS mount failure

Submitted by Anonymous
on November 2, 2009 - 2:53pm
Linux

I have a problem I'm hoping someone can help me with...

I have an NFS server/client setup. On the server side, I export a romfs filesystem to the NFS. The client computers boot via PXE bootloader and mount their root filesystem via NFS, using tempfs/unionfs to make parts of the filesystem read/write (i.e., /etc, /var, etc)

soft lockup and hang

Submitted by gondim
on October 29, 2009 - 6:02pm
Linux

Hi,

I´m looking for this kernel messages. I would like to know how resolve this problem. Please help me. :)
My system is: CentOS 5.4 - kernel version 2.6.18-164.el5PAE
Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz
eth0: Ethernet controller: Intel Corporation 82566DC-2 Gigabit Network Connection (rev 02)

checking TSC synchronization

Submitted by handrix
on September 14, 2009 - 9:38am
Linux

Im' trying to upgrade a Linux kernel from 2.6.18 to 2.6.30.
The machine is a HP Proliant with Intel Xeon processor double six core. The 2.6.18 kernel work fine.

When booting with the 2.6.30, the kernel can not identify more than 8 CPU and i get this message :

checking TSC synchronization [CPU#0 -> CPU#8]: passed.

And the kernel could not boot.
Please can you help me with this

GDT assignment

Submitted by Anonymous
on September 12, 2009 - 5:22pm
Linux

In older kernels, GDT is set in arch/i386/kernel/head.S. Now, in newer kernels everything moved to arch/x86, bud in arch/x86/kernel/head_32.S i can't find where is GDT set. I mean something like


.quad 0x0000000000000000 /* NULL descriptor */
.quad 0x0000000000000000 /* not used */