September 2012 - kernelnewbies@kernelnewbies.org

Performance of /dev/urandom and /dev/random
by Mihir Pandya Sept. 28, 2012

Sept. 28, 2012

Hey guys, I'm interested to know why generating cryptographically secure random numbers takes so long, particularly in my case. I have two Java classes that are generating a million random numbers. One is generating random numbers regularly while the other is using SecureRandom to generate the numbers. The SecureRandom method is taking almost double the time and I realised that it's because the method reads from /dev/urandom (not sure which) and obtains its random numbers from the entropy pool there. I wanted to know where I can read about the exact mechanism used to generate the cryptographically secure random numbers through SecureRandom and if there is any way to make it faster. -- Mihir Pandya

4 4

Compiling the source RPM file.
by K Arun Kumar Sept. 28, 2012

Sept. 28, 2012

I have a rpm for a module which I extracted like - rpm2cpio numactl-2.0.7-3.el6.src.rpm | cpio -idmv then I extracted the source like - tar -xzvf numactl-2.0.7-3.tar.gz then I tried to apply all the patches to available with this rpm to the extracted source which results in Hunks and rejections. I want to apply all patches and compile the source and get an exe. I am not much aware of the rpm build procedure I want to do similar thing to kernel source also Please help me... Regards, Arun

4 4

RE: Linux Kernel not responding to ARP requests
by Nitin Yadav Sept. 28, 2012

Sept. 28, 2012

Thanks for reply, Below are the ARP variables & the values set as default. arp_accept = 0 arp_ignore = 0 arp_announce = 0 arp_filter = 0 proxy_arp = 0 Thanks, Nitin Yadav -----Original Message----- From: Jeff Haran [mailto:Jeff.Haran@citrix.com] Sent: Friday, September 28, 2012 6:35 AM To: 'Mulyadi Santosa'; Nitin Yadav Subject: RE: Linux Kernel not responding to ARP requests > -----Original Message----- > From: kernelnewbies-bounces(a)kernelnewbies.org [mailto:kernelnewbies- > bounces(a)kernelnewbies.org] On Behalf Of Mulyadi Santosa > Sent: Thursday, September 27, 2012 5:57 PM > To: Nitin Yadav > Cc: kernelnewbies(a)kernelnewbies.org > Subject: Re: Linux Kernel not responding to ARP requests > > On Thu, Sep 27, 2012 at 1:51 PM, Nitin Yadav <Nitin.Yadav(a)mphasis.com> > wrote: > > Is there any way Kernel is dropping the ARP requests (from the Stand > > by Router to the Kernel)? > > > > If it's not dropping, is there any other reason for not replying to ARP? > > First thing first: are you sure there are no iptables rules that > drop/reject certain arp type? That would be arptables rules. iptables only works on IP packets. Jeff Haran Information transmitted by this e-mail is proprietary to MphasiS, its associated companies and/ or its customers and is intended for use only by the individual or entity to which it is addressed, and may contain information that is privileged, confidential or exempt from disclosure under applicable law. If you are not the intended recipient or it appears that this mail has been forwarded to you without proper authority, you are notified that any use or dissemination of this information in any manner is strictly prohibited. In such cases, please notify us immediately at mailmaster(a)mphasis.com and delete this mail from your records.

1 0

Linux Kernel not responding to ARP requests
by Nitin Yadav Sept. 28, 2012

Sept. 28, 2012

From: Nitin Yadav Sent: Thursday, September 27, 2012 12:16 PM To: 'netdev(a)vger.kernel.org' Subject: RE: Linux Kernel not responding to ARP requests Hi All, I am facing loss of connectivity between Linux system (2.6.18 kernel) & Cisco switch (6509) when HSRP is enabled. The Cisco switches (STAND BY) ARP queue were flushed and new MAC address were requests, but the kernel did not answer to this requests. After investigation I found out that the Cisco switch (STAND BY) is flushing the MAC address of the kernel port. Based on Cisco they are flush the MAC address of inactive port every 8 minutes. A small note about the protocol (HSRP):- Hot Standby Router Protocol (HSRP) is a Cisco proprietary redundancy protocol for establishing a fault-tolerant default gateway, and has been described in detail in RFC 2281. The protocol establishes a framework between network routers in order to achieve default gateway failover if the primary gateway becomes inaccessible,[1] in close association with a rapid-converging routing protocol like EIGRP or OSPF. By multicasting packets, HSRP sends its hello messages to the multicast address 224.0.0.2 (all routers) for version 1, or 224.0.0.102 for version 2[2], using UDP port 1985, to other HSRP-enabled routers, defining priority between the routers. The primary router with the highest configured priority will act as a virtual router with a pre-defined gateway IP address and will respond to the ARP request from machines connected to the LAN with the MAC address 0000.0c07.acXX where XX is the group ID in hex. If the primary router should fail, the router with the next-highest priority would take over the gateway IP address and answer ARP requests with the same mac address, thus achieving transparent default gateway fail-over. A HSRP Basics Simulation visualizes Active/Standby election and link failover with Hello, Coup, ARP Reply packets and timers. My queries :- Is there any way Kernel is dropping the ARP requests (from the Stand by Router to the Kernel)? If it's not dropping, is there any other reason for not replying to ARP? Thanks! Nitin Yadav Information transmitted by this e-mail is proprietary to MphasiS, its associated companies and/ or its customers and is intended for use only by the individual or entity to which it is addressed, and may contain information that is privileged, confidential or exempt from disclosure under applicable law. If you are not the intended recipient or it appears that this mail has been forwarded to you without proper authority, you are notified that any use or dissemination of this information in any manner is strictly prohibited. In such cases, please notify us immediately at mailmaster(a)mphasis.com and delete this mail from your records.

2 1

Questions about hrtimer interrupt handling
by John Gamboa Sept. 27, 2012

Sept. 27, 2012

Hello... I'm reading the code of the "hrtimer" [1] in order to use them in something I've been programming lately. I read the tutorial in [2] and [3] about them, but it seems that the code has changed a little since when it was written. Thus, I have two questions: 1 - How can I know what CPU is handling a timer interruption (while inside of the interruption function)? Is it possible to a timer be started by a core and have its interruption handled by another core? (if the answer is "no", then I could just make sure I know what is the core starting the timer) 2 - In [2] it is said that there is a "void *data" pointer in the hrtimer API. The code changed in newer kernel versions and I didn't find such a pointer anymore. Supposing there is still such a pointer, what would be the "equivalent" of the "void *data" pointer there? I mean: is there still a way to pass some data to the hrtimer callback function? If the answer is "yes", then how? Thanks for your attention... [1] http://www.cs.fsu.edu/~baker/devices/lxr/http/source/linux/include/linux/hr… http://www.cs.fsu.edu/~baker/devices/lxr/http/source/linux/kernel/hrtimer.c [2] http://lwn.net/Articles/167897/ [3] http://www.ibm.com/developerworks/linux/library/l-timers-list/ -- John Gamboa rabanetescebolas.blogspot.com

1 0

an online kernel crash book, and wondering what's deprecated
by Robert P. J. Day Sept. 27, 2012

Sept. 27, 2012

in my online travels yesterday, i ran across this gem, "Linux Kernel Crash Book": http://www.dedoimedo.com/computers/crash-book.html and am now wondering about what would constitute a reasonable (and minimal?) list of canonical kernel debugging tools. first, while the book above covers Linux Kernel Crash Dump (LKCD), the author freely admits that it's been pretty much obsoleted by the more recent and flexible kdump, so there seems to be little value in digging into LKCD (or, in my case, adding any coverage of it to a kernel debugging course, which i am currently designing). next, someone else's course i'm teaching next week has a kernel debugging chapter which opens with netdump and diskdump before moving onto kdump and kexec, but those earlier utilities are *also* deprecated these days, http://serverfault.com/questions/181554/how-should-i-capture-linux-kernel-p… so i would be tempted to skip any coverage of netdump and diskdump in favour of additional and more advanced coverage of kdump and kexec. along those lines, i'm just digging into ftrace and was wondering if it in any way obsoleted systemtap, but i've heard from more than one source that while ftrace is allegedly more powerful, systemtap still has its place and is worth talking about. so ... if one was going to put together a (small) toolbox of kernel debugging tools, what's worth covering? interested in any feedback. rday -- ======================================================================== Robert P. J. Day Ottawa, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================

4 4

Re: starting to patch kernel
by Constantine Shulyupin Sept. 26, 2012

Sept. 26, 2012

Most of work in drivers is very HW or application specific and unfortunately is not suitable for up-streaming . A few of my experience I've converted to LDT - Linux Driver Template: https://github.com/makelinux/ldt/blob/master/README.md I would be very glad to receive your feedback. BTW, the code passed checkpatch.pl. Thank you! On Wed, Sep 26, 2012 at 10:37 PM, Greg Kroah-Hartman <greg(a)kroah.com> wrote: > On Wed, Sep 26, 2012 at 10:21:44PM +0200, Constantine Shulyupin wrote: >> Hi >> >> I am experienced embedded Linux driver developer. I have some spare >> time which I want to contribute to mainstream Linux kernel. > > How about working to get some of the drivers you have worked on upstream > properly? That would be a good task that you would know a lot about, > right? > > good luck, > > greg k-h -- Constantine Shulyupin http://www.MakeLinux.com/ Embedded Linux Systems, Device Drivers, TI DaVinci

1 0

starting to patch kernel
by Constantine Shulyupin Sept. 26, 2012

Sept. 26, 2012

Hi I am experienced embedded Linux driver developer. I have some spare time which I want to contribute to mainstream Linux kernel. Unfortunately http://kernelnewbies.org/KernelJanitors/Todo is very old. Can you please suggest small tasks good for start? It can be code clean up, API updates. Thank you -- Constantine Shulyupin http://www.MakeLinux.com/ Embedded Linux Systems, Device Drivers, TI DaVinci

2 1

Missing __versions section on .ko
by Pritam Bankar Sept. 26, 2012

Sept. 26, 2012

Any idea what causes __versions section to get added in *.ko module. A simple hello.ko on ubuntu 12.04 i386 machine gets loaded successfully. Other module just doesn't get loaded giving error " Invalid module ". When I compared both modules with objdump -h I found __versions section was missing from second module and that can be the reason its not getting loaded. How can I add this section to other .ko ? What causes __versions to get added in ko ? Thanks, Pritam Bankar

3 6

getxattr field not initialized when trying to execute /init during boot
by stl Sept. 26, 2012

Sept. 26, 2012

Hello all, I am facing a problem when trying to boot linux 2.6.37 on a new architecture. At the end of the boot, it tries to launch /init by executing a sys_execve() Here is the what it does: sys_execve() |-> do_execve() |-> prepare_binprm() |->security_bprm_set_creds() |->cap_bprm_set_creds() |->get_file_caps() |->get_vfs_caps_from_disk() In the get_vfs_caps_from_disk(), these is a verification to know if inode->i_op->getxattr has been initialized. In my case, it returns and error value, because this field is not initialized. So I am wondering by who and where this field is initialized? And what is the aim of getxattr? Thanks in advance!

2 1