August 2013 - kernelnewbies@kernelnewbies.org

multiple skbs
by A B Aug. 8, 2013

Aug. 8, 2013

Hi, I'm working on a way to combine multiple SKBs that are chained, into a single SKB. Does anyone has any ideas on how to do this without using a copy operation. I heard about frag_list but quite couldn't understand the usage of this. Any documentation / online links would be helpful.

1 0

Has anyone worked with PF_RING
by Robert Clove Aug. 8, 2013

Aug. 8, 2013

Hi ALL, I have to understand the PF_RING part. I mean why it is being used,is it available in kernel freely? What are the benefits of PF_RING and what are the disadvantages of using it? Why we normally don't use it? If anyone has any clues please share..... Thanks

2 1

Network Driver From Scrach
by Robert Clove Aug. 7, 2013

Aug. 7, 2013

Hi All, I want to first understand and then write a network driver for simple Ethernet device (actually instead of inbuilt network driver i want to run my network driver). Actually the new driver should have PF_RING enable/support and Net filter usage. Although we can copy the driver too but i want to understand some driver and write my own so that i get clear understanding about network driver. Now can you please help me how should i proceed? First should i see how the packet flows in the kernel or how should i go and please provide some good links to study. Thanks

3 2

Not able to get crash dump
by neha naik Aug. 6, 2013

Aug. 6, 2013

Hi All, I have loaded the linux crashdump on ubuntu machine. I can manually generate the crashdump by the 'echo c > /proc/sysrq-trigger'. However, i am having a panic in a module i have written, which is not generating a core dump. I simply see the stack in the console and it kind of hangs there. I have to manually power it off and power it on ... Can someone explain why this happens? Is it because the kernel has gone into such a state that it cannot even follow the procedure for crash dump. Regards, Neha

4 5

Re: submit_bio() flush to disk immediately
by Rajat Sharma Aug. 6, 2013

Aug. 6, 2013

On Tue, Aug 6, 2013 at 1:20 PM, Prashant Shah <pshah.mumbai(a)gmail.com> wrote: > Hi, > > On Tue, Aug 6, 2013 at 1:15 PM, Rajat Sharma <fs.rajat(a)gmail.com> wrote: >> How are you reading and writing? What is your test utility? 'dd' has >> an option to specify iflag/oflag=direct, which would bypass buffer >> cache. But anyways this problem should not happen even if data is >> written to cache, you should read it back from cache. > > I am using cat command. Its a block that belongs to a file and I am > directly writing to the block. On reading the file I am getting stale > data. > Most likely you are getting stale data from filesystem page-cache. Please don't use cat, use dd with iflag/oflag=direct. > Regards. -Rajat

2 13

How to send the patch of a bug
by Iker Pedrosa Aug. 6, 2013

Aug. 6, 2013

Hi guys, I've solved a bug reported in bugzilla (https://bugzilla.kernel.org/show_bug.cgi?id=60401) But now I don't know what I have to do now. Do I have to send the patch by email as if it was a normal patch? Or do I have to upload the patch as an attachment to the previous web page? I have also seen that there is a command called git-send-bugzilla, can I use it? I hope that there is someone that can help me. -- Regards, Iker Pedrosa <ikerpedrosam(a)gmail.com>

3 2

Pushd and Propd issue in filesystem
by Saket Sinha Aug. 6, 2013

Aug. 6, 2013

I have to support pushd and popd from userspace in a filesystem that I have designed. In my filesystem driver, to implement it I have done the following for popd - the driver implementation is https://github.com/disdi/hepunion/blob/master/fs/hepunion/hepunion.h#L386 #define pop_root() \ current->fsuid = context->uid; \ current->fsgid = context->gid; \ recursive_mutex_unlock(&context->id_lock) for pushd - the driver implementation is https://github.com/disdi/hepunion/blob/master/fs/hepunion/hepunion.h#L393 #define push_root() \ recursive_mutex_lock(&context->id_lock); \ context->uid = current->fsuid; \ context->gid = current->fsgid; \ current->fsuid = 0; \ current->fsgid = 0 But this implementation is failing at run time. This is the script with which I am testing this driver. All pushd and propd are not working in this script. After trying many other alternatives, I think overriding commits would be a better choice. So then I would try overriding commits as the below link http://lxr.free-electrons.com/source/fs/open.c?v=3.8#L312 Regards, Saket Sinha Can somebody help

2 1

A question about -mm tree and -next tree.
by Zhan Jianyu Aug. 6, 2013

Aug. 6, 2013

I know that -next tree is doing these things daily: 1. reset HEAD to HEAD of linus's tree 2. merge about 221 subsytem trees automatically And I know that -mm tree : 1. contains patches which are going to applied into latest Linus release (3.x or 3.x-rcY) 2. most of the patches are included in -next (this is the original purpose of the -next tree - to reduce Andres Mortan's stress and make the patches more tested) So here comes my questions: I found that most of the -mm patches are in -next, which means they will be tested more by all developers , but there are still some patches are not included in -next tree. So why are they special and how they make their way into Linus 's mainline code ? -- Regards, Zhan Jianyu

2 1

Kernel Debugging
by Saket Sinha Aug. 6, 2013

Aug. 6, 2013

I am facing an issue where I have been developing a filesystem driver(over x86) which has become "unstable" by which I mean , at compile time it is building but during runtime it fails when I call the corresponding APIs from the user-space. In order to find out exactly where it is failing, I need to debug the driver. I am very new to kernel development. I have heard about KGDB and KDB but it takes quite a lot to get them working so waht are the other tools available? Regards, Saket Sinha

2 1

submit_bio() flush to disk immediately
by Prashant Shah Aug. 6, 2013

Aug. 6, 2013

Hi, I am trying to write a bio to a disk block directly using submit_bio(), it works without any errors but on read it shows the old data until I reboot the machine. Is there any way I can flush the particular bio/dirty page immediately or force it to flush before the next read ? code : https://github.com/prashants/km/blob/master/writeblock/writeblock.c Regards.

2 2