November 2014 - kernelnewbies@kernelnewbies.org

skb headroom and tailroom
by Varun Sharma Nov. 19, 2014

Nov. 19, 2014

Hi, I have few basic question regarding headroom and tailroom in skb. 1. At time of skb allocation sk->sk_prot->max_header value is 272 . If we add tcp header+ip header+mac header outcome will always less than 272 .Then why allocate more ? Is it for future case ? 2. Why space left in headroom and tailroom ? Regards Varun

2 1

Advantages in Scalability of Softriqs versus Tasklets
by Nick Krause Nov. 19, 2014

Nov. 19, 2014

I am curious about reading through LKD, when to start using softriqs other tasklets in terms of when the limit of scalability of tasklets are hit. I am assuming after reading LKD it seems to happen when multiple cores would be of great advantage in scaling the bottom half of my or our people's code. In addition in seems to also happen when timing of interrupts needs to be in the range of under milliseconds to process the bottom half and this seems very important for the networking subsystem. Cheers and Thanks for Any Answers, Nick

1 0

Re: gso vs tso
by Varun Sharma Nov. 19, 2014

Nov. 19, 2014

If both are ON then which one works tso or gso ? On Wed, Nov 19, 2014 at 12:22 PM, Varun Sharma <vsdssd(a)gmail.com> wrote: > If both are ON then which one works tso or gso ? > > > > On Wed, Nov 19, 2014 at 12:07 PM, Raj Ravi <mekaviraj(a)gmail.com> wrote: >> TSO is a hardware feature whereas GSO is a software feature. >> >> In the sense, TSO needs "the device" to break the frame into MTU sized segments. >> BUT, GSO is a strategy followed by kernel to avoid processing smaller >> packets throughout the stack. >> >> Refer this: >> http://blogs.gnome.org/markmc/2008/05/28/checksums-scatter-gather-io-and-se… >> >> <<< >> GSO is a generalisation of this in the kernel. The idea is that you >> delay segmenting a packet until the latest possible moment. In the >> case where a device doesn’t support TSO, this would be just before >> passing the skb to the driver. If the device does support TSO, the >> unsegmented skb would be passed to the driver. >> <<< >> >> >> On Wed, Nov 19, 2014 at 11:40 AM, Varun Sharma <vsdssd(a)gmail.com> wrote: >>> Hi, >>> >>> I have basic question about gso(generic segmentation offload) vs >>> tso(TCP segmentation offload) >>> >>> 1. How both are different to each other? >>> >>> 2. If both are ON using ethtool which one is perform segmentation gso or tso ? >>> >>> >>> Regards >>> varun >>> >>> _______________________________________________ >>> Kernelnewbies mailing list >>> Kernelnewbies(a)kernelnewbies.org >>> http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies

1 0

gso vs tso
by Varun Sharma Nov. 19, 2014

Nov. 19, 2014

Hi, I have basic question about gso(generic segmentation offload) vs tso(TCP segmentation offload) 1. How both are different to each other? 2. If both are ON using ethtool which one is perform segmentation gso or tso ? Regards varun

2 1

sched_clock generic for all architectures
by Hemanth Kumar Nov. 18, 2014

Nov. 18, 2014

Hi Gerg, I want know why this two commit id is not backported v3.10 1 . sched_clock: Make ARM's sched_clock generic for all architectures 2 . arch_timer: Move to generic sched_clock framework where as the 1 commit is already backported v3.11 I back ported these two commit id 3.10 arm cortex-a9 boot's fine, Is there any issue with this commit's if it is backported to v3.10. Regards,

5 5

Re: what is CONFIG_OF
by manty kuma Nov. 18, 2014

Nov. 18, 2014

Thank you all. I got it. Generally naming in kernel is very precise but in this case I feel it is little confusing. Thank you again. On Tue, Nov 18, 2014 at 7:20 PM, sanjeev sharma <sanjeevsharmaengg(a)gmail.com > wrote: > This option will let you know whether you would like to leverage support > of Device tree or not ? > > Regards > Sanjeev Sharma > > On Tue, Nov 18, 2014 at 1:38 PM, Robert P. J. Day <rpjday(a)crashcourse.ca> > wrote: > >> On Tue, 18 Nov 2014, manty kuma wrote: >> >> > Hi, >> > >> > What is CONFIG_OF? I see that it is being widely used at many places >> > but could not understand in what scenarios i need to use it. There >> > is no help documentation available in kconfig for this. >> >> Open Firmware, aka Device trees. >> >> rday >> >> -- >> >> ======================================================================== >> Robert P. J. Day Ottawa, Ontario, CANADA >> http://crashcourse.ca >> >> Twitter: http://twitter.com/rpjday >> LinkedIn: http://ca.linkedin.com/in/rpjday >> ======================================================================== >> >> _______________________________________________ >> Kernelnewbies mailing list >> Kernelnewbies(a)kernelnewbies.org >> http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies >> > >

1 0

what is CONFIG_OF
by manty kuma Nov. 18, 2014

Nov. 18, 2014

Hi, What is CONFIG_OF? I see that it is being widely used at many places but could not understand in what scenarios i need to use it. There is no help documentation available in kconfig for this. Best Regards, Manty

4 3

mapping of inode and file structure in Open System call
by lokesh kumar Nov. 18, 2014

Nov. 18, 2014

Hi, I have a question regarding open systemcall "open(dev_name, O_RDWR, 0)" -->userspace suppose devname is /dev/my_device Now when we look into open file operation, we have prototype "int (*open) (struct inode *, struct file *);" could anyone tell me how inode and file structure pointer are filled. I think logic is written in fs/open.c but i got confused in callbacks used in it.could anyone please tell me which topics should i cover to understand open system call. Thanks.

3 3

Doubt in major and minor numbers
by me storage Nov. 16, 2014

Nov. 16, 2014

Hi All, I am reading chapter 3 in LDD3 i checked same topics in LDD2 so i got some confusions.According to my understanding Major number is used by kernel to identify the driver i.e major number tells about driver . Minor number is used by the driver module to know about the device which is using this module i.e minor number related to device number can i any one please tell am i correct? if wrong please explain me a bit more Thanks

2 1

do_coredump (elf_core_dump) and vm_file
by Sudharsan Vijayaraghavan Nov. 15, 2014

Nov. 15, 2014

Hi, I had to debug a coredump for a user process. I found the entire /proc/<pid>/maps data in core file, except the file names (vma->vm_file) as part of core generated. I was surprised to see this vma->vm_file is not dumped into core. Infact this is very much needed to debug cores. Why is that vma->vm_file not dumped into coredump? Any inputs will be of great help Thanks, Sudharsan

1 0