March 2013 - kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

question about
by Kevin Wilson March 25, 2013

March 25, 2013

Hi, Any idea which methods adds the following entries: /sys/devices/system/clocksource/clocksource0 /sys/devices/system/clocksource/clocksource0/available_clocksource ? Best, KevinW

3 2

feature removal schedule is removed ?
by David Shwatrz March 25, 2013

March 25, 2013

Hi, Once upon a time there was a feature removal schedule in Documentation: http://lxr.free-electrons.com/source/Documentation/feature-removal-schedule… Was it removed ? I cannot find it in recent kernels (like 3.9.0-rc4) Rgs DS

3 2

kernel build error
by Kumar amit mehta March 25, 2013

March 25, 2013

I'm facing a kernel build issue with the latest kernel on my machine. It seem like a configuration issue, and I haven't been able to resolve it for sometime, hence need your help. The error message is as below: <error snip> ERROR: "copy_from_user_overflow" [net/core/pktgen.ko] undefined! ERROR: "copy_from_user_overflow" [fs/binfmt_misc.ko] undefined! ERROR: "copy_from_user_overflow" [drivers/staging/cxt1e1/cxt1e1.ko] undefined! ERROR: "copy_from_user_overflow" [drivers/isdn/icn/icn.ko] undefined! ERROR: "copy_from_user_overflow" [drivers/isdn/hardware/avm/b1.ko] undefined! <error snip> The subroutine "copy_from_user_overflow" is defined in lib/usercopy.c, and the 'Makefile' in the lib directory looks something like this: <snip from lib/Makefile> lib-$(CONFIG_ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS) += usercopy.o <snip from lib/Makefile> I think that this config option should be a built-in ('y') and I see that this config option is indeed set to 'y' in the .config file. <snip from .config> amit@ubuntu:~/linux-next/linux-next$ grep CONFIG_ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS .config CONFIG_ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS=y <snip from .config> So what am I missing here ? Some information about my git tree: amit@ubuntu:~/linux-next/linux-next$ git remote show origin * remote origin Fetch URL: git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git amit@ubuntu:~/linux-next/linux-next$ git tag -l next-*|tail -n 1 next-20130314 <--- latest tag amit@ubuntu:~/linux-next/linux-next$ git branch * 15-March <--- -- branch based on the latest tag 22-Feb-13 -Amit

5 15

Why we Use both mm_users and mm_count in struct mm_struct{ }
by Niroj Pokhrel March 25, 2013

March 25, 2013

Hi all, I have been going through the Address Space in the linux and came across two variables in the struct mm_struct and I'm a bit confused about the two: struct mm_struct { .......... atomic_t mm_users; atomic_t mm_count; ............ } Basically, after reading through I came to understand that mm_users are used to store the number of processes or threads using the memory so depending upon the number of users it is going to be set. But, I am confused with mm_count, it is said the mm_count is increment by one for all the mm_users and when all the mm_users value is reduced to zero then mm_count is reduced. So, my question is can the value of mm_count be ever greater than one because all the mm_users are equivalent to mm_count . So, if not then why are we using the mm_count as we can simply remove the memory areas whenever the mm_users count reduce to zero. May be the explanation is simple but I'm lost. Thanking all of you in advance. Regards, Niroj Pokhrel

3 5

initramfs.cpio
by ishare March 25, 2013

March 25, 2013

Hi : I find that my initramfs_data.cpio generated by gcc does not contain init files ,which should be executed by terminal initialization. My initramfs_data.cpio only contains these : /dev /dev/consol /root . where to search the init file? thanks

2 2

optimization in kernel compile
by ishare March 24, 2013

March 24, 2013

Is it needed or must to compile fs and driver with -O2 option when compiling kernel ? thanks!

2 5

time: Questions about kernel function timekeeping_bigadjust()
by Dong ZHu March 23, 2013

March 23, 2013

Hi All, Could anybody explain the following kernel code for me in detail ? I am royally confused on it. kernel/time/timekeeping.c timekeeping_bigadjust() { /* * Use the current error value to determine how much to look ahead. * The larger the error the slower we adjust for it to avoid problems * with losing too many ticks, otherwise we would overadjust and * produce an even larger error. The smaller the adjustment the * faster we try to adjust for it, as lost ticks can do less harm * here. This is tuned so that an error of about 1 msec is adjusted * within about 1 sec (or 2^20 nsec in 2^SHIFT_HZ ticks). */ error2 = tk->ntp_error >> (NTP_SCALE_SHIFT + 22 - 2 * SHIFT_HZ); } Thanks in advance. -- Best Regards, Dong Zhu MySite: http://bluezd.info

1 0

[Need Help] time: questions about kernel function timekeeping_bigadjust()
by Dong Zhu March 22, 2013

March 22, 2013

Hi All, Could anybody explain the following kernel code for me in detail ? I am royally confused on it. kernel/time/timekeeping.c timekeeping_bigadjust() { /* * Use the current error value to determine how much to look ahead. * The larger the error the slower we adjust for it to avoid problems * with losing too many ticks, otherwise we would overadjust and * produce an even larger error. The smaller the adjustment the * faster we try to adjust for it, as lost ticks can do less harm * here. This is tuned so that an error of about 1 msec is adjusted * within about 1 sec (or 2^20 nsec in 2^SHIFT_HZ ticks). */ error2 = tk->ntp_error >> (NTP_SCALE_SHIFT + 22 - 2 * SHIFT_HZ); } Thanks in advance. -- Best Regards, Dong Zhu MySite: http://bluezd.info

1 0

Confusion in the usage of linux key management APIs
by Ranjan Sinha March 22, 2013

March 22, 2013

Hi, I need to store some crypto keys in the kernel where each key is related to a path on the disk. A restricted set of users can create files on such a path but the crypto keys are shared by all such user. I am thinking of using linux kernel key management facilities for my project. I have a usermode application that manages key and a kernel module. However the add_key interface is not clear to me. How do I create a global keyring that is accessible to everyone? Man page of add_key has this to say about add_key -----------------------8<------------------------------------8<---------------------- key_serial_t add_key(const char *type, const char *description, const void *payload, size_t plen, key_serial_t keyring); DESCRIPTION add_key() asks the kernel to create or update a key of the given type and description, Instantiate it with the payload of length plen, and to attach it to the nominated keyring and to return its serial number. The destination keyring serial number may be that of a valid keyring to which the caller has write permission, or it may be a special keyring ID: KEY_SPEC_THREAD_KEYRING This specifies the caller's thread-specific keyring. KEY_SPEC_PROCESS_KEYRING This specifies the caller's process-specific keyring. KEY_SPEC_SESSION_KEYRING This specifies the caller's session-specific keyring. KEY_SPEC_USER_KEYRING This specifies the caller's UID-specific keyring. KEY_SPEC_USER_SESSION_KEYRING This specifies the caller's UID-session keyring. -----------------------8<------------------------------------8<---------------------- I understand that I may need to register my own "type" in the kernel using register_key_type API in the kernel. However I do not understand what value should be passed to the keyring parameter. Also what is meant by possessor in assigning permissions? Also what should be passed to uid and gid parameters in the call to kernel API key_alloc() -- Regards, Ranjan

1 0

Userspace access to PCIe - SATA device memory BAR
by John Beard March 21, 2013

March 21, 2013

Hi, I have a PCIe - SATA bridge device, and I would like to access the I2C control registers in the PCI memory region (BAR5) - these use just two addresses in the BAR. I have tried, in userspace, to mmap() the /sys/bus/pci/devices/..../resource5 file and read from that, but all I get is 0xFF in every byte. I used pcimem (https://github.com/billfarrow/pcimem.git) to try to read the BAR. The device is already matched by the 'ahci' driver, and the attached disks are visible and mountable. If I unbind the ahci driver to use uio, I can of course no longer use the device as a SATA bridge. In other PCI devices which are also matched by ahci, I can mmap and read published BARs fine without unbinding the ahci driver. What is the correct way to fiddle with the memory in the BARs of a PCI device, while still letting the normal driver run the device? I am not doing anything related to AHCI or SATA, just reading and writing to the "side-channel" communication hardware. Best regards, John

1 0