- kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

Re: Query about a PCI device error detection.
by Nilesh July 13, 2011

July 13, 2011

Resending the email to kernelnewbies list as well (have sent the same to edac mailing list). Appreciate any pointers on below query, if any one has worked on edac. On Wednesday 06 July 2011 03:56 PM, nilesh wrote: > Hi, > > I am fairly new to the edac driver and have some basic query regarding > the error detection in case of PCI devices. I could see the module for > memory controller's error detection is done s.t. UE/CE/controller > specific errors are scanned and detected, also the PCI errors are also > scanned. > > E.g. in case of i5400 : > i5400_edac.c : > i5400_probe1() > { > > 1285 /* Set the function pointer to an actual operation function > */ > 1286 mci->edac_check = i5400_check_error; > ... > 1312 /* allocating generic PCI control info */ > 1313 i5400_pci = edac_pci_create_generic_ctl(&pdev->dev, > EDAC_MOD_STR); > } > > And we check the PCI status registers for the errors. > edac_pci_generic_check() --> edac_pci_do_parity_check() --> > edac_pci_dev_parity_test() > > So, i5400_check_error - detect UE/CE. > edac_pci_generic_check - detect PCI errors. > > Query: > ------ > * If one wants to get the errors harvested for non-MC device (e.g. > ethernet controller etc.) is it going to be done through > edac_pci_generic_check() automatically - as we already scan through all > PCI devices in a loop? Is it required to add the separate module for > that device? > -- Thanks, Nilesh

1 0

How much control does a block driver have over the buffer cache?
by Neil Baylis July 13, 2011

July 13, 2011

I'm trying to clarify my picture of how the buffer cache works. I notice that if I do the following: dd if=/dev/sda7 of=/dev/null bs=10k count=100 The first time will be slow, and the second and subsequent times will be very fast because the data are now in the buffer cache. However, this is only true if there's a filesystem mounted on the device. This leads me to think of the buffer cache as associated with the filesystem rather than the device. For example, would it be possible to create a block driver that does not use the buffer cache? I think the answer must be 'no' because the caching happens before the block driver gets any requests to move data.

2 1

Kernel panic - not syncing: Attempted to kill init!
by 史星星(研六福州) July 13, 2011

July 13, 2011

Hi, I got a problem when trying to boot linux 2.6.32.13 on my Octeon CN5650 board. Booting logs below: ...omit…. EXT3-fs warning: maximal mount count reached, running e2fsck is recommended kjournald starting. Commit interval 5 seconds EXT3 FS on mtdblock1, internal journal EXT3-fs: recovery complete. EXT3-fs: mounted filesystem with ordered data mode. VFS: Mounted root (ext3 filesystem) on device 31:1. Freeing unused kernel memory: 176k freed Kernel panic - not syncing: Attempted to kill init! <-- kernel panic here What I do: 1、The busybox init is OK in another version kernel .I replace init with “Hello world” prog or dead loop prog, kernel panic still. 2、 run_init_process(“/sbin/init”) --> kernel_execve --> sys_execve so I add some debug msgs in sys_execve. asmlinkage int sys_execve(nabi_no_regargs struct pt_regs regs) { int error; char * filename; printk("enter sys_execve\n"); filename = getname((char __user *) (long)regs.regs[4]); error = PTR_ERR(filename); if (IS_ERR(filename)) goto out; error = do_execve(filename, (char __user *__user *) (long)regs.regs[5], (char __user *__user *) (long)regs.regs[6], &regs); putname(filename); printk("out sys_execve\n"); out: return error; } Boot logs: VFS: Mounted root (ext3 filesystem) on device 31:1. Freeing unused kernel memory: 176k freed enter sys_execve out sys_execve Kernel panic - not syncing: Attempted to kill init! <-- kernel panic here My question: 1.the kernel panic happened when return to userspace prog? 2.how to solve this problem? Thank you!

2 2

data transfer from kernel to user space
by Amit Nagal July 12, 2011

July 12, 2011

Hi , Normally we use copy_to_user / copy_from_user api's to transfer data to/from user space from/to kernel space . Can i be suggested with any other data transfer mechanism b/w kernel - user space for a char device , which are much faster than copy_xxx ? Regards Amit Nagal

4 4

faster way to find task
by loody July 11, 2011

July 11, 2011

Dear all: Is there faster way to find a give name of task? for example, I have a thread, named "test", is there faster way to get the task struct by passing "test" to some kernel api? -- Regards,

2 3

Re: Need basics of booting process
by Rigved Rakshit July 11, 2011

July 11, 2011

> > I want to know the basics linux booting process. I googled for this but i > didn't find the detailed things anywhere. I want to know the boot process > including the hardware work flow. Can anyone give me the detailed tutorial > to learn about this. > > First, when you power up the system the bios goes through a POST, which is > power-on self test. This just checks that hardware is working, etc. Then it > looks for a boot sektor, which must be 512 bytes long and end with a magic > number, and loads the kernel. From there, you can look at the head.s > assembly, and I think someone posted a tutorial here a while back--maybe > they'll post that again. > 1. The POST (Power-On-Self-Test) is done. POST checks all the processors that are connected, the amount of RAM available, disks that are connected, network cards available etc. 2. Then the BIOS (Basic Input/Output Services) program is loaded. The BIOS searches for IDE, SCSI, PATA drives. Then it searches for active (bootable) partition(s). Using this knowledge, it checks the information in the Master Boot Record (MBR) of the drive (that has the active partition(s)). Then, it loads the bootloader (mostly GRUB - Grand Unified Bootloader) from the MBR. 3. Grub knows how to read all kinds of partitions. So, it can boot from any file system. 1. It checks and loads the settings in /boot/grub/grub.conf. This file contains path for root = (hd0,0), kernel = vmlinuz, initrd = initrd.img (cpio archive). 2. The Linux kernel is loaded into the RAM. It knows how to unpack the cpio archive which was specified by initrd parameter in grub.conf. The kernel loads this unpacked initial RAM disk and root file system (also called initramfs) into the RAM. Then it starts executing the /etc/init file present in this initramfs. 1. initrd: It contains the initial RAM disk and the initial root file system. These are loaded into the RAM before the actual root file system from the hard disk is loaded. 2. ntldr: In case another bootloader is required to be loaded, then the 'chainloader+1' command tells grub to load the bootloader mentioned. ntldr is the Windows bootloader. In Windows, all drives are loaded in user-space. 4. init starts populating /dev with the devices that are connected. init inserts kernel modules (.ko files) to detect the connected hardware. 1. init loads the hard disk. Then, init "switches root" to the root file system in the hard disk. Finally, before terminating, this init file of the initramfs passes control to the init file in the hard disk. 2. init loads the partitions which are mentioned in /etc/fstab. 3. Then, all the files up to and including the required "init level" are executed. 1. These include rc.sysinit, rc.local (directory) and rcX.d (directory - where X refers to a whole number). 2. If you check the /etc/rc0.d directory, you will find symlinks starting with either a 'K' or a 'S'. This is followed by a number. This convention is followed purposely. All symlinks with a starting K are 'kill' commands. All symlinks with a starting S are start commands. All the symlinks are followed in lexicographic order. So, all the kill commands are executed first (K00*, K01* etc.), followed by the start commands (S00*, S01* etc.). 3. So, if you want to boot to init level 1, then all the commands in /etc/rc1.d directory are executed. If you want a Graphical Environment (i.e., init level 5), then all commands from rc1.d up to rc5.d are executed. 4. init level 0 is for shutdown and init level 6 is for restart (i think; not sure about init level 6). This is a brief introduction to the boot process. Modern OSs require much more than these basic things. Also, this boot process refers to the SysVInit model. Then is another model used by Upstart, which Ubuntu uses. Watch this video to learn more about Upstart: http://www.youtube.com/watch?v=ssnZBQqnze8 Hope this helps! Rigved Rakshit

2 1

Re:Interview question
by Bond July 11, 2011

July 11, 2011

On Tue, Jul 5, 2011 at 9:33 PM, Greg Freemyer <greg.freemyer(a)gmail.com> wrote: > On Tue, Jul 5, 2011 at 9:43 AM, Mandeep Sandhu > <mandeepsandhu.chd(a)gmail.com> wrote: >>> Let me simplify the question. >> >> I'll attempt to answer your questions...for my own edification! :) >> >>> >>> 1) What are the FIVE classic system calls for interfacing with a >>> character device. (ie. If it did not exist in 1970, don't list it). >> >> open/close/read/write/seek?...and the infamous ioctl. > > You listed 6. Historically character devices don't seek. So drop > that one to get to 5. > > (And yes, you can find char devices that implement seek, but ioctl is > far more common. And when I learned UNIX 30 years ago, it was just > the basic 5 I'm pretty sure that actually have matching driver code..) > > That is, drivers don't implement seek. seek() just sets a offset > variable tracked in the kernel. The driver is not even woken up when > the happens. > > That is if userspace did: > > open() > seek(1) > seek(2) > seek(3) > write() > > The driver would get invoked for open, and again for write. The thing which he wanted to go in depth was why and how will the drivers open() ,write() get invoked. I tried my attempt that depending upon vendor and device id register_chrdev will have registered etc. My guess is he probably wanted to know the system call routine that will get executed with these open and close in kernel space. i.e. for example __NR_open() ,__NR_read() and the system call numbers that would get involved in this process.

1 0

Linux Scheduler Statistics
by Vipin July 11, 2011

July 11, 2011

Hi All, Posting it from my parallel question at LQ http://www.linuxquestions.org/questions/linux-kernel-70/proc-schedstats-sta… I am trying to understand the statistics posted by /proc/schedstat. The documentation I could find on the subject was taken from here: Linux Scheduler Statistics http://eaglet.rain.com/rick/linux/sc...format-15.html I am running a FC14 2.6.35.6-45.fc14.x86_64 Below is a dump for my statistics version 15 timestamp 4356814703 cpu0 46540 0 96049852 39682767 50849563 28504181 6402127420973 1547256962507 55119801 domain0 00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000003 22495 21765 675 1037826 94 33 0 21765 5797 5345 416 1801124 55 0 1 5344 757236 702579 46725 114900414 9212 244 119 702460 44 13 31 0 0 0 0 0 0 375257 21494 0 domain1 00000000,00000000,00000000,00000000,00000000,00000000,00000000,0000000f 5593 4668 737 2285778 273 25 7 4661 5790 5750 34 266075 19 0 7 2801 374455 353235 19119 49007529 2851 30 4416 348819 17 4 13 0 0 0 0 0 0 93063 4192 0 ... Removed the rest of the cpu/domains Can somebody please explain me that why my cpu fields (2-4) are not zero as suggested in the document ? Am I doing something wrong ? or are the results different for x86 & x86_64 boxes? Thanks. (P.S. I am running an i5 with 4 cores) Linux Boot process - somebody asked for it as I checked the archives http://www.opensourceproject.org.cn/article.php?id=182 Read Appendix A Thanks, Vipin Jain

1 0

Problem in compiling a kernel module
by Ponkumaran Annadurai July 10, 2011

July 10, 2011

I am trying to compile a sample hello world module, but when issue make command, it says that "make: Nothing to be done for `all'.". Also I tried the command "make hello", then it is showing that "linux/module.h no such file". But I am in the newly compiled kernel. What might be the problem? -- thanks, kumaran

3 9

timer problem
by Prashant Shah July 10, 2011

July 10, 2011

Hi, I am test out a simple schedule_timeout() code. When I insert the module kernel goes in a infinite loop printing this in the log : Jul 6 22:04:35 ubuntupc kernel: [ 5738.281659] Pid: 0, comm: swapper Not tainted 2.6.35-22-generic #33-Ubuntu Jul 6 22:04:35 ubuntupc kernel: [ 5738.281663] Call Trace: Jul 6 22:04:35 ubuntupc kernel: [ 5738.281671] [<ffffffff8104d15a>] dequeue_task_idle+0x3a/0x50 Jul 6 22:04:35 ubuntupc kernel: [ 5738.281679] [<ffffffff81052d6a>] dequeue_task+0x9a/0xb0 Jul 6 22:04:35 ubuntupc kernel: [ 5738.281687] [<ffffffff81052dae>] deactivate_task+0x2e/0x40 Jul 6 22:04:35 ubuntupc kernel: [ 5738.281693] [<ffffffff81586ba9>] schedule+0x4a9/0x830 Jul 6 22:04:35 ubuntupc kernel: [ 5738.281702] [<ffffffff81008ddb>] cpu_idle+0xeb/0x110 Jul 6 22:04:35 ubuntupc kernel: [ 5738.281710] [<ffffffff815832ad>] start_secondary+0x100/0x102 Any ideas on what is going on ? ---------------------------------- <testmod.c> ---------------------------------- #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> #include <linux/timer.h> #include <linux/delay.h> #include <linux/sched.h> struct timer_list tim; void timfunc(unsigned long data) { printk("hi this is a timer %lu\n", data); set_current_state(TASK_INTERRUPTIBLE); schedule_timeout(10 * HZ); printk("hi again this is a timer after timeout %lu\n", data); } static int __init init_testmod(void) { init_timer(&tim); tim.expires = jiffies + HZ*5; tim.data = 1000; tim.function = timfunc; add_timer(&tim); return 0; } static void __exit exit_testmod(void) { printk("end of module\n"); return; } module_init(init_testmod); module_exit(exit_testmod); MODULE_LICENSE("GPL"); ------------------------------ </testmod.c> ---------------------------------- ---------------------------------- <Makefile> ---------------------------------- obj-m += testmod.o KERNELDIR ?= /lib/modules/$(shell uname -r)/build PWD := $(shell pwd) default: $(MAKE) -C $(KERNELDIR) M=$(PWD) clean: rm -rf *.o *~ core .depend .*.cmd *.ko *.mod.c .tmp_versions depend .depend dep: $(CC) $(CFLAGS) -M *.c > .depend ifeq (.depend,$(wildcard .depend)) include .depend endif ---------------------------------- </Makefile> ----------------------------------

4 6