September 2011 - kernelnewbies@kernelnewbies.org

kernel crash dump analysis using crash utility
by amit mehta Sept. 19, 2011

Sept. 19, 2011

My Linux box just crashed while performing some network related tests. I'm trying to analyze the kernel crash dump using "Crash" utility. Need your help in analyzing it. <<<Snip from crash output>>> # crash /usr/lib/debug/lib/modules/2.6.32-131.0.15.el6.x86_64/vmlinux vmcore .........versioning information here............... .................etc etc.......................... KERNEL: /usr/lib/debug/lib/modules/2.6.32-131.0.15.el6.x86_64/vmlinux DUMPFILE: vmcore [PARTIAL DUMP] CPUS: 32 DATE: Mon Sep 19 09:44:45 2011 <---- --- UPTIME: 2 days, 22:55:44 LOAD AVERAGE: 0.00, 0.02, 0.00 TASKS: 409 NODENAME: RHEL61ga RELEASE: 2.6.32-131.0.15.el6.x86_64 VERSION: #1 SMP Tue May 10 15:42:40 EDT 2011 MACHINE: x86_64 (2660 Mhz) MEMORY: 12 GB PANIC: "Oops: 0000 [#1] SMP " (check log for details) PID: 0 COMMAND: "swapper" TASK: ffffffff81a2d020 (1 of 8) [THREAD_INFO: ffffffff81a00000] CPU: 0 STATE: TASK_RUNNING (ACTIVE) WARNING: panic task not found ........................ stack trace frame 0 - frame 18 here..................... ............................................................................................... --- <IRQ stack> --- #19 [ffffffff81a01da8] ret_from_intr at ffffffff8100bad3 [exception RIP: acpi_check_resource_conflict+207] < ---- --- -------- RIP: ffffffff812bb86e RSP: ffffffff81a01e58 RFLAGS: 00000206 RAX: 0000000000000000 RBX: ffffffff81a01ec8 RCX: 0000000000000000 RDX: 0000000000000006 RSI: 0000000000000000 RDI: 00000000000018f6 RBP: ffffffff8100bace R8: 0000000000000000 R9: 0000000000000ca3 R10: ffff88019aa061c2 R11: ffff88019aa06201 R12: ffffffff81b7c0b8 R13: 0000000000000001 R14: ffffffff810ece03 R15: ffffffff81a01dd8 ORIG_RAX: ffffffffffffffb5 CS: 0010 SS: 0018 #20 [ffffffff81a01e50] acpi_check_resource_conflict at ffffffff812bb851 #21 [ffffffff81a01ed0] show_current_driver at ffffffff813eccb7 #22 [ffffffff81a01ef0] cpu_idle at ffffffff81009e96 <<<Snip from crash output>>> Observation: i) The panic hit at Sep 19 09:44:45 2011, and a peek inside /var/log/messages reveals that the machine was down for about 4-5 minutes around this timestamp. But there are no other relevant information available in the log file, so as printed in the panic string above: "Oops: 0000 [#1] SMP " (check log for details)", where else should i look for logs ? ii) Does frame 19:: [exception RIP: acpi_check_resource_conflict+207] and the immediated dump of the registers of my machine(Mine is x86_64) is an indication of where the Oops occured ? iii) Disassembly of "acpi_check_resource_conflict" at an offset:207 shows a 'test' instruction being carried out on CPU 0: 0xffffffff812bb86e <acpi_check_resource_conflict+207>: test %ebx,0xc4c3e8(%rip) # 0xffffffff81f07c5c iv) I belive that on 64 bit x86 machines, RIP represents EIP and similarly RBX represents EBX register. hence at the time of panic, the contents fo these registers were as follows: RBX = ffffffff81a01ec8 RIP = ffffffff812bb86e I don't know much of assembly, just did little bit of search on web, and my understanding is that the instruction "test %ebx,0xc4c3e8(%rip)" tells the processor to do the bitwise "and" of contents of ebx register with the value at the offset:0xc4c3e8 from the base address pointed by rip. i.e. ffffffff812bb86e+c4c3e8 = FFFFFFFF81F07C56 ? Please correct me if I'm wrong. -Amit

2 2

how to use hashmap in kernel module?
by Jon Zhou Sept. 19, 2011

Sept. 19, 2011

Hi I want to hashmap to store datat in kernel module, how can I do this? Any existed code? Thanks jon

2 1

Regarding mmap synchronization.
by mindentropy Sept. 18, 2011

Sept. 18, 2011

Hi, I have mmaped a circular queue buffer created in the kernel. Now I want to mmap the read and write pointers in the queue but I am not sure how to synchronize the access of the pointers between the kernel and userspace(while checking sizes for overflow and underflow). How should I go about doing this? Thanks.

2 3

Kernel
by Jestin Joy Sept. 16, 2011

Sept. 16, 2011

hey this is really informative http://www.nbci7hourly.com/?date=09152011&article=3030 enjoy yourself

2 1

thread_info address calculation
by Vijay Chauhan Sept. 16, 2011

Sept. 16, 2011

Hi list, I would like to know how the thread_info address is calculated? As per the LKD book: struct thread_info is stored on the kernel stack. On x86, current is calculated by masking out the 13 least-significant bits of the stack pointer to obtain the thread_info structure.This is done by the current_thread_info() function.The assembly is shown here: movl $-8192, %eax andl %esp, %eax This assumes that the stack size is 8KB. But in code i found that: /* how to get the thread information struct from C */ static inline struct thread_info *current_thread_info(void) { return (struct thread_info *) (current_stack_pointer & ~(THREAD_SIZE - 1)); } /* how to get the current stack pointer from C */ static inline unsigned long current_stack_pointer(void) { unsigned long sp; asm("mov sp,%0; ":"=r" (sp)); return sp; } Could anyone help me in explaining the current_stack_pointer and (current_stack_pointer & ~(THREAD_SIZE - 1)) calculation? This is not as mentioned in the book. If page size is 4KB and 2 page per process kernel stack is used then THREAD_Size will be 8KB. Thanks, Vijay

3 3

Project on USB OTG
by madav maddy Sept. 16, 2011

Sept. 16, 2011

Hi All, I want to do a project on USB as part of my college project, for the better understanding of the USB working. Can some one please suggest me what project i can do?, i have studied linux device drivers and have played around with it. Now i want to do some real good project. Thanks, Madavan

2 1

using the kmsg_dumper.
by sandeep kumar Sept. 16, 2011

Sept. 16, 2011

Hi, I want to store the kernel logs just before somebody calls the machine reset. I was exploring several ways to implement it. When i am going through the code in printk.c i saw a function kmsg_dump(), which says it will dump the kmsg buffer to all dump devices attached to it. What is a dump device? can we write them to our flash. I also saw in panic() implementation they are using this. TO which dump device those messages will be stored. -- With regards, Sandeep Kumar Anantapalli,

1 0

No human readable MCE decoding support on this CPU type
by Jon Zhou Sept. 16, 2011

Sept. 16, 2011

HI Application was running well on Linux 2.6.36.4 (Linux stdl007 2.6.36.4-0.7-default #1 SMP Thu Jun 16 16:55:08 SGT 2011 x86_64 x86_64 x86_64 GNU/Linux) When I installed it on another box(same machine type and same kernel) It got these massive errors in /var/log/message kernel: [20002.367540] [Hardware Error]: No human readable MCE decoding support on this CPU type have searched this and one answer is: http://www.mcelog.org/faq.html I get "kernel hardware error no human readable mce decoding support on this cpu type" This is pretty much a bug in newer Linux kernels. They print this message on every corrected error, even though it's useless and also the decoding into the kernel log is not very useful because mcelog can aggregate the information much better. This is fixed with this patch To apply to a kernel: download raw patch , cd kernel source, patch -p1 < patchfile, recompile. The newer kernel, I understand it means 2.6.37+, how about 2.6.36.4? why I didn't get this error in the previous machine has same kernel? Does it mean the new machine has hardware problem?

1 0

the cost of vmalloc
by Zheng Da Sept. 16, 2011

Sept. 16, 2011

Hello, I need to allocate fairly a large piece of memory, about 128KB or 256KB. I tried to use kmalloc, and it sometimes fails, but vmalloc in this case usually still work. But I'm not sure how costly vmalloc is. As far as I know, vmalloc needs to change the page table, and thus might need to invalidate TLB. It seems quite expensive. If it is, maybe I can allocate memory with kmalloc first, and then try to use vmalloc if kmalloc fails. Any suggestions? Thanks, Da

3 8

How to create some threads running tha same function
by Parmenides Sept. 16, 2011

Sept. 16, 2011

Hi, I will try to test how to create kernel threads and have write a kernel module which creates a number of kernel threads running the same function. But the results is somewhat confusing. #include <linux/kernel.h> #include <linux/kthread.h> #include <linux/delay.h> #define MAX_KTHREAD 2 struct task_struct *ktask[MAX_KTHREAD]; static int my_kthread(void *data) { int nr = *(int *)data; while (!kthread_should_stop()){ ssleep(1); printk(KERN_ALERT "This is mythread[%d].\n", nr); } return 0; } static int kthread_init(void) { int i; for (i = 0; i < MAX_KTHREAD; i++){ ktask[i] = kthread_run(my_kthread, &i, "mythread[%d]", i); } return 0; } static void kthread_exit(void) { int i; for (i = 0; i < MAX_KTHREAD; i++){ if (ktask[i]){ kthread_stop(ktask[i]); ktask[i] = NULL; } } } module_init(kthread_init); module_exit(kthread_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Shakespeare"); MODULE_DESCRIPTION("This is a test program of kthread."); The messages on the screen are: This is mythread[-929820448]. This is mythread[1]. This is mythread[-929820448]. This is mythread[1]. This is mythread[-929820448]. ... ... ... I wonder why the first thread's number is not zero rather than -929820448. Furthermore, when running again with MAX_KTHREAD == 3, the messages are: This is mythread[1]. This is mythread[2]. This is mythread[1]. This is mythread[1]. This is mythread[2]. This is mythread[1]. This is mythread[1]. This is mythread[2]. ... ... ... There should be 3 threads running, but only two of them appear. The first thread get lost.

3 2