May 2012 - kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

Query regarding ESP payload in 3DES-CBC
by Mukesh Yadav May 17, 2012

May 17, 2012

Hi, First 12 bytes getting decrypted wrong in 3Des-CBC... Using 24 bytes key and IV is 8 octet. Whole ESP payload is decypted fine except first 12 bytes... I am able to decrypt received packets on machine B send by M/c A and comparing it with original packet prior to encryption on M/c A shows that first 12 bytes are not same.. Does first block in ESP payload need some special handling apart from rest of payload. Any pointer for same will be helpful... Details are as below. I am trying to achieve Ipsec functionality using fast-path application which will do encryption/decryption using some hardware(Cavium) specific API. This application will by-pass the IP layer of kernel.. Keys for start-up are pre-shared. Communication is done between two machine A and B. On Machine A running i386 linux, SA/SP database are updated using setkey utility and packets is encrypted/decrypted using kernel Ipsec. On Machine B Cavium h/w, keys are pre-shared to application performing Ipsec functionlity... Example: M/c A configuration: add 50.50.50.51 50.50.50.53 esp 15701 -E 3des-cbc "0123456789abcdef12345678"; spdadd 10.10.10.20 10.10.10.21 any -P out ipsec esp/tunnel/50.50.50.51 50.50.50.53/require -- Cheers Mukesh

1 0

device crashing while unpacking ramdisk.img
by sandeep kumar May 16, 2012

May 16, 2012

Dear All, I am facing a situation where, the device is crashing while doing the following, err = unpack_to_rootfs((char *)initrd_start, initrd_end - initrd_start); /* in initramfs.c */ >From Trace32 we could observe the following things, 1) initrd_start has the ramdisk.img which we have flashed onto the flash. And it is succesfully been copied to RAM.( We verified that image existense) 2) Just after exiting that a bus error is generated,(panic is also not seen) We are wondering what exactly unpack_to_rootfs will do? and Where(the destination address) does it copy the unzipped ramdisk.img? With regards, Sandeep Kumar Anantapalli, -- With regards, Sandeep Kumar Anantapalli,

1 0

[bug] shrink_slab shrinkersize handling
by solmac john May 16, 2012

May 16, 2012

Hi All, During mm performance testing sometimes I observed below kernel messages [ 80.776000] shrink_slab: ashmem_shrink+0x0/0x114 negative objects to delete nr=-2133936901 [ 80.784000] shrink_slab: ashmem_shrink+0x0/0x114 negative objects to delete nr=-2139256767 [ 80.796000] shrink_slab: ashmem_shrink+0x0/0x114 negative objects to delete nr=-2079333971 [ 80.804000] shrink_slab: ashmem_shrink+0x0/0x114 negative objects to delete nr=-2096156269 [ 80.812000] shrink_slab: ashmem_shrink+0x0/0x114 negative objects to delete nr=-20658392 After debugging is I fount these prints from mm/vmscan.c Kernel Error msg point unsigned long total_scan; unsigned long max_pass; max_pass = do_shrinker_shrink(shrinker, shrink, 0); delta = (4 * nr_pages_scanned) / shrinker->seeks; delta *= max_pass; do_div(delta, lru_pages + 1); shrinker->nr += delta; if (shrinker->nr < 0) { printk(KERN_ERR "shrink_slab: %pF negative objects to " "delete nr=%ld\n", shrinker->shrink, shrinker->nr); shrinker->nr = max_pass; } I found one patch http://lkml.org/lkml/2011/8/22/80 for this fix Please let me know reason why I am getting above error and above is really fix for this problem. ? I am working on ARM cortex A9 linux-3.0.20 kernel. Thanks, John

1 0

Kernel module to dump Used heap/stack of a user process
by Dhyan May 15, 2012

May 15, 2012

Dear All, I am writing a kernel module which will accept user process id (PID) as input and dump all the address space of that user process,like stack,heap segment. I took mm->start_brk as start of heap ,mm->brk as end of heap,then i am seeing, even though user process is not using 33 pages heap size shows as 33 pages (mm->brk -mm->start_brk) . (This i verified using /proc/pid/maps) .Stack also behaves in similar way, *So is there any way i can find out the really used heap staring from mm->start_brk,So that i can dump only the used heap* -- Thanks Dhyan

3 3

rootfs issue in kernel porting - FPGA board [/bin/sh: can't access tty; job control turned off]
by pcuser p May 15, 2012

May 15, 2012

Hi, I am doing linux kernel porting for FPGA board. I am facing issue in getting the shell prompt at the end of kernel booting. Shell prompt is not taking any commands I’m typing. I can place break point in Tx & Rx ISR with Jtag debugger.ISR functions seems working till shell is entered (echo message in the /etc/inittab script & mounting /proc,/devtmpfs and ls command (added to check)are working, I added ls,cat command in the /etc/init.d/rcS file to check uart Tx / Rx) Once the shell is entered with the below message, UART ISR is NOT invoked (using JTAG to put breakpoints) ,if I type any command & I can’t get any message. *Err message: /bin/sh: can't access tty; job control turned off / #* Kindly share any suggestion to fix this issue. * Below are the /etc/initab & /etc/init.d/rcS * /etc/inittab: # This is run first script ::sysinit:/etc/init.d/rcS # Start an "askfirst" shell on the console ttySAC0::askfirst:-/bin/sh ttySAC0::respawn:-/sbin/getty 115200 ttySAC0 # Stuff to do when restarting the init process ::restart:/sbin/init # Stuff to do before rebooting ::ctrlaltdel:/sbin/reboot ::shutdown:/bin/umount a r */etc/init.d/rcS* #!/bin/sh echo -e "\n in rcS script\n" echo -e "\n Mounting proc FS\n" mount t proc none /proc echo -e "\nMounting sysfs \n" mount t sysfs none /sys ls -alh / echo -e "\n\n" ls -alh /dev/ttyS* echo -e "\n" cat /proc/interrupts echo -e "\n ttySAC0 respawning \n" ttySAC0::respawn:-/sbin/getty 115200 ttySAC0 echo -e "\n ttySAC0 respawning finished \n" echo -e "\n exec /bin/sh\n" exec /bin/sh ls #ls -alh /dev *Below are the console message of boot up.* Linux version 3.1.6-00011-g82af784-dirty (pcuser@pcuser-linux) (gcc version 4.6.1 (Ubuntu/Linaro 4.6.1-9ubuntu3) ) #66 Fri May 11 16:21:22 IST 2012 CPU: ARM926EJ-S [41069265] revision 5 (ARMv5TEJ), cr=00053177 CPU: VIVT data cache, VIVT instruction cache Machine: FPGA Board Memory policy: ECC disabled, Data cache writeback Available lowmem size : 32 MB Built 1 zonelists in Zone order, mobility grouping on. Total pages: 8128 Kernel command line: mem=32M console=ttySAC0,115200 root=/dev/ram0 init=/sbin/init initrd=0x80800000,8M user_debug=31 PID hash table entries: 128 (order: -3, 512 bytes) Dentry cache hash table entries: 4096 (order: 2, 16384 bytes) Inode-cache hash table entries: 2048 (order: 1, 8192 bytes) Memory: 32MB = 32MB total Memory: 22456k/22456k available, 10312k reserved, 0K highmem Virtual kernel memory layout: vector : 0xffff0000 - 0xffff1000 ( 4 kB) fixmap : 0xfff00000 - 0xfffe0000 ( 896 kB) DMA : 0xffc00000 - 0xffe00000 ( 2 MB) vmalloc : 0xc2800000 - 0xf6000000 ( 824 MB) lowmem : 0xc0000000 - 0xc2000000 ( 32 MB) modules : 0xbf000000 - 0xc0000000 ( 16 MB) .text : 0xc0008000 - 0xc018c000 (1552 kB) .init : 0xc018c000 - 0xc01a1000 ( 84 kB) .data : 0xc01a2000 - 0xc01afe00 ( 56 kB) .bss : 0xc01afe24 - 0xc01c0f24 ( 69 kB) SLUB: Genslabs=13, HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1 NR_IRQS:32 Console: colour dummy device 80x30 console [ttySAC0] enabled Calibrating delay loop... 1.86 BogoMIPS (lpj=9344) pid_max: default: 32768 minimum: 301 Mount-cache hash table entries: 512 CPU: Testing write buffer coherency: ok devtmpfs: initialized exceptions_init Switching to clocksource timer0 Trying to unpack rootfs image as initramfs... rootfs image is not initramfs (junk in compressed archive); looks like an initrd Freeing initrd memory: 8192K msgmni has been set to 59 io scheduler noop registered (default) brd: module loaded loop: module loaded mousedev: PS/2 mouse device common for all mice RAMDISK: ext2 filesystem found at block 0 RAMDISK: Loading 3072KiB [1 disk] into ram disk... \ done. VFS: Mounted root (ext2 filesystem) readonly on device 1:0. devtmpfs: mounted Freeing init memory: 84K *echo messages from /etc/rcS & /etc/inintab* in rcS script Mounting proc FS Mounting sysfs drwxrwxr-x 12 0 0 1.0K May 9 2012 . drwxrwxr-x 12 0 0 1.0K May 9 2012 .. drwxrwxr-x 2 0 0 2.0K May 11 2012 bin drwxr-xr-x 3 0 0 11.9K Jan 1 00:00 dev drwxr-xr-x 3 0 0 1.0K May 11 2012 etc drwxr-xr-x 2 0 0 1.0K May 9 2012 lib lrwxrwxrwx 1 0 0 11 May 8 2012 linuxrc -> bin/busybox drwx------ 2 0 0 12.0K May 8 2012 lost+found drwxr-xr-x 2 0 0 1.0K May 8 2012 proc drwxrwxr-x 2 0 0 1.0K May 8 2012 sbin drwxr-xr-x 2 0 0 1.0K May 8 2012 sys drwxrwxr-x 4 0 0 1.0K May 8 2012 usr drwxr-xr-x 2 0 0 1.0K May 9 2012 var crw------- 1 0 0 204, 64 Jan 1 00:00 /dev/ttySAC0 ttySAC0 respawning /etc/init.d/rcS: line 20: ttySAC0::respawn:-/sbin/getty: not found ttySAC0 respawning finished -rwxr-xr-x 1 0 0 1.0K May 11 2012 /bin/hello calling hello world /etc/init.d/rcS: line 24: ./hello: not found /etc/init.d/rcS: line 25: hello: not found /etc/init.d/rcS: line 26: ./hello: not found /etc/init.d/rcS: line 27: /bin/hello: not found exec /bin/sh /bin/sh: can't access tty; job control turned off / # I can't type any commad at the shell prompt, shell is not accepting it.

1 0

Linux shutdown/reboot process
by Asutosh Das May 15, 2012

May 15, 2012

Hi List I have a query about the linux shutdown or reboot process. Please can you let me know if the below sequence is correct: When reboot command is given on the shell: 1. all user-space process is frozen (by sending some signal ?? ) 2. all file-systems are remounted as read-only - (sysrq handler does this ?? ) 3. all drivers, whose shutdown handlers have been defined, are called in the reverse order in which these drivers were initialized. I am not sure what happens after this. TIA. -- Thanks & Regards ~/asd

4 5

Kernelnewbies Digest, Vol 18, Issue 17
by jeshwanth Kumar N K May 14, 2012

May 14, 2012

Hello All I am facing some dificulty please help me, actually I want to access the register in beagle bone. I want to write it in C, and I got to know that it s done through mem character file. So how can I access the mem char driver for it and can you please send me a link that explains more about mem character file ? Thank you so much. Sent from my HTC ----- Reply message ----- From: kernelnewbies-request(a)kernelnewbies.org Date: Sun, May 13, 2012 9:30 pm Subject: Kernelnewbies Digest, Vol 18, Issue 17 To: <kernelnewbies(a)kernelnewbies.org> Send Kernelnewbies mailing list submissions to kernelnewbies(a)kernelnewbies.org To subscribe or unsubscribe via the World Wide Web, visit http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies or, via email, send a message with subject or body 'help' to kernelnewbies-request(a)kernelnewbies.org You can reach the person managing the list at kernelnewbies-owner(a)kernelnewbies.org When replying, please edit your Subject line so it is more specific than "Re: Contents of Kernelnewbies digest..." Today's Topics: 1. Re: interface for a hardware trigger driver (richard -rw- weinberger) ---------------------------------------------------------------------- Message: 1 Date: Sat, 12 May 2012 22:15:47 +0200 From: richard -rw- weinberger <richard.weinberger(a)gmail.com> Subject: Re: interface for a hardware trigger driver To: Andre Haupt <andre(a)bitwigglers.org> Cc: kernelnewbies(a)kernelnewbies.org Message-ID: <CAFLxGvzvec_Mg70s9v6hD-tqTQtQYfcdUSJ7uPT2XNBHTf9JhA(a)mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1 On Thu, May 10, 2012 at 11:09 AM, Andre Haupt <andre(a)bitwigglers.org> wrote: > Do you have any hints to the right approach for such a driver? > Should i use some sysfs interface? If yes, which? Or does > such a driver already exist? Can it be one completely in > user space? > You can use UIO. -- Thanks, //richard ------------------------------ _______________________________________________ Kernelnewbies mailing list Kernelnewbies(a)kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies End of Kernelnewbies Digest, Vol 18, Issue 17 *********************************************

3 2

interface for a hardware trigger driver
by Andre Haupt May 12, 2012

May 12, 2012

Hi folks, Its been a while since i last did some kernel related stuff, so please bear with me if this sounds strange to your ears. What i want to achieve: I want to implement a hardware trigger that a user space process can react on. I need a driver that blocks a process/thread until a sepcific hardware interrupt occurs. The process should call a kernel interface and then should get blocked until another process/thread calls another kernel interface to stop waiting for the irq or an interrupt actually occurs. What i have: Back in the days i wrote a little character driver which implemented 2 ioctl commands. One command (IOCTL_TRIGGER_WAIT_IRQ) put the calling process to sleep using wait_evemt_interruptible() and the other command (IOCTL_TRIGGER_STOP_WAIT_IRQ) woke a processes again by calling wake_up_interruptible(). Now ioctls are frowned upon and i do not want to mess with majors and minors anymore either. Do you have any hints to the right approach for such a driver? Should i use some sysfs interface? If yes, which? Or does such a driver already exist? Can it be one completely in user space? puzzled. cheers, Andre

3 4

linker error: section __modver vma 0x6028c000 overlaps previous
by michi1＠michaelblizek.twilightparadox.com May 11, 2012

May 11, 2012

Hi! I am facing a weird problem: $ ARCH=um make linux make[1]: `arch/um/sys-x86_64/user-offsets.s' is up to date. CHK include/linux/version.h CHK include/generated/utsrelease.h CALL scripts/checksyscalls.sh CHK include/generated/compile.h QUOTE arch/um/kernel/config.tmp QUOTE arch/um/kernel/config.c CC arch/um/kernel/config.o LD arch/um/kernel/built-in.o VDSOSYM arch/um/sys-x86_64/vdso/vdso-syms.lds LD arch/um/sys-x86_64/vdso/built-in.o LD arch/um/sys-x86_64/built-in.o CHK kernel/config_data.h LD vmlinux.o MODPOST vmlinux.o GEN .version CHK include/generated/compile.h UPD include/generated/compile.h CC init/version.o LD init/built-in.o LD .tmp_vmlinux1 KSYM .tmp_kallsyms1.S AS .tmp_kallsyms1.o LD .tmp_vmlinux2 /usr/bin/ld: .tmp_vmlinux2: section __modver vma 0x6028c000 overlaps previous sections KSYM .tmp_kallsyms2.S AS .tmp_kallsyms2.o LD vmlinux /usr/bin/ld: vmlinux: section __modver vma 0x6028c000 overlaps previous sections SYSMAP System.map SYSMAP .tmp_System.map LINK linux This is triggered by changing files in my network protocol code for no apparent reason. My code is at available at http://repo.or.cz/w/cor.git?a=tree This change triggers it: diff --git a/net/cor/credits.c b/net/cor/credits.c index e0fb39a..776d197 100644 --- a/net/cor/credits.c +++ b/net/cor/credits.c @@ -556,7 +556,14 @@ void connreset_credits(struct conn *cn) { unsigned long iflags; - __u32 unlockhints = credits_lock_conn(cn); + if (cn->is_client) { + spin_lock_bh(&(cn->rcv_lock)); + spin_lock_bh(&(cn->reversedir->rcv_lock)); + } else { + spin_lock_bh(&(cn->reversedir->rcv_lock)); + spin_lock_bh(&(cn->rcv_lock)); + } + if (cn->in_credit_list) { spin_lock_irqsave(&credits_list_lock, iflags); list_del(&(cn->credit_list)); @@ -582,13 +589,26 @@ void connreset_credits(struct conn *cn) cn->crate_out, 0); spin_unlock_irqrestore(&(nb->credits_lock), iflags); } - credits_unlock_conn(cn, unlockhints); + + if (cn->is_client) { + spin_unlock_bh(&(cn->rcv_lock)); + spin_unlock_bh(&(cn->reversedir->rcv_lock)); + } else { + spin_unlock_bh(&(cn->reversedir->rcv_lock)); + spin_unlock_bh(&(cn->rcv_lock)); + } } void set_conn_in_decaytime(struct neighbor *nb, __u32 conn_id, struct conn *src_in, __u8 decaytime_seqno, __u16 decaytime) { - __u32 unlockhints = credits_lock_conn(src_in); + if (src_in->is_client) { + spin_lock_bh(&(src_in->rcv_lock)); + spin_lock_bh(&(src_in->reversedir->rcv_lock)); + } else { + spin_lock_bh(&(src_in->reversedir->rcv_lock)); + spin_lock_bh(&(src_in->rcv_lock)); + } if (unlikely(is_conn_in(src_in, nb, conn_id) == 0)) goto out; @@ -608,7 +628,13 @@ set: refresh_conn_credits(src_in, 0, 1); out: - credits_unlock_conn(src_in, unlockhints); + if (src_in->is_client) { + spin_unlock_bh(&(src_in->rcv_lock)); + spin_unlock_bh(&(src_in->reversedir->rcv_lock)); + } else { + spin_unlock_bh(&(src_in->reversedir->rcv_lock)); + spin_unlock_bh(&(src_in->rcv_lock)); + } } static void background_refresh_credits(struct work_struct *work) However, if only one instead of both credits_(un)lock_conn changes are made, the error/warning is gone. By searching on the internet it looks like the linker runs out of ressources, maybe because of certain properties of the build environment. However, I did not find a solution yet. Any ideas? My system is debian stable/squeeze. -Michi

2 6

Update kernel without boot!?!? From oracle/linux site
by Daniel Hilst May 11, 2012

May 11, 2012

From oracle/linux[1] site, they're bragging about having the unique OS that can have kernel updated on the fly.. I have two questions: 1) Is that possible, and if is, can this be safe??? On PDF they say about a module that does the trick.. 2) Is there any open software project trying this? would be good to have that feature! Cheers! Hilst [1] http://www.oracle.com/us/technologies/linux/index.html -- Follow the white rabbit!

4 3