- kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

Update kernel source to latest?
by habisravi March 30, 2014

March 30, 2014

I cloned the linux kernel source from "git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git". When i type "git tag" in the source folder i see version up to "v3.9-rc8", but kernel.org shows latest version is 3.14-rc8. How can i update my source repository to the latest.

4 3

Could I send a patch which I'm not sure it was correct?
by wei zhang March 30, 2014

March 30, 2014

Hi I found a bug and send a trivial patch to fix it. The kind maintainer gave me some advice about this patch. I'm not familier with it and this is a challenge to me, I want to do it as my best, but I'm afraid that it is impolite to send a patch, which I'm not sure it was correct! This is my first time to send a patch to the kernel, any advice is appreciated! Regards, Wei Zhang

2 2

how to assign IRQ for devices..
by Chan Kim March 29, 2014

March 29, 2014

Hi, In our system, I assigned Ethernet, SD card and USB interrupts to sparc processor's IRQ pin 7, 10 and 8 at my choice. As far as the interrupt is concerned, they worked fine so far. But I found in arch/sparc/kernel/head_32.S that some H/W interrupts have their supposed IRQ pins. t_irq1: TRAP_ENTRY_INTERRUPT(1) /* IRQ Software/SBUS Level 1 */ t_irq2: TRAP_ENTRY_INTERRUPT(2) /* IRQ SBUS Level 2 */ t_irq3: TRAP_ENTRY_INTERRUPT(3) /* IRQ SCSI/DMA/SBUS Level 3 */ t_irq4: TRAP_ENTRY_INTERRUPT(4) /* IRQ Software Level 4 */ t_irq5: TRAP_ENTRY_INTERRUPT(5) /* IRQ SBUS/Ethernet Level 5 */ t_irq6: TRAP_ENTRY_INTERRUPT(6) /* IRQ Software Level 6 */ t_irq7: TRAP_ENTRY_INTERRUPT(7) /* IRQ Video/SBUS Level 5 */ t_irq8: TRAP_ENTRY_INTERRUPT(8) /* IRQ SBUS Level 6 */ t_irq9: TRAP_ENTRY_INTERRUPT(9) /* IRQ SBUS Level 7 */ t_irq10:TRAP_ENTRY_INTERRUPT(10) /* IRQ Timer #1 (one we use) */ t_irq11:TRAP_ENTRY_INTERRUPT(11) /* IRQ Floppy Intr. */ t_irq12:TRAP_ENTRY_INTERRUPT(12) /* IRQ Zilog serial chip */ t_irq13:TRAP_ENTRY_INTERRUPT(13) /* IRQ Audio Intr. */ t_irq14:TRAP_ENTRY_INTERRUPT(14) /* IRQ Timer #2 */ Connecting external interrupt to one of processor's IRQ pin is my job so I did it, but I explicitly assigned the IRQ number (at my choice) in the platform device resource structure and passed it to request_irq() during the driver initialization. This way I could use whatever number . Is this what people usually do? (besides, above code doesn't have USB or SD card anyway) or do I have to change Ethernet IRQ to 5 as above? If so, what about SD or USB? Thanks. regards, Chan

1 0

Finally got busybox shell on LCD but keyboard stalls..
by Chan Kim March 29, 2014

March 29, 2014

Hi all, I managed to make the busybox shell appear on the LCD. I changed the frame buffer allocation method : from (method 1) fb->fb.screen_base = dma_alloc_writecombine(&pdev->dev, framesize, &fbpaddr, GFP_KERNEL); to (method 2) fb->fb.screen_base = alloc_pages_exact(framesize, GFP_KERNEL | __GFP_ZERO); fbpaddr = virt_to_phys(fb->fb.screen_base); framesize = PAGE_ALIGN(framesize); When the texts scroll up, it's somewhat bleared. like bit lines are not drawn hurridly..When the scroll is slow, it maintains the character form. Of course when it's not moving everything is clean.. After the first 'enter' key, it stops. At this state, interrupt masks are all on.(all interrupts blocked) I have some questions. Q1 : what's the difference of method 1 and 2? I understand in method 1, it tells the kernel that 'this region is for using DMA. so don't use cache.' and method 2 is telling 'just allocate this page aligned region'. Having written it, I reallize the new frame buffer area is using cache. That's maybe causing the blur. Anybody who can clear things a bit more? using method 1, the page fault handler fails.(probably due to not being able to expand stack?? I'm not sure what it's doing in do_sparc_fault function.) Q2 : After the 'enter a key to activate this console' message, when I (first time very excitedly..) enter the Enter key, it stops. The interrupts are all masked. In what case does linux masks all the interupts? I had similar experience twice before (when there was cpu read/write timing error for a chip, and after dma_alloc_writecombine). This time it seems it's related to poke_blanked_console function.. Any hint will be deeply appreciated. Thanks, Chan

1 0

in sparc linux system call entry assembly code
by 김찬 March 29, 2014

March 29, 2014

Hi, below is a code snippet from arch/sparc/kernel/entry.S. I can't understand a part of it (I think I once understood it but forgot..). #define LINUX_SYSCALL_TRAP \ sethi %hi(sys_call_table), %l7; \ or %l7, %lo(sys_call_table), %l7; \ b linux_sparc_syscall; \ rd %psr, %l0; .... linux_sparc_syscall: sethi %hi(PSR_SYSCALL), %l4 or %l0, %l4, %l0 /* Direct access to user regs, must faster. */ cmp %g1, NR_syscalls bgeu linux_sparc_ni_syscall sll %g1, 2, %l4 ld [%l7 + %l4], %l7 andcc %l7, 1, %g0 bne linux_fast_syscall /* Just do first insn from SAVE_ALL in the delay slot */ when linux system call trap is entered, the sys_call_table address is stored in %l7 and %psr in %l0 and it branches to linux_sparc_syscall. (%psr is read in branch delayed slot). From the linux_sparc_syscall, the the %l0 (PSR value) is set with system call bit (to indicate it's in system call later), and if the system call number is compared with the max value, if it is greater or equal, it branches to linux_sparc_ni_syscall (not implemented). The part I can't understand follows. %l7 contains sys_call_table, so it loads the system call address with corresponding offset (nubmer * 4, or 2 bit shift left). Then what is this "andcc %l7, 1, %g0" doing? Thanks! Chan

3 2

Total Number of lines in the kernel code
by Kaushal Shriyan March 28, 2014

March 28, 2014

Hi, How many lines of code and files and directories in the latest stable kernel 3.13.7? Regards, Kaushal

3 4

Unable to handle kernel paging request at virtual address ... while writing to frame buffer..
by Chan Kim March 28, 2014

March 28, 2014

Hi, I have this 'unable to handle kernel paging request at virtual address ...' when writing to the frame buffer. Has anybody seen similar situation? and hopeful can tell me what the problem was? Thanks! Chan

4 5

Building nvidia driver with dkms fails on debian jessie/sid 3.14-rc8 (git)
by Tyrell Keene March 27, 2014

March 27, 2014

Hello everyone, Not sure if I should send this here since this deals with a proprietary driver but i'll give it a try. I'm trying to build the nvidia driver with: $ sudo dkms build -m nvidia-current -v 331.49 -k 3.14.0-rc8-00011-gf217c44 it fails with output: Building module: cleaning build area.... env CCACHE_DISABLE=1 make -C .SYSSRC=/lib/modules/3.14.0-rc8-00011-gf217c44/build....(bad exit status:2) Error! Bad return status for module build on kernel: 3.14.0-rc8-00011-gf217c44 (x86_64) Consult /var/lib/dkms/nvidia-current/331.49/build/make.log for more information. make.log (attached) tells me the kernel config is invalid and to make oldconfig && make prepare I do so but I still get the same error. Any ideas? Thanks, Tyrell

3 2

Storing GUIDs in linux [filesystems]
by Sankar P March 27, 2014

March 27, 2014

Hi, http://en.wikipedia.org/wiki/Globally_unique_identifier is one way to uniquely identify users. It is used by Windows and Active Directory too. The size of a GUID is 128 bits. I want to store the GUID of an user on my experimental filesystem. The size of the "uid" in UNIX is just 16 bits. So what is the way used by the Linux [filesystems] to store the GUID in linux to uniquely identify an user ? Were there any proposals in the past, to increase the size of the unix uid to match that of a GUID ? -- Sankar P http://psankar.blogspot.com

2 1

nf_conntrack: table full .. leads to reboot
by Pietro Paolini March 26, 2014

March 26, 2014

Hello, I am doing some investigation regarding an error I am experiencing using my linux embedded device as router in my home network. If I try to push the number the connections to around 20K I will se a lot of messages from the kernel saying : nf_conntrack: table full, dropping packet. nf_conntrack: table full, dropping packet. nf_conntrack: table full, dropping packet. nf_conntrack: table full, dropping packet. nf_conntrack: table full, dropping packet. nf_conntrack: table full, dropping packet. nf_conntrack: table full, dropping packet. Actually this is fine for me since the number of allowed conntrack is /proc/sys/net/netfilter/nf_conntrack_max is around 7844, the strange thing is that the device will reboot after a while - actually my watchdog reboots the device since is not able to allocate memory - I have a try with slabtop running and I have seen that but monitoring the /proc/slabinfo: 3968 nf_conntrack_c0d4eacc 4392 size-8192 6360 skbuff_head_cache 15356 size-4096 31808 size-1024 Well size-1024 is quite high, do you know where is that from and how is that related to the packets coming in ? Even point me at the right folder in the source code could be great since I am really lost here. Thanks in advance, Pietro.

1 0