- kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

[ARM_LINUX] ioremap() allowing to map system memory...
by sandeep kumar March 4, 2013

March 4, 2013

Hi All I am using ARM based board. In mine, i did the following... void __iomem *tcpm_base = ioremap_nocache(0x03B00000, 10*SZ_3MB); Actually i didnt reserve the 30MB memory @ 0x3B00000. But still the call is succesful and i am able to read the memory. In the logs it is just showing a warning, to fix my driver as i am calling ioremap() on system memory. However if i try to write something on that memory, then only it is calling panic().. Don't you think it should throw panic()while calling the ioremap() itself. Because this sounds like a serious violation... What say? -- With regards, Sandeep Kumar Anantapalli,

5 8

How to measure the RAM read/write performance
by sandeep kumar March 4, 2013

March 4, 2013

Hi All In performance benchmark tools, When we profile read/write timings mostly, those read/writes are done to cache only. I want to measure my DDR(RAM chip) performance. So i want to make sure, every read/write should happen to DDR RAM chip only. How can i achieve this...Any ideas/suggestions...? -- With regards, Sandeep Kumar Anantapalli,

9 24

how to trace tcp protocol stack ?
by ishare March 3, 2013

March 3, 2013

hi:) Is there mothod to look up the call stack of tcp protocol solution? thanks!

3 2

Why vmlinux.bin are changed from raw image to elf for x86 ?
by Jacky March 2, 2013

March 2, 2013

Hi All, Recently, I found that x86 vmlinux.bin are changed from raw image to elf, we can got it from the following objcopy flags: linux-3.7.x/arch/x86/boot/compressed/Makefile： OBJCOPYFLAGS_vmlinux.bin := -R .comment -S $(obj)/vmlinux.bin: vmlinux FORCE $(call if_changed,objcopy) we can see that kbuild pass no "-O binary" to objcopy. And I double check the vmlinux.bin as following: linux-3.7.x/arch/x86/boot/compressed$ file vmlinux.bin vmlinux.bin: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), statically linked, BuildID[sha1]=0x7ad967473a9b36bc529551eecd326b5dfd15a426, stripped Make sure the vmlinux.bin is elf, not raw binary. Anyone knows the rootcause ? Thanks, Jacky

2 2

wby mapping beyond end by "MAPPING_BEYOND_END" ?
by Jacky March 2, 2013

March 2, 2013

Hi All, For x86, in arch/x86/kernel/head_32.S, for Initializing page tables, why mapping beyond end by "MAPPING_BEYOND_END": /* * End condition: we must map up to the end + MAPPING_BEYOND_END. */ movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp cmpl %ebp,%eax jb 10b According vmlinux.lds.S, the paging table is between __brk_base and _end. So why map MAPPING_BEYOND_END after _end ? Regards, Jacky

1 0

Doubts about vt tty driver
by Chinmay Bapat March 2, 2013

March 2, 2013

Hi I am trying to understand the vt tty driver. I have been trying to read the code and have been referring to Linux Device Drivers book. I have read this http://www.linusakesson.net/programming/ttyto get an idea of what ttys are etc. I have many doubts and would be very greatful if someone could help me out. I am not sure where this driver fits into the rest of things. I mean, from where are the functions in this driver called, and what does the driver actually do? Does it interact with the screen hardware? If it does, how? >From what I understand, the vt driver does the work of the 'Terminal emulator' in this http://www.linusakesson.net/programming/tty/case3.png . Am I right? The LDD3 book says that after the driver registers itself, it registers the devices it controls through the tty_register_device function. But I cannot see this being done in the vt.c file. Why? Also, there are two functions con_write and vt_console_print that write. I was told that vt_console_print is high level. It is called by the console code. con_write is low level. It writes to the tty. But I still cannot understand what each one does. Also I do not know how to find out from where these functions are called: con_write is assigned to .write of the struct tty_operations con_ops, vt_console_print .write of struct console vt_console_driver. But from where are these .write s called? This is the first time I am working with a driver or anything to do with the kernel, so even if my doubts are stupid please help me out. Thank you. -- Regards, Chinmay Bapat

1 0

[filesystem] struct of m_inode
by lx March 2, 2013

March 2, 2013

Hi, I'm learning the kernel of 0.12, and the function in inode.c: #################################################################### static int _bmap(struct m_inode * inode,int block,int create) { struct buffer_head * bh; int i; if (block<0) panic("_bmap: block<0"); if (block >= 7+512+512*512) panic("_bmap: block>big"); if (block<7) { if (create && !inode->i_zone[block]) if (inode->i_zone[block]=new_block(inode->i_dev)) { inode->i_ctime=CURRENT_TIME; inode->i_dirt=1; } return inode->i_zone[block]; } block -= 7; if (block<512) { if (create && !inode->i_zone[7]) if (inode->i_zone[7]=new_block(inode->i_dev)) { inode->i_dirt=1; inode->i_ctime=CURRENT_TIME; } if (!inode->i_zone[7]) return 0; if (!(bh = bread(inode->i_dev,inode->i_zone[7]))) return 0; i = ((unsigned short *) (bh->b_data))[block]; if (create && !i) if (i=new_block(inode->i_dev)) { ((unsigned short *) (bh->b_data))[block]=i; bh->b_dirt=1; } brelse(bh); return i; } block -= 512; if (create && !inode->i_zone[8]) if (inode->i_zone[8]=new_block(inode->i_dev)) { inode->i_dirt=1; inode->i_ctime=CURRENT_TIME; } if (!inode->i_zone[8]) return 0; if (!(bh=bread(inode->i_dev,inode->i_zone[8]))) return 0; i = ((unsigned short *)bh->b_data)[block>>9]; if (create && !i) if (i=new_block(inode->i_dev)) { ((unsigned short *) (bh->b_data))[block>>9]=i; bh->b_dirt=1; } brelse(bh); if (!i) return 0; if (!(bh=bread(inode->i_dev,i))) return 0; i = ((unsigned short *)bh->b_data)[block&511]; if (create && !i) if (i=new_block(inode->i_dev)) { ((unsigned short *) (bh->b_data))[block&511]=i; bh->b_dirt=1; } brelse(bh); return i; } ######################################################################## "if (block >= 7+512+512*512)" because the i_zone[9]. But the question is why the i_zone[7] can repesent 512 , and i_zone[8] can repesent 512*512 ? Thank you.

2 1

relocatable modules' symbols
by horseriver March 1, 2013

March 1, 2013

hi: I have built vmlinux at the top dir of kernel source ,then I use objdump to look into its section information.I find the statup_32 which is the start routine of kernel , locats at 0xc0100000. I know the 0x100000 is the defined address for locating protect-mode code .But which I can not understand is the 0xc prefix .where does it come from? thanks!

6 7

Examples of Per-CPU Variables in Kernel source
by Shraddha Kamat March 1, 2013

March 1, 2013

I was reading about Per-CPU Variables from UTLK. Where are such variables used in the kernel source code - example. -- Shraddha

3 3

atomic operations
by Shraddha Kamat March 1, 2013

March 1, 2013

what is the relation between atomic operations and memory alignment ? I read from UTLK that "an unaligned memory access is not atomic" please explain me , I am not able to get the relationship between memory alignment and atomicity of the operation.

5 4