- kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

Design Patterns in Linux Kernel: Fancy Tricks With Linked Lists
by Arlie Stephens March 20, 2013

March 20, 2013

Hi Folks, I'm trying to understand the linux way of doing things, so as to write code which will be easier for experienced linux kernel people to understand. Yesterday I wound up in a 3 engineer discussion about something conceptually simple, that just didn't want to be done with the tools at hand; what's worse, the only ways I could see to do it were (a) do my own linked lists or (b) write some routines that relied kind of heavily on the never-mentioned-in-documentation details of linux/list.h I'll probably go with option (b), as that seems less insane, but I thought I might as well ask here as well. Here's the problem. I have a collection of resources. Call them A,B,C,D but note that resources get added at run time, so I don't know how many to expect. I want to go through them in a round robin fashion, but not all at once. On the first request, I use A. On a later request I use B, etc. This seems to me to map naturally to a circular linked list, with a "cursor" - a pointer of some kind to the next one I want to use. To make my specific situation more interesting, I actually have multiple cursors - that may or may not happen to point to the same node at any given time, but usually will not. This is where I wish I could draw pictures in ascii emails ;-) Perhaps the following will come through halfway sanely: C1 C2 C3 V / V A<->B<->C-<->D ^------------^ list.h implements a circular linked list - see for example http://www.makelinux.net/books/lkd2/app01lev1sec2 - so I thought this would be easy and natural. But then I discovered there was no such thing as a list_next(). OK, I can write it - Something like: cursor->resource = list_entry(cursor->resource->link.next, struct resource, link); But the fact there was no interface made me ask advice from colleagues. And that's when the debate erupted. First of all, it's unclear whether that would even work. If the list is initialized following the standard pardigm, there may be a "head" element embedded in the list, which all the existing interfaces know to ignore. I.e. LIST_HEAD(resources); add_resource(...) new_resource = kmalloc(sizeof(struct resource), GFP_KERNEL); new_resource->ID = ... INIT_LIST_HEAD(&new_resource->link); list_add(&new_resource->link, &resources) It looks as if the circular list is likely to include the variable named 'resources', so new_resource->link.next might point to resources, and list_entry(new_resource->link.next, struct resource, link) might point to whatever unrelated variable happens to be somewhere near 'resources'. It's certain that the implementation makes no attempt to clarify what kind of list it's using ... not when we had 3 experienced engineers draw 2 different conclusions, one changing his mind midway. Obviously I can figure out what the link.h implementation actually does, and code accordingly. That's my plan for this morning. I'm just not sure it's going to produce code that other engineers will be comfortable maintaining. And if they aren't comfortable, there will be uneccessary bugs. So, how would a dyed-in-the-wool linux kernel engineer approach this task? I.e. what's the natural design pattern here, within linux? I think it's pretty clear that the one I've come up with seems so unnatural to the authors of list.h that they never even imagined it. Thanks for any insights, -- Arlie (Arlie Stephens arlie(a)worldash.org)

5 10

Wake_lock in linux kernel
by Ben Wu March 20, 2013

March 20, 2013

Dear All: I now study power driver in linux source code, but didn't find some doc about liux,all the doc was for android, can some help me?BTW,how /sys/power interactive with linux kernel?? Many thanks

1 0

do_initcalls
by ishare March 19, 2013

March 19, 2013

for (call = &__initcall_start; call < &__initcall_end; call++) (*call)(); I want to know where are these function point assigned by specified function ? thanks!

2 5

Re: Design Patterns in Linux Kernel: Fancy Tricks With Linked Lists
by Victor Buciuc March 19, 2013

March 19, 2013

On Tue, Mar 19, 2013 at 8:50 PM, Victor Buciuc <victor.buciuc(a)gmail.com> wrote: > On Tue, Mar 19, 2013 at 5:34 PM, Arlie Stephens <arlie(a)worldash.org> wrote: >> Hi Folks, >> >> I'm trying to understand the linux way of doing things, so as to write >> code which will be easier for experienced linux kernel people to >> understand. Yesterday I wound up in a 3 engineer discussion about >> something conceptually simple, that just didn't want to be done with >> the tools at hand; what's worse, the only ways I could see to do it >> were (a) do my own linked lists or (b) write some routines that relied >> kind of heavily on the never-mentioned-in-documentation details of >> linux/list.h I'll probably go with option (b), as that seems less >> insane, but I thought I might as well ask here as well. >> >> Here's the problem. I have a collection of resources. Call them >> A,B,C,D but note that resources get added at run time, so I don't know >> how many to expect. I want to go through them in a round robin >> fashion, but not all at once. On the first request, I use A. On a >> later request I use B, etc. This seems to me to map naturally to a >> circular linked list, with a "cursor" - a pointer of some kind to the >> next one I want to use. To make my specific situation more >> interesting, I actually have multiple cursors - that may or may not >> happen to point to the same node at any given time, but usually will >> not. >> >> This is where I wish I could draw pictures in ascii emails ;-) >> Perhaps the following will come through halfway sanely: >> >> C1 C2 C3 >> V / V >> A<->B<->C-<->D >> ^------------^ >> > > Hi, > > Another option might be to not iterate through the list but to simply > move the element you're processing to the end of the list using > list_rotate_left[1]. > And you could use list_next_entry[2] to get the next entry after the > head. You'll > have to check that the list is empty, tough. > > This would cost you some extra operations but you'll use the existing interface. > > [1] http://lxr.linux.no/linux+v3.8.3/include/linux/list.h#L214 > [2] http://lxr.linux.no/linux+v3.8.3/include/linux/list.h#L361 > > Regards, > Victor Buciuc. Forgot to send to the list. Regards, Victor Buciuc.

1 0

for interested folks, i'll announce new kernel tutorials via twitter
by Robert P. J. Day March 19, 2013

March 19, 2013

as i don't want to seem like i'm spamming this list with notes about new, posted kernel tutorials, people who want to stay on top of these are welcome to follow me on twitter at @rpjday. everything written and posted will be freely available (creative commons license), and even after it's incorporated into future training courses, even those courses will be released under the CC so you'll never lose access to any of it. so feel free to play along. rday -- ======================================================================== Robert P. J. Day Ottawa, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================

10 9

Kernel warning messages- protocol 0000 is buggy, dev eth0
by Sengottuvelan S March 19, 2013

March 19, 2013

Hi I am getting kernel messages = "protocol 0000 is buggy, dev eth0" in my kernel while calling dev_queue_xmit(skb);. The packet is successfully tranmitted on the wire but these messages are keep coming on the console. I am not sure what is the exact modification needed to avoid these kernel messages. Can you someone help on this. Exact code I have in my driver code: tx_len is length of the packet to be transmitted tx_packet is the pointer to packet container. skb = dev_alloc_skb(tx_len); if (skb == NULL) return -1; skb_put(skb,tx_len); memcpy(skb->data, tx_packet, tx_len); skb->dev = dev_my_device; skb->len = tx_len; dev_queue_xmit(skb); dev_put ( dev_my_device); return 0; Thank you. -- Regards, S. Sengottuvelan.

2 1

only for test
by Ben Wu March 19, 2013

March 19, 2013

FYI

1 0

Re: programme header
by ishare March 19, 2013

March 19, 2013

On Mon, Mar 18, 2013 at 03:02:02PM -0700, Ian Lance Taylor wrote: > On Sun, Mar 17, 2013 at 1:54 AM, ishare <june.tune.sea(a)gmail.com> wrote: > > > > > > when product a .so liberary , if there is not a programme header in the linker script file, > > will the linker automaticlly generate a deault one for use ? > > Yes. > > > If I write the programme header manully , need I write it before writing section command ? > > No. I am linking my kernel by a link script. its contens is as below: ********************************************************************* VSYSCALL_BASE = 0xffffe000; SECTIONS { . = VSYSCALL_BASE + SIZEOF_HEADERS; .hash : { *(.hash) } :text .dynsym : { *(.dynsym) } .dynstr : { *(.dynstr) } .gnu.version : { *(.gnu.version) } .gnu.version_d : { *(.gnu.version_d) } .gnu.version_r : { *(.gnu.version_r) } /* This linker script is used both with -r and with -shared. For the layouts to match, we need to skip more than enough space for the dynamic symbol table et al. If this amount is insufficient, ld -shared will barf. Just increase it here. */ . = VSYSCALL_BASE + 0x400; .text : { *(.text) } :text =0x90909090 .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr .eh_frame : { KEEP (*(.eh_frame)) } :text .dynamic : { *(.dynamic) } :text :dynamic .useless : { *(.got.plt) *(.got) *(.data .data.* .gnu.linkonce.d.*) *(.dynbss) *(.bss .bss.* .gnu.linkonce.b.*) } :text } PHDRS { text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */ } ******************************************************************************* I think it will work ,but ld report that "No enough room for programme header",what is the reason? what should I do ? thanks! > Ian

2 3

signals handling: kill() successful, but nothing delivered
by michi1＠michaelblizek.twilightparadox.com March 18, 2013

March 18, 2013

Hi! I am trying to test for signal handling race conditions (specifically, I suspect the kernel side of connect() does interesting things if the server response arrives while the program executes a signal handler) and want to flood a program with lots of signals. So I have run this: --- #!/bin/bash while [ 1 -eq 1 ]; do killall -s SIGURG wget 2> /dev/null done --- ...and then then executed: strace wget -O - "http://kernel.org" ... connect(3, {sa_family=AF_INET, sin_port=htons(80), sin_addr=inet_addr("198.145.20.140")}, 16) = ? ERESTARTSYS (To be restarted) --- SIGURG (Urgent I/O condition) @ 0 (0) --- connect(3, {sa_family=AF_INET, sin_port=htons(80), sin_addr=inet_addr("198.145.20.140")}, 16) = ? ERESTARTSYS (To be restarted) --- SIGURG (Urgent I/O condition) @ 0 (0) --- connect(3, {sa_family=AF_INET, sin_port=htons(80), sin_addr=inet_addr("198.145.20.140")}, 16) = 0 ... It shows that signals arrive, but no interesting behaviour yet. I suspect this is because the rate signals are sent is too low. So I come up with this: --- #include <sys/types.h> #include <signal.h> #include <stdio.h> int main(int argc, char **argv) { int pid; int rc = 0; if (argc < 2) return 1; pid = atoi(argv[1]); if (pid <= 1 || pid >= 65536) return 1; printf("kill %d %d\n", pid, SIGURG); while (kill(pid, SIGURG) == 0) { printf("signal sent\n"); /* probably slow, remove later */ } perror("killloop end"); return 0; } --- #!/bin/bash while [ 1 -eq 1 ]; do ./a.out `ps a|grep wget|grep -v grep|sed "s/^[^0-9]*$[0-9]*$.*$/\1/"` done --- The output looks like this: kill 15574 23 signal sent signal sent signal sent ... signal sent signal sent signal sent killloop end: No such process However, no signal arrives. But if I change SIGURG to SIGTERM, they arrive! Why does SIGURG arrive when sent with "killall -s SIGURG wget", but not when sent with my test prog? -Michi -- programing a layer 3+4 network protocol for mesh networks see http://michaelblizek.twilightparadox.com

2 2

creat all the dentry objects in memory when mount a fs?‏
by RS March 18, 2013

March 18, 2013

when a new unlike-Unix filesystem mounted on my Linux system , will it creat all the dentry objects in the meory , or will it creat only needed, and is the same to inode objects ? So , if all the dentry objects created , I think it must consume a large memory.

2 1