July 2014 - kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

some project
by Oscar Salvador July 14, 2014

July 14, 2014

Hi to everybody! How are you? I'm new here, and I was looking the kernelnewbies website to search some information explaining how a newbie should start in this area. The problem is I think that http://kernelnewbies.org/KernelJanitors/Todo and http://kernelnewbies.org/KernelProjects are outdated, so I don't know which project I can take, which not, or if I should start with other thing. Maybe you have received a lots of questions about that, in this case I apologize for that, was only for learn, not to disturb. Thank you very much. Best Regards Oscar

3 3

Question on path mtu and IP fragmentation in Linux kernel 3.0
by Sudharsan Vijayaraghavan July 13, 2014

July 13, 2014

Hi, i find that even when /proc/sys/net/ipv4/ip_no_pmtu_disc is 0 (default value), that is path mtu is enabled, IP layer fragments the packets if it exceeds MTU For example MTU is 1500, i do ping -s 15000 <ip addr>, IP layer nicely fragments the packets, how is that possible? Does that mean PATH MTU is excercised earlier for that route and lowest MTU is already negotiated for the destination IP address, thats dst_mtu is already calculated Please give expert comments Thanks, Sudharsna

2 1

linux-header files.
by Anand Moon July 13, 2014

July 13, 2014

Hi All, I would like to know a procedure to extract linux-headers from linux src tree. -Anand Moon

2 1

Re: [PATCH] reformat_with_checkpatch: Add automation to checkpatch
by Greg KH July 13, 2014

July 13, 2014

On Sat, Jul 12, 2014 at 11:29:37AM -0700, Joe Perches wrote: > On Sat, 2014-07-12 at 10:55 -0700, Greg KH wrote: > > Yes, the warnings are all due to line-length, but Joe, you shouldn't add > > a patch that causes more checkpatch warnings than before :) > > Yeah, that was intentional though. > > This script does a series of relatively discrete changes. > > Lindent would more or less work, but it's _horrible_ at > wrapping overlong lines and merges all types of changes > together. Oh I agree, I don't want to see Lindent, but maybe, if the patch adds checkpatch warnings, it should be at least "flagged" as maybe a problem? > > While I always want to see more developers get involved with kernel > > development, there should be a minimum barrier to entry. And that > > barrier is the knowledge of the C language, and knowledge of how to edit > > a text file, and use git. This script takes that barrier away, for > > whitespace cleanups, with not much real use overall. > > > > So, I'll keep my local copy of this script now, just to have fun with at > > times when I'm bored. But I don't think it should be merged, as-is. > > Dunno, I still think it's useful. For you, and me, but the world? Would you want to be on the receiving end of this patch script? I don't, and I'm willing to take almost any patch cleanup for staging code. I think that says something :) > Maybe when you get new code, you might run it through a > script like this before committing it. I will keep it for me, like I said. thanks, greg k-h

2 1

Re: [PATCH] reformat_with_checkpatch: Add automation to checkpatch
by Dan Carpenter July 12, 2014

July 12, 2014

On Fri, Jul 11, 2014 at 06:40:16PM -0700, Joe Perches wrote: > On Fri, 2014-07-11 at 18:34 -0700, Greg KH wrote: > > On Fri, Jul 11, 2014 at 06:21:27PM -0700, Joe Perches wrote: > > > A simple script to run checkpatch --fix for various types of > > > of cleanups. > > > > > > This script is useful primarily for staging. > > > > > > This reformats code to a more CodingStyle conforming style, > > > compiles it, verifies that the object code hasn't changed, > > > and git commits it too. > > > > And 'git commits' it? > > The thought I had was to made it easier to > submit "my first kernel patch" correctly. > like that tuxradar article you wrote a few > years ago. > > http://www.tuxradar.com/content/newbies-guide-hacking-linux-kernel > > > Heh, I should just run this myself to clean up > > staging and beat everyone else to it... > > At least for the whitespace noise, you could > but I hope it'll encourage a few more people > to try kernel patching instead. I have always hate the idea of automated patches from random people because I don't trust them to not be malicious so I have to review them manually. There is a story that back in the day the US government paid someone to send tons of checkpatch style patches to BSD. The guy thought he was "cleaning up the code" but actually he was part of a larger scheme to flood the maintainers with patches so another person could slip in some malicious code. It's better if someone just ran this on all new staging code before adding it to the tree. regards, dan carpenter

3 2

Re: [PATCH] reformat_with_checkpatch: Add automation to checkpatch
by Greg KH July 12, 2014

July 12, 2014

On Fri, Jul 11, 2014 at 06:21:27PM -0700, Joe Perches wrote: > A simple script to run checkpatch --fix for various types of > of cleanups. > > This script is useful primarily for staging. > > This reformats code to a more CodingStyle conforming style, > compiles it, verifies that the object code hasn't changed, > and git commits it too. > > You must have the necessary development tools, git, and a > recent git tree. Ideally use Greg KH's staging-next, which > can be retrieved via these commands: > > git clone git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git > git checkout staging-next > > To use this script try a sequence of commands like: > > cd <linux_repository> > git checkout -b <your_branch> > make allyesconfig > mkdir patches > ./scripts/reformat_with_checkpatch.sh drivers/staging/<dir>/*.[ch] > git format-patch --cover-letter -o patches/<your_branch> staging-next > git send-email patches/<your_branch> When running this on drivers/base/bus.c, it says that the .o files are different, when the diffstat for what makes them different is only whitespace. I did the following: $ scripts/reformat_with_checkpatch.sh drivers/base/bus.c Ignore the first set of things it tries to commit by answering N to the "Would you like to commit these changes". Then the second thing it tries to change in the file says that there is a .o file difference. Yet the diff is below, I don't see how this happens. Is this due to there being some old temp file around because I did not accept the first set of changes? thanks, greg k-h ------------- diff: drivers/base/bus.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 83e910a57563..3546d02b46f0 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -32,7 +32,6 @@ static struct kset *system_kset; #define to_drv_attr(_attr) container_of(_attr, struct driver_attribute, attr) - static int __must_check bus_rescan_devices_helper(struct device *dev, void *data); @@ -128,6 +127,7 @@ static const struct sysfs_ops bus_sysfs_ops = { int bus_create_file(struct bus_type *bus, struct bus_attribute *attr) { int error; + if (bus_get(bus)) { error = sysfs_create_file(&bus->p->subsys.kobj, &attr->attr); bus_put(bus); @@ -817,6 +817,7 @@ EXPORT_SYMBOL_GPL(device_reprobe); struct bus_type *find_bus(char *name) { struct kobject *k = kset_find_obj(bus_kset, name); + return k ? to_bus(k) : NULL; } #endif /* 0 */ -------------- What the script complained about: Comparing objects... --- drivers/base/bus.o.new 2014-07-12 01:16:32.984755945 -0700 +++ drivers/base/bus.o.old 2014-07-12 01:16:31.924755967 -0700 @@ -2449,13 +2449,13 @@ 0000000000000000 <descriptor.17493>: ... -: bf 03 00 00 00 mov $0x3,%edi +: be 03 00 00 00 mov $0x3,%esi : 00 00 add %al,(%rax) ... 0000000000000028 <descriptor.17483>: ... -: a2 03 00 00 00 00 00 movabs %al,0x3 +: a1 03 00 00 00 00 00 movabs 0x3,%eax : 00 00 0000000000000050 <descriptor.17406>: @@ -2468,7 +2468,7 @@ 0000000000000078 <descriptor.17073>: ... -: 56 push %rsi +: 57 push %rdi : 00 00 add %al,(%rax) : 00 00 add %al,(%rax) : 00 00 add %al,(%rax) Object differences exist! - Verify changes before commit!

2 1

Re: [PATCH] reformat_with_checkpatch: Add automation to checkpatch
by Greg KH July 12, 2014

July 12, 2014

On Fri, Jul 11, 2014 at 06:57:24PM -0700, Joe Perches wrote: > On Fri, 2014-07-11 at 18:53 -0700, Greg KH wrote: > > On Fri, Jul 11, 2014 at 06:21:27PM -0700, Joe Perches wrote: > > > A simple script to run checkpatch --fix for various types of > > > of cleanups. > [] > > Anyway, try running this script on > > drivers/staging/lustre/lnet/lnet/acceptor.c to see how this build fails. > > lustre doesn't use "normal" kernel makefiles. Ok, I've fixed all of the lustre Makefile crud up and checked it into my staging-next branch. In doing so, it exposed just what a horrid mess the lustre include files are, which is good, hopefully people will notice and start fixing it up. And with the fixes, this script now properly can build and test lustre files. thanks, greg k-h

1 0

Eudyptula Challenge
by Narendra Joshi July 12, 2014

July 12, 2014

I am waiting for the results of my task03 submission for almost a week now. Has anyone been getting replies meanwhile? -- Narendra Joshi

3 2

question about mmc_rescan function
by Chan Kim July 12, 2014

July 12, 2014

Hi, I found the mmc_rescan function is called periodically. and the only place I can see the name is this line below in the mmc_alloc_host function INIT_DELAYED_WORK(&host->detect, mmc_rescan); so it is regsitered as a delayed work in mmc_host. And at the end of mmc_rescan, I see mmc_schedule_delayed_work(&host->detect, HZ); Q1 : Is this how it is called periodically every second? I guess so.. In the SD host driver, in initial slot initialization, there is mcc_host_add which eventually schedules mmc_rescan. (mmc_add_host -> mmc_start_host -> mmc_detect_change -> mmc_schedule_delayed_work(&host->detect, delay)) and towards the end of this initial slot initialization, there is queue_work(my_workqueue, &host->card_work); And the card_work is a function supposed to be called when a card is inserted or removed. In the card_work, there is at the end mmc_detect_change(slot->mmc, msecs_to_jiffies(host->pdata->detect_delay_ms)); and inside mmc_detect_change, mmc_schedule_delayed_work(&host->detect, delay); is called which is mmc_rescan. Q2 : Is there something wrong (mmc_rescan is started and schedule itself by mmc_add_host, but it is started by another card_work)? I tried removing queue_work and it still seems to work.. can anybody confirm or tell me anything? best regards, Chan

1 0

Re: [PATCH] reformat_with_checkpatch: Add automation to checkpatch
by Greg KH July 12, 2014

July 12, 2014

On Fri, Jul 11, 2014 at 06:40:16PM -0700, Joe Perches wrote: > On Fri, 2014-07-11 at 18:34 -0700, Greg KH wrote: > > On Fri, Jul 11, 2014 at 06:21:27PM -0700, Joe Perches wrote: > > > A simple script to run checkpatch --fix for various types of > > > of cleanups. > > > > > > This script is useful primarily for staging. > > > > > > This reformats code to a more CodingStyle conforming style, > > > compiles it, verifies that the object code hasn't changed, > > > and git commits it too. > > > > And 'git commits' it? > > The thought I had was to made it easier to > submit "my first kernel patch" correctly. > like that tuxradar article you wrote a few > years ago. > > http://www.tuxradar.com/content/newbies-guide-hacking-linux-kernel Yeah, but it's good for people to actually understand the change they are making, and edit it themselves. Or at least I think so, but I know others don't, so I don't mind this script. thanks, greg k-h

1 0