- kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

Re: Kernel thread scheduling
by Ruben Safir April 13, 2015

April 13, 2015

On 04/12/2015 12:36 AM, Mohammed Ghriga wrote: > In addition to the suggestions that were offered, I recommend you try reading Chapter 16: https://www.google.com/search?tbo=p&tbm=bks&q=isbn:0596554672 (pages: 267-272). > http://it-ebooks.info/book/481/ FWIW I found this lovely wait example in the other kernel text under the scheduling section Sleeping and Waking Up page 58 in Robert Love, Linux Kernel Development http://www.it-ebooks.info/book/819/

2 1

Problem with SPI RTC DS3234 on RaspberryPi 2
by John Whitmore April 12, 2015

April 12, 2015

I've had a problem for weeks with a development board I have used in the past on the Original RaspberryPi. The expansion board has two spi devices connected, a CAN Bus controller (MCP2515) and a Real Time Clock chip DS3234. On the old RPi everything works great with two devices in: /sys/bus/spi/devices pi@can-rtc:~$ cat /sys/bus/spi/devices/spi0.0/modalias spi:mcp2515 pi@can-rtc:~$ cat /sys/bus/spi/devices/spi0.1/modalias spi:ds3234 This output is exactly the same on the RPi-2 but the RPi-2 finds the RTC time invalid: sudo hwclock -r hwclock: The Hardware Clock registers contain values that are either invalid (e.g. 50th day of month) or beyond. The CAN Bus Controller does work on both versions of the RPi, it's just the RTC that has a problem. If connect up the Ethernet interface to get the correct time, from the network, I can write it to the device with "hwclock -w". Then remove the expansion board and power it up on the original RPi and it's got the correct time which is sets as the system time. If I now remove it from the original RPi and connect to RPi-2 it can't read a valid time. but it all seems the same. cat /sys/bus/spi/devices/spi0.1/rtc/rtc0/name ds3234 In both cases I've compiled the ds3234 and mcp2515 drivers into the kernel. In the original RPi the board file is edited to add the two devices but in the RPi-2 a hardware overlay file is added. Like I say everything appears to be exactly the same in both cases. Just the result is totally different. Can anybody explain why there should be a difference? The GPIO headers in both versions of the RPi are the same for the first set of pins. Original RPi kenel - 3.15.5 RPi-2 kernel - 3.18.8

2 2

confusion about locking [ Pls help ]
by Er Krishna April 12, 2015

April 12, 2015

Hi All, Can somebody help me here to understand following basic scenario about locking. Sorry if I am wrong in putting my points in correct manner. 1. Can we take semaphore (binary semaphore or mutex) if the critical section data is shared across two different threads/processes which are running on two different processors simultaneously/parallely ? I know spin_lock() can handle this scenario, but I was trying to understand can semaphore also handle this scenario, assuming there is no interrupt and only process context is there ? I think yes. Pls confirm me on this and if not then why not ? If any pross and cons are involved pls tell that also. 2. If the critical section data is shared across process and interrupt ( consider this scenario on uniprocessor machine and I have not taken local_irq_save() or any spin_lock_irqsave ) and in the currently executing process critical section is protected by preempt_disable() and suddenly an interrupt occurs which also wants to access same critical section data, so what will happen ? Will in this case Interrupt handler acquire the critical section data and corrupt the kernel or it will sleep and corrupt the kernel or it will keep doing busy waiting ? I think interrupt handler will keep doing busy waiting and machine will stuck, kind of deadlock. Pls confirm me. 3. Does Preemption gets disable ( on local processor ) in case of binary semaphore or mutex ? I think no. Pls confirm. Second, Lets say critical section is shared across process and interrupt and it is locked by semaphore (binary semaphore or mutex) in process context on uniprocessor machine. If Interrupt comes then what will happen ? Will it preempt the process and acquire the critical section of data and corrupt it or interrupt will go into sleep mode and machine can crash/stuck. Pls explain. Best Regards, Krishna

2 1

Re: Kernel thread scheduling
by Ruben Safir April 12, 2015

April 12, 2015

On 04/10/2015 09:09 AM, nick wrote: > > > On 2015-04-09 11:37 PM, Ruben Safir wrote: >> On 04/09/2015 10:52 PM, nick wrote: >>> Before asking questions again like this please look into either using lxr or ctags >>> to navigate the kernel tree for answers as can be faster then waiting for me or >>> someone else to respond. >> >> >> well, I reading the text is ctags aren't much value there. >> > Ctags is useful for searching the code, which is why I am recommending it. > Nick I have it built into gvim, but you can't use it from a textbook. I'm finding it is not as useful as it could be for the kernel code. There are stacks of tags to get around. Another 2 days to learn to get around tags in vi is not in the agenda right now. It is the tool I have so I'll have to live with it right now. I also have a question that is not obvious from the code I'm looking at. I'm not sure how these structs are attached together. Or more specifically, I'm not sure how pulling the correct sched_entity gets one the coresponding task_entity You have struct task_struct with a struct sched_entity struct sched_enitities are nodes in the RB tree which are a "container" for "struct rb_node run_node". So a look at sched_entity ... is in ../linux/sched.h 1161 struct sched_entity { 1162 struct load_weight load; /* for load-balancing */ 1163 struct rb_node run_node; 1164 struct list_head group_node; 1165 unsigned int on_rq; 1166 1167 u64 exec_start; 1168 u64 sum_exec_runtime; 1169 u64 vruntime; 1170 u64 prev_sum_exec_runtime; 1171 1172 u64 nr_migrations; 1173 1174 #ifdef CONFIG_SCHEDSTATS 1175 struct sched_statistics statistics; 1176 #endif 1177 1178 #ifdef CONFIG_FAIR_GROUP_SCHED 1179 int depth; 1180 struct sched_entity *parent; 1181 /* rq on which this entity is (to be) queued: */ 1182 struct cfs_rq *cfs_rq; 1183 /* rq "owned" by this entity/group: */ 1184 struct cfs_rq *my_q; 1185 #endif 1186 1187 #ifdef CONFIG_SMP 1188 /* Per-entity load-tracking */ 1189 struct sched_avg avg; 1190 #endif 1191 }; I see no means of referencing a specific task from this struct that forms the node. So when you pull the node with the smallest vruntime from the left most postion of the RB tree, by calling pick_next_task(), static struct sched_entity *__pick_next_entity(struct sched_entity *se) { struct rb_node *next = rb_next(&se->run_node); if (!next) return NULL; return rb_entry(next, struct sched_entity, run_node); } how do we know what task we are attached to? Ruben >> _______________________________________________ >> Kernelnewbies mailing list >> Kernelnewbies(a)kernelnewbies.org >> http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies >> > >

2 3

Kernel thread scheduling
by Vincenzo Scotti April 10, 2015

April 10, 2015

Hello, I am actually studying kernel threads, and I have some doubts about them. Let's take for example this snippet of code static int thread_function(void *data) { while (!kthread_should_stop()) { schedule(); } pr_err("Stopped"); return 0; } This way it works just fine, and waits until I call kthread_stop on it. But if I comment out that schedule() call, it just hangs my system when I load it (it is part of a module). I see that the loop-schedule-wakeup pattern is used among all the others kernel threads. But I don't get why I need to call the scheduler explicitly. I know that the kernel is fully preemptible, and in my interpretation I thought that it could stop every running thread, even in kernel space, using a timer-based interrupt handler, to give cpu to other threads. Doesn't this pattern resemble a voluntary preemption model? Where am I wrong?

5 7

Re: Out of Date Project Ideas on Btrfs
by nick April 9, 2015

April 9, 2015

On 2015-04-09 06:07 PM, Robert P. J. Day wrote: > On Thu, 9 Apr 2015, nick wrote: > >> Greetings All, > >> I have been searching for work related to the btrfs file system. I >> am wondering if there is any that is still valid on the Project >> Ideas page of the wiki as most of it seems out of date to me. The >> only one that seems to be still valid is the Rename syscall new >> parameters needing to be supported by btrfs as most of the others >> including fs_info rather then root object to certain functions as >> parameters seem very out of date to me. > > why do you keep asking about btrfs ***here***? go talk to the btrfs > maintainers, fer gods sake. if you're utterly incapable of using > google, here, let me help: > > https://btrfs.wiki.kernel.org/index.php/Main_Page > > rday > That's fine,I will ask on the right list. Thanks, Nick > p.s. now do you understand the concept of "spoon feeding?" sheesh ... >

1 0

Out of Date Project Ideas on Btrfs
by nick April 9, 2015

April 9, 2015

Greetings All, I have been searching for work related to the btrfs file system. I am wondering if there is any that is still valid on the Project Ideas page of the wiki as most of it seems out of date to me. The only one that seems to be still valid is the Rename syscall new parameters needing to be supported by btrfs as most of the others including fs_info rather then root object to certain functions as parameters seem very out of date to me. Thanks Nick

2 1

Re: Question about shared interrupts in devicetree
by Stefan Wahren April 9, 2015

April 9, 2015

Hi Geert, > Geert Uytterhoeven <geert(a)linux-m68k.org> hat am 9. April 2015 um 09:37 > geschrieben: > > [...] > > > > > Could you please explain the benefit / reason of this approach? > > Since you have different logical modules in the subsystem, this allows to > model > the subsystem as separate modules and a separate interrupt controller, and > have separate drivers for all modules. > > Look at da9063 for an example (there are more according to "git grep irq_chip > -- drivers/mfd", but not all of them may have DT bindings). > thanks for the hint, but this makes my patch series not really smaller. I invested a lot of time (started in September 2014) and i want to submit at least 1 of the 3 drivers soon as possible. I think the best approch would be to submit my series with option B as a proposal. In the discussion about the patch it would be easier to decide if a interrupt controller is really necessary. > Gr{oetje,eeting}s, > > Geert > > -- > Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- > geert(a)linux-m68k.org > > In personal conversations with technical people, I call myself a hacker. But > when I'm talking to journalists I just say "programmer" or something like > that. > -- Linus Torvalds Stefan

1 0

Spoon feeding
by Nicholas Krause April 9, 2015

April 9, 2015

Greetings All, After thinking for a few days I have come to the conclusion that the communities definition of spoon feeding and my are different. If someone would like to give me the community definition so there is no more misunderstandings and time wasting that would be great. Nick -- Sent from my Android device with K-9 Mail. Please excuse my brevity.

3 2

Re: Question about shared interrupts in devicetree
by Stefan Wahren April 9, 2015

April 9, 2015

Hi Geert, Am 08.04.2015 um 21:20 schrieb Geert Uytterhoeven: > On Tue, Apr 7, 2015 at 7:06 PM, Stefan Wahren <stefan.wahren(a)i2se.com> wrote: >> [...] >> >> thanks for the good explanation. After looking into the reference manual [1] of >> the i.MX28 i still can't decide if the subsystem generate the interrupts as a >> whole or >> as a logical group. I only found this para in chapter 11.11: >> >> The VDDA_BO_IRQ, VDDD_BO_IRQ, VDDIO_BO_IRQ, and BATT_BO_IRQ each >> have their own interrupt line back to the interrupt collector. >> However, the remaining five interrupts—VDD5V_GT_VDDIO_IRQ, DC_OK_IRQ, >> VBUSVALID_IRQ, LINREG_OK_IRQ and PSWITCH_IRQ—all share a single >> interrupt line. In this case, software must read the interrupt status >> bits to discover which event caused the interrupt. >> >> In my case DC_OK_IRQ and PSWITCH_IRQ are relevant. >> >> Maybe someone else has a idea? > Perhaps you can implement an interrupt-controller to handle the multiplexing > of the 5 remaining interrupts? Could you please explain the benefit / reason of this approach? > > Can they be disabled/enabled individually? Yes. What are the consequences? Thanks Stefan > >> Thanks Stefan >> >> [1] - http://cache.freescale.com/files/dsp/doc/ref_manual/MCIMX28RM.pdf >> >> Document Number: MCIMX28RM >> Rev 2, 08/2013 >> >> > Gr{oetje,eeting}s, > > Geert > > -- > Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert(a)linux-m68k.org > > In personal conversations with technical people, I call myself a hacker. But > when I'm talking to journalists I just say "programmer" or something like that. > -- Linus Torvalds

2 1