October 2011 - kernelnewbies@kernelnewbies.org

What is the significance of __releases & __acquires in rpm_callback() of power/runtime.c ?
by venu Oct. 11, 2011

Oct. 11, 2011

While going thro’ run time power management code in K39 power/runtime.c file, am seeing a function definition as given below. Any idea, what “__releases” & “__acquires” are doing in this function definition? static int rpm_callback(int (*cb)(struct device *), struct device *dev) __releases(&dev->power.lock) __acquires(&dev->power.lock) { int retval; …. return retval; }

2 1

userspace address to bus address
by Florian Bauer Oct. 10, 2011

Oct. 10, 2011

Hi, I'm currently writing my first kernel module, but i got stuck with address translations. I want my pcie device to access a buffer which was allocated in userspace (page aligned, currently smaller than one page to ignore scattering). I tried to translate the userspace address to a bus address in my ioctrl handler like this: void __user *data = arg; down_read(&current->mm->mmap_sem); get_user_pages(current, current->mm, (unsigned long)data, len/*1*/, 1, 0, &pages, NULL); up_read(&current->mm->mmap_sem); void* kvaddr = kmap(pages[0]); unsigned long baddr = virt_to_bus(kaddr); kunmap(pages[0]); page_cache_release(pages[0]); //... data is page-aligned and was locked in userspace with mlock. There is some address in baddr, but the device does not write to the correct location and sometimes causes kernel freezes. Can someone tell me, what is wrong with this code?TheI device works with a address, that was received in different way (closed source driver). I use 32 bit Ubuntu with an 2.6.32 kernel.

2 2

Enabling USB OTG
by madav maddy Oct. 10, 2011

Oct. 10, 2011

Hi Experts, I am newbie and trying to enable usb OTG support in Android linux kernel, can someone please give me the pointers to proceed with this. The hardware on which android kernel boots has OTG harware, so inorder to enable it at the kenel level i enabled the following configs, CONFIG_USB_SUPPORT=y CONFIG_USB_ARCH_HAS_HCD=y CONFIG_USB_ARCH_HAS_OHCI=y CONFIG_USB_ARCH_HAS_EHCI=y CONFIG_USB=y CONFIG_USB_DEBUG=y . . CONFIG_USB_EHCI_HCD=m CONFIG_USB_OHCI_HCD=m . . CONFIG_USB_OTG=y CONFIG_USB_MUSB_OTG=y CONFIG_USB_OTG_UTILS=y am using the following configuration, musb <--- usb converter <--- usb harware device. But still the hardware doesnt seems to work properly (as no new entry is created in /dev, when a usb harware device is connected to the board), Inorder to confirm that the base board drivers are working properly, i tried loading the gadget driver and connected it to the PC host, where i was able to access the device mass storage successfully. Do i need to do anything else to enable the musb to work as OTG port. Thanks, Madavan

2 1

When is to preempt safe?
by Parmenides Oct. 9, 2011

Oct. 9, 2011

Hi, Preemption has two cases: user preemption and kernel preemption. I have tow questions about them. 1. According to Love, "If the kernel is returning to user-space, it knows it is in a safe quiescent state. In other words, if it is safe to continue executing the current task, it is also safe to pick a new task to execute." What's the meaning of user preemption's safety? How can we deduce safety of schedule from the current task going on? 2. Another statement from Love is that "the kernel can preempt a task running in the kernel so long as it does not hold a lock". Why is it not safe while kernel code holding lock?

7 13

problem mapping physical address from /dev/mem
by Vaibhav Jain Oct. 7, 2011

Oct. 7, 2011

Hi, I am trying to run a program that scans memory from a given physical address using /dev/mem. It uses mmap to map physical address from /dev/mem. So to start with I used /proc/iomem to look up the physical memory mapping and found the address 00010000 to be the starting address for System ram. But whenever I provide this address to the program it throws an error of "Operation not permitted". Browsing through some articles I found the option of compiling the kernel with CONFIG_STRICT_DEVMEM option disabled. I have tried this but this isn't working either. Please help me figure out the problem. Thanks Vaibhav Jain

4 9

Re: [lm-sensors] [PATCH] hwmon class driver registration with a device number
by Himanshu Chauhan Oct. 7, 2011

Oct. 7, 2011

On Thu, Oct 06, 2011 at 12:07:52PM -0700, Guenter Roeck wrote: > And why, and what for. The initial idea of posting to kernelnewbies was to get a hint on how the patch would be taken as. I wanted to know if developers will like the idea behind it or not. I guess, Guenter is not convinced with any of my reasoning. I am willing to clean it up further only if I get a positive hint. But it doesn't seem to be going anywhere. Thanks Guenter, Alan, and Greg for taking your time and reviewing it. Greg: To answer your last question, if this was taken positively, I was thinking of having functionality similar to misc device registration. -Himanshu

2 3

Why do processes with higher priority to be allocated more timeslice?
by Parmenides Oct. 7, 2011

Oct. 7, 2011

Hi, It seems that Linux's scheduler tends to allocate longer timeslice for processes with higher priority. Actually, the CFS scheduler which is a new scheduler in Linux kernel also does the same thing. But, I think this way does not fit with scheduler's principle. The goal chased by a scheduler is low latency and high thoughput. Normally, a I/O-bound process has higher priority, while a CPU-bound process has lower priority. So, a I/O-bound process (which has enough timeslice) can preempt a CPU-bound process easily. This way ensures lower latency. It is also necessary that CPU-bound processes are to be allocated longer timeslice to improve throughput owing to less process switch costs. That means lower priority processes (CPU-bound) should be allocated longer timeslice, whichs obviously conflicts with the actual practice taken by the Linux's scheduler. Any explanation? Thanks.

4 8

How to open and write to a block driver within a character driver?
by Yang Xiang Oct. 7, 2011

Oct. 7, 2011

I am taksed to develop a Block device driver which will be used within a character device driver so the write to the character driver will be also formatted, converted and duplicated to the block drvice. I am new at this. And I am doing all this within kernel space so I really don't know I can open a block device without using "open" and "write" syscalls. I know there is a generic_make_request call to block driver but it requires the bio vec to be filled out with data and I don't know where to start on that. Any quick solution to this plase? Thanks so much in advance, yx

3 6

[PATCH] hwmon class driver registration with a device number
by Himanshu Chauhan Oct. 6, 2011

Oct. 6, 2011

This patch adds a way to register a hwmon class driver with a device number rather than the default MK_DEV(0,0). This would help in creating required "dev" file under sysfs which in turn can be used in populating /dev tree during bootup (helpful to driver which want to show up a char interface along with sysfs). Signed-off-by: Himanshu Chauhan <hschauhan(a)nulltrace.org> --- drivers/hwmon/hwmon.c | 28 ++++++++++++++++++++++------ include/linux/hwmon.h | 1 + 2 files changed, 23 insertions(+), 6 deletions(-) diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c index a61e781..36961c5 100644 --- a/drivers/hwmon/hwmon.c +++ b/drivers/hwmon/hwmon.c @@ -31,15 +31,17 @@ static DEFINE_IDR(hwmon_idr); static DEFINE_SPINLOCK(idr_lock); /** - * hwmon_device_register - register w/ hwmon - * @dev: the device to register + * hwmon_device_register_numbered - register hwmon class dev + * with specific device number. + * @dev: Device to register. + * @dev_id: device number to register. * - * hwmon_device_unregister() must be called when the device is no - * longer needed. + * hwmon_device_unregister() must be called when the device + * is no longer needed. * * Returns the pointer to the new device. */ -struct device *hwmon_device_register(struct device *dev) +struct device* hwmon_device_register_numbered(struct device *dev, dev_t dev_id) { struct device *hwdev; int id, err; @@ -58,7 +60,7 @@ again: return ERR_PTR(err); id = id & MAX_ID_MASK; - hwdev = device_create(hwmon_class, dev, MKDEV(0, 0), NULL, + hwdev = device_create(hwmon_class, dev, dev_id, NULL, HWMON_ID_FORMAT, id); if (IS_ERR(hwdev)) { @@ -71,6 +73,20 @@ again: } /** + * hwmon_device_register - register w/ hwmon + * @dev: the device to register + * + * hwmon_device_unregister() must be called when the device is no + * longer needed. + * + * Returns the pointer to the new device. + */ +struct device *hwmon_device_register(struct device *dev) +{ + return hwmon_device_register_numbered(dev, MK_DEV(0,0)); +} + +/** * hwmon_device_unregister - removes the previously registered class device * * @dev: the class device to destroy diff --git a/include/linux/hwmon.h b/include/linux/hwmon.h index 6b6ee70..d816cbe 100644 --- a/include/linux/hwmon.h +++ b/include/linux/hwmon.h @@ -16,6 +16,7 @@ #include <linux/device.h> +struct device *hwmon_device_register_numbered(struct device *dev, dev_t dev_id); struct device *hwmon_device_register(struct device *dev); void hwmon_device_unregister(struct device *dev); -- 1.7.4.1

2 3

understand the result of lockstat
by Zheng Da Oct. 6, 2011

Oct. 6, 2011

Hello, I try to analyze the lock contention in the linux FS, so I enable lockstats in the kernel. But I have some troubles to understand the result in lockstats. When I read 4GB data randomly from a file on a ramdisk with 8 threads (I have 8 cores), and it takes 21 seconds. The most contended lock is class name con-bounces contentions waittime-min waittime-max waittime-total acq-bounces acquisitions holdtime-min holdtime-max holdtime-total &(&mapping->tree_lock)->rlock 104380 104615 0 0 61705.23 169228 2862470 0 0 1010557.22 &(&zone->lru_lock)->rlock 1233 1245 0 0 2632.71 21834 255752 0 3.86 402904.1 xtime_lock 722 731 0 0 1916.15 1342 2901 0 0 4812.51 What is unit of the waittime? in nanoseconds? I look into the code of x86, and it seems the clock time the code reads is nanoseconds. But if it's in nanoseconds, the wait time of mapping->tree_lock is very short, only 62us in total? Also, the result shows that there are 104615 contentions out of 2862470 acquisitions. I guess it's not very highly contended. Thanks, Da

1 1