Probe function in Linux USB device driver in brief

-
The probe function is called when a device is installed that the USB core thinks this driver should handle; the probe function should perform checks on the information passed to it about the device and decide whether the driver is really appropriate for that device.


In the probe function callback, the USB driver should initialize any local structures that it might use to manage the USB device. It should also save any information that it needs about the device to the local structure, as it is usually easier to do so at this time. As an example, USB drivers usually want to detect what the endpoint address and buffer sizes are for the device, as they are needed in order to communicate with the device. 

The probe function has following prototype,

static int test_probe(struct usb_interface *interface, const struct usb_device_id *id)

Typical flow of detecting the endpoints' information, driver:
  • loops over every endpoint that is present in provided interface 
  • looks for endpoint direction(IN or OUT) which can be tested by seeing whether the bitmask USB_DIR_* is contained in bEndPointAddress 
  • determines whether the endpoint type is what driver expects by first masking off the bmAttributes variable with the USB_ENDPOINT_XFERTYPE_MASK bitmask, and then checking if it matches the value of expected endpoint type USB_ENDPOINT_XFER_* 
If all of these tests are true, the driver knows it found the proper type of endpoint and can save the information about the endpoint that it will later need to communicate over it in a local structure.

After identifying appropriate endpoint, driver: 
  • saves it's local data structure into usb_interface using usb_set_intfdata and this local data structure can be retrieved using usb_get_intfdata. usb_get_intfdata is usually called in driver's open and disconnect function 
  • registers itself with usb core using usb_register_dev at last
References: 

Comments

Post a Comment

Popular posts from this blog

MBR partitioning with 'parted' utility

-
Image
As you might know, linux provides a very handy utility i.e. parted to create, destroy, resize, move and copy(and much more) partitions. So, in this post I will explain how parted is used to create partitions for MBR scheme. Let's commence with some pre-requisites, Machine running linux to perform operations Any storage device like HDD, USB etc. on which we will perform various operations I am using USB stick for this purpose and will carry out all operations on that. I strongly recommend you to use a external disk which doesn't have any important data and if it has, you must take backup before performing any of the following operations. If you do not have much idea about MBR partitioning schemes, Please refer this post first to get basic understanding of disk partitioning. NOTE: You need to have root permission to carry out all the following operations. First thing we do here is to check what's there inside the storage disk. We insert USB drive t...
Read more

Disk Partitioning: MBR vs GPT

-
Image
What is disk partitioning? Disk Partitioning is a process of creating different regions or blocks, commonly referred as partitions of mass storage devices like fixed disks and removable disks. Why disk partitioning is required? The whole purpose of Disk Partitioning is to let Operating System(OS) manage information/data in each region/partition separately. There are two different disk partitioning schemes as follows: MBR (Master Boot Record) GPT (GUID Partition Table) MBR (Master Boot Record) MBR is a special type of boot sector resides in partitioned mass storage device. The concept of MBR was publicly introduced in 1983 with PC DOS 2.0 (From Wikipedia). Conventionally, sector size is 512 bytes and first sector of any storage drive contains MBR if that drive is partitioned with MBR scheme. MBR holds minimum boot loader code, primary/logical/ extended partition entry information, file system type, boot signature etc. The structure of classical MB...
Read more

Replace default splash screen in Yocto

-
Image
On boot, if you have observed a splash screen with yocto logo appears in core-image-sato. This post is all about to change that splash screen with your own customized image.  Splash screen is handled by a recipe called psplash found under "poky/meta-poky/recipes-core" directory of the source tree.  Since psplash expects an image to be in header file format, first you need to convert your image into a header file format by using script called "make-image-header.sh" Here is what  make-image-header.sh contains, source -  http://git.yoctoproject.org/cgit/cgit.cgi/psplash/tree/make-image-header.sh You could replace your generated image header in  poky/meta-poky/recipes-core/psplash/files with existing one but I would recommend to create a recipe in your own layer and add this image file there. Steps to be followed, Get your own splash screen image(I took .jpg image) Convert that into header file by using script described above ./ mak...
Read more