Take the 2-minute tour ×
Ask Ubuntu is a question and answer site for Ubuntu users and developers. It's 100% free, no registration required.

I would like to understand the flow of the USB events form the Kernel space to the User space (just out of my curiosity, in knowing how things are implemented).

To be more clear, I would like to know how does that pop-up comes in my desktop when I plug a USB drive to my system and how does the drive gets mounted.

I would also like to know how it finds out if some images are present in my USB and if "yes" , how it asks me whether I need to open it in GIMP or some other software?

I know its a very big and wide question, but please guide me with some pointers on how to understand the whole idea behind it.

I have not done large code browsings, but I assure you that I have no problem in doing so if I can be guided correctly.

share|improve this question

1 Answer 1

up vote 12 down vote accepted
  1. Device is plugged in, and the software that manages the hardware bus for that device receives an interrupt (or other notification at the hardware level), and the bus driver enumerates attached devices, or performs other bus-specific hardware actions to identify the device.
  2. Kernel requests to load a driver for the new hardware by calling /sbin/modprobe with hardware's bus/device/etc identifier.
  3. In user-space, modprobe tries to find a matching driver-specified alias. (See /lib/modules/$(uname -r)/modules.alias for the complete list.) These will look different based on the hardware interface. For example, pci:v0000102Bd00002527sv*sd*bc*sc*i* for a PCI device vendor 102B, device 2527, and anything for subvendor, etc, or USB: usb:v2040p4982d*dc*dsc*dp*ic*isc*ip*.
  4. once the device driver is loaded (or a new device that already had a driver is initialized), the driver in the kernel sends a notification of the loaded device to udev in userspace.
  5. udev matches the notification against its list of rules in /lib/udev/rules.d/ and /etc/udev/rules.d. From here, the behavior is extremely specialized, based on the rules.
  6. In the case of a USB disk, the 80-udisks.rules file is likely the best place to work from. These rules will use things like blkid and other helpers to probe the type and contents of a disk, populating all sort of configuration values including things like ENV{UDISKS_PRESENTATION_HIDE}="1" to ignore a disk for some reason. See "man 7 udisks" for details.
  7. The udisks-daemon watches for devices to appear in the udev database, and presents them them as a discoverable list of devices over DBus. (See "udisks --enumerate".)
  8. Various actions are configured in udisks, and the policy for allowing those actions can be seen in the policy file /usr/share/polkit-1/actions/org.freedesktop.udisks.policy. (Who can mount, umount, etc.)
  9. Services that are interested in devices will listen for DBus events from udisks, and take actions when they see certain conditions. For example, GNOME's Nautilus (via gvfs volume-monitor) will request automounting for devices (via udisks, which will check its policy, mentioned above).
  10. Once a filesystem has been mounted, those same listening services will take more actions. For example, Nautilus will ask if you want to open F-Spot when the common photo storage directory DCIM is found on a filesystem.
share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.