This document is part of the Driver Development Kit tutorial documentation.
Writing a device driver is often viewed as a daunting task, fraught with complexities and requiring arcane knowledge of little-known kernel secrets.
The goal of this section is to demystify the process; you'll learn everything you need to know about how to write device drivers, starting with what they do, how they work, and how they fit into the overall system.
At the highest level, a device driver‘s job is to provide a uniform interface to a particular device, while hiding details specific to the device’s implementation.
Two different ethernet drivers, for example, both allow a client to send packets out an interface, using the exact same C language function. Each driver is responsible for managing its own hardware in a way that makes the client interfaces identical, even though the hardware is different.
Note that the interfaces that are provided by the driver may be “intermediate” — that is, they might not necessarily represent the “final” device in the chain.
Consider a PCI-based ethernet device. First, a base PCI driver is required that understands how to talk to the PCI bus itself. This driver doesn't know anything about ethernet, but it does know how to deal with the specific PCI chipset present on the machine.
It enumerates the devices on that bus, collects information from the various registers on each device, and provides functions that allow its clients (such as the PCI-based ethernet driver) to perform PCI operations like allocating an interrupt or a DMA channel.
Thus, this base PCI driver provides services to the ethernet driver, allowing the ethernet driver to manage its associated hardware.
At the same time, other devices (such as a video card) could also use the base PCI driver in a similar manner to manage their hardware.
In order to provide maximum flexibility, drivers in the Zircon world are allowed to bind to matching “parent” devices, and publish “children” of their own. This hierarchy extends as required: one driver might publish a child, only to have another driver consider that child their parent, with the second driver publishing its own children, and so on.
In order to understand how this works, let's follow the PCI-based ethernet example.
The system starts by providing a special “PCI root” parent. Effectively, it‘s saying "I know that there’s a PCI bus on this system, when you find it, bind it here."
The “Advanced Topics” section below has more details about this process.
Drivers are evaluated by the system (a directory is searched), and drivers that match are automatically bound.
In this case, a driver that binds to a “PCI root” parent is found, and bound.
This is the base PCI driver. It's job is to configure the PCI bus, and enumerate the peripherals on the bus.
The PCI bus has specific conventions for how peripherals are identified: a combination of a Vendor ID (VID) and Device ID (DID) uniquely identifies all possible PCI devices. During enumeration, these values are read from the peripheral, and new parent nodes are published containing the detected VID and DID (and a host of other information).
Every time a new device is published, the same process as described above (for the initial PCI root device publication) repeats; that is, drivers are evaluated by the system, searching for drivers that match up with the new parents' characteristics.
Whereas with the PCI root device we were searching for a driver that matched a certain kind of functionality (called a “protocol,” we‘ll see this shortly), in this case, however, we’re searching for drivers that match a different protocol, namely one that satisfies the requirements of “is a PCI device and has a given VID and DID.”
If a suitable driver is found (one that matches the required protocol, VID and DID), it's bound to the parent.
As part of binding, we initialize the driver — this involves such operations as setting up the card for operation, bringing up the interface(s), and publishing a child or children of this device. In the case of the PCI ethernet driver, it publishes the “ethernet” interface, which conforms to yet another protocol, called the “ethernet implementation” protocol. This protocol represents a common protocol that‘s close to the functions that clients use (but is one step removed; we’ll come back to this).
We mentioned three protocols above:
The names in brackets are the C language constants corresponding to the protocols, for reference.
So what is a protocol?
A protocol is a strict interface definition.
The ethernet driver published an interface that conforms to
ZX_PROTOCOL_ETHERNET_IMPL. This means that it must provide a set of functions defined in a data structure (in this case,
These functions are common to all devices implementing the protocol — for example, all ethernet devices must provide a function that queries the MAC address of the interface.
Other protocols will of course have different requirements for the functions they must provide. For example a block device will publish an interface that conforms to the “block implementation protocol” (
ZX_PROTOCOL_BLOCK_IMPL) and provide functions defined by
block_protocol_ops_t. This protocol includes a function that returns the size of the device in blocks, for example.
We'll examine these protocols in the following chapters.
The above has presented a big picture view of Zircon drivers, with a focus on protocols.
In this section, we'll examine some advanced topics, such as platform dependent and platform independent code decoupling, the “miscellaneous” protocol, and how protocols and processes are mapped.
Above, we mentioned that
ZX_PROTOCOL_ETHERNET_IMPL was “close to” the functions used by the client, but one step removed. That‘s because there’s one more protocol,
ZX_PROTOCOL_ETHERNET, that sits between the client and the driver. This additional protocol is in place to handle functionality common to all ethernet drivers (in order to avoid code duplication). Such functionality includes buffer management, status reporting, and administrative functions.
This is effectively a “platform dependent” vs “platform independent” decoupling; common code exists in the platform independent part (once), and driver-specific code is implemented in the platform dependent part.
This architecture is repeated in multiple places. With block devices, for example, the hardware driver binds to the bus (e.g., PCI) and provides a
ZX_PROTOCOL_BLOCK_IMPL protocol. The platform independent driver binds to
ZX_PROTOCOL_BLOCK_IMPL, and publishes the client-facing protocol,
You'll also see this with the display controllers, I2C bus, and serial drivers.
In simple drivers, we show the code for several drivers that illustrate basic functionality, but don't provide services related to a specific protocol (i.e., they are not “ethernet” or “block” devices). These drivers are bound to
@@@ More content?
In order to keep the discussions above simple, we didn‘t talk about process separation as it relates to the drivers. To understand the issues, let’s see how other operating systems deal with them, and compare that to the Zircon approach.
In a monolithic kernel, such as Linux, many drivers are implemented within the kernel. This means that they share the same address space, and effectively live in the same “process.”
The major problem with this approach is fault isolation / exploitation. A bad driver can take out the entire kernel, because it lives in the same address space and thus has privileged access to all kernel memory and resources. A compromised driver can present a security threat for the same reason.
The other extreme, that is, putting each and every driver service into its own process, is used by some microkernel operating systems. Its major drawback is that if one driver relies on the services of another driver, the kernel must effect at least a context switch operation (if not a data transfer as well) between the two driver processes. While microkernel operating systems are usually designed to be fast at these kinds of operations, performing them at high frequency is undesirable.
The approach taken by Zircon is based on the concept of a device host (devhost). A devhost is a process that contains a protocol stack — that is, one or more protocols that work together. The devhost loads drivers from ELF shared libraries (called Dynamic Shared Objects, or DSOs). In the simple drivers section, we‘ll see the meta information that’s contained in the DSO to facilitate the discovery process.
The protocol stack effectively allows the creation of a complete “driver” for a device, consisting of platform dependent and platform independent components, in a self-contained process container.
For the advanced reader, take a look at the
dm dump command available from the Zircon command line. It displays a tree of devices, and shows you the process ID, DSO name, and other useful information.
Here's a highly-edited version showing just the PCI ethernet driver parts:
1. [root] 2. [sys] 3. <sys> pid=1416 /boot/driver/bus-acpi.so 4. [acpi] pid=1416 /boot/driver/bus-acpi.so 5. [pci] pid=1416 /boot/driver/bus-acpi.so ... 6. [00:02:00] pid=1416 /boot/driver/bus-pci.so 7. <00:02:00> pid=2052 /boot/driver/bus-pci.proxy.so 8. [e1000] pid=2052 /boot/driver/e1000.so 9. [ethernet] pid=2052 /boot/driver/ethernet.so
From the above, you can see that process ID
1416 (lines 3 through 6) is the Advanced Configuration and Power Interface (ACPI) driver, implemented by the DSO
During primary enumeration, the ACPI DSO detected a PCI bus. This caused the publication of a parent with
ZX_PROTOCOL_PCI_ROOT (line 5, causing the appearance of the
[pci] entry), which then caused the devhost to load the
bus-pci.so DSO and bind to it. That DSO is the “base PCI driver” to which we've been referring throughout the discussions above.
During its binding, the base PCI driver enumerated the PCI bus, and found an ethernet card (line 6 detects bus 0, device 2, function 0, shown as
[00:02:00]). (Of course, many other devices were found as well, but we've removed them from the above listing for simplicity).
The detection of this device then caused the base PCI driver to publish a new parent with
ZX_PROTOCOL_PCI and the device's VID and DID. Additionally, a new devhost (process ID
2052) was created and loaded with the
bus-pci.proxy.so DSO (line 7). This proxy serves as the interface from the new devhost (pid
2052) to the base PCI driver (pid
This is where the decision was made to “sever” the device driver into its own process — the new devhost and the base PCI driver now live in two different processes.
The new devhost
2052 then finds a matching child (the
e1000.so DSO on line 8; it‘s considered a match because it has
ZX_PROTOCOL_PCI and the correct VID and DID). That DSO publishes a
ZX_PROTOCOL_ETHERNET_IMPL, which binds to a matching child (the
ethernet.so DSO on line 9; it’s considered a match because it has a
What‘s not shown by this chain is that the final DSO (
ethernet.so) publishes a
ZX_PROTOCOL_ETHERNET — that’s the piece that clients can use, so of course there's no further “device” binding involved.