How To Write Linux PCI Drivers by Martin Mares on 07-Feb-2000 updated by Grant Grundler on 26-Jul-2006 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The world of PCI is vast and it's full of (mostly unpleasant) surprises. Since each CPU architecture implements different chipsets and PCI devices have different requirements (erm, "features"), the result is the PCI support in the Linux kernel is not as trivial as one would wish. This short paper tries to introduce all potential driver authors to Linux APIs for PCI device drivers. A more complete resource is the third edition of "Linux Device Drivers" by Jonathan Corbet, Alessandro Rubini, and Greg Kroah-Hartman. LDD3 is available for free (under Creative Commons License) from: http://lwn.net/Kernel/LDD3/ However, this document and the LDD3 are both subject to "bit rot". Refer to the source code if things are not working as described here. Please send questions/comments/patches about Linux PCI API to the "Linux PCI" mailing list. 0. Structure of PCI drivers ~~~~~~~~~~~~~~~~~~~~~~~~~~~ PCI drivers "discover" PCI devices in a system via pci_register_driver(). Actually, it's the other way around. The PCI generic code will notify the driver at every the PCI device which match a "description" advertised by the driver. Details on this below. pci_register_driver() leaves most of the probing for devices to the PCI layer and supports online insertion/removal of devices [thus supporting PCI, hot-pluggable PCI and CardBus in a single driver]. pci_register_driver() call requires passing in a table of function calls and thus dictates the high level structure of a driver. Once the driver knows about a PCI device and takes ownership, the driver generally needs to perform the following initialization: request MMIO/IOP resources Enable the device set the DMA mask size (for both coherent and streaming DMA) allocate and initialize shared control data (pci_allocate_coherent()) Access device configuration space (if needed) register IRQ handler (request_irq()) Initialize non-PCI (ie LAN/SCSI/etc parts of the chip) enable DMA/processing engines. When done using the device, and perhaps the module needs to be unloaded, the driver needs to take the follow steps: disable the device from generating IRQs release the IRQ (free_irq()) stop all DMA activity release DMA buffers (both streaming and coherent) unregister from other subsystems (e.g. scsi or netdev) Disable the device Most of these topics are covered in the following sections. For the rest look at LDD3 or . If the PCI subsystem is not configured (CONFIG_PCI is not set), most of the PCI functions described below are defined as inline functions either completely empty or just returning an appropriate error codes to avoid lots of ifdefs in the drivers. 1. pci_register_driver() call ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ New-style drivers call pci_register_driver() during their initialization with a pointer to a structure describing the driver (struct pci_driver): field name Description ---------- ------------------------------------------------------ id_table Pointer to table of device ID's the driver is interested in. Most drivers should export this table using MODULE_DEVICE_TABLE(pci,...). probe This probing function gets called (during execution of pci_register_driver() for already existing devices or later if a new device gets inserted) for all PCI devices which match the ID table and are not "owned" by the other drivers yet. This function gets passed a "struct pci_dev *" for each device whose entry in the ID table matches the device. The probe function returns zero when the driver chooses to take "ownership" of the device or an error code (negative number) otherwise. The probe function always gets called from process context, so it can sleep. remove The remove() function gets called whenever a device being handled by this driver is removed (either during deregistration of the driver or when it's manually pulled out of a hot-pluggable slot). The remove function always gets called from process context, so it can sleep. save_state Save a device's state before it is suspended. suspend Put device into low power state. resume Wake device from low power state. enable_wake Enable device to generate wake events from a low power state. (Please see Documentation/power/pci.txt for descriptions of PCI Power Management and the related functions) The ID table is an array of struct pci_device_id ending with an all-zero entry. Each entry consists of: vendor, device Vendor and device ID to match (or PCI_ANY_ID) subvendor, Subsystem vendor and device ID to match (or PCI_ANY_ID) subdevice, class Device class, subclass, and "interface" to match. See Appendix D of the PCI Local Bus Spec or include/linux/pci_ids.h for a full list of classes. Most drivers do not need to specify class/class_mask as vendor/device is normally sufficient. class_mask limit which sub-fields of the class field are compared. See drivers/scsi/sym53c8xx_2/ for example of usage. driver_data Data private to the driver. Most drivers don't need to use driver_data field. Best practice is to use driver_data as an index into a static list of equivalent device types, instead of using it as a pointer. Have a table entry {PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID} to have probe() called for every PCI device known to the system. New PCI IDs may be added to a device driver pci_ids table at runtime as shown below: echo "vendor device subvendor subdevice class class_mask driver_data" > \ /sys/bus/pci/drivers/{driver}/new_id All fields are passed in as hexadecimal values (no leading 0x). Users need pass only as many fields as necessary: o vendor, device, subvendor, and subdevice fields default to PCI_ANY_ID (FFFFFFFF), o class and classmask fields default to 0 o driver_data defaults to 0UL. Once added, the driver probe routine will be invoked for any unclaimed PCI devices listed in it's (newly updated) pci_ids list. Device drivers must initialize use_driver_data in the dynids struct in their pci_driver struct prior to calling pci_register_driver in order for the driver_data field to get passed to the driver. Otherwise, only a 0 is passed in that field. When the driver exits, it just calls pci_unregister_driver() and the PCI layer automatically calls the remove hook for all devices handled by the driver. Please mark the initialization and cleanup functions where appropriate (the corresponding macros are defined in ): __init Initialization code. Thrown away after the driver initializes. __exit Exit code. Ignored for non-modular drivers. __devinit Device initialization code. Identical to __init if the kernel is not compiled with CONFIG_HOTPLUG, normal function otherwise. __devexit The same for __exit. Tips on marks: o The module_init()/module_exit() functions (and all initialization functions called _only_ from these) should be marked __init/exit. o The struct pci_driver shouldn't be marked with any of these tags. o The ID table array should be marked __devinitdata. o The probe() and remove() functions (and all initialization functions called only from these) should be marked __devinit/exit. o If the driver is not a hotplug driver then use only __init/exit and __initdata/exitdata. o Pointers to functions marked as __devexit must be created using __devexit_p(function_name). That will generate the function name or NULL if the __devexit function will be discarded. 2. How to find PCI devices manually ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ PCI drivers should have a really good reason for not using the pci_register_driver() interface to search for PCI devices. A manual search may be performed using the following constructs. Searching by vendor and device ID: struct pci_dev *dev = NULL; while (dev = pci_get_device(VENDOR_ID, DEVICE_ID, dev)) configure_device(dev); Searching by class ID (iterate in a similar way): pci_get_class(CLASS_ID, dev) Searching by both vendor/device and subsystem vendor/device ID: pci_get_subsys(VENDOR_ID, DEVICE_ID, SUBSYS_VENDOR_ID, SUBSYS_DEVICE_ID, dev). You can use the constant PCI_ANY_ID as a wildcard replacement for VENDOR_ID or DEVICE_ID. This allows searching for any device from a specific vendor, for example. These functions are hotplug-safe. They increment the reference count on the pci_dev that they return. You must eventually (possibly at module unload) decrement the reference count on these devices by calling pci_dev_put(). 3. Device Initialization Steps ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ As noted in introduction, most PCI drivers need the following steps for device initialization: request MMIO/IOP resources Enable the device enable/program PCI Config capabilities (e.g. MSI) set the DMA mask size (for both coherent and streaming DMA) allocate and initialize shared control data (pci_allocate_coherent()) register IRQ handler (request_irq()) "other" device specific initialization (start DMA/Processing engine) The driver can access PCI config space registers at any time. (Well, almost. When running BIST, config space can go away...but that will just result in a PCI Bus Master Abort). 3.1 Request MMIO/IOP resources ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Memory (MMIO), and I/O port addresses numbers should NOT be read directly from the PCI device config space. Use the values in the pci_dev structure as the PCI "bus address" might have been remapped to a "host physical" address by the kernel. See Documentation/IO-mapping.txt for how to access device registers or device memory. The device driver needs to call pci_request_region() to make sure no other device is already using the same address resource. The driver is expected to determine MMIO and IO Port resource availability _before_ calling pci_enable_device(). Conversely, drivers should call pci_release_region() AFTER calling pci_disable_device(). The idea is to prevent two devices colliding on the same address range. Generic flavors of pci_request_region() are request_mem_region() (for MMIO ranges) and request_region() (for IO Port ranges). Use these for address resources that are not described by "normal" PCI interfaces (e.g. BAR). 3.2 Enable the PCI device ~~~~~~~~~~~~~~~~~~~~~~~~~ Before touching any device registers, you need to enable the PCI device by calling pci_enable_device(). This will: o wakes up the device if it was in suspended state, o enable I/O and memory regions of the device, o allocates an IRQ (if BIOS did not), NOTE: pci_enable_device() can fail! Check the return value. pci_set_master() will enable DMA by setting the bus master bit in PCI_COMMAND register. It also fixes the latency timer value if it's set to something bogus by the BIOS. If the PCI device can use the PCI Memory-Write-Invalidate transaction, call pci_set_mwi(). This enables the PCI_COMMAND bit for Mem-Wr-Inval and also ensures that the cache line size register is set correctly. Check the return value of pci_set_mwi() as not all architectures or chipsets may support Memory-Write-Invalidate. 3.2 Set the DMA mask size ~~~~~~~~~~~~~~~~~~~~~~~~~ [ If anything below doesn't make sense, it means you need to read Documentation/DMA-API.txt. This section is just a reminder that drivers need to indicate DMA capabilities of the device and is not an authoritative source for DMA interfaces. ] While all drivers should explictly indicate the DMA capability (e.g. 32 or 64 bit) of the PCI bus master, devices with more than 32-bit bus master capability for streaming data need the driver to "register" this capability by calling pci_set_dma_mask() with appropriate parameters. In general this allows more efficient DMA on systems where System RAM exists above 4G _physical_ address. Drivers for all PCI-X and PCIe compliant devices must call pci_set_dma_mask(). All PCI-X and PCIe compliant devices are 64-bit DMA devices. Similarly, drivers must also "register" this capability if the device can directly address "consistent" in System RAM above 4G physical address. Drivers for PCI-X and PCIe compliant devices should also call pci_set_consistent_dma_mask(). 3.3 Setup shared control data ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Once the DMA masks are set, the driver can allocate "consistent" (aka shared) memory. See Documentation/DMA-API.txt for a full description of the DMA APIs. This section is just a reminder that it needs to be done before enabling DMA on the device. 3.4 initialize device registers ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Some drivers will need specific "capability" fields programmed or other "vendor specific" register initialized or reset. E.g. clearing pending interrupts. 3.5 register IRQ handler ~~~~~~~~~~~~~~~~~~~~~~~~ While calling request_irq() is the the last step describe here, this is often just another intermediate step to initializing a device. This step can often be deferred until the device is opened for use. All interrupt handlers should be registered with IRQF_SHARED and use the devid to map IRQs to devices (remember that all PCI interrupts are shared). request_irq() will associate a interrupt handler and device handle with an interrupt number. Historically interrupt numbers represent IRQ lines which run from the PCI device to the Interrupt controller. With MSI and MSI-X (more below) the interrupt number is a CPU "vector". request_irq() also enables the interrupt. Make sure the device is quiesced and does not have any interrupts pending before registering the interrupt handler. MSI and MSI-X are PCI capabilities. Both are "Message Signaled Interrupts" which deliver interrupts to the CPU via a DMA write to a Local APIC. The fundemental difference between MSI and MSI-X are how multiple "vectors" get allocated. MSI requires contiguous blocks of vectors while MSI-X can allocate several individual ones. MSI capability can be enabled by calling pci_enable_msi() or pci_enable_msix() before calling request_irq(). This causes the PCI support to program CPU vector data into the PCI device capability registers. If your PCI device supports both, try to enable MSI-X first. Only one can be enabled at a time. Many architectures, chipsets, or BIOSs do NOT support MSI or MSI-X and the call to pci_enable_msi/msix will fail. This is important to note since many drivers have two (or more) interrupt handlers: one for MSI/MSI-X and another for IRQs. They choose which handler to register with request_irq() based on the return value from pci_enable_msi/msix(). There are (at least) two really good reasons for using MSI: 1) MSI is an exclusive interrupt vector by definition. This means the interrupt handler doesn't have to verify that it's device caused the interrupt. 2) MSI avoids DMA/IRQ race conditions. DMA to host memory is guaranteed to be visible to the host CPU(s) when the MSI is delivered. This is important for both data coherency and avoiding stale control data. This guarantee allows the driver to omit MMIO reads to flush the DMA stream. See drivers/infiniband/hw/mthca/ or drivers/net/tg3.c for examples of MSI/MSI-X usage. 4 PCI device shutdown ~~~~~~~~~~~~~~~~~~~~~~ When a PCI device driver is being unloaded, most of the follow steps need to be performed: disable the device from generating IRQs release the IRQ (free_irq()) stop all DMA activity release DMA buffers (both streaming and consistent) unregister from other subsystems (e.g. scsi or netdev) Disable device from responding to MMIO/IO Port addresses release MMIO/IO Port resource(s) 4.1 Stop IRQs on the device ~~~~~~~~~~~~~~~~~~~~~~~~~~~ How to do this is chip/device specific. If it's not done, it opens the possibility of a "screaming interrupt" if (and only if) the IRQ is shared with another device. When the shared IRQ handler is "unhoooked", the remaining devices using the same IIRQ still need the IRQ enabled. Thus if the "unhooked" device asserts IRQ signal, the system wil respond assuming it was one of the remaining devices asserted the IRQ line. Since none of the other devices will handle the IRQ, the system will "hang" until it decides the IRQ isn't going to get handled and masks the IRQ (100,000 iterations later). Once the shared IRQ is masked, the remaining devices will stop functioning properly. Not a nice situation. This is another reason to use MSI or MSI-X if it's available. MSI and MSI-X are defined to be exclusive interrupts and thus are not susceptible to the "screaming interrupt" problem. 4.2 release the IRQ ~~~~~~~~~~~~~~~~~~~ Once the device is quiesced (no more IRQs), one can call free_irq(). This function will return control once any pending IRQs are handled, "unhook" the drivers IRQ handler from that IRQ, and finally release the IRQ if no one else is using it. 4.3 stop all DMA activity ~~~~~~~~~~~~~~~~~~~~~~~~~ It's extremely important to stop all DMA operations BEFORE attempting to deallocate resources. Failure to do so can result in memory corruption, hangs, and on some architectures with IOMMUs, hard crash. Stopping DMA after stopping the IRQs helps avoid races where the IRQ handler might restart DMA engines. While this step sounds obvious and trivial, several "mature" drivers didn't get this step right in the past. 4.4 release DMA buffers ~~~~~~~~~~~~~~~~~~~~~~~ Once DMA is stopped, clean up streaming DMA first. ie. unmap data buffers and return buffers to "upstream" owners if there is one. Then clean up "consistent" buffers which contain the control data. See Documentation/DMA-API.txt for details on unmapping interfaces. 4.5 unregister from other subsystems ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Most low level PCI device drivers support some other subsystem like USB, ALSA, SCSI, NetDev, Infiniband, etc. Make sure your driver isn't losing resources from that other subsystem. If this happens, typically the symptom is an Oops (panic) when the subsystem attempts to call into a driver that has been unloaded. 4.6 Disable device from responding to MMIO/IO Port addresses ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This is just a call to pci_disable_device(). This is the symetric opposite of pci_enable_device(). Do not do anything with the device after calling pci_disable_device(). 4.7 release MMIO/IO Port resource(s) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Call pci_release_region() to mark the MMIO or IO Port range as available. Failure to do so usually results in the inability to reload the driver. 5. How to access PCI config space ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ You can use pci_(read|write)_config_(byte|word|dword) to access the config space of a device represented by struct pci_dev *. All these functions return 0 when successful or an error code (PCIBIOS_...) which can be translated to a text string by pcibios_strerror. Most drivers expect that accesses to valid PCI devices don't fail. If you don't have a struct pci_dev available, you can call pci_bus_(read|write)_config_(byte|word|dword) to access a given device and function on that bus. If you access fields in the standard portion of the config header, please use symbolic names of locations and bits declared in . If you need to access Extended PCI Capability registers, just call pci_find_capability() for the particular capability and it will find the corresponding register block for you. 6. Other interesting functions ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ pci_find_slot() Find pci_dev corresponding to given bus and slot numbers. pci_set_power_state() Set PCI Power Management state (0=D0 ... 3=D3) pci_find_capability() Find specified capability in device's capability list. pci_module_init() Inline helper function for ensuring correct pci_driver initialization and error handling. pci_resource_start() Returns bus start address for a given PCI region pci_resource_end() Returns bus end address for a given PCI region pci_resource_len() Returns the byte length of a PCI region pci_set_drvdata() Set private driver data pointer for a pci_dev pci_get_drvdata() Return private driver data pointer for a pci_dev pci_set_mwi() Enable Memory-Write-Invalidate transactions. pci_clear_mwi() Disable Memory-Write-Invalidate transactions. 7. Miscellaneous hints ~~~~~~~~~~~~~~~~~~~~~~ When displaying PCI device names to the user (for example when a driver wants to tell the user what card has it found), please use pci_name(pci_dev) for this purpose. Always refer to the PCI devices by a pointer to the pci_dev structure. All PCI layer functions use this identification and it's the only reasonable one. Don't use bus/slot/function numbers except for very special purposes -- on systems with multiple primary buses their semantics can be pretty complex. Don't try to turn on Fast Back to Back writes in your driver. All devices on the bus need to be capable of doing it, so this is something which needs to be handled by platform and generic code, not individual drivers. 8. Vendor and device identifications ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Please do not add new device ids to include/linux/pci_ids.h. Use PCI_VENDOR_ID_xxx for vendors, and a hex constant for device ids. PCI_VENDOR_ID_xxx constants are re-used. The device ids are arbitrary hex numbers (vendor controlled) and normally used only in a single location, the pci_device_id table. Please DO submit new vendor/device ids to pciids.sourceforge.net project. 9. Obsolete functions ~~~~~~~~~~~~~~~~~~~~~ There are several functions which you might come across when trying to port an old driver to the new PCI interface. They are no longer present in the kernel as they aren't compatible with hotplug or PCI domains or having sane locking. pci_find_device() Superseded by pci_get_device() pci_find_subsys() Superseded by pci_get_subsys() pci_find_slot() Superseded by pci_get_slot() The alternative is the traditional PCI device driver that walks PCI device lists. This is still possible but discouraged.