Differences
This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision | ||
soc:2008:mdeck:notes:gpxe_driver_api [2008/05/28 14:18] mdeck |
soc:2008:mdeck:notes:gpxe_driver_api [2009/11/03 08:58] (current) meteger |
||
---|---|---|---|
Line 1: | Line 1: | ||
====== gPXE Driver API Documentation ====== | ====== gPXE Driver API Documentation ====== | ||
- | :!: Under Construction | + | A gPXE network driver may incorporate elements of the following: |
- | ===== Driver Exported Functions ===== | + | * [[#gpxe_pci_device_driver_api|gPXE PCI Device Driver API]] |
- | The following functions should be defined in any network driver: | + | * [[#gpxe_network_driver_api|gPXE Network Driver API]] |
+ | * [[#non-volatile_storage_api|Non-Volatile Storage API]] | ||
+ | |||
+ | Note the [[:dev:netdriverapi|previous driver model]] of Etherboot is deprecated. | ||
+ | Existing Etherboot PCI drivers are temporarily supported via the compatibility layer in src/drivers/net/legacy.c | ||
+ | Drivers currently conforming to the gPXE Network Driver API are: | ||
+ | * 3c90x | ||
+ | * ath5k | ||
+ | * atl1e | ||
+ | * b44 | ||
+ | * e1000 | ||
+ | * etherfabric | ||
+ | * mtnic | ||
+ | * natsemi | ||
+ | * phantom | ||
+ | * pnic | ||
+ | * r8169 | ||
+ | * rtl8139 | ||
+ | * rtl818x | ||
+ | * sis190 | ||
+ | * sky2 | ||
+ | |||
+ | |||
+ | ===== gPXE PCI Device Driver API ===== | ||
+ | A PCI driver provides its API routines to gPXE via a ''struct pci_driver''. For example, in natsemi.c: | ||
+ | <code c> | ||
+ | struct pci_driver natsemi_driver __pci_driver = { | ||
+ | .ids = natsemi_nics, | ||
+ | .id_count = (sizeof (natsemi_nics) / sizeof (natsemi_nics[0])), | ||
+ | .probe = natsemi_probe, | ||
+ | .remove = natsemi_remove, | ||
+ | }; | ||
+ | </code> | ||
+ | The ''.ids'' and ''.id_count'' members list the vendor & device IDs of supported devices. | ||
+ | The functions natsemi_probe & natsemi_remove are driver implementations of the required PCI device driver API functions: | ||
* ''[[#probe|static int probe ( struct pci_device* , const struct pci_device_id* )]]'' | * ''[[#probe|static int probe ( struct pci_device* , const struct pci_device_id* )]]'' | ||
* ''[[#remove|static void remove ( struct pci_device* )]]'' | * ''[[#remove|static void remove ( struct pci_device* )]]'' | ||
- | * ''[[#reset|static void reset ( struct net_device* )]]'' | ||
- | * ''[[#open|static int open ( struct net_device* )]]'' | ||
- | * ''[[#close|static void close ( struct net_device* )]]'' | ||
- | * ''[[#transmit|static int transmit ( struct net_device*, struct io_buffer* )]]'' | ||
- | * ''[[#poll|static void poll ( struct net_device* )]]'' | ||
- | * ''[[#irq|static void irq ( struct net_device*, int enable )]]'' | ||
- | In addition, your driver should define a ''struct pci_driver'' where | ||
- | >''.ids = '' an array of ''struct pci_device_id''s | ||
- | >''.id_count = '' array count | ||
- | >''.probe = '' the driver ''probe'' function | ||
- | >''.remove = '' the driver ''remove'' function | ||
==== probe ==== | ==== probe ==== | ||
''static int probe ( struct pci_device* , const struct pci_device_id* )''\\ | ''static int probe ( struct pci_device* , const struct pci_device_id* )''\\ | ||
- | This function is called first to initialize the card. Here a typical driver will: | + | This function is called [[:soc:2008:mdeck:notes:initialization|first]] to initialize the card. Here a typical network driver will: |
- Allocate a ''struct net_device'' with associated private data using ''alloc_etherdev()''. | - Allocate a ''struct net_device'' with associated private data using ''alloc_etherdev()''. | ||
- Associate the driver functions with the ''net_device'' via ''netdev_init()''. | - Associate the driver functions with the ''net_device'' via ''netdev_init()''. | ||
Line 28: | Line 51: | ||
- Initialize [[#EEPROM|EEPROM]]. | - Initialize [[#EEPROM|EEPROM]]. | ||
- Read the MAC address from EEPROM. | - Read the MAC address from EEPROM. | ||
- | - Mark the ''net_device'' as having a link up with ''netdev_link_up()'', as we don't yet handle the link state. | + | - Check the link state and report ''netdev_link_up()'' if connected. Many drivers don't yet handle the link state and simply assume the link is up. |
- Name the device and add it to the list of network devices via ''register_netdev()''. | - Name the device and add it to the list of network devices via ''register_netdev()''. | ||
- Possibly setup a non-volatile stored options block with ''nvo_init()'' & ''register_nvo()''. | - Possibly setup a non-volatile stored options block with ''nvo_init()'' & ''register_nvo()''. | ||
Line 42: | Line 65: | ||
- Decrement reference count of ''net_device'' with ''netdev_put()''. | - Decrement reference count of ''net_device'' with ''netdev_put()''. | ||
- | ==== reset ==== | ||
- | ''static void reset ( struct net_device* )''\\ | ||
- | This function issues a hardware reset and waits for completion. A typical driver might: | ||
- | - Mask off interrupts. | ||
- | - Clear pending transmits. | ||
- | - Trigger hardware reset. | ||
- | - Wait for completion. | ||
- | ==== open ==== | + | |
- | ''static int open ( struct net_device* )''\\ | + | ===== gPXE Network Driver API ===== |
- | In this function, a driver might: | + | A network driver in gPXE provides its API routines to the system via a ''struct net_device_operations'' during the initial ''probe()'' call |
- | - Program MAC address to device. | + | (see [[:soc:2008:mdeck:notes:initialization|initialization of a network driver]]). For example, in natsemi.c: |
- | - Setup TX & RX rings. | + | <code c> |
- | - Perform other configuration (e.g. filters, bursts, interrupts). | + | static struct net_device_operations natsemi_operations = { |
- | - Enable RX and TX. | + | .open = natsemi_open, |
+ | .close = natsemi_close, | ||
+ | .transmit = natsemi_transmit, | ||
+ | .poll = natsemi_poll, | ||
+ | .irq = natsemi_irq, | ||
+ | }; | ||
+ | </code> | ||
+ | Here, natsemi_open/close/etc are driver implementations of the required network driver API functions: | ||
+ | * ''[[#close|static void close ( struct net_device* )]]'' | ||
+ | * ''[[#open|static int open ( struct net_device* )]]'' | ||
+ | * ''[[#transmit|static int transmit ( struct net_device*, struct io_buffer* )]]'' | ||
+ | * ''[[#poll|static void poll ( struct net_device* )]]'' | ||
+ | * ''[[#irq|static void irq ( struct net_device*, int enable )]]'' | ||
==== close ==== | ==== close ==== | ||
''static void close ( struct net_device* )''\\ | ''static void close ( struct net_device* )''\\ | ||
- | In this function, a typical driver might: | + | This function is called if the device is open in ''autoboot()'' during [[:soc:2008:mdeck:notes:initialization|initialization]], after a failed or successful attempt to boot the network device. In this routine, a typical driver might: |
- Acknowledge interrupts. | - Acknowledge interrupts. | ||
- Disable irq, receives. | - Disable irq, receives. | ||
- Reset the device. | - Reset the device. | ||
- Free any used resources (e.g. rx/tx rings, dma buffers, etc). | - Free any used resources (e.g. rx/tx rings, dma buffers, etc). | ||
+ | |||
+ | ==== open ==== | ||
+ | ''static int open ( struct net_device* )''\\ | ||
+ | This function is first called in ''netboot()'' when attempting a device boot during [[:soc:2008:mdeck:notes:initialization|initialization]], after ''close()'' of any previous device attempt is called. A driver would: | ||
+ | - Program MAC address to device. | ||
+ | - Setup TX & RX rings. | ||
+ | - Perform other configuration (e.g. filters, bursts, interrupts). | ||
+ | - Enable RX and TX. | ||
==== transmit ==== | ==== transmit ==== | ||
''static int transmit ( struct net_device*, struct io_buffer* )''\\ | ''static int transmit ( struct net_device*, struct io_buffer* )''\\ | ||
- | In this function, a typical driver might: | + | A data transmission is actuated with this routine. A typical driver might: |
- Check for tx overflow. | - Check for tx overflow. | ||
- Save buffer pointer for later tx completion reference. | - Save buffer pointer for later tx completion reference. | ||
Line 77: | Line 113: | ||
==== poll ==== | ==== poll ==== | ||
''static void poll ( struct net_device* )''\\ | ''static void poll ( struct net_device* )''\\ | ||
- | In this function, a typical driver might: | + | This function is called periodically by the network stack to process tx completions and rx packets. A typical driver would: |
- Acknowledge interrupts. | - Acknowledge interrupts. | ||
- | - Feed tx completions to ''netdev_tx_complete()'' or ''netdev_tx_complete_err()''. | + | - Check hardware and feed tx completions to ''netdev_tx_complete()'' or ''netdev_tx_complete_err()''. |
- | - Add good received packets to receive queue with ''netdev_rx()'', or feed corrupted packets to ''netdev_rx_err()'' | + | - Add good received packets to receive queue with ''netdev_rx()'', or report corrupted packets to ''netdev_rx_err()'' |
+ | - Check link state occasionally, and report changes with ''netdev_link_up()'' or ''netdev_link_down()'' | ||
==== irq ==== | ==== irq ==== | ||
Line 86: | Line 123: | ||
In this function, a typical driver will: | In this function, a typical driver will: | ||
- Enable interrupts if the int parameter is non-zero | - Enable interrupts if the int parameter is non-zero | ||
+ | Note the //force interrupt// behavior from Etherboot is deprecated. | ||
+ | |||
+ | |||
- | ===== EEPROM ===== | + | ===== Non-Volatile Storage API ===== |
- | The EEPROM can be accessed via direct port access or using nvs functions. | + | The nvs API may be used to access non-volatile storage that conforms to a number of supported SPI variants. |
- | ==== Non-volatile storage (nvs) ==== | + | |
- | The nvs functions may be used to access non-volatile storage that conforms to a number of supported SPI protocols. | + | |
* To initialize nvs support: | * To initialize nvs support: | ||
- The read_bit() and write_bit() function pointers are stored in a ''struct spi_bit_basher''. | - The read_bit() and write_bit() function pointers are stored in a ''struct spi_bit_basher''. | ||
Line 97: | Line 135: | ||
- A ''struct spi_device'' is initialized (EEPROM-model dependent). | - A ''struct spi_device'' is initialized (EEPROM-model dependent). | ||
- Bus is copied: ''spidev.bus = &spibb.bus'' | - Bus is copied: ''spidev.bus = &spibb.bus'' | ||
- | - EITHER | + | - If non-volatile options will be utilized: |
+ | - Call ''nvo_init(&nvob, &spidev.nvs, nvof, ..)'' where ''nvof'' is an array of ''struct nvo_fragment''s that specify the usable regions. This will also initialize a settings block. | ||
+ | - Else, if non-volatile options will //not// be used: | ||
- Initialize a ''struct nvo_block'' with ''nvob.nvs = &spidev.nvs'' | - Initialize a ''struct nvo_block'' with ''nvob.nvs = &spidev.nvs'' | ||
- Assign usable regions via ''nvob.fragments = nvof'' where ''nvof'' is an array of ''struct nvo_fragment''s that specify the usable regions. | - Assign usable regions via ''nvob.fragments = nvof'' where ''nvof'' is an array of ''struct nvo_fragment''s that specify the usable regions. | ||
- | - OR | ||
- | - Call ''nvo_init(&nvob, &spidev.nvs, nvof, ..)'' where ''nvof'' is an array of ''struct nvo_fragment''s that specify the usable regions. | ||
* Use ''nvs_read()'' to perform a serial read @ a specific address. | * Use ''nvs_read()'' to perform a serial read @ a specific address. | ||
* Use ''nvs_write()'' to perform a serial write @ a specific address. | * Use ''nvs_write()'' to perform a serial write @ a specific address. |