1Accessing PCI device resources through sysfs 2-------------------------------------------- 3 4sysfs, usually mounted at /sys, provides access to PCI resources on platforms 5that support it. For example, a given bus might look like this: 6 7 /sys/devices/pci0000:17 8 |-- 0000:17:00.0 9 | |-- class 10 | |-- config 11 | |-- device 12 | |-- enable 13 | |-- irq 14 | |-- local_cpus 15 | |-- remove 16 | |-- resource 17 | |-- resource0 18 | |-- resource1 19 | |-- resource2 20 | |-- rom 21 | |-- subsystem_device 22 | |-- subsystem_vendor 23 | `-- vendor 24 `-- ... 25 26The topmost element describes the PCI domain and bus number. In this case, 27the domain number is 0000 and the bus number is 17 (both values are in hex). 28This bus contains a single function device in slot 0. The domain and bus 29numbers are reproduced for convenience. Under the device directory are several 30files, each with their own function. 31 32 file function 33 ---- -------- 34 class PCI class (ascii, ro) 35 config PCI config space (binary, rw) 36 device PCI device (ascii, ro) 37 enable Whether the device is enabled (ascii, rw) 38 irq IRQ number (ascii, ro) 39 local_cpus nearby CPU mask (cpumask, ro) 40 remove remove device from kernel's list (ascii, wo) 41 resource PCI resource host addresses (ascii, ro) 42 resource0..N PCI resource N, if present (binary, mmap, rw[1]) 43 resource0_wc..N_wc PCI WC map resource N, if prefetchable (binary, mmap) 44 rom PCI ROM resource, if present (binary, ro) 45 subsystem_device PCI subsystem device (ascii, ro) 46 subsystem_vendor PCI subsystem vendor (ascii, ro) 47 vendor PCI vendor (ascii, ro) 48 49 ro - read only file 50 rw - file is readable and writable 51 wo - write only file 52 mmap - file is mmapable 53 ascii - file contains ascii text 54 binary - file contains binary data 55 cpumask - file contains a cpumask type 56 57[1] rw for RESOURCE_IO (I/O port) regions only 58 59The read only files are informational, writes to them will be ignored, with 60the exception of the 'rom' file. Writable files can be used to perform 61actions on the device (e.g. changing config space, detaching a device). 62mmapable files are available via an mmap of the file at offset 0 and can be 63used to do actual device programming from userspace. Note that some platforms 64don't support mmapping of certain resources, so be sure to check the return 65value from any attempted mmap. The most notable of these are I/O port 66resources, which also provide read/write access. 67 68The 'enable' file provides a counter that indicates how many times the device 69has been enabled. If the 'enable' file currently returns '4', and a '1' is 70echoed into it, it will then return '5'. Echoing a '0' into it will decrease 71the count. Even when it returns to 0, though, some of the initialisation 72may not be reversed. 73 74The 'rom' file is special in that it provides read-only access to the device's 75ROM file, if available. It's disabled by default, however, so applications 76should write the string "1" to the file to enable it before attempting a read 77call, and disable it following the access by writing "0" to the file. Note 78that the device must be enabled for a rom read to return data successfully. 79In the event a driver is not bound to the device, it can be enabled using the 80'enable' file, documented above. 81 82The 'remove' file is used to remove the PCI device, by writing a non-zero 83integer to the file. This does not involve any kind of hot-plug functionality, 84e.g. powering off the device. The device is removed from the kernel's list of 85PCI devices, the sysfs directory for it is removed, and the device will be 86removed from any drivers attached to it. Removal of PCI root buses is 87disallowed. 88 89Accessing legacy resources through sysfs 90---------------------------------------- 91 92Legacy I/O port and ISA memory resources are also provided in sysfs if the 93underlying platform supports them. They're located in the PCI class hierarchy, 94e.g. 95 96 /sys/class/pci_bus/0000:17/ 97 |-- bridge -> ../../../devices/pci0000:17 98 |-- cpuaffinity 99 |-- legacy_io 100 `-- legacy_mem 101 102The legacy_io file is a read/write file that can be used by applications to 103do legacy port I/O. The application should open the file, seek to the desired 104port (e.g. 0x3e8) and do a read or a write of 1, 2 or 4 bytes. The legacy_mem 105file should be mmapped with an offset corresponding to the memory offset 106desired, e.g. 0xa0000 for the VGA frame buffer. The application can then 107simply dereference the returned pointer (after checking for errors of course) 108to access legacy memory space. 109 110Supporting PCI access on new platforms 111-------------------------------------- 112 113In order to support PCI resource mapping as described above, Linux platform 114code must define HAVE_PCI_MMAP and provide a pci_mmap_page_range function. 115Platforms are free to only support subsets of the mmap functionality, but 116useful return codes should be provided. 117 118Legacy resources are protected by the HAVE_PCI_LEGACY define. Platforms 119wishing to support legacy functionality should define it and provide 120pci_legacy_read, pci_legacy_write and pci_mmap_legacy_page_range functions. 121