1 What: /sys/bus/pci/drivers/.../bind
3 Contact: linux-pci@vger.kernel.org
5 Writing a device location to this file will cause
6 the driver to attempt to bind to the device found at
7 this location. This is useful for overriding default
8 bindings. The format for the location is: DDDD:BB:DD.F.
9 That is Domain:Bus:Device.Function and is the same as
10 found in /sys/bus/pci/devices/. For example::
12 # echo 0000:00:19.0 > /sys/bus/pci/drivers/foo/bind
14 (Note: kernels before 2.6.28 may require echo -n).
16 What: /sys/bus/pci/drivers/.../unbind
18 Contact: linux-pci@vger.kernel.org
20 Writing a device location to this file will cause the
21 driver to attempt to unbind from the device found at
22 this location. This may be useful when overriding default
23 bindings. The format for the location is: DDDD:BB:DD.F.
24 That is Domain:Bus:Device.Function and is the same as
25 found in /sys/bus/pci/devices/. For example::
27 # echo 0000:00:19.0 > /sys/bus/pci/drivers/foo/unbind
29 (Note: kernels before 2.6.28 may require echo -n).
31 What: /sys/bus/pci/drivers/.../new_id
33 Contact: linux-pci@vger.kernel.org
35 Writing a device ID to this file will attempt to
36 dynamically add a new device ID to a PCI device driver.
37 This may allow the driver to support more hardware than
38 was included in the driver's static device ID support
39 table at compile time. The format for the device ID is:
40 VVVV DDDD SVVV SDDD CCCC MMMM PPPP. That is Vendor ID,
41 Device ID, Subsystem Vendor ID, Subsystem Device ID,
42 Class, Class Mask, and Private Driver Data. The Vendor ID
43 and Device ID fields are required, the rest are optional.
44 Upon successfully adding an ID, the driver will probe
45 for the device and attempt to bind to it. For example::
47 # echo "8086 10f5" > /sys/bus/pci/drivers/foo/new_id
49 What: /sys/bus/pci/drivers/.../remove_id
51 Contact: Chris Wright <chrisw@sous-sol.org>
53 Writing a device ID to this file will remove an ID
54 that was dynamically added via the new_id sysfs entry.
55 The format for the device ID is:
56 VVVV DDDD SVVV SDDD CCCC MMMM. That is Vendor ID, Device
57 ID, Subsystem Vendor ID, Subsystem Device ID, Class,
58 and Class Mask. The Vendor ID and Device ID fields are
59 required, the rest are optional. After successfully
60 removing an ID, the driver will no longer support the
61 device. This is useful to ensure auto probing won't
62 match the driver to the device. For example::
64 # echo "8086 10f5" > /sys/bus/pci/drivers/foo/remove_id
66 What: /sys/bus/pci/rescan
68 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
70 Writing a non-zero value to this attribute will
71 force a rescan of all PCI buses in the system, and
72 re-discover previously removed devices.
74 What: /sys/bus/pci/devices/.../msi_bus
76 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
78 Writing a zero value to this attribute disallows MSI and
79 MSI-X for any future drivers of the device. If the device
80 is a bridge, MSI and MSI-X will be disallowed for future
81 drivers of all child devices under the bridge. Drivers
82 must be reloaded for the new setting to take effect.
84 What: /sys/bus/pci/devices/.../msi_irqs/
86 Contact: Neil Horman <nhorman@tuxdriver.com>
88 The /sys/devices/.../msi_irqs directory contains a variable set
89 of files, with each file being named after a corresponding msi
90 irq vector allocated to that device.
92 What: /sys/bus/pci/devices/.../msi_irqs/<N>
94 Contact: Neil Horman <nhorman@tuxdriver.com>
96 This attribute indicates the mode that the irq vector named by
97 the file is in (msi vs. msix)
99 What: /sys/bus/pci/devices/.../remove
101 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
103 Writing a non-zero value to this attribute will
104 hot-remove the PCI device and any of its children.
106 What: /sys/bus/pci/devices/.../pci_bus/.../rescan
108 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
110 Writing a non-zero value to this attribute will
111 force a rescan of the bus and all child buses,
112 and re-discover devices removed earlier from this
113 part of the device tree.
115 What: /sys/bus/pci/devices/.../rescan
117 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
119 Writing a non-zero value to this attribute will
120 force a rescan of the device's parent bus and all
121 child buses, and re-discover devices removed earlier
122 from this part of the device tree.
124 What: /sys/bus/pci/devices/.../reset_method
126 Contact: Amey Narkhede <ameynarkhede03@gmail.com>
128 Some devices allow an individual function to be reset
129 without affecting other functions in the same slot.
131 For devices that have this support, a file named
132 reset_method is present in sysfs. Reading this file
133 gives names of the supported and enabled reset methods and
134 their ordering. Writing a space-separated list of names of
135 reset methods sets the reset methods and ordering to be
136 used when resetting the device. Writing an empty string
137 disables the ability to reset the device. Writing
138 "default" enables all supported reset methods in the
141 What: /sys/bus/pci/devices/.../reset
143 Contact: Michael S. Tsirkin <mst@redhat.com>
145 Some devices allow an individual function to be reset
146 without affecting other functions in the same device.
147 For devices that have this support, a file named reset
148 will be present in sysfs. Writing 1 to this file
151 What: /sys/bus/pci/devices/.../vpd
153 Contact: Ben Hutchings <bwh@kernel.org>
155 A file named vpd in a device directory will be a
156 binary file containing the Vital Product Data for the
157 device. It should follow the VPD format defined in
158 PCI Specification 2.1 or 2.2, but users should consider
159 that some devices may have incorrectly formatted data.
160 If the underlying VPD has a writable section then the
161 corresponding section of this file will be writable.
163 What: /sys/bus/pci/devices/.../virtfnN
165 Contact: Yu Zhao <yu.zhao@intel.com>
167 This symbolic link appears when hardware supports the SR-IOV
168 capability and the Physical Function driver has enabled it.
169 The symbolic link points to the PCI device sysfs entry of the
170 Virtual Function whose index is N (0...MaxVFs-1).
172 What: /sys/bus/pci/devices/.../dep_link
174 Contact: Yu Zhao <yu.zhao@intel.com>
176 This symbolic link appears when hardware supports the SR-IOV
177 capability and the Physical Function driver has enabled it,
178 and this device has vendor specific dependencies with others.
179 The symbolic link points to the PCI device sysfs entry of
180 Physical Function this device depends on.
182 What: /sys/bus/pci/devices/.../physfn
184 Contact: Yu Zhao <yu.zhao@intel.com>
186 This symbolic link appears when a device is a Virtual Function.
187 The symbolic link points to the PCI device sysfs entry of the
188 Physical Function this device associates with.
190 What: /sys/bus/pci/slots/.../module
192 Contact: linux-pci@vger.kernel.org
194 This symbolic link points to the PCI hotplug controller driver
195 module that manages the hotplug slot.
197 What: /sys/bus/pci/devices/.../label
199 Contact: Narendra K <narendra_k@dell.com>, linux-bugs@dell.com
201 Reading this attribute will provide the firmware
202 given name (SMBIOS type 41 string or ACPI _DSM string) of
203 the PCI device. The attribute will be created only
204 if the firmware has given a name to the PCI device.
205 ACPI _DSM string name will be given priority if the
206 system firmware provides SMBIOS type 41 string also.
208 Userspace applications interested in knowing the
209 firmware assigned name of the PCI device.
211 What: /sys/bus/pci/devices/.../index
213 Contact: Narendra K <narendra_k@dell.com>, linux-bugs@dell.com
215 Reading this attribute will provide the firmware given instance
216 number of the PCI device. Depending on the platform this can
217 be for example the SMBIOS type 41 device type instance or the
218 user-defined ID (UID) on s390. The attribute will be created
219 only if the firmware has given an instance number to the PCI
220 device and that number is guaranteed to uniquely identify the
221 device in the system.
223 Userspace applications interested in knowing the
224 firmware assigned device type instance of the PCI
225 device that can help in understanding the firmware
226 intended order of the PCI device.
228 What: /sys/bus/pci/devices/.../acpi_index
230 Contact: Narendra K <narendra_k@dell.com>, linux-bugs@dell.com
232 Reading this attribute will provide the firmware
233 given instance (ACPI _DSM instance number) of the PCI device.
234 The attribute will be created only if the firmware has given
235 an instance number to the PCI device. ACPI _DSM instance number
236 will be given priority if the system firmware provides SMBIOS
237 type 41 device type instance also.
239 Userspace applications interested in knowing the
240 firmware assigned instance number of the PCI
241 device that can help in understanding the firmware
242 intended order of the PCI device.
244 What: /sys/bus/pci/devices/.../d3cold_allowed
246 Contact: Huang Ying <ying.huang@intel.com>
248 d3cold_allowed is bit to control whether the corresponding PCI
249 device can be put into D3Cold state. If it is cleared, the
250 device will never be put into D3Cold state. If it is set, the
251 device may be put into D3Cold state if other requirements are
252 satisfied too. Reading this attribute will show the current
253 value of d3cold_allowed bit. Writing this attribute will set
254 the value of d3cold_allowed bit.
256 What: /sys/bus/pci/devices/.../sriov_totalvfs
258 Contact: Donald Dutile <ddutile@redhat.com>
260 This file appears when a physical PCIe device supports SR-IOV.
261 Userspace applications can read this file to determine the
262 maximum number of Virtual Functions (VFs) a PCIe physical
263 function (PF) can support. Typically, this is the value reported
264 in the PF's SR-IOV extended capability structure's TotalVFs
265 element. Drivers have the ability at probe time to reduce the
266 value read from this file via the pci_sriov_set_totalvfs()
269 What: /sys/bus/pci/devices/.../sriov_numvfs
271 Contact: Donald Dutile <ddutile@redhat.com>
273 This file appears when a physical PCIe device supports SR-IOV.
274 Userspace applications can read and write to this file to
275 determine and control the enablement or disablement of Virtual
276 Functions (VFs) on the physical function (PF). A read of this
277 file will return the number of VFs that are enabled on this PF.
278 A number written to this file will enable the specified
279 number of VFs. A userspace application would typically read the
280 file and check that the value is zero, and then write the number
281 of VFs that should be enabled on the PF; the value written
282 should be less than or equal to the value in the sriov_totalvfs
283 file. A userspace application wanting to disable the VFs would
284 write a zero to this file. The core ensures that valid values
285 are written to this file, and returns errors when values are not
286 valid. For example, writing a 2 to this file when sriov_numvfs
287 is not 0 and not 2 already will return an error. Writing a 10
288 when the value of sriov_totalvfs is 8 will return an error.
290 What: /sys/bus/pci/devices/.../driver_override
292 Contact: Alex Williamson <alex.williamson@redhat.com>
294 This file allows the driver for a device to be specified which
295 will override standard static and dynamic ID matching. When
296 specified, only a driver with a name matching the value written
297 to driver_override will have an opportunity to bind to the
298 device. The override is specified by writing a string to the
299 driver_override file (echo pci-stub > driver_override) and
300 may be cleared with an empty string (echo > driver_override).
301 This returns the device to standard matching rules binding.
302 Writing to driver_override does not automatically unbind the
303 device from its current driver or make any attempt to
304 automatically load the specified driver. If no driver with a
305 matching name is currently loaded in the kernel, the device
306 will not bind to any driver. This also allows devices to
307 opt-out of driver binding using a driver_override name such as
308 "none". Only a single driver may be specified in the override,
309 there is no support for parsing delimiters.
311 What: /sys/bus/pci/devices/.../numa_node
313 Contact: Prarit Bhargava <prarit@redhat.com>
315 This file contains the NUMA node to which the PCI device is
316 attached, or -1 if the node is unknown. The initial value
317 comes from an ACPI _PXM method or a similar firmware
318 source. If that is missing or incorrect, this file can be
319 written to override the node. In that case, please report
320 a firmware bug to the system vendor. Writing to this file
321 taints the kernel with TAINT_FIRMWARE_WORKAROUND, which
322 reduces the supportability of your system.
324 What: /sys/bus/pci/devices/.../revision
326 Contact: Emil Velikov <emil.l.velikov@gmail.com>
328 This file contains the revision field of the PCI device.
329 The value comes from device config space. The file is read only.
331 What: /sys/bus/pci/devices/.../sriov_drivers_autoprobe
333 Contact: Bodong Wang<bodong@mellanox.com>
335 This file is associated with the PF of a device that
336 supports SR-IOV. It determines whether newly-enabled VFs
337 are immediately bound to a driver. It initially contains
338 1, which means the kernel automatically binds VFs to a
339 compatible driver immediately after they are enabled. If
340 an application writes 0 to the file before enabling VFs,
341 the kernel will not bind VFs to a driver.
343 A typical use case is to write 0 to this file, then enable
344 VFs, then assign the newly-created VFs to virtual machines.
345 Note that changing this file does not affect already-
346 enabled VFs. In this scenario, the user must first disable
347 the VFs, write 0 to sriov_drivers_autoprobe, then re-enable
350 This is similar to /sys/bus/pci/drivers_autoprobe, but
351 affects only the VFs associated with a specific PF.
353 What: /sys/bus/pci/devices/.../p2pmem/size
355 Contact: Logan Gunthorpe <logang@deltatee.com>
357 If the device has any Peer-to-Peer memory registered, this
358 file contains the total amount of memory that the device
359 provides (in decimal).
361 What: /sys/bus/pci/devices/.../p2pmem/available
363 Contact: Logan Gunthorpe <logang@deltatee.com>
365 If the device has any Peer-to-Peer memory registered, this
366 file contains the amount of memory that has not been
367 allocated (in decimal).
369 What: /sys/bus/pci/devices/.../p2pmem/published
371 Contact: Logan Gunthorpe <logang@deltatee.com>
373 If the device has any Peer-to-Peer memory registered, this
374 file contains a '1' if the memory has been published for
375 use outside the driver that owns the device.
377 What: /sys/bus/pci/devices/.../link/clkpm
378 /sys/bus/pci/devices/.../link/l0s_aspm
379 /sys/bus/pci/devices/.../link/l1_aspm
380 /sys/bus/pci/devices/.../link/l1_1_aspm
381 /sys/bus/pci/devices/.../link/l1_2_aspm
382 /sys/bus/pci/devices/.../link/l1_1_pcipm
383 /sys/bus/pci/devices/.../link/l1_2_pcipm
385 Contact: Heiner Kallweit <hkallweit1@gmail.com>
386 Description: If ASPM is supported for an endpoint, these files can be
387 used to disable or enable the individual power management
388 states. Write y/1/on to enable, n/0/off to disable.
390 What: /sys/bus/pci/devices/.../power_state
392 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
394 This file contains the current PCI power state of the device.
395 The value comes from the PCI kernel device state and can be one
396 of: "unknown", "error", "D0", D1", "D2", "D3hot", "D3cold".
397 The file is read only.
399 What: /sys/bus/pci/devices/.../sriov_vf_total_msix
401 Contact: Leon Romanovsky <leonro@nvidia.com>
403 This file is associated with a SR-IOV physical function (PF).
404 It contains the total number of MSI-X vectors available for
405 assignment to all virtual functions (VFs) associated with PF.
406 The value will be zero if the device doesn't support this
407 functionality. For supported devices, the value will be
408 constant and won't be changed after MSI-X vectors assignment.
410 What: /sys/bus/pci/devices/.../sriov_vf_msix_count
412 Contact: Leon Romanovsky <leonro@nvidia.com>
414 This file is associated with a SR-IOV virtual function (VF).
415 It allows configuration of the number of MSI-X vectors for
416 the VF. This allows devices that have a global pool of MSI-X
417 vectors to optimally divide them between VFs based on VF usage.
419 The values accepted are:
420 * > 0 - this number will be reported as the Table Size in the
421 VF's MSI-X capability
423 * = 0 - will reset to the device default value
425 The file is writable if the PF is bound to a driver that
426 implements ->sriov_set_msix_vec_count().