1 What: /sys/bus/pci/drivers/.../bind 1 What: /sys/bus/pci/drivers/.../bind
2 What: /sys/devices/pciX/.../bind 2 What: /sys/devices/pciX/.../bind
3 Date: December 2003 3 Date: December 2003
4 Contact: linux-pci@vger.kernel.org 4 Contact: linux-pci@vger.kernel.org
5 Description: 5 Description:
6 Writing a device location to t 6 Writing a device location to this file will cause
7 the driver to attempt to bind 7 the driver to attempt to bind to the device found at
8 this location. This is useful 8 this location. This is useful for overriding default
9 bindings. The format for the 9 bindings. The format for the location is: DDDD:BB:DD.F.
10 That is Domain:Bus:Device.Func 10 That is Domain:Bus:Device.Function and is the same as
11 found in /sys/bus/pci/devices/ 11 found in /sys/bus/pci/devices/. For example::
12 12
13 # echo 0000:00:19.0 > /sys/b 13 # echo 0000:00:19.0 > /sys/bus/pci/drivers/foo/bind
14 14
15 (Note: kernels before 2.6.28 m 15 (Note: kernels before 2.6.28 may require echo -n).
16 16
17 What: /sys/bus/pci/drivers/.../unbin 17 What: /sys/bus/pci/drivers/.../unbind
18 What: /sys/devices/pciX/.../unbind 18 What: /sys/devices/pciX/.../unbind
19 Date: December 2003 19 Date: December 2003
20 Contact: linux-pci@vger.kernel.org 20 Contact: linux-pci@vger.kernel.org
21 Description: 21 Description:
22 Writing a device location to t 22 Writing a device location to this file will cause the
23 driver to attempt to unbind fr 23 driver to attempt to unbind from the device found at
24 this location. This may be us 24 this location. This may be useful when overriding default
25 bindings. The format for the 25 bindings. The format for the location is: DDDD:BB:DD.F.
26 That is Domain:Bus:Device.Func 26 That is Domain:Bus:Device.Function and is the same as
27 found in /sys/bus/pci/devices/ 27 found in /sys/bus/pci/devices/. For example::
28 28
29 # echo 0000:00:19.0 > /sys/b 29 # echo 0000:00:19.0 > /sys/bus/pci/drivers/foo/unbind
30 30
31 (Note: kernels before 2.6.28 m 31 (Note: kernels before 2.6.28 may require echo -n).
32 32
33 What: /sys/bus/pci/drivers/.../new_i 33 What: /sys/bus/pci/drivers/.../new_id
34 What: /sys/devices/pciX/.../new_id 34 What: /sys/devices/pciX/.../new_id
35 Date: December 2003 35 Date: December 2003
36 Contact: linux-pci@vger.kernel.org 36 Contact: linux-pci@vger.kernel.org
37 Description: 37 Description:
38 Writing a device ID to this fi 38 Writing a device ID to this file will attempt to
39 dynamically add a new device I 39 dynamically add a new device ID to a PCI device driver.
40 This may allow the driver to s 40 This may allow the driver to support more hardware than
41 was included in the driver's s 41 was included in the driver's static device ID support
42 table at compile time. The fo 42 table at compile time. The format for the device ID is:
43 VVVV DDDD SVVV SDDD CCCC MMMM 43 VVVV DDDD SVVV SDDD CCCC MMMM PPPP. That is Vendor ID,
44 Device ID, Subsystem Vendor ID 44 Device ID, Subsystem Vendor ID, Subsystem Device ID,
45 Class, Class Mask, and Private 45 Class, Class Mask, and Private Driver Data. The Vendor ID
46 and Device ID fields are requi 46 and Device ID fields are required, the rest are optional.
47 Upon successfully adding an ID 47 Upon successfully adding an ID, the driver will probe
48 for the device and attempt to 48 for the device and attempt to bind to it. For example::
49 49
50 # echo "8086 10f5" > /sys/bu 50 # echo "8086 10f5" > /sys/bus/pci/drivers/foo/new_id
51 51
52 What: /sys/bus/pci/drivers/.../remov 52 What: /sys/bus/pci/drivers/.../remove_id
53 What: /sys/devices/pciX/.../remove_i 53 What: /sys/devices/pciX/.../remove_id
54 Date: February 2009 54 Date: February 2009
55 Contact: Chris Wright <chrisw@sous-sol.o 55 Contact: Chris Wright <chrisw@sous-sol.org>
56 Description: 56 Description:
57 Writing a device ID to this fi 57 Writing a device ID to this file will remove an ID
58 that was dynamically added via 58 that was dynamically added via the new_id sysfs entry.
59 The format for the device ID i 59 The format for the device ID is:
60 VVVV DDDD SVVV SDDD CCCC MMMM. 60 VVVV DDDD SVVV SDDD CCCC MMMM. That is Vendor ID, Device
61 ID, Subsystem Vendor ID, Subsy 61 ID, Subsystem Vendor ID, Subsystem Device ID, Class,
62 and Class Mask. The Vendor ID 62 and Class Mask. The Vendor ID and Device ID fields are
63 required, the rest are optiona 63 required, the rest are optional. After successfully
64 removing an ID, the driver wil 64 removing an ID, the driver will no longer support the
65 device. This is useful to ens 65 device. This is useful to ensure auto probing won't
66 match the driver to the device 66 match the driver to the device. For example::
67 67
68 # echo "8086 10f5" > /sys/bu 68 # echo "8086 10f5" > /sys/bus/pci/drivers/foo/remove_id
69 69
70 What: /sys/bus/pci/rescan 70 What: /sys/bus/pci/rescan
71 Date: January 2009 71 Date: January 2009
72 Contact: Linux PCI developers <linux-pci 72 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
73 Description: 73 Description:
74 Writing a non-zero value to th 74 Writing a non-zero value to this attribute will
75 force a rescan of all PCI buse 75 force a rescan of all PCI buses in the system, and
76 re-discover previously removed 76 re-discover previously removed devices.
77 77
78 What: /sys/bus/pci/devices/.../msi_b 78 What: /sys/bus/pci/devices/.../msi_bus
79 Date: September 2014 79 Date: September 2014
80 Contact: Linux PCI developers <linux-pci 80 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
81 Description: 81 Description:
82 Writing a zero value to this a 82 Writing a zero value to this attribute disallows MSI and
83 MSI-X for any future drivers o 83 MSI-X for any future drivers of the device. If the device
84 is a bridge, MSI and MSI-X wil 84 is a bridge, MSI and MSI-X will be disallowed for future
85 drivers of all child devices u 85 drivers of all child devices under the bridge. Drivers
86 must be reloaded for the new s 86 must be reloaded for the new setting to take effect.
87 87
88 What: /sys/bus/pci/devices/.../msi_i 88 What: /sys/bus/pci/devices/.../msi_irqs/
89 Date: September, 2011 89 Date: September, 2011
90 Contact: Neil Horman <nhorman@tuxdriver. 90 Contact: Neil Horman <nhorman@tuxdriver.com>
91 Description: 91 Description:
92 The /sys/devices/.../msi_irqs 92 The /sys/devices/.../msi_irqs directory contains a variable set
93 of files, with each file being 93 of files, with each file being named after a corresponding msi
94 irq vector allocated to that d 94 irq vector allocated to that device.
95 95
96 What: /sys/bus/pci/devices/.../msi_i 96 What: /sys/bus/pci/devices/.../msi_irqs/<N>
97 Date: September 2011 97 Date: September 2011
98 Contact: Neil Horman <nhorman@tuxdriver. 98 Contact: Neil Horman <nhorman@tuxdriver.com>
99 Description: 99 Description:
100 This attribute indicates the m 100 This attribute indicates the mode that the irq vector named by
101 the file is in (msi vs. msix) 101 the file is in (msi vs. msix)
102 102
103 What: /sys/bus/pci/devices/.../irq 103 What: /sys/bus/pci/devices/.../irq
104 Date: August 2021 104 Date: August 2021
105 Contact: Linux PCI developers <linux-pci 105 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
106 Description: 106 Description:
107 If a driver has enabled MSI (n 107 If a driver has enabled MSI (not MSI-X), "irq" contains the
108 IRQ of the first MSI vector. O 108 IRQ of the first MSI vector. Otherwise "irq" contains the
109 IRQ of the legacy INTx interru 109 IRQ of the legacy INTx interrupt.
110 110
111 "irq" being set to 0 indicates 111 "irq" being set to 0 indicates that the device isn't
112 capable of generating legacy I 112 capable of generating legacy INTx interrupts.
113 113
114 What: /sys/bus/pci/devices/.../remov 114 What: /sys/bus/pci/devices/.../remove
115 Date: January 2009 115 Date: January 2009
116 Contact: Linux PCI developers <linux-pci 116 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
117 Description: 117 Description:
118 Writing a non-zero value to th 118 Writing a non-zero value to this attribute will
119 hot-remove the PCI device and 119 hot-remove the PCI device and any of its children.
120 120
121 What: /sys/bus/pci/devices/.../pci_b 121 What: /sys/bus/pci/devices/.../pci_bus/.../rescan
122 Date: May 2011 122 Date: May 2011
123 Contact: Linux PCI developers <linux-pci 123 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
124 Description: 124 Description:
125 Writing a non-zero value to th 125 Writing a non-zero value to this attribute will
126 force a rescan of the bus and 126 force a rescan of the bus and all child buses,
127 and re-discover devices remove 127 and re-discover devices removed earlier from this
128 part of the device tree. 128 part of the device tree.
129 129
130 What: /sys/bus/pci/devices/.../resca 130 What: /sys/bus/pci/devices/.../rescan
131 Date: January 2009 131 Date: January 2009
132 Contact: Linux PCI developers <linux-pci 132 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
133 Description: 133 Description:
134 Writing a non-zero value to th 134 Writing a non-zero value to this attribute will
135 force a rescan of the device's 135 force a rescan of the device's parent bus and all
136 child buses, and re-discover d 136 child buses, and re-discover devices removed earlier
137 from this part of the device t 137 from this part of the device tree.
138 138
139 What: /sys/bus/pci/devices/.../reset 139 What: /sys/bus/pci/devices/.../reset_method
140 Date: August 2021 140 Date: August 2021
141 Contact: Amey Narkhede <ameynarkhede03@g 141 Contact: Amey Narkhede <ameynarkhede03@gmail.com>
142 Description: 142 Description:
143 Some devices allow an individu 143 Some devices allow an individual function to be reset
144 without affecting other functi 144 without affecting other functions in the same slot.
145 145
146 For devices that have this sup 146 For devices that have this support, a file named
147 reset_method is present in sys 147 reset_method is present in sysfs. Reading this file
148 gives names of the supported a 148 gives names of the supported and enabled reset methods and
149 their ordering. Writing a spa 149 their ordering. Writing a space-separated list of names of
150 reset methods sets the reset m 150 reset methods sets the reset methods and ordering to be
151 used when resetting the device 151 used when resetting the device. Writing an empty string
152 disables the ability to reset 152 disables the ability to reset the device. Writing
153 "default" enables all supporte 153 "default" enables all supported reset methods in the
154 default ordering. 154 default ordering.
155 155
156 What: /sys/bus/pci/devices/.../reset 156 What: /sys/bus/pci/devices/.../reset
157 Date: July 2009 157 Date: July 2009
158 Contact: Michael S. Tsirkin <mst@redhat. 158 Contact: Michael S. Tsirkin <mst@redhat.com>
159 Description: 159 Description:
160 Some devices allow an individu 160 Some devices allow an individual function to be reset
161 without affecting other functi 161 without affecting other functions in the same device.
162 For devices that have this sup 162 For devices that have this support, a file named reset
163 will be present in sysfs. Wri 163 will be present in sysfs. Writing 1 to this file
164 will perform reset. 164 will perform reset.
165 165
166 What: /sys/bus/pci/devices/.../vpd 166 What: /sys/bus/pci/devices/.../vpd
167 Date: February 2008 167 Date: February 2008
168 Contact: Ben Hutchings <bwh@kernel.org> 168 Contact: Ben Hutchings <bwh@kernel.org>
169 Description: 169 Description:
170 A file named vpd in a device d 170 A file named vpd in a device directory will be a
171 binary file containing the Vit 171 binary file containing the Vital Product Data for the
172 device. It should follow the 172 device. It should follow the VPD format defined in
173 PCI Specification 2.1 or 2.2, 173 PCI Specification 2.1 or 2.2, but users should consider
174 that some devices may have inc 174 that some devices may have incorrectly formatted data.
175 If the underlying VPD has a wr 175 If the underlying VPD has a writable section then the
176 corresponding section of this 176 corresponding section of this file will be writable.
177 177
178 What: /sys/bus/pci/devices/.../virtf 178 What: /sys/bus/pci/devices/.../virtfn<N>
179 Date: March 2009 179 Date: March 2009
180 Contact: Yu Zhao <yu.zhao@intel.com> 180 Contact: Yu Zhao <yu.zhao@intel.com>
181 Description: 181 Description:
182 This symbolic link appears whe 182 This symbolic link appears when hardware supports the SR-IOV
183 capability and the Physical Fu 183 capability and the Physical Function driver has enabled it.
184 The symbolic link points to th 184 The symbolic link points to the PCI device sysfs entry of the
185 Virtual Function whose index i 185 Virtual Function whose index is N (0...MaxVFs-1).
186 186
187 What: /sys/bus/pci/devices/.../dep_l 187 What: /sys/bus/pci/devices/.../dep_link
188 Date: March 2009 188 Date: March 2009
189 Contact: Yu Zhao <yu.zhao@intel.com> 189 Contact: Yu Zhao <yu.zhao@intel.com>
190 Description: 190 Description:
191 This symbolic link appears whe 191 This symbolic link appears when hardware supports the SR-IOV
192 capability and the Physical Fu 192 capability and the Physical Function driver has enabled it,
193 and this device has vendor spe 193 and this device has vendor specific dependencies with others.
194 The symbolic link points to th 194 The symbolic link points to the PCI device sysfs entry of
195 Physical Function this device 195 Physical Function this device depends on.
196 196
197 What: /sys/bus/pci/devices/.../physf 197 What: /sys/bus/pci/devices/.../physfn
198 Date: March 2009 198 Date: March 2009
199 Contact: Yu Zhao <yu.zhao@intel.com> 199 Contact: Yu Zhao <yu.zhao@intel.com>
200 Description: 200 Description:
201 This symbolic link appears whe 201 This symbolic link appears when a device is a Virtual Function.
202 The symbolic link points to th 202 The symbolic link points to the PCI device sysfs entry of the
203 Physical Function this device 203 Physical Function this device associates with.
204 204
205 What: /sys/bus/pci/devices/.../modal 205 What: /sys/bus/pci/devices/.../modalias
206 Date: May 2005 206 Date: May 2005
207 Contact: Greg Kroah-Hartman <gregkh@linu 207 Contact: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
208 Description: 208 Description:
209 This attribute indicates the P 209 This attribute indicates the PCI ID of the device object.
210 210
211 That is in the format: 211 That is in the format:
212 pci:vXXXXXXXXdXXXXXXXXsvXXXXXX 212 pci:vXXXXXXXXdXXXXXXXXsvXXXXXXXXsdXXXXXXXXbcXXscXXiXX,
213 where: 213 where:
214 214
215 - vXXXXXXXX contains the v 215 - vXXXXXXXX contains the vendor ID;
216 - dXXXXXXXX contains the d 216 - dXXXXXXXX contains the device ID;
217 - svXXXXXXXX contains the 217 - svXXXXXXXX contains the sub-vendor ID;
218 - sdXXXXXXXX contains the 218 - sdXXXXXXXX contains the subsystem device ID;
219 - bcXX contains the device 219 - bcXX contains the device class;
220 - scXX contains the device 220 - scXX contains the device subclass;
221 - iXX contains the device 221 - iXX contains the device class programming interface.
222 222
223 What: /sys/bus/pci/slots/.../module 223 What: /sys/bus/pci/slots/.../module
224 Date: June 2009 224 Date: June 2009
225 Contact: linux-pci@vger.kernel.org 225 Contact: linux-pci@vger.kernel.org
226 Description: 226 Description:
227 This symbolic link points to t 227 This symbolic link points to the PCI hotplug controller driver
228 module that manages the hotplu 228 module that manages the hotplug slot.
229 229
230 What: /sys/bus/pci/devices/.../label 230 What: /sys/bus/pci/devices/.../label
231 Date: July 2010 231 Date: July 2010
232 Contact: Narendra K <narendra_k@dell.com 232 Contact: Narendra K <narendra_k@dell.com>, linux-bugs@dell.com
233 Description: 233 Description:
234 Reading this attribute will pr 234 Reading this attribute will provide the firmware
235 given name (SMBIOS type 41 str 235 given name (SMBIOS type 41 string or ACPI _DSM string) of
236 the PCI device. The attribute 236 the PCI device. The attribute will be created only
237 if the firmware has given a na 237 if the firmware has given a name to the PCI device.
238 ACPI _DSM string name will be 238 ACPI _DSM string name will be given priority if the
239 system firmware provides SMBIO 239 system firmware provides SMBIOS type 41 string also.
240 Users: 240 Users:
241 Userspace applications interes 241 Userspace applications interested in knowing the
242 firmware assigned name of the 242 firmware assigned name of the PCI device.
243 243
244 What: /sys/bus/pci/devices/.../index 244 What: /sys/bus/pci/devices/.../index
245 Date: July 2010 245 Date: July 2010
246 Contact: Narendra K <narendra_k@dell.com 246 Contact: Narendra K <narendra_k@dell.com>, linux-bugs@dell.com
247 Description: 247 Description:
248 Reading this attribute will pr 248 Reading this attribute will provide the firmware given instance
249 number of the PCI device. Dep 249 number of the PCI device. Depending on the platform this can
250 be for example the SMBIOS type 250 be for example the SMBIOS type 41 device type instance or the
251 user-defined ID (UID) on s390. 251 user-defined ID (UID) on s390. The attribute will be created
252 only if the firmware has given 252 only if the firmware has given an instance number to the PCI
253 device and that number is guar 253 device and that number is guaranteed to uniquely identify the
254 device in the system. 254 device in the system.
255 Users: 255 Users:
256 Userspace applications interes 256 Userspace applications interested in knowing the
257 firmware assigned device type 257 firmware assigned device type instance of the PCI
258 device that can help in unders 258 device that can help in understanding the firmware
259 intended order of the PCI devi 259 intended order of the PCI device.
260 260
261 What: /sys/bus/pci/devices/.../acpi_ 261 What: /sys/bus/pci/devices/.../acpi_index
262 Date: July 2010 262 Date: July 2010
263 Contact: Narendra K <narendra_k@dell.com 263 Contact: Narendra K <narendra_k@dell.com>, linux-bugs@dell.com
264 Description: 264 Description:
265 Reading this attribute will pr 265 Reading this attribute will provide the firmware
266 given instance (ACPI _DSM inst 266 given instance (ACPI _DSM instance number) of the PCI device.
267 The attribute will be created 267 The attribute will be created only if the firmware has given
268 an instance number to the PCI 268 an instance number to the PCI device. ACPI _DSM instance number
269 will be given priority if the 269 will be given priority if the system firmware provides SMBIOS
270 type 41 device type instance a 270 type 41 device type instance also.
271 Users: 271 Users:
272 Userspace applications interes 272 Userspace applications interested in knowing the
273 firmware assigned instance num 273 firmware assigned instance number of the PCI
274 device that can help in unders 274 device that can help in understanding the firmware
275 intended order of the PCI devi 275 intended order of the PCI device.
276 276
277 What: /sys/bus/pci/devices/.../d3col 277 What: /sys/bus/pci/devices/.../d3cold_allowed
278 Date: July 2012 278 Date: July 2012
279 Contact: Huang Ying <ying.huang@intel.co 279 Contact: Huang Ying <ying.huang@intel.com>
280 Description: 280 Description:
281 d3cold_allowed is bit to contr 281 d3cold_allowed is bit to control whether the corresponding PCI
282 device can be put into D3Cold 282 device can be put into D3Cold state. If it is cleared, the
283 device will never be put into 283 device will never be put into D3Cold state. If it is set, the
284 device may be put into D3Cold 284 device may be put into D3Cold state if other requirements are
285 satisfied too. Reading this a 285 satisfied too. Reading this attribute will show the current
286 value of d3cold_allowed bit. 286 value of d3cold_allowed bit. Writing this attribute will set
287 the value of d3cold_allowed bi 287 the value of d3cold_allowed bit.
288 288
289 What: /sys/bus/pci/devices/.../sriov 289 What: /sys/bus/pci/devices/.../sriov_totalvfs
290 Date: November 2012 290 Date: November 2012
291 Contact: Donald Dutile <ddutile@redhat.c 291 Contact: Donald Dutile <ddutile@redhat.com>
292 Description: 292 Description:
293 This file appears when a physi 293 This file appears when a physical PCIe device supports SR-IOV.
294 Userspace applications can rea 294 Userspace applications can read this file to determine the
295 maximum number of Virtual Func 295 maximum number of Virtual Functions (VFs) a PCIe physical
296 function (PF) can support. Typ 296 function (PF) can support. Typically, this is the value reported
297 in the PF's SR-IOV extended ca 297 in the PF's SR-IOV extended capability structure's TotalVFs
298 element. Drivers have the abi 298 element. Drivers have the ability at probe time to reduce the
299 value read from this file via 299 value read from this file via the pci_sriov_set_totalvfs()
300 function. 300 function.
301 301
302 What: /sys/bus/pci/devices/.../sriov 302 What: /sys/bus/pci/devices/.../sriov_numvfs
303 Date: November 2012 303 Date: November 2012
304 Contact: Donald Dutile <ddutile@redhat.c 304 Contact: Donald Dutile <ddutile@redhat.com>
305 Description: 305 Description:
306 This file appears when a physi 306 This file appears when a physical PCIe device supports SR-IOV.
307 Userspace applications can rea 307 Userspace applications can read and write to this file to
308 determine and control the enab 308 determine and control the enablement or disablement of Virtual
309 Functions (VFs) on the physica 309 Functions (VFs) on the physical function (PF). A read of this
310 file will return the number of 310 file will return the number of VFs that are enabled on this PF.
311 A number written to this file 311 A number written to this file will enable the specified
312 number of VFs. A userspace app 312 number of VFs. A userspace application would typically read the
313 file and check that the value 313 file and check that the value is zero, and then write the number
314 of VFs that should be enabled 314 of VFs that should be enabled on the PF; the value written
315 should be less than or equal t 315 should be less than or equal to the value in the sriov_totalvfs
316 file. A userspace application 316 file. A userspace application wanting to disable the VFs would
317 write a zero to this file. The 317 write a zero to this file. The core ensures that valid values
318 are written to this file, and 318 are written to this file, and returns errors when values are not
319 valid. For example, writing a 319 valid. For example, writing a 2 to this file when sriov_numvfs
320 is not 0 and not 2 already wil 320 is not 0 and not 2 already will return an error. Writing a 10
321 when the value of sriov_totalv 321 when the value of sriov_totalvfs is 8 will return an error.
322 322
323 What: /sys/bus/pci/devices/.../drive 323 What: /sys/bus/pci/devices/.../driver_override
324 Date: April 2014 324 Date: April 2014
325 Contact: Alex Williamson <alex.williamso 325 Contact: Alex Williamson <alex.williamson@redhat.com>
326 Description: 326 Description:
327 This file allows the driver fo 327 This file allows the driver for a device to be specified which
328 will override standard static 328 will override standard static and dynamic ID matching. When
329 specified, only a driver with 329 specified, only a driver with a name matching the value written
330 to driver_override will have a 330 to driver_override will have an opportunity to bind to the
331 device. The override is speci 331 device. The override is specified by writing a string to the
332 driver_override file (echo pci 332 driver_override file (echo pci-stub > driver_override) and
333 may be cleared with an empty s 333 may be cleared with an empty string (echo > driver_override).
334 This returns the device to sta 334 This returns the device to standard matching rules binding.
335 Writing to driver_override doe 335 Writing to driver_override does not automatically unbind the
336 device from its current driver 336 device from its current driver or make any attempt to
337 automatically load the specifi 337 automatically load the specified driver. If no driver with a
338 matching name is currently loa 338 matching name is currently loaded in the kernel, the device
339 will not bind to any driver. 339 will not bind to any driver. This also allows devices to
340 opt-out of driver binding usin 340 opt-out of driver binding using a driver_override name such as
341 "none". Only a single driver 341 "none". Only a single driver may be specified in the override,
342 there is no support for parsin 342 there is no support for parsing delimiters.
343 343
344 What: /sys/bus/pci/devices/.../numa_ 344 What: /sys/bus/pci/devices/.../numa_node
345 Date: Oct 2014 345 Date: Oct 2014
346 Contact: Prarit Bhargava <prarit@redhat. 346 Contact: Prarit Bhargava <prarit@redhat.com>
347 Description: 347 Description:
348 This file contains the NUMA no 348 This file contains the NUMA node to which the PCI device is
349 attached, or -1 if the node is 349 attached, or -1 if the node is unknown. The initial value
350 comes from an ACPI _PXM method 350 comes from an ACPI _PXM method or a similar firmware
351 source. If that is missing or 351 source. If that is missing or incorrect, this file can be
352 written to override the node. 352 written to override the node. In that case, please report
353 a firmware bug to the system v 353 a firmware bug to the system vendor. Writing to this file
354 taints the kernel with TAINT_F 354 taints the kernel with TAINT_FIRMWARE_WORKAROUND, which
355 reduces the supportability of 355 reduces the supportability of your system.
356 356
357 What: /sys/bus/pci/devices/.../revis 357 What: /sys/bus/pci/devices/.../revision
358 Date: November 2016 358 Date: November 2016
359 Contact: Emil Velikov <emil.l.velikov@gm 359 Contact: Emil Velikov <emil.l.velikov@gmail.com>
360 Description: 360 Description:
361 This file contains the revisio 361 This file contains the revision field of the PCI device.
362 The value comes from device co 362 The value comes from device config space. The file is read only.
363 363
364 What: /sys/bus/pci/devices/.../sriov 364 What: /sys/bus/pci/devices/.../sriov_drivers_autoprobe
365 Date: April 2017 365 Date: April 2017
366 Contact: Bodong Wang<bodong@mellanox.com 366 Contact: Bodong Wang<bodong@mellanox.com>
367 Description: 367 Description:
368 This file is associated with t 368 This file is associated with the PF of a device that
369 supports SR-IOV. It determine 369 supports SR-IOV. It determines whether newly-enabled VFs
370 are immediately bound to a dri 370 are immediately bound to a driver. It initially contains
371 1, which means the kernel auto 371 1, which means the kernel automatically binds VFs to a
372 compatible driver immediately 372 compatible driver immediately after they are enabled. If
373 an application writes 0 to the 373 an application writes 0 to the file before enabling VFs,
374 the kernel will not bind VFs t 374 the kernel will not bind VFs to a driver.
375 375
376 A typical use case is to write 376 A typical use case is to write 0 to this file, then enable
377 VFs, then assign the newly-cre 377 VFs, then assign the newly-created VFs to virtual machines.
378 Note that changing this file d 378 Note that changing this file does not affect already-
379 enabled VFs. In this scenario 379 enabled VFs. In this scenario, the user must first disable
380 the VFs, write 0 to sriov_driv 380 the VFs, write 0 to sriov_drivers_autoprobe, then re-enable
381 the VFs. 381 the VFs.
382 382
383 This is similar to /sys/bus/pc 383 This is similar to /sys/bus/pci/drivers_autoprobe, but
384 affects only the VFs associate 384 affects only the VFs associated with a specific PF.
385 385
386 What: /sys/bus/pci/devices/.../p2pme 386 What: /sys/bus/pci/devices/.../p2pmem/size
387 Date: November 2017 387 Date: November 2017
388 Contact: Logan Gunthorpe <logang@deltate 388 Contact: Logan Gunthorpe <logang@deltatee.com>
389 Description: 389 Description:
390 If the device has any Peer-to- 390 If the device has any Peer-to-Peer memory registered, this
391 file contains the total amount 391 file contains the total amount of memory that the device
392 provides (in decimal). 392 provides (in decimal).
393 393
394 What: /sys/bus/pci/devices/.../p2pme 394 What: /sys/bus/pci/devices/.../p2pmem/available
395 Date: November 2017 395 Date: November 2017
396 Contact: Logan Gunthorpe <logang@deltate 396 Contact: Logan Gunthorpe <logang@deltatee.com>
397 Description: 397 Description:
398 If the device has any Peer-to- 398 If the device has any Peer-to-Peer memory registered, this
399 file contains the amount of me 399 file contains the amount of memory that has not been
400 allocated (in decimal). 400 allocated (in decimal).
401 401
402 What: /sys/bus/pci/devices/.../p2pme 402 What: /sys/bus/pci/devices/.../p2pmem/published
403 Date: November 2017 403 Date: November 2017
404 Contact: Logan Gunthorpe <logang@deltate 404 Contact: Logan Gunthorpe <logang@deltatee.com>
405 Description: 405 Description:
406 If the device has any Peer-to- 406 If the device has any Peer-to-Peer memory registered, this
407 file contains a '1' if the mem 407 file contains a '1' if the memory has been published for
408 use outside the driver that ow 408 use outside the driver that owns the device.
409 409
410 What: /sys/bus/pci/devices/.../p2pme 410 What: /sys/bus/pci/devices/.../p2pmem/allocate
411 Date: August 2022 411 Date: August 2022
412 Contact: Logan Gunthorpe <logang@deltate 412 Contact: Logan Gunthorpe <logang@deltatee.com>
413 Description: 413 Description:
414 This file allows mapping p2pme 414 This file allows mapping p2pmem into userspace. For each
415 mmap() call on this file, the 415 mmap() call on this file, the kernel will allocate a chunk
416 of Peer-to-Peer memory for use 416 of Peer-to-Peer memory for use in Peer-to-Peer transactions.
417 This memory can be used in O_D 417 This memory can be used in O_DIRECT calls to NVMe backed
418 files for Peer-to-Peer copies. 418 files for Peer-to-Peer copies.
419 419
420 What: /sys/bus/pci/devices/.../link/ 420 What: /sys/bus/pci/devices/.../link/clkpm
421 /sys/bus/pci/devices/.../link/ 421 /sys/bus/pci/devices/.../link/l0s_aspm
422 /sys/bus/pci/devices/.../link/ 422 /sys/bus/pci/devices/.../link/l1_aspm
423 /sys/bus/pci/devices/.../link/ 423 /sys/bus/pci/devices/.../link/l1_1_aspm
424 /sys/bus/pci/devices/.../link/ 424 /sys/bus/pci/devices/.../link/l1_2_aspm
425 /sys/bus/pci/devices/.../link/ 425 /sys/bus/pci/devices/.../link/l1_1_pcipm
426 /sys/bus/pci/devices/.../link/ 426 /sys/bus/pci/devices/.../link/l1_2_pcipm
427 Date: October 2019 427 Date: October 2019
428 Contact: Heiner Kallweit <hkallweit1@gma 428 Contact: Heiner Kallweit <hkallweit1@gmail.com>
429 Description: If ASPM is supported for an en 429 Description: If ASPM is supported for an endpoint, these files can be
430 used to disable or enable the 430 used to disable or enable the individual power management
431 states. Write y/1/on to enable 431 states. Write y/1/on to enable, n/0/off to disable.
432 432
433 What: /sys/bus/pci/devices/.../power 433 What: /sys/bus/pci/devices/.../power_state
434 Date: November 2020 434 Date: November 2020
435 Contact: Linux PCI developers <linux-pci 435 Contact: Linux PCI developers <linux-pci@vger.kernel.org>
436 Description: 436 Description:
437 This file contains the current 437 This file contains the current PCI power state of the device.
438 The value comes from the PCI k 438 The value comes from the PCI kernel device state and can be one
439 of: "unknown", "error", "D0", 439 of: "unknown", "error", "D0", D1", "D2", "D3hot", "D3cold".
440 The file is read only. 440 The file is read only.
441 441
442 What: /sys/bus/pci/devices/.../sriov 442 What: /sys/bus/pci/devices/.../sriov_vf_total_msix
443 Date: January 2021 443 Date: January 2021
444 Contact: Leon Romanovsky <leonro@nvidia. 444 Contact: Leon Romanovsky <leonro@nvidia.com>
445 Description: 445 Description:
446 This file is associated with a 446 This file is associated with a SR-IOV physical function (PF).
447 It contains the total number o 447 It contains the total number of MSI-X vectors available for
448 assignment to all virtual func 448 assignment to all virtual functions (VFs) associated with PF.
449 The value will be zero if the 449 The value will be zero if the device doesn't support this
450 functionality. For supported d 450 functionality. For supported devices, the value will be
451 constant and won't be changed 451 constant and won't be changed after MSI-X vectors assignment.
452 452
453 What: /sys/bus/pci/devices/.../sriov 453 What: /sys/bus/pci/devices/.../sriov_vf_msix_count
454 Date: January 2021 454 Date: January 2021
455 Contact: Leon Romanovsky <leonro@nvidia. 455 Contact: Leon Romanovsky <leonro@nvidia.com>
456 Description: 456 Description:
457 This file is associated with a 457 This file is associated with a SR-IOV virtual function (VF).
458 It allows configuration of the 458 It allows configuration of the number of MSI-X vectors for
459 the VF. This allows devices th 459 the VF. This allows devices that have a global pool of MSI-X
460 vectors to optimally divide th 460 vectors to optimally divide them between VFs based on VF usage.
461 461
462 The values accepted are: 462 The values accepted are:
463 * > 0 - this number will be r 463 * > 0 - this number will be reported as the Table Size in the
464 VF's MSI-X capability 464 VF's MSI-X capability
465 * < 0 - not valid 465 * < 0 - not valid
466 * = 0 - will reset to the dev 466 * = 0 - will reset to the device default value
467 467
468 The file is writable if the PF 468 The file is writable if the PF is bound to a driver that
469 implements ->sriov_set_msix_ve 469 implements ->sriov_set_msix_vec_count().
470 470
471 What: /sys/bus/pci/devices/.../resou 471 What: /sys/bus/pci/devices/.../resourceN_resize
472 Date: September 2022 472 Date: September 2022
473 Contact: Alex Williamson <alex.williamso 473 Contact: Alex Williamson <alex.williamson@redhat.com>
474 Description: 474 Description:
475 These files provide an interfa 475 These files provide an interface to PCIe Resizable BAR support.
476 A file is created for each BAR 476 A file is created for each BAR resource (N) supported by the
477 PCIe Resizable BAR extended ca 477 PCIe Resizable BAR extended capability of the device. Reading
478 each file exposes the bitmap o 478 each file exposes the bitmap of available resource sizes:
479 479
480 # cat resource1_resize 480 # cat resource1_resize
481 00000000000001c0 481 00000000000001c0
482 482
483 The bitmap represents supporte 483 The bitmap represents supported resource sizes for the BAR,
484 where bit0 = 1MB, bit1 = 2MB, 484 where bit0 = 1MB, bit1 = 2MB, bit2 = 4MB, etc. In the above
485 example the device supports 64 485 example the device supports 64MB, 128MB, and 256MB BAR sizes.
486 486
487 When writing the file, the use 487 When writing the file, the user provides the bit position of
488 the desired resource size, for 488 the desired resource size, for example:
489 489
490 # echo 7 > resource1_resize 490 # echo 7 > resource1_resize
491 491
492 This indicates to set the size 492 This indicates to set the size value corresponding to bit 7,
493 128MB. The resulting size is 493 128MB. The resulting size is 2 ^ (bit# + 20). This definition
494 matches the PCIe specification 494 matches the PCIe specification of this capability.
495 495
496 In order to make use of resour 496 In order to make use of resource resizing, all PCI drivers must
497 be unbound from the device and 497 be unbound from the device and peer devices under the same
498 parent bridge may need to be s 498 parent bridge may need to be soft removed. In the case of
499 VGA devices, writing a resize 499 VGA devices, writing a resize value will remove low level
500 console drivers from the devic 500 console drivers from the device. Raw users of pci-sysfs
501 resourceN attributes must be t 501 resourceN attributes must be terminated prior to resizing.
502 Success of the resizing operat 502 Success of the resizing operation is not guaranteed.
503 503
504 What: /sys/bus/pci/devices/.../leds/ 504 What: /sys/bus/pci/devices/.../leds/*:enclosure:*/brightness
505 What: /sys/class/leds/*:enclosure:*/ 505 What: /sys/class/leds/*:enclosure:*/brightness
506 Date: August 2024 506 Date: August 2024
507 KernelVersion: 6.12 507 KernelVersion: 6.12
508 Description: 508 Description:
509 LED indications on PCIe storag 509 LED indications on PCIe storage enclosures which are controlled
510 through the NPEM interface (Na 510 through the NPEM interface (Native PCIe Enclosure Management,
511 PCIe r6.1 sec 6.28) are access 511 PCIe r6.1 sec 6.28) are accessible as led class devices, both
512 below /sys/class/leds and belo 512 below /sys/class/leds and below NPEM-capable PCI devices.
513 513
514 Although these led class devic 514 Although these led class devices could be manipulated manually,
515 in practice they are typically 515 in practice they are typically manipulated automatically by an
516 application such as ledmon(8). 516 application such as ledmon(8).
517 517
518 The name of a led class device 518 The name of a led class device is as follows:
519 <bdf>:enclosure:<indication> 519 <bdf>:enclosure:<indication>
520 where: 520 where:
521 521
522 - <bdf> is the domain, bus, de 522 - <bdf> is the domain, bus, device and function number
523 (e.g. 10000:02:05.0) 523 (e.g. 10000:02:05.0)
524 - <indication> is a short desc 524 - <indication> is a short description of the LED indication
525 525
526 Valid indications per PCIe r6. 526 Valid indications per PCIe r6.1 table 6-27 are:
527 527
528 - ok (drive is functioning nor 528 - ok (drive is functioning normally)
529 - locate (drive is being ident 529 - locate (drive is being identified by an admin)
530 - fail (drive is not functioni 530 - fail (drive is not functioning properly)
531 - rebuild (drive is part of an 531 - rebuild (drive is part of an array that is rebuilding)
532 - pfa (drive is predicted to f 532 - pfa (drive is predicted to fail soon)
533 - hotspare (drive is marked to 533 - hotspare (drive is marked to be used as a replacement)
534 - ica (drive is part of an arr 534 - ica (drive is part of an array that is degraded)
535 - ifa (drive is part of an arr 535 - ifa (drive is part of an array that is failed)
536 - idt (drive is not the right 536 - idt (drive is not the right type for the connector)
537 - disabled (drive is disabled, 537 - disabled (drive is disabled, removal is safe)
538 - specific0 to specific7 (encl 538 - specific0 to specific7 (enclosure-specific indications)
539 539
540 Broadly, the indications fall 540 Broadly, the indications fall into one of these categories:
541 541
542 - to signify drive state (ok, 542 - to signify drive state (ok, locate, fail, idt, disabled)
543 - to signify drive role or sta 543 - to signify drive role or state in a software RAID array
544 (rebuild, pfa, hotspare, ica 544 (rebuild, pfa, hotspare, ica, ifa)
545 - to signify any other role or 545 - to signify any other role or state (specific0 to specific7)
546 546
547 Mandatory indications per PCIe 547 Mandatory indications per PCIe r6.1 sec 7.9.19.2 comprise:
548 ok, locate, fail, rebuild. All 548 ok, locate, fail, rebuild. All others are optional.
549 A led class device is only vis 549 A led class device is only visible if the corresponding
550 indication is supported by the 550 indication is supported by the device.
551 551
552 To manipulate the indications, 552 To manipulate the indications, write 0 (LED_OFF) or 1 (LED_ON)
553 to the "brightness" file. Note 553 to the "brightness" file. Note that manipulating an indication
554 may implicitly manipulate othe 554 may implicitly manipulate other indications at the vendor's
555 discretion. E.g. when the user 555 discretion. E.g. when the user lights up the "ok" indication,
556 the vendor may choose to autom 556 the vendor may choose to automatically turn off the "fail"
557 indication. The current state 557 indication. The current state of an indication can be
558 retrieved by reading its "brig 558 retrieved by reading its "brightness" file.
559 559
560 The PCIe Base Specification al 560 The PCIe Base Specification allows vendors leeway to choose
561 different colors or blinking p 561 different colors or blinking patterns for the indications,
562 but they typically follow the 562 but they typically follow the IBPI standard. E.g. the "locate"
563 indication is usually presente 563 indication is usually presented as one or two LEDs blinking at
564 4 Hz frequency: 564 4 Hz frequency:
565 https://en.wikipedia.org/wiki/ 565 https://en.wikipedia.org/wiki/International_Blinking_Pattern_Interpretation
566 566
567 PCI Firmware Specification r3. 567 PCI Firmware Specification r3.3 sec 4.7 defines a DSM interface
568 to facilitate shared access by 568 to facilitate shared access by operating system and platform
569 firmware to a device's NPEM re 569 firmware to a device's NPEM registers. The kernel will use
570 this DSM interface where avail 570 this DSM interface where available, instead of accessing NPEM
571 registers directly. The DSM in 571 registers directly. The DSM interface does not support the
572 enclosure-specific indications 572 enclosure-specific indications "specific0" to "specific7",
573 hence the corresponding led cl 573 hence the corresponding led class devices are unavailable if
574 the DSM interface is used. 574 the DSM interface is used.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.