1 =================== 1 ===================
2 Fallback mechanisms 2 Fallback mechanisms
3 =================== 3 ===================
4 4
5 A fallback mechanism is supported to allow to 5 A fallback mechanism is supported to allow to overcome failures to do a direct
6 filesystem lookup on the root filesystem or wh 6 filesystem lookup on the root filesystem or when the firmware simply cannot be
7 installed for practical reasons on the root fi 7 installed for practical reasons on the root filesystem. The kernel
8 configuration options related to supporting th 8 configuration options related to supporting the firmware fallback mechanism are:
9 9
10 * CONFIG_FW_LOADER_USER_HELPER: enables buil 10 * CONFIG_FW_LOADER_USER_HELPER: enables building the firmware fallback
11 mechanism. Most distributions enable this 11 mechanism. Most distributions enable this option today. If enabled but
12 CONFIG_FW_LOADER_USER_HELPER_FALLBACK is d 12 CONFIG_FW_LOADER_USER_HELPER_FALLBACK is disabled, only the custom fallback
13 mechanism is available and for the request 13 mechanism is available and for the request_firmware_nowait() call.
14 * CONFIG_FW_LOADER_USER_HELPER_FALLBACK: for 14 * CONFIG_FW_LOADER_USER_HELPER_FALLBACK: force enables each request to
15 enable the kobject uevent fallback mechani 15 enable the kobject uevent fallback mechanism on all firmware API calls
16 except request_firmware_direct(). Most dis 16 except request_firmware_direct(). Most distributions disable this option
17 today. The call request_firmware_nowait() 17 today. The call request_firmware_nowait() allows for one alternative
18 fallback mechanism: if this kconfig option 18 fallback mechanism: if this kconfig option is enabled and your second
19 argument to request_firmware_nowait(), uev 19 argument to request_firmware_nowait(), uevent, is set to false you are
20 informing the kernel that you have a custo 20 informing the kernel that you have a custom fallback mechanism and it will
21 manually load the firmware. Read below for 21 manually load the firmware. Read below for more details.
22 22
23 Note that this means when having this configur 23 Note that this means when having this configuration:
24 24
25 CONFIG_FW_LOADER_USER_HELPER=y 25 CONFIG_FW_LOADER_USER_HELPER=y
26 CONFIG_FW_LOADER_USER_HELPER_FALLBACK=n 26 CONFIG_FW_LOADER_USER_HELPER_FALLBACK=n
27 27
28 the kobject uevent fallback mechanism will nev 28 the kobject uevent fallback mechanism will never take effect even
29 for request_firmware_nowait() when uevent is s 29 for request_firmware_nowait() when uevent is set to true.
30 30
31 Justifying the firmware fallback mechanism 31 Justifying the firmware fallback mechanism
32 ========================================== 32 ==========================================
33 33
34 Direct filesystem lookups may fail for a varie 34 Direct filesystem lookups may fail for a variety of reasons. Known reasons for
35 this are worth itemizing and documenting as it 35 this are worth itemizing and documenting as it justifies the need for the
36 fallback mechanism: 36 fallback mechanism:
37 37
38 * Race against access with the root filesystem 38 * Race against access with the root filesystem upon bootup.
39 39
40 * Races upon resume from suspend. This is reso 40 * Races upon resume from suspend. This is resolved by the firmware cache, but
41 the firmware cache is only supported if you 41 the firmware cache is only supported if you use uevents, and its not
42 supported for request_firmware_into_buf(). 42 supported for request_firmware_into_buf().
43 43
44 * Firmware is not accessible through typical m 44 * Firmware is not accessible through typical means:
45 <<
46 * It cannot be installed into the root 45 * It cannot be installed into the root filesystem
47 * The firmware provides very unique de 46 * The firmware provides very unique device specific data tailored for
48 the unit gathered with local informa 47 the unit gathered with local information. An example is calibration
49 data for WiFi chipsets for mobile de 48 data for WiFi chipsets for mobile devices. This calibration data is
50 not common to all units, but tailore 49 not common to all units, but tailored per unit. Such information may
51 be installed on a separate flash par 50 be installed on a separate flash partition other than where the root
52 filesystem is provided. 51 filesystem is provided.
53 52
54 Types of fallback mechanisms 53 Types of fallback mechanisms
55 ============================ 54 ============================
56 55
57 There are really two fallback mechanisms avail 56 There are really two fallback mechanisms available using one shared sysfs
58 interface as a loading facility: 57 interface as a loading facility:
59 58
60 * Kobject uevent fallback mechanism 59 * Kobject uevent fallback mechanism
61 * Custom fallback mechanism 60 * Custom fallback mechanism
62 61
63 First lets document the shared sysfs loading f 62 First lets document the shared sysfs loading facility.
64 63
65 Firmware sysfs loading facility 64 Firmware sysfs loading facility
66 =============================== 65 ===============================
67 66
68 In order to help device drivers upload firmwar 67 In order to help device drivers upload firmware using a fallback mechanism
69 the firmware infrastructure creates a sysfs in 68 the firmware infrastructure creates a sysfs interface to enable userspace
70 to load and indicate when firmware is ready. T 69 to load and indicate when firmware is ready. The sysfs directory is created
71 via fw_create_instance(). This call creates a 70 via fw_create_instance(). This call creates a new struct device named after
72 the firmware requested, and establishes it in 71 the firmware requested, and establishes it in the device hierarchy by
73 associating the device used to make the reques 72 associating the device used to make the request as the device's parent.
74 The sysfs directory's file attributes are defi 73 The sysfs directory's file attributes are defined and controlled through
75 the new device's class (firmware_class) and gr !! 74 the new device's class (firmare_class) and group (fw_dev_attr_groups).
76 This is actually where the original firmware_c !! 75 This is actually where the original firmware_class.c file name comes from,
77 given that originally the only firmware loadin !! 76 as originally the only firmware loading mechanism available was the
78 mechanism we now use as a fallback mechanism, !! 77 mechanism we now use as a fallback mechanism.
79 firmware_class. Because the attributes exposed <<
80 module name firmware_class cannot be renamed i <<
81 compatibility with old userspace. <<
82 78
83 To load firmware using the sysfs interface we 79 To load firmware using the sysfs interface we expose a loading indicator,
84 and a file upload firmware into: 80 and a file upload firmware into:
85 81
86 * /sys/$DEVPATH/loading 82 * /sys/$DEVPATH/loading
87 * /sys/$DEVPATH/data 83 * /sys/$DEVPATH/data
88 84
89 To upload firmware you will echo 1 onto the lo 85 To upload firmware you will echo 1 onto the loading file to indicate
90 you are loading firmware. You then write the f !! 86 you are loading firmware. You then cat the firmware into the data file,
91 and you notify the kernel the firmware is read 87 and you notify the kernel the firmware is ready by echo'ing 0 onto
92 the loading file. 88 the loading file.
93 89
94 The firmware device used to help load firmware 90 The firmware device used to help load firmware using sysfs is only created if
95 direct firmware loading fails and if the fallb 91 direct firmware loading fails and if the fallback mechanism is enabled for your
96 firmware request, this is set up with :c:func: !! 92 firmware request, this is set up with fw_load_from_user_helper(). It is
97 important to re-iterate that no device is crea 93 important to re-iterate that no device is created if a direct filesystem lookup
98 succeeded. 94 succeeded.
99 95
100 Using:: 96 Using::
101 97
102 echo 1 > /sys/$DEVPATH/loading 98 echo 1 > /sys/$DEVPATH/loading
103 99
104 Will clean any previous partial load at once a 100 Will clean any previous partial load at once and make the firmware API
105 return an error. When loading firmware the fir 101 return an error. When loading firmware the firmware_class grows a buffer
106 for the firmware in PAGE_SIZE increments to ho 102 for the firmware in PAGE_SIZE increments to hold the image as it comes in.
107 103
108 firmware_data_read() and firmware_loading_show 104 firmware_data_read() and firmware_loading_show() are just provided for the
109 test_firmware driver for testing, they are not 105 test_firmware driver for testing, they are not called in normal use or
110 expected to be used regularly by userspace. 106 expected to be used regularly by userspace.
111 107
112 firmware_fallback_sysfs <<
113 ----------------------- <<
114 .. kernel-doc:: drivers/base/firmware_loader/f <<
115 :functions: firmware_fallback_sysfs <<
116 <<
117 Firmware kobject uevent fallback mechanism 108 Firmware kobject uevent fallback mechanism
118 ========================================== 109 ==========================================
119 110
120 Since a device is created for the sysfs interf 111 Since a device is created for the sysfs interface to help load firmware as a
121 fallback mechanism userspace can be informed o 112 fallback mechanism userspace can be informed of the addition of the device by
122 relying on kobject uevents. The addition of th 113 relying on kobject uevents. The addition of the device into the device
123 hierarchy means the fallback mechanism for fir 114 hierarchy means the fallback mechanism for firmware loading has been initiated.
124 For details of implementation refer to fw_load !! 115 For details of implementation refer to _request_firmware_load(), in particular
125 on the use of dev_set_uevent_suppress() and ko 116 on the use of dev_set_uevent_suppress() and kobject_uevent().
126 117
127 The kernel's kobject uevent mechanism is imple 118 The kernel's kobject uevent mechanism is implemented in lib/kobject_uevent.c,
128 it issues uevents to userspace. As a supplemen 119 it issues uevents to userspace. As a supplement to kobject uevents Linux
129 distributions could also enable CONFIG_UEVENT_ 120 distributions could also enable CONFIG_UEVENT_HELPER_PATH, which makes use of
130 core kernel's usermode helper (UMH) functional 121 core kernel's usermode helper (UMH) functionality to call out to a userspace
131 helper for kobject uevents. In practice though 122 helper for kobject uevents. In practice though no standard distribution has
132 ever used the CONFIG_UEVENT_HELPER_PATH. If CO 123 ever used the CONFIG_UEVENT_HELPER_PATH. If CONFIG_UEVENT_HELPER_PATH is
133 enabled this binary would be called each time 124 enabled this binary would be called each time kobject_uevent_env() gets called
134 in the kernel for each kobject uevent triggere 125 in the kernel for each kobject uevent triggered.
135 126
136 Different implementations have been supported 127 Different implementations have been supported in userspace to take advantage of
137 this fallback mechanism. When firmware loading 128 this fallback mechanism. When firmware loading was only possible using the
138 sysfs mechanism the userspace component "hotpl 129 sysfs mechanism the userspace component "hotplug" provided the functionality of
139 monitoring for kobject events. Historically th 130 monitoring for kobject events. Historically this was superseded be systemd's
140 udev, however firmware loading support was rem 131 udev, however firmware loading support was removed from udev as of systemd
141 commit be2ea723b1d0 ("udev: remove userspace f 132 commit be2ea723b1d0 ("udev: remove userspace firmware loading support")
142 as of v217 on August, 2014. This means most Li 133 as of v217 on August, 2014. This means most Linux distributions today are
143 not using or taking advantage of the firmware 134 not using or taking advantage of the firmware fallback mechanism provided
144 by kobject uevents. This is specially exacerba 135 by kobject uevents. This is specially exacerbated due to the fact that most
145 distributions today disable CONFIG_FW_LOADER_U 136 distributions today disable CONFIG_FW_LOADER_USER_HELPER_FALLBACK.
146 137
147 Refer to do_firmware_uevent() for details of t 138 Refer to do_firmware_uevent() for details of the kobject event variables
148 setup. The variables currently passed to users !! 139 setup. Variables passwdd with a kobject add event:
149 event are: <<
150 140
151 * FIRMWARE=firmware name 141 * FIRMWARE=firmware name
152 * TIMEOUT=timeout value 142 * TIMEOUT=timeout value
153 * ASYNC=whether or not the API request was asy 143 * ASYNC=whether or not the API request was asynchronous
154 144
155 By default DEVPATH is set by the internal kern 145 By default DEVPATH is set by the internal kernel kobject infrastructure.
156 Below is an example simple kobject uevent scri 146 Below is an example simple kobject uevent script::
157 147
158 # Both $DEVPATH and $FIRMWARE are alre 148 # Both $DEVPATH and $FIRMWARE are already provided in the environment.
159 MY_FW_DIR=/lib/firmware/ 149 MY_FW_DIR=/lib/firmware/
160 echo 1 > /sys/$DEVPATH/loading 150 echo 1 > /sys/$DEVPATH/loading
161 cat $MY_FW_DIR/$FIRMWARE > /sys/$DEVPA 151 cat $MY_FW_DIR/$FIRMWARE > /sys/$DEVPATH/data
162 echo 0 > /sys/$DEVPATH/loading 152 echo 0 > /sys/$DEVPATH/loading
163 153
164 Firmware custom fallback mechanism 154 Firmware custom fallback mechanism
165 ================================== 155 ==================================
166 156
167 Users of the request_firmware_nowait() call ha 157 Users of the request_firmware_nowait() call have yet another option available
168 at their disposal: rely on the sysfs fallback 158 at their disposal: rely on the sysfs fallback mechanism but request that no
169 kobject uevents be issued to userspace. The or 159 kobject uevents be issued to userspace. The original logic behind this
170 was that utilities other than udev might be re 160 was that utilities other than udev might be required to lookup firmware
171 in non-traditional paths -- paths outside of t 161 in non-traditional paths -- paths outside of the listing documented in the
172 section 'Direct filesystem lookup'. This optio 162 section 'Direct filesystem lookup'. This option is not available to any of
173 the other API calls as uevents are always forc 163 the other API calls as uevents are always forced for them.
174 164
175 Since uevents are only meaningful if the fallb 165 Since uevents are only meaningful if the fallback mechanism is enabled
176 in your kernel it would seem odd to enable uev 166 in your kernel it would seem odd to enable uevents with kernels that do not
177 have the fallback mechanism enabled in their k 167 have the fallback mechanism enabled in their kernels. Unfortunately we also
178 rely on the uevent flag which can be disabled 168 rely on the uevent flag which can be disabled by request_firmware_nowait() to
179 also setup the firmware cache for firmware req 169 also setup the firmware cache for firmware requests. As documented above,
180 the firmware cache is only set up if uevent is 170 the firmware cache is only set up if uevent is enabled for an API call.
181 Although this can disable the firmware cache f 171 Although this can disable the firmware cache for request_firmware_nowait()
182 calls, users of this API should not use it for 172 calls, users of this API should not use it for the purposes of disabling
183 the cache as that was not the original purpose 173 the cache as that was not the original purpose of the flag. Not setting
184 the uevent flag means you want to opt-in for t 174 the uevent flag means you want to opt-in for the firmware fallback mechanism
185 but you want to suppress kobject uevents, as y 175 but you want to suppress kobject uevents, as you have a custom solution which
186 will monitor for your device addition into the 176 will monitor for your device addition into the device hierarchy somehow and
187 load firmware for you through a custom path. 177 load firmware for you through a custom path.
188 178
189 Firmware fallback timeout 179 Firmware fallback timeout
190 ========================= 180 =========================
191 181
192 The firmware fallback mechanism has a timeout. 182 The firmware fallback mechanism has a timeout. If firmware is not loaded
193 onto the sysfs interface by the timeout value 183 onto the sysfs interface by the timeout value an error is sent to the
194 driver. By default the timeout is set to 60 se 184 driver. By default the timeout is set to 60 seconds if uevents are
195 desirable, otherwise MAX_JIFFY_OFFSET is used 185 desirable, otherwise MAX_JIFFY_OFFSET is used (max timeout possible).
196 The logic behind using MAX_JIFFY_OFFSET for no 186 The logic behind using MAX_JIFFY_OFFSET for non-uevents is that a custom
197 solution will have as much time as it needs to 187 solution will have as much time as it needs to load firmware.
198 188
199 You can customize the firmware timeout by echo 189 You can customize the firmware timeout by echo'ing your desired timeout into
200 the following file: 190 the following file:
201 191
202 * /sys/class/firmware/timeout 192 * /sys/class/firmware/timeout
203 193
204 If you echo 0 into it means MAX_JIFFY_OFFSET w 194 If you echo 0 into it means MAX_JIFFY_OFFSET will be used. The data type
205 for the timeout is an int. 195 for the timeout is an int.
206 <<
207 EFI embedded firmware fallback mechanism <<
208 ======================================== <<
209 <<
210 On some devices the system's EFI code / ROM ma <<
211 of firmware for some of the system's integrate <<
212 the peripheral's Linux device-driver needs to <<
213 <<
214 Device drivers which need such firmware can us <<
215 firmware_request_platform() function for this, <<
216 separate fallback mechanism from the other fal <<
217 this does not use the sysfs interface. <<
218 <<
219 A device driver which needs this can describe <<
220 using an efi_embedded_fw_desc struct: <<
221 <<
222 .. kernel-doc:: include/linux/efi_embedded_fw. <<
223 :functions: efi_embedded_fw_desc <<
224 <<
225 The EFI embedded-fw code works by scanning all <<
226 segments for an eight byte sequence matching p <<
227 then does a sha256 over length bytes and if th <<
228 bytes and adds that to its list with found fir <<
229 <<
230 To avoid doing this somewhat expensive scan on <<
231 used. Drivers are expected to export a dmi_sys <<
232 driver_data pointing to an efi_embedded_fw_des <<
233 <<
234 To register this array with the efi-embedded-f <<
235 <<
236 1. Always be builtin to the kernel or store th <<
237 separate object file which always gets buil <<
238 <<
239 2. Add an extern declaration for the dmi_syste <<
240 include/linux/efi_embedded_fw.h. <<
241 <<
242 3. Add the dmi_system_id array to the embedded <<
243 drivers/firmware/efi/embedded-firmware.c wr <<
244 the driver is being builtin. <<
245 <<
246 4. Add "select EFI_EMBEDDED_FIRMWARE if EFI_ST <<
247 <<
248 The firmware_request_platform() function will <<
249 with the specified name directly from the disk <<
250 always be overridden by placing a file under / <<
251 <<
252 Note that: <<
253 <<
254 1. The code scanning for EFI embedded-firmware <<
255 of start_kernel(), just before calling rest <<
256 subsystems using subsys_initcall() to regis <<
257 matter. This means that code running earlie <<
258 embedded-firmware. <<
259 <<
260 2. At the moment the EFI embedded-fw code assu <<
261 an offset which is a multiple of 8 bytes, i <<
262 send in a patch to fix this. <<
263 <<
264 3. At the moment the EFI embedded-fw code only <<
265 free EFI_BOOT_SERVICES_CODE before the EFI <<
266 scan it. <<
267 <<
268 4. The current brute-force scanning of EFI_BOO <<
269 brute-force solution. There has been discus <<
270 Initialization (PI) spec's Firmware Volume <<
271 because the FV Protocol relies on *internal <<
272 1. The PI spec does not define peripheral f <<
273 2. The internal interfaces of the PI spec d <<
274 compatibility. Any implementation details i <<
275 and may vary system to system. Supporting t <<
276 difficult as it is purposely ambiguous. <<
277 <<
278 Example how to check for and extract embedded <<
279 ---------------------------------------------- <<
280 <<
281 To check for, for example Silead touchscreen c <<
282 do the following: <<
283 <<
284 1. Boot the system with efi=debug on the kerne <<
285 <<
286 2. cp /sys/kernel/debug/efi/boot_services_code <<
287 <<
288 3. Open the boot_services_code? files in a hex <<
289 magic prefix for Silead firmware: F0 00 00 <<
290 the beginning address of the firmware insid <<
291 <<
292 4. The firmware has a specific pattern, it sta <<
293 typically F0 00 00 00 02 00 00 00 for the f <<
294 word-address + 32-bit value pairs. With the <<
295 bytes (1 word) for each pair until a page i <<
296 followed by a new page-address, followed by <<
297 leads to a very distinct pattern. Scroll do <<
298 this gives you the end of the firmware insi <<
299 <<
300 5. "dd if=boot_services_code? of=firmware bs=1 <<
301 will extract the firmware for you. Inspect <<
302 hexeditor to make sure you got the dd param <<
303 <<
304 6. Copy it to /lib/firmware under the expected <<
305 <<
306 7. If the extracted firmware works, you can us <<
307 efi_embedded_fw_desc struct to describe it, <<
308 to get the sha256sum to put in the sha256 f <<
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.