1 # SPDX-License-Identifier: GPL-2.0
2 menu "Generic Driver Options"
5 bool "Support for uevent helper"
8 The uevent helper program is forked by the kernel for
10 Before the switch to the netlink-based uevent source, this was
11 used to hook hotplug scripts into kernel device events. It
12 usually pointed to a shell script at /sbin/hotplug.
13 This should not be used today, because usual systems create
14 many events at bootup or device discovery in a very short time
15 frame. One forked process per event can create so many processes
16 that it creates a high system load, or on smaller systems
17 it is known to create out-of-memory situations during bootup.
19 config UEVENT_HELPER_PATH
20 string "path to uevent helper"
21 depends on UEVENT_HELPER
24 To disable user space helper program execution at by default
25 specify an empty string here. This setting can still be altered
26 via /proc/sys/kernel/hotplug or via /sys/kernel/uevent_helper
30 bool "Maintain a devtmpfs filesystem to mount at /dev"
32 This creates a tmpfs/ramfs filesystem instance early at bootup.
33 In this filesystem, the kernel driver core maintains device
34 nodes with their default names and permissions for all
35 registered devices with an assigned major/minor number.
36 Userspace can modify the filesystem content as needed, add
37 symlinks, and apply needed permissions.
38 It provides a fully functional /dev directory, where usually
39 udev runs on top, managing permissions and adding meaningful
41 In very limited environments, it may provide a sufficient
42 functional /dev without any further help. It also allows simple
43 rescue systems, and reliably handles dynamic major/minor numbers.
45 Notice: if CONFIG_TMPFS isn't enabled, the simpler ramfs
46 file system will be used instead.
49 bool "Automount devtmpfs at /dev, after the kernel mounted the rootfs"
52 This will instruct the kernel to automatically mount the
53 devtmpfs filesystem at /dev, directly after the kernel has
54 mounted the root filesystem. The behavior can be overridden
55 with the commandline parameter: devtmpfs.mount=0|1.
56 This option does not affect initramfs based booting, here
57 the devtmpfs filesystem always needs to be mounted manually
58 after the rootfs is mounted.
59 With this option enabled, it allows to bring up a system in
60 rescue mode with init=/bin/sh, even when the /dev directory
61 on the rootfs is completely empty.
64 bool "Select only drivers that don't need compile-time external firmware"
67 Select this option if you don't have magic firmware for drivers that
72 config PREVENT_FIRMWARE_BUILD
73 bool "Prevent firmware from being built"
76 Say yes to avoid building firmware. Firmware is usually shipped
77 with the driver and only when updating the firmware should a
79 If unsure, say Y here.
82 tristate "Userspace firmware loading support" if EXPERT
85 This option is provided for the case where none of the in-tree modules
86 require userspace firmware loading support, but a module built
89 config FIRMWARE_IN_KERNEL
90 bool "Include in-kernel firmware blobs in kernel binary"
94 Various drivers in the kernel source tree may require firmware,
95 which is generally available in your distribution's linux-firmware
98 The linux-firmware package should install firmware into
99 /lib/firmware/ on your system, so they can be loaded by userspace
102 Enabling this option will build each required firmware blob
103 specified by EXTRA_FIRMWARE into the kernel directly, where
104 request_firmware() will find them without having to call out to
105 userspace. This may be useful if your root file system requires a
106 device that uses such firmware and you do not wish to use an
109 This single option controls the inclusion of firmware for
110 every driver that uses request_firmware(), which avoids a
111 proliferation of 'Include firmware for xxx device' options.
113 Say 'N' and let firmware be loaded from userspace.
115 config EXTRA_FIRMWARE
116 string "External firmware blobs to build into the kernel binary"
119 This option allows firmware to be built into the kernel for the case
120 where the user either cannot or doesn't want to provide it from
121 userspace at runtime (for example, when the firmware in question is
122 required for accessing the boot device, and the user doesn't want to
125 This option is a string and takes the (space-separated) names of the
126 firmware files -- the same names that appear in MODULE_FIRMWARE()
127 and request_firmware() in the source. These files should exist under
128 the directory specified by the EXTRA_FIRMWARE_DIR option, which is
129 by default the firmware subdirectory of the kernel source tree.
131 For example, you might set CONFIG_EXTRA_FIRMWARE="usb8388.bin", copy
132 the usb8388.bin file into the firmware directory, and build the kernel.
133 Then any request_firmware("usb8388.bin") will be satisfied internally
134 without needing to call out to userspace.
136 WARNING: If you include additional firmware files into your binary
137 kernel image that are not available under the terms of the GPL,
138 then it may be a violation of the GPL to distribute the resulting
139 image since it combines both GPL and non-GPL work. You should
140 consult a lawyer of your own before distributing such an image.
142 config EXTRA_FIRMWARE_DIR
143 string "Firmware blobs root directory"
144 depends on EXTRA_FIRMWARE != ""
145 default "/lib/firmware"
147 This option controls the directory in which the kernel build system
148 looks for the firmware files listed in the EXTRA_FIRMWARE option.
150 config FW_LOADER_USER_HELPER
153 config FW_LOADER_USER_HELPER_FALLBACK
154 bool "Fallback user-helper invocation for firmware loading"
156 select FW_LOADER_USER_HELPER
158 This option enables / disables the invocation of user-helper
159 (e.g. udev) for loading firmware files as a fallback after the
160 direct file loading in kernel fails. The user-mode helper is
161 no longer required unless you have a special firmware file that
162 resides in a non-standard path. Moreover, the udev support has
163 been deprecated upstream.
165 If you are unsure about this, say N here.
167 config WANT_DEV_COREDUMP
170 Drivers should "select" this option if they desire to use the
171 device coredump mechanism.
173 config ALLOW_DEV_COREDUMP
174 bool "Allow device coredump" if EXPERT
177 This option controls if the device coredump mechanism is available or
178 not; if disabled, the mechanism will be omitted even if drivers that
179 can use it are enabled.
180 Say 'N' for more sensitive systems or systems that don't want
181 to ever access the information to not have the code, nor keep any
188 default y if WANT_DEV_COREDUMP
189 depends on ALLOW_DEV_COREDUMP
192 bool "Driver Core verbose debug messages"
193 depends on DEBUG_KERNEL
195 Say Y here if you want the Driver core to produce a bunch of
196 debug messages to the system log. Select this if you are having a
197 problem with the driver core and want to see more of what is
200 If you are unsure about this, say N here.
203 bool "Managed device resources verbose debug messages"
204 depends on DEBUG_KERNEL
206 This option enables kernel parameter devres.log. If set to
207 non-zero, devres debug messages are printed. Select this if
208 you are having a problem with devres or want to debug
209 resource management for a managed device. devres.log can be
210 switched on and off from sysfs node.
212 If you are unsure about this, Say N here.
214 config DEBUG_TEST_DRIVER_REMOVE
215 bool "Test driver remove calls during probe (UNSTABLE)"
216 depends on DEBUG_KERNEL
218 Say Y here if you want the Driver core to test driver remove functions
219 by calling probe, remove, probe. This tests the remove path without
220 having to unbind the driver or unload the driver module.
222 This option is expected to find errors and may render your system
223 unusable. You should say N here unless you are explicitly looking to
224 test this functionality.
226 source "drivers/base/test/Kconfig"
228 config SYS_HYPERVISOR
232 config GENERIC_CPU_DEVICES
236 config GENERIC_CPU_AUTOPROBE
239 config GENERIC_CPU_VULNERABILITIES
246 source "drivers/base/regmap/Kconfig"
248 config DMA_SHARED_BUFFER
254 This option enables the framework for buffer-sharing between
255 multiple drivers. A buffer is associated with a file using driver
256 APIs extension; the file's descriptor can then be passed on to other
259 config DMA_FENCE_TRACE
260 bool "Enable verbose DMA_FENCE_TRACE messages"
261 depends on DMA_SHARED_BUFFER
263 Enable the DMA_FENCE_TRACE printks. This will add extra
264 spam to the console log, but will make it easier to diagnose
265 lockup related problems for dma-buffers shared across multiple
269 bool "DMA Contiguous Memory Allocator"
270 depends on HAVE_DMA_CONTIGUOUS && CMA
272 This enables the Contiguous Memory Allocator which allows drivers
273 to allocate big physically-contiguous blocks of memory for use with
274 hardware components that do not support I/O map nor scatter-gather.
276 You can disable CMA by specifying "cma=0" on the kernel's command
279 For more information see <include/linux/dma-contiguous.h>.
283 comment "Default contiguous memory area size:"
285 config CMA_SIZE_MBYTES
286 int "Size in Mega Bytes"
287 depends on !CMA_SIZE_SEL_PERCENTAGE
291 Defines the size (in MiB) of the default memory area for Contiguous
292 Memory Allocator. If the size of 0 is selected, CMA is disabled by
293 default, but it can be enabled by passing cma=size[MG] to the kernel.
296 config CMA_SIZE_PERCENTAGE
297 int "Percentage of total memory"
298 depends on !CMA_SIZE_SEL_MBYTES
302 Defines the size of the default memory area for Contiguous Memory
303 Allocator as a percentage of the total memory in the system.
304 If 0 percent is selected, CMA is disabled by default, but it can be
305 enabled by passing cma=size[MG] to the kernel.
308 prompt "Selected region size"
309 default CMA_SIZE_SEL_MBYTES
311 config CMA_SIZE_SEL_MBYTES
312 bool "Use mega bytes value only"
314 config CMA_SIZE_SEL_PERCENTAGE
315 bool "Use percentage value only"
317 config CMA_SIZE_SEL_MIN
318 bool "Use lower value (minimum)"
320 config CMA_SIZE_SEL_MAX
321 bool "Use higher value (maximum)"
326 int "Maximum PAGE_SIZE order of alignment for contiguous buffers"
330 DMA mapping framework by default aligns all buffers to the smallest
331 PAGE_SIZE order which is greater than or equal to the requested buffer
332 size. This works well for buffers up to a few hundreds kilobytes, but
333 for larger buffers it just a memory waste. With this parameter you can
334 specify the maximum PAGE_SIZE order for contiguous buffers. Larger
335 buffers will be aligned only to this specified order. The order is
336 expressed as a power of two multiplied by the PAGE_SIZE.
338 For example, if your system defaults to 4KiB pages, the order value
339 of 8 means that the buffers will be aligned up to 1MiB only.
341 If unsure, leave the default value "8".
345 config GENERIC_ARCH_TOPOLOGY
348 Enable support for architectures common topology code: e.g., parsing
349 CPU capacity information from DT, usage of such information for
350 appropriate scaling, sysfs interface for changing capacity values at