1 menu "Generic Driver Options"
4 bool "Support for uevent helper"
7 The uevent helper program is forked by the kernel for
9 Before the switch to the netlink-based uevent source, this was
10 used to hook hotplug scripts into kernel device events. It
11 usually pointed to a shell script at /sbin/hotplug.
12 This should not be used today, because usual systems create
13 many events at bootup or device discovery in a very short time
14 frame. One forked process per event can create so many processes
15 that it creates a high system load, or on smaller systems
16 it is known to create out-of-memory situations during bootup.
18 config UEVENT_HELPER_PATH
19 string "path to uevent helper"
20 depends on UEVENT_HELPER
23 To disable user space helper program execution at by default
24 specify an empty string here. This setting can still be altered
25 via /proc/sys/kernel/hotplug or via /sys/kernel/uevent_helper
29 bool "Maintain a devtmpfs filesystem to mount at /dev"
31 This creates a tmpfs/ramfs filesystem instance early at bootup.
32 In this filesystem, the kernel driver core maintains device
33 nodes with their default names and permissions for all
34 registered devices with an assigned major/minor number.
35 Userspace can modify the filesystem content as needed, add
36 symlinks, and apply needed permissions.
37 It provides a fully functional /dev directory, where usually
38 udev runs on top, managing permissions and adding meaningful
40 In very limited environments, it may provide a sufficient
41 functional /dev without any further help. It also allows simple
42 rescue systems, and reliably handles dynamic major/minor numbers.
44 Notice: if CONFIG_TMPFS isn't enabled, the simpler ramfs
45 file system will be used instead.
48 bool "Automount devtmpfs at /dev, after the kernel mounted the rootfs"
51 This will instruct the kernel to automatically mount the
52 devtmpfs filesystem at /dev, directly after the kernel has
53 mounted the root filesystem. The behavior can be overridden
54 with the commandline parameter: devtmpfs.mount=0|1.
55 This option does not affect initramfs based booting, here
56 the devtmpfs filesystem always needs to be mounted manually
57 after the rootfs is mounted.
58 With this option enabled, it allows to bring up a system in
59 rescue mode with init=/bin/sh, even when the /dev directory
60 on the rootfs is completely empty.
63 bool "Select only drivers that don't need compile-time external firmware"
66 Select this option if you don't have magic firmware for drivers that
71 config PREVENT_FIRMWARE_BUILD
72 bool "Prevent firmware from being built"
75 Say yes to avoid building firmware. Firmware is usually shipped
76 with the driver and only when updating the firmware should a
78 If unsure, say Y here.
81 tristate "Userspace firmware loading support" if EXPERT
84 This option is provided for the case where none of the in-tree modules
85 require userspace firmware loading support, but a module built
88 config FIRMWARE_IN_KERNEL
89 bool "Include in-kernel firmware blobs in kernel binary"
93 The kernel source tree includes a number of firmware 'blobs'
94 that are used by various drivers. The recommended way to
95 use these is to run "make firmware_install", which, after
96 converting ihex files to binary, copies all of the needed
97 binary files in firmware/ to /lib/firmware/ on your system so
98 that they can be loaded by userspace helpers on request.
100 Enabling this option will build each required firmware blob
101 into the kernel directly, where request_firmware() will find
102 them without having to call out to userspace. This may be
103 useful if your root file system requires a device that uses
104 such firmware and do not wish to use an initrd.
106 This single option controls the inclusion of firmware for
107 every driver that uses request_firmware() and ships its
108 firmware in the kernel source tree, which avoids a
109 proliferation of 'Include firmware for xxx device' options.
111 Say 'N' and let firmware be loaded from userspace.
113 config EXTRA_FIRMWARE
114 string "External firmware blobs to build into the kernel binary"
117 This option allows firmware to be built into the kernel for the case
118 where the user either cannot or doesn't want to provide it from
119 userspace at runtime (for example, when the firmware in question is
120 required for accessing the boot device, and the user doesn't want to
123 This option is a string and takes the (space-separated) names of the
124 firmware files -- the same names that appear in MODULE_FIRMWARE()
125 and request_firmware() in the source. These files should exist under
126 the directory specified by the EXTRA_FIRMWARE_DIR option, which is
127 by default the firmware subdirectory of the kernel source tree.
129 For example, you might set CONFIG_EXTRA_FIRMWARE="usb8388.bin", copy
130 the usb8388.bin file into the firmware directory, and build the kernel.
131 Then any request_firmware("usb8388.bin") will be satisfied internally
132 without needing to call out to userspace.
134 WARNING: If you include additional firmware files into your binary
135 kernel image that are not available under the terms of the GPL,
136 then it may be a violation of the GPL to distribute the resulting
137 image since it combines both GPL and non-GPL work. You should
138 consult a lawyer of your own before distributing such an image.
140 config EXTRA_FIRMWARE_DIR
141 string "Firmware blobs root directory"
142 depends on EXTRA_FIRMWARE != ""
145 This option controls the directory in which the kernel build system
146 looks for the firmware files listed in the EXTRA_FIRMWARE option.
147 The default is firmware/ in the kernel source tree, but by changing
148 this option you can point it elsewhere, such as /lib/firmware/ or
149 some other directory containing the firmware files.
151 config FW_LOADER_USER_HELPER
154 config FW_LOADER_USER_HELPER_FALLBACK
155 bool "Fallback user-helper invocation for firmware loading"
157 select FW_LOADER_USER_HELPER
159 This option enables / disables the invocation of user-helper
160 (e.g. udev) for loading firmware files as a fallback after the
161 direct file loading in kernel fails. The user-mode helper is
162 no longer required unless you have a special firmware file that
163 resides in a non-standard path. Moreover, the udev support has
164 been deprecated upstream.
166 If you are unsure about this, say N here.
168 config WANT_DEV_COREDUMP
171 Drivers should "select" this option if they desire to use the
172 device coredump mechanism.
174 config ALLOW_DEV_COREDUMP
175 bool "Allow device coredump" if EXPERT
178 This option controls if the device coredump mechanism is available or
179 not; if disabled, the mechanism will be omitted even if drivers that
180 can use it are enabled.
181 Say 'N' for more sensitive systems or systems that don't want
182 to ever access the information to not have the code, nor keep any
189 default y if WANT_DEV_COREDUMP
190 depends on ALLOW_DEV_COREDUMP
193 bool "Driver Core verbose debug messages"
194 depends on DEBUG_KERNEL
196 Say Y here if you want the Driver core to produce a bunch of
197 debug messages to the system log. Select this if you are having a
198 problem with the driver core and want to see more of what is
201 If you are unsure about this, say N here.
204 bool "Managed device resources verbose debug messages"
205 depends on DEBUG_KERNEL
207 This option enables kernel parameter devres.log. If set to
208 non-zero, devres debug messages are printed. Select this if
209 you are having a problem with devres or want to debug
210 resource management for a managed device. devres.log can be
211 switched on and off from sysfs node.
213 If you are unsure about this, Say N here.
215 config SYS_HYPERVISOR
219 config GENERIC_CPU_DEVICES
223 config GENERIC_CPU_AUTOPROBE
229 source "drivers/base/regmap/Kconfig"
231 config DMA_SHARED_BUFFER
236 This option enables the framework for buffer-sharing between
237 multiple drivers. A buffer is associated with a file using driver
238 APIs extension; the file's descriptor can then be passed on to other
242 bool "Enable verbose FENCE_TRACE messages"
243 depends on DMA_SHARED_BUFFER
245 Enable the FENCE_TRACE printks. This will add extra
246 spam to the console log, but will make it easier to diagnose
247 lockup related problems for dma-buffers shared across multiple
251 bool "DMA Contiguous Memory Allocator"
252 depends on HAVE_DMA_CONTIGUOUS && CMA
254 This enables the Contiguous Memory Allocator which allows drivers
255 to allocate big physically-contiguous blocks of memory for use with
256 hardware components that do not support I/O map nor scatter-gather.
258 You can disable CMA by specifying "cma=0" on the kernel's command
261 For more information see <include/linux/dma-contiguous.h>.
265 comment "Default contiguous memory area size:"
267 config CMA_SIZE_MBYTES
268 int "Size in Mega Bytes"
269 depends on !CMA_SIZE_SEL_PERCENTAGE
273 Defines the size (in MiB) of the default memory area for Contiguous
274 Memory Allocator. If the size of 0 is selected, CMA is disabled by
275 default, but it can be enabled by passing cma=size[MG] to the kernel.
278 config CMA_SIZE_PERCENTAGE
279 int "Percentage of total memory"
280 depends on !CMA_SIZE_SEL_MBYTES
284 Defines the size of the default memory area for Contiguous Memory
285 Allocator as a percentage of the total memory in the system.
286 If 0 percent is selected, CMA is disabled by default, but it can be
287 enabled by passing cma=size[MG] to the kernel.
290 prompt "Selected region size"
291 default CMA_SIZE_SEL_MBYTES
293 config CMA_SIZE_SEL_MBYTES
294 bool "Use mega bytes value only"
296 config CMA_SIZE_SEL_PERCENTAGE
297 bool "Use percentage value only"
299 config CMA_SIZE_SEL_MIN
300 bool "Use lower value (minimum)"
302 config CMA_SIZE_SEL_MAX
303 bool "Use higher value (maximum)"
308 int "Maximum PAGE_SIZE order of alignment for contiguous buffers"
312 DMA mapping framework by default aligns all buffers to the smallest
313 PAGE_SIZE order which is greater than or equal to the requested buffer
314 size. This works well for buffers up to a few hundreds kilobytes, but
315 for larger buffers it just a memory waste. With this parameter you can
316 specify the maximum PAGE_SIZE order for contiguous buffers. Larger
317 buffers will be aligned only to this specified order. The order is
318 expressed as a power of two multiplied by the PAGE_SIZE.
320 For example, if your system defaults to 4KiB pages, the order value
321 of 8 means that the buffers will be aligned up to 1MiB only.
323 If unsure, leave the default value "8".