1 # SPDX-License-Identifier: GPL-2.0
4 select ARCH_HAS_SG_CHAIN
5 select ARCH_HAS_SYNC_DMA_FOR_CPU
6 select ARCH_HAS_SYNC_DMA_FOR_DEVICE
7 select ARCH_NO_COHERENT_DMA_MMAP if !MMU
8 select ARCH_WANT_FRAME_POINTERS
9 select ARCH_WANT_IPC_PARSE_VERSION
10 select BUILDTIME_EXTABLE_SORT
11 select CLONE_BACKWARDS
14 select GENERIC_ATOMIC64
15 select GENERIC_CLOCKEVENTS
16 select GENERIC_IRQ_SHOW
17 select GENERIC_PCI_IOMAP
18 select GENERIC_SCHED_CLOCK
19 select GENERIC_STRNCPY_FROM_USER if KASAN
20 select HAVE_ARCH_KASAN if MMU
21 select HAVE_DEBUG_KMEMLEAK
22 select HAVE_DMA_CONTIGUOUS
23 select HAVE_EXIT_THREAD
24 select HAVE_FUNCTION_TRACER
25 select HAVE_FUTEX_CMPXCHG if !MMU
26 select HAVE_HW_BREAKPOINT if PERF_EVENTS
27 select HAVE_IRQ_TIME_ACCOUNTING
30 select HAVE_PERF_EVENTS
31 select HAVE_STACKPROTECTOR
33 select MODULES_USE_ELF_RELA
34 select PERF_USE_VMALLOC
37 Xtensa processors are 32-bit RISC machines designed by Tensilica
38 primarily for embedded systems. These processors are both
39 configurable and extensible. The Linux port to the Xtensa
40 architecture supports all processor configurations and extensions,
41 with reasonable minimum requirements. The Xtensa Linux project has
42 a home page at <http://www.linux-xtensa.org/>.
44 config RWSEM_XCHGADD_ALGORITHM
47 config GENERIC_HWEIGHT
50 config ARCH_HAS_ILOG2_U32
53 config ARCH_HAS_ILOG2_U64
63 config LOCKDEP_SUPPORT
66 config STACKTRACE_SUPPORT
69 config TRACE_IRQFLAGS_SUPPORT
75 config HAVE_XTENSA_GPIO32
78 config KASAN_SHADOW_OFFSET
82 menu "Processor type and features"
85 prompt "Xtensa Processor Configuration"
86 default XTENSA_VARIANT_FSF
88 config XTENSA_VARIANT_FSF
89 bool "fsf - default (not generic) configuration"
92 config XTENSA_VARIANT_DC232B
93 bool "dc232b - Diamond 232L Standard Core Rev.B (LE)"
95 select HAVE_XTENSA_GPIO32
97 This variant refers to Tensilica's Diamond 232L Standard core Rev.B (LE).
99 config XTENSA_VARIANT_DC233C
100 bool "dc233c - Diamond 233L Standard Core Rev.C (LE)"
102 select HAVE_XTENSA_GPIO32
104 This variant refers to Tensilica's Diamond 233L Standard core Rev.C (LE).
106 config XTENSA_VARIANT_CUSTOM
107 bool "Custom Xtensa processor configuration"
108 select HAVE_XTENSA_GPIO32
110 Select this variant to use a custom Xtensa processor configuration.
111 You will be prompted for a processor variant CORENAME.
114 config XTENSA_VARIANT_CUSTOM_NAME
115 string "Xtensa Processor Custom Core Variant Name"
116 depends on XTENSA_VARIANT_CUSTOM
118 Provide the name of a custom Xtensa processor variant.
119 This CORENAME selects arch/xtensa/variant/CORENAME.
120 Dont forget you have to select MMU if you have one.
122 config XTENSA_VARIANT_NAME
124 default "dc232b" if XTENSA_VARIANT_DC232B
125 default "dc233c" if XTENSA_VARIANT_DC233C
126 default "fsf" if XTENSA_VARIANT_FSF
127 default XTENSA_VARIANT_CUSTOM_NAME if XTENSA_VARIANT_CUSTOM
129 config XTENSA_VARIANT_MMU
130 bool "Core variant has a Full MMU (TLB, Pages, Protection, etc)"
131 depends on XTENSA_VARIANT_CUSTOM
135 Build a Conventional Kernel with full MMU support,
136 ie: it supports a TLB with auto-loading, page protection.
138 config XTENSA_VARIANT_HAVE_PERF_EVENTS
139 bool "Core variant has Performance Monitor Module"
140 depends on XTENSA_VARIANT_CUSTOM
143 Enable if core variant has Performance Monitor Module with
144 External Registers Interface.
148 config XTENSA_FAKE_NMI
149 bool "Treat PMM IRQ as NMI"
150 depends on XTENSA_VARIANT_HAVE_PERF_EVENTS
153 If PMM IRQ is the only IRQ at EXCM level it is safe to
154 treat it as NMI, which improves accuracy of profiling.
156 If there are other interrupts at or above PMM IRQ priority level
157 but not above the EXCM level, PMM IRQ still may be treated as NMI,
158 but only if these IRQs are not used. There will be a build warning
159 saying that this is not safe, and a bugcheck if one of these IRQs
164 config XTENSA_UNALIGNED_USER
165 bool "Unaligned memory access in use space"
167 The Xtensa architecture currently does not handle unaligned
168 memory accesses in hardware but through an exception handler.
169 Per default, unaligned memory accesses are disabled in user space.
171 Say Y here to enable unaligned memory access in user space.
174 bool "System Supports SMP (MX)"
175 depends on XTENSA_VARIANT_CUSTOM
178 This option is use to indicate that the system-on-a-chip (SOC)
179 supports Multiprocessing. Multiprocessor support implemented above
180 the CPU core definition and currently needs to be selected manually.
182 Multiprocessor support in implemented with external cache and
183 interrupt controllers.
185 The MX interrupt distributer adds Interprocessor Interrupts
186 and causes the IRQ numbers to be increased by 4 for devices
187 like the open cores ethernet driver and the serial interface.
189 You still have to select "Enable SMP" to enable SMP on this SOC.
192 bool "Enable Symmetric multi-processing support"
194 select GENERIC_SMP_IDLE_THREAD
196 Enabled SMP Software; allows more than one CPU/CORE
197 to be activated during startup.
201 int "Maximum number of CPUs (2-32)"
206 bool "Enable CPU hotplug support"
209 Say Y here to allow turning CPUs off and on. CPUs can be
210 controlled through /sys/devices/system/cpu.
212 Say N if you want to disable CPU hotplug.
214 config INITIALIZE_XTENSA_MMU_INSIDE_VMLINUX
215 bool "Initialize Xtensa MMU inside the Linux kernel code"
216 depends on !XTENSA_VARIANT_FSF && !XTENSA_VARIANT_DC232B
217 default y if XTENSA_VARIANT_DC233C || XTENSA_VARIANT_CUSTOM
219 Earlier version initialized the MMU in the exception vector
220 before jumping to _startup in head.S and had an advantage that
221 it was possible to place a software breakpoint at 'reset' and
222 then enter your normal kernel breakpoints once the MMU was mapped
223 to the kernel mappings (0XC0000000).
225 This unfortunately won't work for U-Boot and likely also wont
226 work for using KEXEC to have a hot kernel ready for doing a
229 So now the MMU is initialized in head.S but it's necessary to
230 use hardware breakpoints (gdb 'hbreak' cmd) to break at _startup.
231 xt-gdb can't place a Software Breakpoint in the 0XD region prior
232 to mapping the MMU and after mapping even if the area of low memory
233 was mapped gdb wouldn't remove the breakpoint on hitting it as the
234 PC wouldn't match. Since Hardware Breakpoints are recommended for
235 Linux configurations it seems reasonable to just assume they exist
236 and leave this older mechanism for unfortunate souls that choose
237 not to follow Tensilica's recommendation.
239 Selecting this will cause U-Boot to set the KERNEL Load and Entry
240 address at 0x00003000 instead of the mapped std of 0xD0003000.
244 config MEMMAP_CACHEATTR
245 hex "Cache attributes for the memory address space"
249 These cache attributes are set up for noMMU systems. Each hex digit
250 specifies cache attributes for the corresponding 512MB memory
251 region: bits 0..3 -- for addresses 0x00000000..0x1fffffff,
252 bits 4..7 -- for addresses 0x20000000..0x3fffffff, and so on.
254 Cache attribute values are specific for the MMU type, so e.g.
255 for region protection MMUs: 2 is cache bypass, 4 is WB cached,
256 1 is WT cached, f is illegal. For ful MMU: bit 0 makes it executable,
257 bit 1 makes it writable, bits 2..3 meaning is 0: cache bypass,
258 1: WB cache, 2: WT cache, 3: special (c and e are illegal, f is
262 hex "Physical address of the KSEG mapping"
263 depends on INITIALIZE_XTENSA_MMU_INSIDE_VMLINUX && MMU
266 This is the physical address where KSEG is mapped. Please refer to
267 the chosen KSEG layout help for the required address alignment.
268 Unpacked kernel image (including vectors) must be located completely
270 Physical memory below this address is not available to linux.
272 If unsure, leave the default value here.
274 config KERNEL_LOAD_ADDRESS
275 hex "Kernel load address"
276 default 0x60003000 if !MMU
277 default 0x00003000 if MMU && INITIALIZE_XTENSA_MMU_INSIDE_VMLINUX
278 default 0xd0003000 if MMU && !INITIALIZE_XTENSA_MMU_INSIDE_VMLINUX
280 This is the address where the kernel is loaded.
281 It is virtual address for MMUv2 configurations and physical address
282 for all other configurations.
284 If unsure, leave the default value here.
286 config VECTORS_OFFSET
287 hex "Kernel vectors offset"
290 This is the offset of the kernel image from the relocatable vectors
293 If unsure, leave the default value here.
298 default XTENSA_KSEG_MMU_V2
300 config XTENSA_KSEG_MMU_V2
301 bool "MMUv2: 128MB cached + 128MB uncached"
303 MMUv2 compatible kernel memory map: TLB way 5 maps 128MB starting
304 at KSEG_PADDR to 0xd0000000 with cache and to 0xd8000000
306 KSEG_PADDR must be aligned to 128MB.
308 config XTENSA_KSEG_256M
309 bool "256MB cached + 256MB uncached"
310 depends on INITIALIZE_XTENSA_MMU_INSIDE_VMLINUX
312 TLB way 6 maps 256MB starting at KSEG_PADDR to 0xb0000000
313 with cache and to 0xc0000000 without cache.
314 KSEG_PADDR must be aligned to 256MB.
316 config XTENSA_KSEG_512M
317 bool "512MB cached + 512MB uncached"
318 depends on INITIALIZE_XTENSA_MMU_INSIDE_VMLINUX
320 TLB way 6 maps 512MB starting at KSEG_PADDR to 0xa0000000
321 with cache and to 0xc0000000 without cache.
322 KSEG_PADDR must be aligned to 256MB.
327 bool "High Memory Support"
330 Linux can use the full amount of RAM in the system by
331 default. However, the default MMUv2 setup only maps the
332 lowermost 128 MB of memory linearly to the areas starting
333 at 0xd0000000 (cached) and 0xd8000000 (uncached).
334 When there are more than 128 MB memory in the system not
335 all of it can be "permanently mapped" by the kernel.
336 The physical memory that's not permanently mapped is called
339 If you are compiling a kernel which will never run on a
340 machine with more than 128 MB total physical RAM, answer
345 config FAST_SYSCALL_XTENSA
346 bool "Enable fast atomic syscalls"
349 fast_syscall_xtensa is a syscall that can make atomic operations
350 on UP kernel when processor has no s32c1i support.
352 This syscall is deprecated. It may have issues when called with
353 invalid arguments. It is provided only for backwards compatibility.
354 Only enable it if your userspace software requires it.
358 config FAST_SYSCALL_SPILL_REGISTERS
359 bool "Enable spill registers syscall"
362 fast_syscall_spill_registers is a syscall that spills all active
363 register windows of a calling userspace task onto its stack.
365 This syscall is deprecated. It may have issues when called with
366 invalid arguments. It is provided only for backwards compatibility.
367 Only enable it if your userspace software requires it.
373 config XTENSA_CALIBRATE_CCOUNT
376 On some platforms (XT2000, for example), the CPU clock rate can
377 vary. The frequency can be determined, however, by measuring
378 against a well known, fixed frequency, such as an UART oscillator.
380 config SERIAL_CONSOLE
383 menu "Platform options"
386 prompt "Xtensa System Type"
387 default XTENSA_PLATFORM_ISS
389 config XTENSA_PLATFORM_ISS
391 select XTENSA_CALIBRATE_CCOUNT
392 select SERIAL_CONSOLE
394 ISS is an acronym for Tensilica's Instruction Set Simulator.
396 config XTENSA_PLATFORM_XT2000
400 XT2000 is the name of Tensilica's feature-rich emulation platform.
401 This hardware is capable of running a full Linux distribution.
403 config XTENSA_PLATFORM_XTFPGA
405 select ETHOC if ETHERNET
406 select PLATFORM_WANT_DEFAULT_MEM if !MMU
407 select SERIAL_CONSOLE
408 select XTENSA_CALIBRATE_CCOUNT
410 XTFPGA is the name of Tensilica board family (LX60, LX110, LX200, ML605).
411 This hardware is capable of running a full Linux distribution.
415 config PLATFORM_NR_IRQS
417 default 3 if XTENSA_PLATFORM_XT2000
420 config XTENSA_CPU_CLOCK
421 int "CPU clock rate [MHz]"
422 depends on !XTENSA_CALIBRATE_CCOUNT
425 config GENERIC_CALIBRATE_DELAY
426 bool "Auto calibration of the BogoMIPS value"
428 The BogoMIPS value can easily be derived from the CPU frequency.
431 bool "Default bootloader kernel arguments"
434 string "Initial kernel command string"
435 depends on CMDLINE_BOOL
436 default "console=ttyS0,38400 root=/dev/ram"
438 On some architectures (EBSA110 and CATS), there is currently no way
439 for the boot loader to pass arguments to the kernel. For these
440 architectures, you should supply some command-line options at build
441 time by entering them here. As a minimum, you should specify the
442 memory size and the root device (e.g., mem=64M root=/dev/nfs).
445 bool "Flattened Device Tree support"
447 select OF_EARLY_FLATTREE
448 select OF_RESERVED_MEM
450 Include support for flattened device tree machine descriptions.
453 string "DTB to build into the kernel image"
456 config PARSE_BOOTPARAM
457 bool "Parse bootparam block"
460 Parse parameters passed to the kernel from the bootloader. It may
461 be disabled if the kernel is known to run without the bootloader.
465 config BLK_DEV_SIMDISK
466 tristate "Host file-based simulated block device support"
468 depends on XTENSA_PLATFORM_ISS && BLOCK
470 Create block devices that map to files in the host file system.
471 Device binding to host file may be changed at runtime via proc
472 interface provided the device is not in use.
474 config BLK_DEV_SIMDISK_COUNT
475 int "Number of host file-based simulated block devices"
477 depends on BLK_DEV_SIMDISK
480 This is the default minimal number of created block devices.
481 Kernel/module parameter 'simdisk_count' may be used to change this
482 value at runtime. More file names (but no more than 10) may be
483 specified as parameters, simdisk_count grows accordingly.
485 config SIMDISK0_FILENAME
486 string "Host filename for the first simulated device"
487 depends on BLK_DEV_SIMDISK = y
490 Attach a first simdisk to a host file. Conventionally, this file
491 contains a root file system.
493 config SIMDISK1_FILENAME
494 string "Host filename for the second simulated device"
495 depends on BLK_DEV_SIMDISK = y && BLK_DEV_SIMDISK_COUNT != 1
498 Another simulated disk in a host file for a buildroot-independent
501 config FORCE_MAX_ZONEORDER
502 int "Maximum zone order"
505 The kernel memory allocator divides physically contiguous memory
506 blocks into "zones", where each zone is a power of two number of
507 pages. This option selects the largest power of two that the kernel
508 keeps in the memory allocator. If you need to allocate very large
509 blocks of physically contiguous memory, then you may need to
512 This config option is actually maximum order plus one. For example,
513 a value of 11 means that the largest free memory block is 2^10 pages.
515 config PLATFORM_WANT_DEFAULT_MEM
518 config DEFAULT_MEM_START
520 prompt "PAGE_OFFSET/PHYS_OFFSET" if !MMU && PLATFORM_WANT_DEFAULT_MEM
521 default 0x60000000 if PLATFORM_WANT_DEFAULT_MEM
524 This is the base address used for both PAGE_OFFSET and PHYS_OFFSET
525 in noMMU configurations.
527 If unsure, leave the default value here.
530 bool "Enable XTFPGA LCD driver"
531 depends on XTENSA_PLATFORM_XTFPGA
534 There's a 2x16 LCD on most of XTFPGA boards, kernel may output
535 progress messages there during bootup/shutdown. It may be useful
536 during board bringup.
540 config XTFPGA_LCD_BASE_ADDR
541 hex "XTFPGA LCD base address"
542 depends on XTFPGA_LCD
545 Base address of the LCD controller inside KIO region.
546 Different boards from XTFPGA family have LCD controller at different
547 addresses. Please consult prototyping user guide for your board for
548 the correct address. Wrong address here may lead to hardware lockup.
550 config XTFPGA_LCD_8BIT_ACCESS
551 bool "Use 8-bit access to XTFPGA LCD"
552 depends on XTFPGA_LCD
555 LCD may be connected with 4- or 8-bit interface, 8-bit access may
556 only be used with 8-bit interface. Please consult prototyping user
557 guide for your board for the correct interface width.
561 menu "Power management options"
563 source "kernel/power/Kconfig"