1 menu "Memory management options"
7 bool "Support for memory management hardware"
11 Some SH processors (such as SH-2/SH-2A) lack an MMU. In order to
12 boot on these systems, this option must not be set.
14 On other systems (such as the SH-3 and 4) where an MMU exists,
15 turning this off will boot the kernel on these machines with the
16 MMU implicitly switched off.
20 default "0x80000000" if MMU && SUPERH32
21 default "0x20000000" if MMU && SUPERH64
24 config FORCE_MAX_ZONEORDER
25 int "Maximum zone order"
26 range 9 64 if PAGE_SIZE_16KB
27 default "9" if PAGE_SIZE_16KB
28 range 7 64 if PAGE_SIZE_64KB
29 default "7" if PAGE_SIZE_64KB
34 The kernel memory allocator divides physically contiguous memory
35 blocks into "zones", where each zone is a power of two number of
36 pages. This option selects the largest power of two that the kernel
37 keeps in the memory allocator. If you need to allocate very large
38 blocks of physically contiguous memory, then you may need to
41 This config option is actually maximum order plus one. For example,
42 a value of 11 means that the largest free memory block is 2^10 pages.
44 The page size is not necessarily 4KB. Keep this in mind when
45 choosing a value for this option.
48 hex "Physical memory start address"
51 Computers built with Hitachi SuperH processors always
52 map the ROM starting at address zero. But the processor
53 does not specify the range that RAM takes.
55 The physical memory (RAM) start address will be automatically
56 set to 08000000. Other platforms, such as the Solution Engine
57 boards typically map RAM at 0C000000.
59 Tweak this only when porting to a new machine which does not
60 already have a defconfig. Changing it from the known correct
61 value on any of the known systems will only lead to disaster.
64 hex "Physical memory size"
67 This sets the default memory size assumed by your SH kernel. It can
68 be overridden as normal by the 'mem=' argument on the kernel command
69 line. If unsure, consult your board specifications or just leave it
70 as 0x04000000 which was the default value before this became
73 # Physical addressing modes
84 bool "Support 32-bit physical addressing through PMB"
85 depends on MMU && EXPERIMENTAL && CPU_SH4A
88 If you say Y here, physical addressing will be extended to
89 32-bits through the SH-4A PMB. If this is not set, legacy
90 29-bit physical addressing will be used.
93 prompt "PMB handling type"
99 depends on MMU && EXPERIMENTAL && CPU_SH4A
101 If you say Y here, physical addressing will be extended to
102 32-bits through the SH-4A PMB. If this is not set, legacy
103 29-bit physical addressing will be used.
107 depends on MMU && EXPERIMENTAL && CPU_SH4A
110 If this option is enabled, fixed PMB mappings are inherited
111 from the boot loader, and the kernel does not attempt dynamic
112 management. This is the closest to legacy 29-bit physical mode,
113 and allows systems to support up to 512MiB of system memory.
118 bool "Enable extended TLB mode"
119 depends on (CPU_SHX2 || CPU_SHX3) && MMU && EXPERIMENTAL
121 Selecting this option will enable the extended mode of the SH-X2
122 TLB. For legacy SH-X behaviour and interoperability, say N. For
123 all of the fun new features and a willingless to submit bug reports,
127 bool "Support vsyscall page"
128 depends on MMU && (CPU_SH3 || CPU_SH4)
131 This will enable support for the kernel mapping a vDSO page
132 in process space, and subsequently handing down the entry point
133 to the libc through the ELF auxiliary vector.
135 From the kernel side this is used for the signal trampoline.
136 For systems with an MMU that can afford to give up a page,
137 (the default value) say Y.
140 bool "Non Uniform Memory Access (NUMA) Support"
141 depends on MMU && SYS_SUPPORTS_NUMA && EXPERIMENTAL
144 Some SH systems have many various memories scattered around
145 the address space, each with varying latencies. This enables
146 support for these blocks by binding them to nodes and allowing
147 memory policies to be used for prioritizing and controlling
148 allocation behaviour.
152 default "3" if CPU_SUBTYPE_SHX3
154 depends on NEED_MULTIPLE_NODES
156 config ARCH_FLATMEM_ENABLE
160 config ARCH_SPARSEMEM_ENABLE
162 select SPARSEMEM_STATIC
164 config ARCH_SPARSEMEM_DEFAULT
167 config MAX_ACTIVE_REGIONS
169 default "6" if (CPU_SUBTYPE_SHX3 && SPARSEMEM)
170 default "2" if SPARSEMEM && (CPU_SUBTYPE_SH7722 || \
174 config ARCH_POPULATES_NODE_MAP
177 config ARCH_SELECT_MEMORY_MODEL
180 config ARCH_ENABLE_MEMORY_HOTPLUG
182 depends on SPARSEMEM && MMU
184 config ARCH_ENABLE_MEMORY_HOTREMOVE
186 depends on SPARSEMEM && MMU
188 config ARCH_MEMORY_PROBE
190 depends on MEMORY_HOTPLUG
193 prompt "Kernel page size"
194 default PAGE_SIZE_8KB if X2TLB
195 default PAGE_SIZE_4KB
199 depends on !MMU || !X2TLB
201 This is the default page size used by all SuperH CPUs.
205 depends on !MMU || X2TLB
207 This enables 8kB pages as supported by SH-X2 and later MMUs.
209 config PAGE_SIZE_16KB
213 This enables 16kB pages on MMU-less SH systems.
215 config PAGE_SIZE_64KB
217 depends on !MMU || CPU_SH4 || CPU_SH5
219 This enables support for 64kB pages, possible on all SH-4
225 prompt "HugeTLB page size"
226 depends on HUGETLB_PAGE
227 default HUGETLB_PAGE_SIZE_1MB if PAGE_SIZE_64KB
228 default HUGETLB_PAGE_SIZE_64K
230 config HUGETLB_PAGE_SIZE_64K
232 depends on !PAGE_SIZE_64KB
234 config HUGETLB_PAGE_SIZE_256K
238 config HUGETLB_PAGE_SIZE_1MB
241 config HUGETLB_PAGE_SIZE_4MB
245 config HUGETLB_PAGE_SIZE_64MB
249 config HUGETLB_PAGE_SIZE_512MB
258 bool "Multi-core scheduler support"
262 Multi-core scheduler support improves the CPU scheduler's decision
263 making when dealing with multi-core CPU chips at a cost of slightly
264 increased overhead in some places. If unsure say N here.
268 menu "Cache configuration"
270 config SH7705_CACHE_32KB
271 bool "Enable 32KB cache size for SH7705"
272 depends on CPU_SUBTYPE_SH7705
277 default CACHE_WRITEBACK if CPU_SH2A || CPU_SH3 || CPU_SH4 || CPU_SH5
278 default CACHE_WRITETHROUGH if (CPU_SH2 && !CPU_SH2A)
280 config CACHE_WRITEBACK
283 config CACHE_WRITETHROUGH
286 Selecting this option will configure the caches in write-through
287 mode, as opposed to the default write-back configuration.
289 Since there's sill some aliasing issues on SH-4, this option will
290 unfortunately still require the majority of flushing functions to
291 be implemented to deal with aliasing.