Commit 0c02a20ff7695f9c54cc7c013dda326270ccdac8
1 parent
64de5af000
Exists in
master
and in
7 other branches
intel-iommu: Kill DMAR_BROKEN_GFX_WA option.
Just make it depend on BROKEN for now, in case people scream really loud about it (and because we might want to keep some of this logic for an upcoming BIOS workaround, so I don't just want to rip it out entirely just yet). But for graphics devices, it really ought to be unnecessary. Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Showing 2 changed files with 2 additions and 6 deletions Inline Diff
Documentation/Intel-IOMMU.txt
1 | Linux IOMMU Support | 1 | Linux IOMMU Support |
2 | =================== | 2 | =================== |
3 | 3 | ||
4 | The architecture spec can be obtained from the below location. | 4 | The architecture spec can be obtained from the below location. |
5 | 5 | ||
6 | http://www.intel.com/technology/virtualization/ | 6 | http://www.intel.com/technology/virtualization/ |
7 | 7 | ||
8 | This guide gives a quick cheat sheet for some basic understanding. | 8 | This guide gives a quick cheat sheet for some basic understanding. |
9 | 9 | ||
10 | Some Keywords | 10 | Some Keywords |
11 | 11 | ||
12 | DMAR - DMA remapping | 12 | DMAR - DMA remapping |
13 | DRHD - DMA Engine Reporting Structure | 13 | DRHD - DMA Engine Reporting Structure |
14 | RMRR - Reserved memory Region Reporting Structure | 14 | RMRR - Reserved memory Region Reporting Structure |
15 | ZLR - Zero length reads from PCI devices | 15 | ZLR - Zero length reads from PCI devices |
16 | IOVA - IO Virtual address. | 16 | IOVA - IO Virtual address. |
17 | 17 | ||
18 | Basic stuff | 18 | Basic stuff |
19 | ----------- | 19 | ----------- |
20 | 20 | ||
21 | ACPI enumerates and lists the different DMA engines in the platform, and | 21 | ACPI enumerates and lists the different DMA engines in the platform, and |
22 | device scope relationships between PCI devices and which DMA engine controls | 22 | device scope relationships between PCI devices and which DMA engine controls |
23 | them. | 23 | them. |
24 | 24 | ||
25 | What is RMRR? | 25 | What is RMRR? |
26 | ------------- | 26 | ------------- |
27 | 27 | ||
28 | There are some devices the BIOS controls, for e.g USB devices to perform | 28 | There are some devices the BIOS controls, for e.g USB devices to perform |
29 | PS2 emulation. The regions of memory used for these devices are marked | 29 | PS2 emulation. The regions of memory used for these devices are marked |
30 | reserved in the e820 map. When we turn on DMA translation, DMA to those | 30 | reserved in the e820 map. When we turn on DMA translation, DMA to those |
31 | regions will fail. Hence BIOS uses RMRR to specify these regions along with | 31 | regions will fail. Hence BIOS uses RMRR to specify these regions along with |
32 | devices that need to access these regions. OS is expected to setup | 32 | devices that need to access these regions. OS is expected to setup |
33 | unity mappings for these regions for these devices to access these regions. | 33 | unity mappings for these regions for these devices to access these regions. |
34 | 34 | ||
35 | How is IOVA generated? | 35 | How is IOVA generated? |
36 | --------------------- | 36 | --------------------- |
37 | 37 | ||
38 | Well behaved drivers call pci_map_*() calls before sending command to device | 38 | Well behaved drivers call pci_map_*() calls before sending command to device |
39 | that needs to perform DMA. Once DMA is completed and mapping is no longer | 39 | that needs to perform DMA. Once DMA is completed and mapping is no longer |
40 | required, device performs a pci_unmap_*() calls to unmap the region. | 40 | required, device performs a pci_unmap_*() calls to unmap the region. |
41 | 41 | ||
42 | The Intel IOMMU driver allocates a virtual address per domain. Each PCIE | 42 | The Intel IOMMU driver allocates a virtual address per domain. Each PCIE |
43 | device has its own domain (hence protection). Devices under p2p bridges | 43 | device has its own domain (hence protection). Devices under p2p bridges |
44 | share the virtual address with all devices under the p2p bridge due to | 44 | share the virtual address with all devices under the p2p bridge due to |
45 | transaction id aliasing for p2p bridges. | 45 | transaction id aliasing for p2p bridges. |
46 | 46 | ||
47 | IOVA generation is pretty generic. We used the same technique as vmalloc() | 47 | IOVA generation is pretty generic. We used the same technique as vmalloc() |
48 | but these are not global address spaces, but separate for each domain. | 48 | but these are not global address spaces, but separate for each domain. |
49 | Different DMA engines may support different number of domains. | 49 | Different DMA engines may support different number of domains. |
50 | 50 | ||
51 | We also allocate guard pages with each mapping, so we can attempt to catch | 51 | We also allocate guard pages with each mapping, so we can attempt to catch |
52 | any overflow that might happen. | 52 | any overflow that might happen. |
53 | 53 | ||
54 | 54 | ||
55 | Graphics Problems? | 55 | Graphics Problems? |
56 | ------------------ | 56 | ------------------ |
57 | If you encounter issues with graphics devices, you can try adding | 57 | If you encounter issues with graphics devices, you can try adding |
58 | option intel_iommu=igfx_off to turn off the integrated graphics engine. | 58 | option intel_iommu=igfx_off to turn off the integrated graphics engine. |
59 | 59 | If this fixes anything, please ensure you file a bug reporting the problem. | |
60 | If it happens to be a PCI device included in the INCLUDE_ALL Engine, | ||
61 | then try enabling CONFIG_DMAR_GFX_WA to setup a 1-1 map. We hear | ||
62 | graphics drivers may be in process of using DMA api's in the near | ||
63 | future and at that time this option can be yanked out. | ||
64 | 60 | ||
65 | Some exceptions to IOVA | 61 | Some exceptions to IOVA |
66 | ----------------------- | 62 | ----------------------- |
67 | Interrupt ranges are not address translated, (0xfee00000 - 0xfeefffff). | 63 | Interrupt ranges are not address translated, (0xfee00000 - 0xfeefffff). |
68 | The same is true for peer to peer transactions. Hence we reserve the | 64 | The same is true for peer to peer transactions. Hence we reserve the |
69 | address from PCI MMIO ranges so they are not allocated for IOVA addresses. | 65 | address from PCI MMIO ranges so they are not allocated for IOVA addresses. |
70 | 66 | ||
71 | 67 | ||
72 | Fault reporting | 68 | Fault reporting |
73 | --------------- | 69 | --------------- |
74 | When errors are reported, the DMA engine signals via an interrupt. The fault | 70 | When errors are reported, the DMA engine signals via an interrupt. The fault |
75 | reason and device that caused it with fault reason is printed on console. | 71 | reason and device that caused it with fault reason is printed on console. |
76 | 72 | ||
77 | See below for sample. | 73 | See below for sample. |
78 | 74 | ||
79 | 75 | ||
80 | Boot Message Sample | 76 | Boot Message Sample |
81 | ------------------- | 77 | ------------------- |
82 | 78 | ||
83 | Something like this gets printed indicating presence of DMAR tables | 79 | Something like this gets printed indicating presence of DMAR tables |
84 | in ACPI. | 80 | in ACPI. |
85 | 81 | ||
86 | ACPI: DMAR (v001 A M I OEMDMAR 0x00000001 MSFT 0x00000097) @ 0x000000007f5b5ef0 | 82 | ACPI: DMAR (v001 A M I OEMDMAR 0x00000001 MSFT 0x00000097) @ 0x000000007f5b5ef0 |
87 | 83 | ||
88 | When DMAR is being processed and initialized by ACPI, prints DMAR locations | 84 | When DMAR is being processed and initialized by ACPI, prints DMAR locations |
89 | and any RMRR's processed. | 85 | and any RMRR's processed. |
90 | 86 | ||
91 | ACPI DMAR:Host address width 36 | 87 | ACPI DMAR:Host address width 36 |
92 | ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed90000 | 88 | ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed90000 |
93 | ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed91000 | 89 | ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed91000 |
94 | ACPI DMAR:DRHD (flags: 0x00000001)base: 0x00000000fed93000 | 90 | ACPI DMAR:DRHD (flags: 0x00000001)base: 0x00000000fed93000 |
95 | ACPI DMAR:RMRR base: 0x00000000000ed000 end: 0x00000000000effff | 91 | ACPI DMAR:RMRR base: 0x00000000000ed000 end: 0x00000000000effff |
96 | ACPI DMAR:RMRR base: 0x000000007f600000 end: 0x000000007fffffff | 92 | ACPI DMAR:RMRR base: 0x000000007f600000 end: 0x000000007fffffff |
97 | 93 | ||
98 | When DMAR is enabled for use, you will notice.. | 94 | When DMAR is enabled for use, you will notice.. |
99 | 95 | ||
100 | PCI-DMA: Using DMAR IOMMU | 96 | PCI-DMA: Using DMAR IOMMU |
101 | 97 | ||
102 | Fault reporting | 98 | Fault reporting |
103 | --------------- | 99 | --------------- |
104 | 100 | ||
105 | DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 | 101 | DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 |
106 | DMAR:[fault reason 05] PTE Write access is not set | 102 | DMAR:[fault reason 05] PTE Write access is not set |
107 | DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 | 103 | DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 |
108 | DMAR:[fault reason 05] PTE Write access is not set | 104 | DMAR:[fault reason 05] PTE Write access is not set |
109 | 105 | ||
110 | TBD | 106 | TBD |
111 | ---- | 107 | ---- |
112 | 108 | ||
113 | - For compatibility testing, could use unity map domain for all devices, just | 109 | - For compatibility testing, could use unity map domain for all devices, just |
114 | provide a 1-1 for all useful memory under a single domain for all devices. | 110 | provide a 1-1 for all useful memory under a single domain for all devices. |
115 | - API for paravirt ops for abstracting functionality for VMM folks. | 111 | - API for paravirt ops for abstracting functionality for VMM folks. |
116 | 112 |
arch/x86/Kconfig
1 | # x86 configuration | 1 | # x86 configuration |
2 | mainmenu "Linux Kernel Configuration for x86" | 2 | mainmenu "Linux Kernel Configuration for x86" |
3 | 3 | ||
4 | # Select 32 or 64 bit | 4 | # Select 32 or 64 bit |
5 | config 64BIT | 5 | config 64BIT |
6 | bool "64-bit kernel" if ARCH = "x86" | 6 | bool "64-bit kernel" if ARCH = "x86" |
7 | default ARCH = "x86_64" | 7 | default ARCH = "x86_64" |
8 | ---help--- | 8 | ---help--- |
9 | Say yes to build a 64-bit kernel - formerly known as x86_64 | 9 | Say yes to build a 64-bit kernel - formerly known as x86_64 |
10 | Say no to build a 32-bit kernel - formerly known as i386 | 10 | Say no to build a 32-bit kernel - formerly known as i386 |
11 | 11 | ||
12 | config X86_32 | 12 | config X86_32 |
13 | def_bool !64BIT | 13 | def_bool !64BIT |
14 | 14 | ||
15 | config X86_64 | 15 | config X86_64 |
16 | def_bool 64BIT | 16 | def_bool 64BIT |
17 | 17 | ||
18 | ### Arch settings | 18 | ### Arch settings |
19 | config X86 | 19 | config X86 |
20 | def_bool y | 20 | def_bool y |
21 | select HAVE_AOUT if X86_32 | 21 | select HAVE_AOUT if X86_32 |
22 | select HAVE_READQ | 22 | select HAVE_READQ |
23 | select HAVE_WRITEQ | 23 | select HAVE_WRITEQ |
24 | select HAVE_UNSTABLE_SCHED_CLOCK | 24 | select HAVE_UNSTABLE_SCHED_CLOCK |
25 | select HAVE_IDE | 25 | select HAVE_IDE |
26 | select HAVE_OPROFILE | 26 | select HAVE_OPROFILE |
27 | select HAVE_IOREMAP_PROT | 27 | select HAVE_IOREMAP_PROT |
28 | select HAVE_KPROBES | 28 | select HAVE_KPROBES |
29 | select ARCH_WANT_OPTIONAL_GPIOLIB | 29 | select ARCH_WANT_OPTIONAL_GPIOLIB |
30 | select ARCH_WANT_FRAME_POINTERS | 30 | select ARCH_WANT_FRAME_POINTERS |
31 | select HAVE_DMA_ATTRS | 31 | select HAVE_DMA_ATTRS |
32 | select HAVE_KRETPROBES | 32 | select HAVE_KRETPROBES |
33 | select HAVE_FTRACE_MCOUNT_RECORD | 33 | select HAVE_FTRACE_MCOUNT_RECORD |
34 | select HAVE_DYNAMIC_FTRACE | 34 | select HAVE_DYNAMIC_FTRACE |
35 | select HAVE_FUNCTION_TRACER | 35 | select HAVE_FUNCTION_TRACER |
36 | select HAVE_FUNCTION_GRAPH_TRACER | 36 | select HAVE_FUNCTION_GRAPH_TRACER |
37 | select HAVE_FUNCTION_GRAPH_FP_TEST | 37 | select HAVE_FUNCTION_GRAPH_FP_TEST |
38 | select HAVE_FUNCTION_TRACE_MCOUNT_TEST | 38 | select HAVE_FUNCTION_TRACE_MCOUNT_TEST |
39 | select HAVE_FTRACE_NMI_ENTER if DYNAMIC_FTRACE | 39 | select HAVE_FTRACE_NMI_ENTER if DYNAMIC_FTRACE |
40 | select HAVE_FTRACE_SYSCALLS | 40 | select HAVE_FTRACE_SYSCALLS |
41 | select HAVE_KVM | 41 | select HAVE_KVM |
42 | select HAVE_ARCH_KGDB | 42 | select HAVE_ARCH_KGDB |
43 | select HAVE_ARCH_TRACEHOOK | 43 | select HAVE_ARCH_TRACEHOOK |
44 | select HAVE_GENERIC_DMA_COHERENT if X86_32 | 44 | select HAVE_GENERIC_DMA_COHERENT if X86_32 |
45 | select HAVE_EFFICIENT_UNALIGNED_ACCESS | 45 | select HAVE_EFFICIENT_UNALIGNED_ACCESS |
46 | select USER_STACKTRACE_SUPPORT | 46 | select USER_STACKTRACE_SUPPORT |
47 | select HAVE_DMA_API_DEBUG | 47 | select HAVE_DMA_API_DEBUG |
48 | select HAVE_KERNEL_GZIP | 48 | select HAVE_KERNEL_GZIP |
49 | select HAVE_KERNEL_BZIP2 | 49 | select HAVE_KERNEL_BZIP2 |
50 | select HAVE_KERNEL_LZMA | 50 | select HAVE_KERNEL_LZMA |
51 | select HAVE_ARCH_KMEMCHECK | 51 | select HAVE_ARCH_KMEMCHECK |
52 | 52 | ||
53 | config OUTPUT_FORMAT | 53 | config OUTPUT_FORMAT |
54 | string | 54 | string |
55 | default "elf32-i386" if X86_32 | 55 | default "elf32-i386" if X86_32 |
56 | default "elf64-x86-64" if X86_64 | 56 | default "elf64-x86-64" if X86_64 |
57 | 57 | ||
58 | config ARCH_DEFCONFIG | 58 | config ARCH_DEFCONFIG |
59 | string | 59 | string |
60 | default "arch/x86/configs/i386_defconfig" if X86_32 | 60 | default "arch/x86/configs/i386_defconfig" if X86_32 |
61 | default "arch/x86/configs/x86_64_defconfig" if X86_64 | 61 | default "arch/x86/configs/x86_64_defconfig" if X86_64 |
62 | 62 | ||
63 | config GENERIC_TIME | 63 | config GENERIC_TIME |
64 | def_bool y | 64 | def_bool y |
65 | 65 | ||
66 | config GENERIC_CMOS_UPDATE | 66 | config GENERIC_CMOS_UPDATE |
67 | def_bool y | 67 | def_bool y |
68 | 68 | ||
69 | config CLOCKSOURCE_WATCHDOG | 69 | config CLOCKSOURCE_WATCHDOG |
70 | def_bool y | 70 | def_bool y |
71 | 71 | ||
72 | config GENERIC_CLOCKEVENTS | 72 | config GENERIC_CLOCKEVENTS |
73 | def_bool y | 73 | def_bool y |
74 | 74 | ||
75 | config GENERIC_CLOCKEVENTS_BROADCAST | 75 | config GENERIC_CLOCKEVENTS_BROADCAST |
76 | def_bool y | 76 | def_bool y |
77 | depends on X86_64 || (X86_32 && X86_LOCAL_APIC) | 77 | depends on X86_64 || (X86_32 && X86_LOCAL_APIC) |
78 | 78 | ||
79 | config LOCKDEP_SUPPORT | 79 | config LOCKDEP_SUPPORT |
80 | def_bool y | 80 | def_bool y |
81 | 81 | ||
82 | config STACKTRACE_SUPPORT | 82 | config STACKTRACE_SUPPORT |
83 | def_bool y | 83 | def_bool y |
84 | 84 | ||
85 | config HAVE_LATENCYTOP_SUPPORT | 85 | config HAVE_LATENCYTOP_SUPPORT |
86 | def_bool y | 86 | def_bool y |
87 | 87 | ||
88 | config FAST_CMPXCHG_LOCAL | 88 | config FAST_CMPXCHG_LOCAL |
89 | bool | 89 | bool |
90 | default y | 90 | default y |
91 | 91 | ||
92 | config MMU | 92 | config MMU |
93 | def_bool y | 93 | def_bool y |
94 | 94 | ||
95 | config ZONE_DMA | 95 | config ZONE_DMA |
96 | def_bool y | 96 | def_bool y |
97 | 97 | ||
98 | config SBUS | 98 | config SBUS |
99 | bool | 99 | bool |
100 | 100 | ||
101 | config GENERIC_ISA_DMA | 101 | config GENERIC_ISA_DMA |
102 | def_bool y | 102 | def_bool y |
103 | 103 | ||
104 | config GENERIC_IOMAP | 104 | config GENERIC_IOMAP |
105 | def_bool y | 105 | def_bool y |
106 | 106 | ||
107 | config GENERIC_BUG | 107 | config GENERIC_BUG |
108 | def_bool y | 108 | def_bool y |
109 | depends on BUG | 109 | depends on BUG |
110 | select GENERIC_BUG_RELATIVE_POINTERS if X86_64 | 110 | select GENERIC_BUG_RELATIVE_POINTERS if X86_64 |
111 | 111 | ||
112 | config GENERIC_BUG_RELATIVE_POINTERS | 112 | config GENERIC_BUG_RELATIVE_POINTERS |
113 | bool | 113 | bool |
114 | 114 | ||
115 | config GENERIC_HWEIGHT | 115 | config GENERIC_HWEIGHT |
116 | def_bool y | 116 | def_bool y |
117 | 117 | ||
118 | config GENERIC_GPIO | 118 | config GENERIC_GPIO |
119 | bool | 119 | bool |
120 | 120 | ||
121 | config ARCH_MAY_HAVE_PC_FDC | 121 | config ARCH_MAY_HAVE_PC_FDC |
122 | def_bool y | 122 | def_bool y |
123 | 123 | ||
124 | config RWSEM_GENERIC_SPINLOCK | 124 | config RWSEM_GENERIC_SPINLOCK |
125 | def_bool !X86_XADD | 125 | def_bool !X86_XADD |
126 | 126 | ||
127 | config RWSEM_XCHGADD_ALGORITHM | 127 | config RWSEM_XCHGADD_ALGORITHM |
128 | def_bool X86_XADD | 128 | def_bool X86_XADD |
129 | 129 | ||
130 | config ARCH_HAS_CPU_IDLE_WAIT | 130 | config ARCH_HAS_CPU_IDLE_WAIT |
131 | def_bool y | 131 | def_bool y |
132 | 132 | ||
133 | config GENERIC_CALIBRATE_DELAY | 133 | config GENERIC_CALIBRATE_DELAY |
134 | def_bool y | 134 | def_bool y |
135 | 135 | ||
136 | config GENERIC_TIME_VSYSCALL | 136 | config GENERIC_TIME_VSYSCALL |
137 | bool | 137 | bool |
138 | default X86_64 | 138 | default X86_64 |
139 | 139 | ||
140 | config ARCH_HAS_CPU_RELAX | 140 | config ARCH_HAS_CPU_RELAX |
141 | def_bool y | 141 | def_bool y |
142 | 142 | ||
143 | config ARCH_HAS_DEFAULT_IDLE | 143 | config ARCH_HAS_DEFAULT_IDLE |
144 | def_bool y | 144 | def_bool y |
145 | 145 | ||
146 | config ARCH_HAS_CACHE_LINE_SIZE | 146 | config ARCH_HAS_CACHE_LINE_SIZE |
147 | def_bool y | 147 | def_bool y |
148 | 148 | ||
149 | config HAVE_SETUP_PER_CPU_AREA | 149 | config HAVE_SETUP_PER_CPU_AREA |
150 | def_bool y | 150 | def_bool y |
151 | 151 | ||
152 | config HAVE_DYNAMIC_PER_CPU_AREA | 152 | config HAVE_DYNAMIC_PER_CPU_AREA |
153 | def_bool y | 153 | def_bool y |
154 | 154 | ||
155 | config HAVE_CPUMASK_OF_CPU_MAP | 155 | config HAVE_CPUMASK_OF_CPU_MAP |
156 | def_bool X86_64_SMP | 156 | def_bool X86_64_SMP |
157 | 157 | ||
158 | config ARCH_HIBERNATION_POSSIBLE | 158 | config ARCH_HIBERNATION_POSSIBLE |
159 | def_bool y | 159 | def_bool y |
160 | 160 | ||
161 | config ARCH_SUSPEND_POSSIBLE | 161 | config ARCH_SUSPEND_POSSIBLE |
162 | def_bool y | 162 | def_bool y |
163 | 163 | ||
164 | config ZONE_DMA32 | 164 | config ZONE_DMA32 |
165 | bool | 165 | bool |
166 | default X86_64 | 166 | default X86_64 |
167 | 167 | ||
168 | config ARCH_POPULATES_NODE_MAP | 168 | config ARCH_POPULATES_NODE_MAP |
169 | def_bool y | 169 | def_bool y |
170 | 170 | ||
171 | config AUDIT_ARCH | 171 | config AUDIT_ARCH |
172 | bool | 172 | bool |
173 | default X86_64 | 173 | default X86_64 |
174 | 174 | ||
175 | config ARCH_SUPPORTS_OPTIMIZED_INLINING | 175 | config ARCH_SUPPORTS_OPTIMIZED_INLINING |
176 | def_bool y | 176 | def_bool y |
177 | 177 | ||
178 | config ARCH_SUPPORTS_DEBUG_PAGEALLOC | 178 | config ARCH_SUPPORTS_DEBUG_PAGEALLOC |
179 | def_bool y | 179 | def_bool y |
180 | 180 | ||
181 | # Use the generic interrupt handling code in kernel/irq/: | 181 | # Use the generic interrupt handling code in kernel/irq/: |
182 | config GENERIC_HARDIRQS | 182 | config GENERIC_HARDIRQS |
183 | bool | 183 | bool |
184 | default y | 184 | default y |
185 | 185 | ||
186 | config GENERIC_HARDIRQS_NO__DO_IRQ | 186 | config GENERIC_HARDIRQS_NO__DO_IRQ |
187 | def_bool y | 187 | def_bool y |
188 | 188 | ||
189 | config GENERIC_IRQ_PROBE | 189 | config GENERIC_IRQ_PROBE |
190 | bool | 190 | bool |
191 | default y | 191 | default y |
192 | 192 | ||
193 | config GENERIC_PENDING_IRQ | 193 | config GENERIC_PENDING_IRQ |
194 | bool | 194 | bool |
195 | depends on GENERIC_HARDIRQS && SMP | 195 | depends on GENERIC_HARDIRQS && SMP |
196 | default y | 196 | default y |
197 | 197 | ||
198 | config USE_GENERIC_SMP_HELPERS | 198 | config USE_GENERIC_SMP_HELPERS |
199 | def_bool y | 199 | def_bool y |
200 | depends on SMP | 200 | depends on SMP |
201 | 201 | ||
202 | config X86_32_SMP | 202 | config X86_32_SMP |
203 | def_bool y | 203 | def_bool y |
204 | depends on X86_32 && SMP | 204 | depends on X86_32 && SMP |
205 | 205 | ||
206 | config X86_64_SMP | 206 | config X86_64_SMP |
207 | def_bool y | 207 | def_bool y |
208 | depends on X86_64 && SMP | 208 | depends on X86_64 && SMP |
209 | 209 | ||
210 | config X86_HT | 210 | config X86_HT |
211 | bool | 211 | bool |
212 | depends on SMP | 212 | depends on SMP |
213 | default y | 213 | default y |
214 | 214 | ||
215 | config X86_TRAMPOLINE | 215 | config X86_TRAMPOLINE |
216 | bool | 216 | bool |
217 | depends on SMP || (64BIT && ACPI_SLEEP) | 217 | depends on SMP || (64BIT && ACPI_SLEEP) |
218 | default y | 218 | default y |
219 | 219 | ||
220 | config X86_32_LAZY_GS | 220 | config X86_32_LAZY_GS |
221 | def_bool y | 221 | def_bool y |
222 | depends on X86_32 && !CC_STACKPROTECTOR | 222 | depends on X86_32 && !CC_STACKPROTECTOR |
223 | 223 | ||
224 | config KTIME_SCALAR | 224 | config KTIME_SCALAR |
225 | def_bool X86_32 | 225 | def_bool X86_32 |
226 | source "init/Kconfig" | 226 | source "init/Kconfig" |
227 | source "kernel/Kconfig.freezer" | 227 | source "kernel/Kconfig.freezer" |
228 | 228 | ||
229 | menu "Processor type and features" | 229 | menu "Processor type and features" |
230 | 230 | ||
231 | source "kernel/time/Kconfig" | 231 | source "kernel/time/Kconfig" |
232 | 232 | ||
233 | config SMP | 233 | config SMP |
234 | bool "Symmetric multi-processing support" | 234 | bool "Symmetric multi-processing support" |
235 | ---help--- | 235 | ---help--- |
236 | This enables support for systems with more than one CPU. If you have | 236 | This enables support for systems with more than one CPU. If you have |
237 | a system with only one CPU, like most personal computers, say N. If | 237 | a system with only one CPU, like most personal computers, say N. If |
238 | you have a system with more than one CPU, say Y. | 238 | you have a system with more than one CPU, say Y. |
239 | 239 | ||
240 | If you say N here, the kernel will run on single and multiprocessor | 240 | If you say N here, the kernel will run on single and multiprocessor |
241 | machines, but will use only one CPU of a multiprocessor machine. If | 241 | machines, but will use only one CPU of a multiprocessor machine. If |
242 | you say Y here, the kernel will run on many, but not all, | 242 | you say Y here, the kernel will run on many, but not all, |
243 | singleprocessor machines. On a singleprocessor machine, the kernel | 243 | singleprocessor machines. On a singleprocessor machine, the kernel |
244 | will run faster if you say N here. | 244 | will run faster if you say N here. |
245 | 245 | ||
246 | Note that if you say Y here and choose architecture "586" or | 246 | Note that if you say Y here and choose architecture "586" or |
247 | "Pentium" under "Processor family", the kernel will not work on 486 | 247 | "Pentium" under "Processor family", the kernel will not work on 486 |
248 | architectures. Similarly, multiprocessor kernels for the "PPro" | 248 | architectures. Similarly, multiprocessor kernels for the "PPro" |
249 | architecture may not work on all Pentium based boards. | 249 | architecture may not work on all Pentium based boards. |
250 | 250 | ||
251 | People using multiprocessor machines who say Y here should also say | 251 | People using multiprocessor machines who say Y here should also say |
252 | Y to "Enhanced Real Time Clock Support", below. The "Advanced Power | 252 | Y to "Enhanced Real Time Clock Support", below. The "Advanced Power |
253 | Management" code will be disabled if you say Y here. | 253 | Management" code will be disabled if you say Y here. |
254 | 254 | ||
255 | See also <file:Documentation/i386/IO-APIC.txt>, | 255 | See also <file:Documentation/i386/IO-APIC.txt>, |
256 | <file:Documentation/nmi_watchdog.txt> and the SMP-HOWTO available at | 256 | <file:Documentation/nmi_watchdog.txt> and the SMP-HOWTO available at |
257 | <http://www.tldp.org/docs.html#howto>. | 257 | <http://www.tldp.org/docs.html#howto>. |
258 | 258 | ||
259 | If you don't know what to do here, say N. | 259 | If you don't know what to do here, say N. |
260 | 260 | ||
261 | config X86_X2APIC | 261 | config X86_X2APIC |
262 | bool "Support x2apic" | 262 | bool "Support x2apic" |
263 | depends on X86_LOCAL_APIC && X86_64 && INTR_REMAP | 263 | depends on X86_LOCAL_APIC && X86_64 && INTR_REMAP |
264 | ---help--- | 264 | ---help--- |
265 | This enables x2apic support on CPUs that have this feature. | 265 | This enables x2apic support on CPUs that have this feature. |
266 | 266 | ||
267 | This allows 32-bit apic IDs (so it can support very large systems), | 267 | This allows 32-bit apic IDs (so it can support very large systems), |
268 | and accesses the local apic via MSRs not via mmio. | 268 | and accesses the local apic via MSRs not via mmio. |
269 | 269 | ||
270 | If you don't know what to do here, say N. | 270 | If you don't know what to do here, say N. |
271 | 271 | ||
272 | config SPARSE_IRQ | 272 | config SPARSE_IRQ |
273 | bool "Support sparse irq numbering" | 273 | bool "Support sparse irq numbering" |
274 | depends on PCI_MSI || HT_IRQ | 274 | depends on PCI_MSI || HT_IRQ |
275 | ---help--- | 275 | ---help--- |
276 | This enables support for sparse irqs. This is useful for distro | 276 | This enables support for sparse irqs. This is useful for distro |
277 | kernels that want to define a high CONFIG_NR_CPUS value but still | 277 | kernels that want to define a high CONFIG_NR_CPUS value but still |
278 | want to have low kernel memory footprint on smaller machines. | 278 | want to have low kernel memory footprint on smaller machines. |
279 | 279 | ||
280 | ( Sparse IRQs can also be beneficial on NUMA boxes, as they spread | 280 | ( Sparse IRQs can also be beneficial on NUMA boxes, as they spread |
281 | out the irq_desc[] array in a more NUMA-friendly way. ) | 281 | out the irq_desc[] array in a more NUMA-friendly way. ) |
282 | 282 | ||
283 | If you don't know what to do here, say N. | 283 | If you don't know what to do here, say N. |
284 | 284 | ||
285 | config NUMA_IRQ_DESC | 285 | config NUMA_IRQ_DESC |
286 | def_bool y | 286 | def_bool y |
287 | depends on SPARSE_IRQ && NUMA | 287 | depends on SPARSE_IRQ && NUMA |
288 | 288 | ||
289 | config X86_MPPARSE | 289 | config X86_MPPARSE |
290 | bool "Enable MPS table" if ACPI | 290 | bool "Enable MPS table" if ACPI |
291 | default y | 291 | default y |
292 | depends on X86_LOCAL_APIC | 292 | depends on X86_LOCAL_APIC |
293 | ---help--- | 293 | ---help--- |
294 | For old smp systems that do not have proper acpi support. Newer systems | 294 | For old smp systems that do not have proper acpi support. Newer systems |
295 | (esp with 64bit cpus) with acpi support, MADT and DSDT will override it | 295 | (esp with 64bit cpus) with acpi support, MADT and DSDT will override it |
296 | 296 | ||
297 | config X86_BIGSMP | 297 | config X86_BIGSMP |
298 | bool "Support for big SMP systems with more than 8 CPUs" | 298 | bool "Support for big SMP systems with more than 8 CPUs" |
299 | depends on X86_32 && SMP | 299 | depends on X86_32 && SMP |
300 | ---help--- | 300 | ---help--- |
301 | This option is needed for the systems that have more than 8 CPUs | 301 | This option is needed for the systems that have more than 8 CPUs |
302 | 302 | ||
303 | if X86_32 | 303 | if X86_32 |
304 | config X86_EXTENDED_PLATFORM | 304 | config X86_EXTENDED_PLATFORM |
305 | bool "Support for extended (non-PC) x86 platforms" | 305 | bool "Support for extended (non-PC) x86 platforms" |
306 | default y | 306 | default y |
307 | ---help--- | 307 | ---help--- |
308 | If you disable this option then the kernel will only support | 308 | If you disable this option then the kernel will only support |
309 | standard PC platforms. (which covers the vast majority of | 309 | standard PC platforms. (which covers the vast majority of |
310 | systems out there.) | 310 | systems out there.) |
311 | 311 | ||
312 | If you enable this option then you'll be able to select support | 312 | If you enable this option then you'll be able to select support |
313 | for the following (non-PC) 32 bit x86 platforms: | 313 | for the following (non-PC) 32 bit x86 platforms: |
314 | AMD Elan | 314 | AMD Elan |
315 | NUMAQ (IBM/Sequent) | 315 | NUMAQ (IBM/Sequent) |
316 | RDC R-321x SoC | 316 | RDC R-321x SoC |
317 | SGI 320/540 (Visual Workstation) | 317 | SGI 320/540 (Visual Workstation) |
318 | Summit/EXA (IBM x440) | 318 | Summit/EXA (IBM x440) |
319 | Unisys ES7000 IA32 series | 319 | Unisys ES7000 IA32 series |
320 | 320 | ||
321 | If you have one of these systems, or if you want to build a | 321 | If you have one of these systems, or if you want to build a |
322 | generic distribution kernel, say Y here - otherwise say N. | 322 | generic distribution kernel, say Y here - otherwise say N. |
323 | endif | 323 | endif |
324 | 324 | ||
325 | if X86_64 | 325 | if X86_64 |
326 | config X86_EXTENDED_PLATFORM | 326 | config X86_EXTENDED_PLATFORM |
327 | bool "Support for extended (non-PC) x86 platforms" | 327 | bool "Support for extended (non-PC) x86 platforms" |
328 | default y | 328 | default y |
329 | ---help--- | 329 | ---help--- |
330 | If you disable this option then the kernel will only support | 330 | If you disable this option then the kernel will only support |
331 | standard PC platforms. (which covers the vast majority of | 331 | standard PC platforms. (which covers the vast majority of |
332 | systems out there.) | 332 | systems out there.) |
333 | 333 | ||
334 | If you enable this option then you'll be able to select support | 334 | If you enable this option then you'll be able to select support |
335 | for the following (non-PC) 64 bit x86 platforms: | 335 | for the following (non-PC) 64 bit x86 platforms: |
336 | ScaleMP vSMP | 336 | ScaleMP vSMP |
337 | SGI Ultraviolet | 337 | SGI Ultraviolet |
338 | 338 | ||
339 | If you have one of these systems, or if you want to build a | 339 | If you have one of these systems, or if you want to build a |
340 | generic distribution kernel, say Y here - otherwise say N. | 340 | generic distribution kernel, say Y here - otherwise say N. |
341 | endif | 341 | endif |
342 | # This is an alphabetically sorted list of 64 bit extended platforms | 342 | # This is an alphabetically sorted list of 64 bit extended platforms |
343 | # Please maintain the alphabetic order if and when there are additions | 343 | # Please maintain the alphabetic order if and when there are additions |
344 | 344 | ||
345 | config X86_VSMP | 345 | config X86_VSMP |
346 | bool "ScaleMP vSMP" | 346 | bool "ScaleMP vSMP" |
347 | select PARAVIRT | 347 | select PARAVIRT |
348 | depends on X86_64 && PCI | 348 | depends on X86_64 && PCI |
349 | depends on X86_EXTENDED_PLATFORM | 349 | depends on X86_EXTENDED_PLATFORM |
350 | ---help--- | 350 | ---help--- |
351 | Support for ScaleMP vSMP systems. Say 'Y' here if this kernel is | 351 | Support for ScaleMP vSMP systems. Say 'Y' here if this kernel is |
352 | supposed to run on these EM64T-based machines. Only choose this option | 352 | supposed to run on these EM64T-based machines. Only choose this option |
353 | if you have one of these machines. | 353 | if you have one of these machines. |
354 | 354 | ||
355 | config X86_UV | 355 | config X86_UV |
356 | bool "SGI Ultraviolet" | 356 | bool "SGI Ultraviolet" |
357 | depends on X86_64 | 357 | depends on X86_64 |
358 | depends on X86_EXTENDED_PLATFORM | 358 | depends on X86_EXTENDED_PLATFORM |
359 | depends on NUMA | 359 | depends on NUMA |
360 | depends on X86_X2APIC | 360 | depends on X86_X2APIC |
361 | ---help--- | 361 | ---help--- |
362 | This option is needed in order to support SGI Ultraviolet systems. | 362 | This option is needed in order to support SGI Ultraviolet systems. |
363 | If you don't have one of these, you should say N here. | 363 | If you don't have one of these, you should say N here. |
364 | 364 | ||
365 | # Following is an alphabetically sorted list of 32 bit extended platforms | 365 | # Following is an alphabetically sorted list of 32 bit extended platforms |
366 | # Please maintain the alphabetic order if and when there are additions | 366 | # Please maintain the alphabetic order if and when there are additions |
367 | 367 | ||
368 | config X86_ELAN | 368 | config X86_ELAN |
369 | bool "AMD Elan" | 369 | bool "AMD Elan" |
370 | depends on X86_32 | 370 | depends on X86_32 |
371 | depends on X86_EXTENDED_PLATFORM | 371 | depends on X86_EXTENDED_PLATFORM |
372 | ---help--- | 372 | ---help--- |
373 | Select this for an AMD Elan processor. | 373 | Select this for an AMD Elan processor. |
374 | 374 | ||
375 | Do not use this option for K6/Athlon/Opteron processors! | 375 | Do not use this option for K6/Athlon/Opteron processors! |
376 | 376 | ||
377 | If unsure, choose "PC-compatible" instead. | 377 | If unsure, choose "PC-compatible" instead. |
378 | 378 | ||
379 | config X86_RDC321X | 379 | config X86_RDC321X |
380 | bool "RDC R-321x SoC" | 380 | bool "RDC R-321x SoC" |
381 | depends on X86_32 | 381 | depends on X86_32 |
382 | depends on X86_EXTENDED_PLATFORM | 382 | depends on X86_EXTENDED_PLATFORM |
383 | select M486 | 383 | select M486 |
384 | select X86_REBOOTFIXUPS | 384 | select X86_REBOOTFIXUPS |
385 | ---help--- | 385 | ---help--- |
386 | This option is needed for RDC R-321x system-on-chip, also known | 386 | This option is needed for RDC R-321x system-on-chip, also known |
387 | as R-8610-(G). | 387 | as R-8610-(G). |
388 | If you don't have one of these chips, you should say N here. | 388 | If you don't have one of these chips, you should say N here. |
389 | 389 | ||
390 | config X86_32_NON_STANDARD | 390 | config X86_32_NON_STANDARD |
391 | bool "Support non-standard 32-bit SMP architectures" | 391 | bool "Support non-standard 32-bit SMP architectures" |
392 | depends on X86_32 && SMP | 392 | depends on X86_32 && SMP |
393 | depends on X86_EXTENDED_PLATFORM | 393 | depends on X86_EXTENDED_PLATFORM |
394 | ---help--- | 394 | ---help--- |
395 | This option compiles in the NUMAQ, Summit, bigsmp, ES7000, default | 395 | This option compiles in the NUMAQ, Summit, bigsmp, ES7000, default |
396 | subarchitectures. It is intended for a generic binary kernel. | 396 | subarchitectures. It is intended for a generic binary kernel. |
397 | if you select them all, kernel will probe it one by one. and will | 397 | if you select them all, kernel will probe it one by one. and will |
398 | fallback to default. | 398 | fallback to default. |
399 | 399 | ||
400 | # Alphabetically sorted list of Non standard 32 bit platforms | 400 | # Alphabetically sorted list of Non standard 32 bit platforms |
401 | 401 | ||
402 | config X86_NUMAQ | 402 | config X86_NUMAQ |
403 | bool "NUMAQ (IBM/Sequent)" | 403 | bool "NUMAQ (IBM/Sequent)" |
404 | depends on X86_32_NON_STANDARD | 404 | depends on X86_32_NON_STANDARD |
405 | select NUMA | 405 | select NUMA |
406 | select X86_MPPARSE | 406 | select X86_MPPARSE |
407 | ---help--- | 407 | ---help--- |
408 | This option is used for getting Linux to run on a NUMAQ (IBM/Sequent) | 408 | This option is used for getting Linux to run on a NUMAQ (IBM/Sequent) |
409 | NUMA multiquad box. This changes the way that processors are | 409 | NUMA multiquad box. This changes the way that processors are |
410 | bootstrapped, and uses Clustered Logical APIC addressing mode instead | 410 | bootstrapped, and uses Clustered Logical APIC addressing mode instead |
411 | of Flat Logical. You will need a new lynxer.elf file to flash your | 411 | of Flat Logical. You will need a new lynxer.elf file to flash your |
412 | firmware with - send email to <Martin.Bligh@us.ibm.com>. | 412 | firmware with - send email to <Martin.Bligh@us.ibm.com>. |
413 | 413 | ||
414 | config X86_VISWS | 414 | config X86_VISWS |
415 | bool "SGI 320/540 (Visual Workstation)" | 415 | bool "SGI 320/540 (Visual Workstation)" |
416 | depends on X86_32 && PCI && X86_MPPARSE && PCI_GODIRECT | 416 | depends on X86_32 && PCI && X86_MPPARSE && PCI_GODIRECT |
417 | depends on X86_32_NON_STANDARD | 417 | depends on X86_32_NON_STANDARD |
418 | ---help--- | 418 | ---help--- |
419 | The SGI Visual Workstation series is an IA32-based workstation | 419 | The SGI Visual Workstation series is an IA32-based workstation |
420 | based on SGI systems chips with some legacy PC hardware attached. | 420 | based on SGI systems chips with some legacy PC hardware attached. |
421 | 421 | ||
422 | Say Y here to create a kernel to run on the SGI 320 or 540. | 422 | Say Y here to create a kernel to run on the SGI 320 or 540. |
423 | 423 | ||
424 | A kernel compiled for the Visual Workstation will run on general | 424 | A kernel compiled for the Visual Workstation will run on general |
425 | PCs as well. See <file:Documentation/sgi-visws.txt> for details. | 425 | PCs as well. See <file:Documentation/sgi-visws.txt> for details. |
426 | 426 | ||
427 | config X86_SUMMIT | 427 | config X86_SUMMIT |
428 | bool "Summit/EXA (IBM x440)" | 428 | bool "Summit/EXA (IBM x440)" |
429 | depends on X86_32_NON_STANDARD | 429 | depends on X86_32_NON_STANDARD |
430 | ---help--- | 430 | ---help--- |
431 | This option is needed for IBM systems that use the Summit/EXA chipset. | 431 | This option is needed for IBM systems that use the Summit/EXA chipset. |
432 | In particular, it is needed for the x440. | 432 | In particular, it is needed for the x440. |
433 | 433 | ||
434 | config X86_ES7000 | 434 | config X86_ES7000 |
435 | bool "Unisys ES7000 IA32 series" | 435 | bool "Unisys ES7000 IA32 series" |
436 | depends on X86_32_NON_STANDARD && X86_BIGSMP | 436 | depends on X86_32_NON_STANDARD && X86_BIGSMP |
437 | ---help--- | 437 | ---help--- |
438 | Support for Unisys ES7000 systems. Say 'Y' here if this kernel is | 438 | Support for Unisys ES7000 systems. Say 'Y' here if this kernel is |
439 | supposed to run on an IA32-based Unisys ES7000 system. | 439 | supposed to run on an IA32-based Unisys ES7000 system. |
440 | 440 | ||
441 | config SCHED_OMIT_FRAME_POINTER | 441 | config SCHED_OMIT_FRAME_POINTER |
442 | def_bool y | 442 | def_bool y |
443 | prompt "Single-depth WCHAN output" | 443 | prompt "Single-depth WCHAN output" |
444 | depends on X86 | 444 | depends on X86 |
445 | ---help--- | 445 | ---help--- |
446 | Calculate simpler /proc/<PID>/wchan values. If this option | 446 | Calculate simpler /proc/<PID>/wchan values. If this option |
447 | is disabled then wchan values will recurse back to the | 447 | is disabled then wchan values will recurse back to the |
448 | caller function. This provides more accurate wchan values, | 448 | caller function. This provides more accurate wchan values, |
449 | at the expense of slightly more scheduling overhead. | 449 | at the expense of slightly more scheduling overhead. |
450 | 450 | ||
451 | If in doubt, say "Y". | 451 | If in doubt, say "Y". |
452 | 452 | ||
453 | menuconfig PARAVIRT_GUEST | 453 | menuconfig PARAVIRT_GUEST |
454 | bool "Paravirtualized guest support" | 454 | bool "Paravirtualized guest support" |
455 | ---help--- | 455 | ---help--- |
456 | Say Y here to get to see options related to running Linux under | 456 | Say Y here to get to see options related to running Linux under |
457 | various hypervisors. This option alone does not add any kernel code. | 457 | various hypervisors. This option alone does not add any kernel code. |
458 | 458 | ||
459 | If you say N, all options in this submenu will be skipped and disabled. | 459 | If you say N, all options in this submenu will be skipped and disabled. |
460 | 460 | ||
461 | if PARAVIRT_GUEST | 461 | if PARAVIRT_GUEST |
462 | 462 | ||
463 | source "arch/x86/xen/Kconfig" | 463 | source "arch/x86/xen/Kconfig" |
464 | 464 | ||
465 | config VMI | 465 | config VMI |
466 | bool "VMI Guest support" | 466 | bool "VMI Guest support" |
467 | select PARAVIRT | 467 | select PARAVIRT |
468 | depends on X86_32 | 468 | depends on X86_32 |
469 | ---help--- | 469 | ---help--- |
470 | VMI provides a paravirtualized interface to the VMware ESX server | 470 | VMI provides a paravirtualized interface to the VMware ESX server |
471 | (it could be used by other hypervisors in theory too, but is not | 471 | (it could be used by other hypervisors in theory too, but is not |
472 | at the moment), by linking the kernel to a GPL-ed ROM module | 472 | at the moment), by linking the kernel to a GPL-ed ROM module |
473 | provided by the hypervisor. | 473 | provided by the hypervisor. |
474 | 474 | ||
475 | config KVM_CLOCK | 475 | config KVM_CLOCK |
476 | bool "KVM paravirtualized clock" | 476 | bool "KVM paravirtualized clock" |
477 | select PARAVIRT | 477 | select PARAVIRT |
478 | select PARAVIRT_CLOCK | 478 | select PARAVIRT_CLOCK |
479 | ---help--- | 479 | ---help--- |
480 | Turning on this option will allow you to run a paravirtualized clock | 480 | Turning on this option will allow you to run a paravirtualized clock |
481 | when running over the KVM hypervisor. Instead of relying on a PIT | 481 | when running over the KVM hypervisor. Instead of relying on a PIT |
482 | (or probably other) emulation by the underlying device model, the host | 482 | (or probably other) emulation by the underlying device model, the host |
483 | provides the guest with timing infrastructure such as time of day, and | 483 | provides the guest with timing infrastructure such as time of day, and |
484 | system time | 484 | system time |
485 | 485 | ||
486 | config KVM_GUEST | 486 | config KVM_GUEST |
487 | bool "KVM Guest support" | 487 | bool "KVM Guest support" |
488 | select PARAVIRT | 488 | select PARAVIRT |
489 | ---help--- | 489 | ---help--- |
490 | This option enables various optimizations for running under the KVM | 490 | This option enables various optimizations for running under the KVM |
491 | hypervisor. | 491 | hypervisor. |
492 | 492 | ||
493 | source "arch/x86/lguest/Kconfig" | 493 | source "arch/x86/lguest/Kconfig" |
494 | 494 | ||
495 | config PARAVIRT | 495 | config PARAVIRT |
496 | bool "Enable paravirtualization code" | 496 | bool "Enable paravirtualization code" |
497 | ---help--- | 497 | ---help--- |
498 | This changes the kernel so it can modify itself when it is run | 498 | This changes the kernel so it can modify itself when it is run |
499 | under a hypervisor, potentially improving performance significantly | 499 | under a hypervisor, potentially improving performance significantly |
500 | over full virtualization. However, when run without a hypervisor | 500 | over full virtualization. However, when run without a hypervisor |
501 | the kernel is theoretically slower and slightly larger. | 501 | the kernel is theoretically slower and slightly larger. |
502 | 502 | ||
503 | config PARAVIRT_SPINLOCKS | 503 | config PARAVIRT_SPINLOCKS |
504 | bool "Paravirtualization layer for spinlocks" | 504 | bool "Paravirtualization layer for spinlocks" |
505 | depends on PARAVIRT && SMP && EXPERIMENTAL | 505 | depends on PARAVIRT && SMP && EXPERIMENTAL |
506 | ---help--- | 506 | ---help--- |
507 | Paravirtualized spinlocks allow a pvops backend to replace the | 507 | Paravirtualized spinlocks allow a pvops backend to replace the |
508 | spinlock implementation with something virtualization-friendly | 508 | spinlock implementation with something virtualization-friendly |
509 | (for example, block the virtual CPU rather than spinning). | 509 | (for example, block the virtual CPU rather than spinning). |
510 | 510 | ||
511 | Unfortunately the downside is an up to 5% performance hit on | 511 | Unfortunately the downside is an up to 5% performance hit on |
512 | native kernels, with various workloads. | 512 | native kernels, with various workloads. |
513 | 513 | ||
514 | If you are unsure how to answer this question, answer N. | 514 | If you are unsure how to answer this question, answer N. |
515 | 515 | ||
516 | config PARAVIRT_CLOCK | 516 | config PARAVIRT_CLOCK |
517 | bool | 517 | bool |
518 | default n | 518 | default n |
519 | 519 | ||
520 | endif | 520 | endif |
521 | 521 | ||
522 | config PARAVIRT_DEBUG | 522 | config PARAVIRT_DEBUG |
523 | bool "paravirt-ops debugging" | 523 | bool "paravirt-ops debugging" |
524 | depends on PARAVIRT && DEBUG_KERNEL | 524 | depends on PARAVIRT && DEBUG_KERNEL |
525 | ---help--- | 525 | ---help--- |
526 | Enable to debug paravirt_ops internals. Specifically, BUG if | 526 | Enable to debug paravirt_ops internals. Specifically, BUG if |
527 | a paravirt_op is missing when it is called. | 527 | a paravirt_op is missing when it is called. |
528 | 528 | ||
529 | config MEMTEST | 529 | config MEMTEST |
530 | bool "Memtest" | 530 | bool "Memtest" |
531 | ---help--- | 531 | ---help--- |
532 | This option adds a kernel parameter 'memtest', which allows memtest | 532 | This option adds a kernel parameter 'memtest', which allows memtest |
533 | to be set. | 533 | to be set. |
534 | memtest=0, mean disabled; -- default | 534 | memtest=0, mean disabled; -- default |
535 | memtest=1, mean do 1 test pattern; | 535 | memtest=1, mean do 1 test pattern; |
536 | ... | 536 | ... |
537 | memtest=4, mean do 4 test patterns. | 537 | memtest=4, mean do 4 test patterns. |
538 | If you are unsure how to answer this question, answer N. | 538 | If you are unsure how to answer this question, answer N. |
539 | 539 | ||
540 | config X86_SUMMIT_NUMA | 540 | config X86_SUMMIT_NUMA |
541 | def_bool y | 541 | def_bool y |
542 | depends on X86_32 && NUMA && X86_32_NON_STANDARD | 542 | depends on X86_32 && NUMA && X86_32_NON_STANDARD |
543 | 543 | ||
544 | config X86_CYCLONE_TIMER | 544 | config X86_CYCLONE_TIMER |
545 | def_bool y | 545 | def_bool y |
546 | depends on X86_32_NON_STANDARD | 546 | depends on X86_32_NON_STANDARD |
547 | 547 | ||
548 | source "arch/x86/Kconfig.cpu" | 548 | source "arch/x86/Kconfig.cpu" |
549 | 549 | ||
550 | config HPET_TIMER | 550 | config HPET_TIMER |
551 | def_bool X86_64 | 551 | def_bool X86_64 |
552 | prompt "HPET Timer Support" if X86_32 | 552 | prompt "HPET Timer Support" if X86_32 |
553 | ---help--- | 553 | ---help--- |
554 | Use the IA-PC HPET (High Precision Event Timer) to manage | 554 | Use the IA-PC HPET (High Precision Event Timer) to manage |
555 | time in preference to the PIT and RTC, if a HPET is | 555 | time in preference to the PIT and RTC, if a HPET is |
556 | present. | 556 | present. |
557 | HPET is the next generation timer replacing legacy 8254s. | 557 | HPET is the next generation timer replacing legacy 8254s. |
558 | The HPET provides a stable time base on SMP | 558 | The HPET provides a stable time base on SMP |
559 | systems, unlike the TSC, but it is more expensive to access, | 559 | systems, unlike the TSC, but it is more expensive to access, |
560 | as it is off-chip. You can find the HPET spec at | 560 | as it is off-chip. You can find the HPET spec at |
561 | <http://www.intel.com/hardwaredesign/hpetspec_1.pdf>. | 561 | <http://www.intel.com/hardwaredesign/hpetspec_1.pdf>. |
562 | 562 | ||
563 | You can safely choose Y here. However, HPET will only be | 563 | You can safely choose Y here. However, HPET will only be |
564 | activated if the platform and the BIOS support this feature. | 564 | activated if the platform and the BIOS support this feature. |
565 | Otherwise the 8254 will be used for timing services. | 565 | Otherwise the 8254 will be used for timing services. |
566 | 566 | ||
567 | Choose N to continue using the legacy 8254 timer. | 567 | Choose N to continue using the legacy 8254 timer. |
568 | 568 | ||
569 | config HPET_EMULATE_RTC | 569 | config HPET_EMULATE_RTC |
570 | def_bool y | 570 | def_bool y |
571 | depends on HPET_TIMER && (RTC=y || RTC=m || RTC_DRV_CMOS=m || RTC_DRV_CMOS=y) | 571 | depends on HPET_TIMER && (RTC=y || RTC=m || RTC_DRV_CMOS=m || RTC_DRV_CMOS=y) |
572 | 572 | ||
573 | # Mark as embedded because too many people got it wrong. | 573 | # Mark as embedded because too many people got it wrong. |
574 | # The code disables itself when not needed. | 574 | # The code disables itself when not needed. |
575 | config DMI | 575 | config DMI |
576 | default y | 576 | default y |
577 | bool "Enable DMI scanning" if EMBEDDED | 577 | bool "Enable DMI scanning" if EMBEDDED |
578 | ---help--- | 578 | ---help--- |
579 | Enabled scanning of DMI to identify machine quirks. Say Y | 579 | Enabled scanning of DMI to identify machine quirks. Say Y |
580 | here unless you have verified that your setup is not | 580 | here unless you have verified that your setup is not |
581 | affected by entries in the DMI blacklist. Required by PNP | 581 | affected by entries in the DMI blacklist. Required by PNP |
582 | BIOS code. | 582 | BIOS code. |
583 | 583 | ||
584 | config GART_IOMMU | 584 | config GART_IOMMU |
585 | bool "GART IOMMU support" if EMBEDDED | 585 | bool "GART IOMMU support" if EMBEDDED |
586 | default y | 586 | default y |
587 | select SWIOTLB | 587 | select SWIOTLB |
588 | select AGP | 588 | select AGP |
589 | depends on X86_64 && PCI | 589 | depends on X86_64 && PCI |
590 | ---help--- | 590 | ---help--- |
591 | Support for full DMA access of devices with 32bit memory access only | 591 | Support for full DMA access of devices with 32bit memory access only |
592 | on systems with more than 3GB. This is usually needed for USB, | 592 | on systems with more than 3GB. This is usually needed for USB, |
593 | sound, many IDE/SATA chipsets and some other devices. | 593 | sound, many IDE/SATA chipsets and some other devices. |
594 | Provides a driver for the AMD Athlon64/Opteron/Turion/Sempron GART | 594 | Provides a driver for the AMD Athlon64/Opteron/Turion/Sempron GART |
595 | based hardware IOMMU and a software bounce buffer based IOMMU used | 595 | based hardware IOMMU and a software bounce buffer based IOMMU used |
596 | on Intel systems and as fallback. | 596 | on Intel systems and as fallback. |
597 | The code is only active when needed (enough memory and limited | 597 | The code is only active when needed (enough memory and limited |
598 | device) unless CONFIG_IOMMU_DEBUG or iommu=force is specified | 598 | device) unless CONFIG_IOMMU_DEBUG or iommu=force is specified |
599 | too. | 599 | too. |
600 | 600 | ||
601 | config CALGARY_IOMMU | 601 | config CALGARY_IOMMU |
602 | bool "IBM Calgary IOMMU support" | 602 | bool "IBM Calgary IOMMU support" |
603 | select SWIOTLB | 603 | select SWIOTLB |
604 | depends on X86_64 && PCI && EXPERIMENTAL | 604 | depends on X86_64 && PCI && EXPERIMENTAL |
605 | ---help--- | 605 | ---help--- |
606 | Support for hardware IOMMUs in IBM's xSeries x366 and x460 | 606 | Support for hardware IOMMUs in IBM's xSeries x366 and x460 |
607 | systems. Needed to run systems with more than 3GB of memory | 607 | systems. Needed to run systems with more than 3GB of memory |
608 | properly with 32-bit PCI devices that do not support DAC | 608 | properly with 32-bit PCI devices that do not support DAC |
609 | (Double Address Cycle). Calgary also supports bus level | 609 | (Double Address Cycle). Calgary also supports bus level |
610 | isolation, where all DMAs pass through the IOMMU. This | 610 | isolation, where all DMAs pass through the IOMMU. This |
611 | prevents them from going anywhere except their intended | 611 | prevents them from going anywhere except their intended |
612 | destination. This catches hard-to-find kernel bugs and | 612 | destination. This catches hard-to-find kernel bugs and |
613 | mis-behaving drivers and devices that do not use the DMA-API | 613 | mis-behaving drivers and devices that do not use the DMA-API |
614 | properly to set up their DMA buffers. The IOMMU can be | 614 | properly to set up their DMA buffers. The IOMMU can be |
615 | turned off at boot time with the iommu=off parameter. | 615 | turned off at boot time with the iommu=off parameter. |
616 | Normally the kernel will make the right choice by itself. | 616 | Normally the kernel will make the right choice by itself. |
617 | If unsure, say Y. | 617 | If unsure, say Y. |
618 | 618 | ||
619 | config CALGARY_IOMMU_ENABLED_BY_DEFAULT | 619 | config CALGARY_IOMMU_ENABLED_BY_DEFAULT |
620 | def_bool y | 620 | def_bool y |
621 | prompt "Should Calgary be enabled by default?" | 621 | prompt "Should Calgary be enabled by default?" |
622 | depends on CALGARY_IOMMU | 622 | depends on CALGARY_IOMMU |
623 | ---help--- | 623 | ---help--- |
624 | Should Calgary be enabled by default? if you choose 'y', Calgary | 624 | Should Calgary be enabled by default? if you choose 'y', Calgary |
625 | will be used (if it exists). If you choose 'n', Calgary will not be | 625 | will be used (if it exists). If you choose 'n', Calgary will not be |
626 | used even if it exists. If you choose 'n' and would like to use | 626 | used even if it exists. If you choose 'n' and would like to use |
627 | Calgary anyway, pass 'iommu=calgary' on the kernel command line. | 627 | Calgary anyway, pass 'iommu=calgary' on the kernel command line. |
628 | If unsure, say Y. | 628 | If unsure, say Y. |
629 | 629 | ||
630 | config AMD_IOMMU | 630 | config AMD_IOMMU |
631 | bool "AMD IOMMU support" | 631 | bool "AMD IOMMU support" |
632 | select SWIOTLB | 632 | select SWIOTLB |
633 | select PCI_MSI | 633 | select PCI_MSI |
634 | depends on X86_64 && PCI && ACPI | 634 | depends on X86_64 && PCI && ACPI |
635 | ---help--- | 635 | ---help--- |
636 | With this option you can enable support for AMD IOMMU hardware in | 636 | With this option you can enable support for AMD IOMMU hardware in |
637 | your system. An IOMMU is a hardware component which provides | 637 | your system. An IOMMU is a hardware component which provides |
638 | remapping of DMA memory accesses from devices. With an AMD IOMMU you | 638 | remapping of DMA memory accesses from devices. With an AMD IOMMU you |
639 | can isolate the the DMA memory of different devices and protect the | 639 | can isolate the the DMA memory of different devices and protect the |
640 | system from misbehaving device drivers or hardware. | 640 | system from misbehaving device drivers or hardware. |
641 | 641 | ||
642 | You can find out if your system has an AMD IOMMU if you look into | 642 | You can find out if your system has an AMD IOMMU if you look into |
643 | your BIOS for an option to enable it or if you have an IVRS ACPI | 643 | your BIOS for an option to enable it or if you have an IVRS ACPI |
644 | table. | 644 | table. |
645 | 645 | ||
646 | config AMD_IOMMU_STATS | 646 | config AMD_IOMMU_STATS |
647 | bool "Export AMD IOMMU statistics to debugfs" | 647 | bool "Export AMD IOMMU statistics to debugfs" |
648 | depends on AMD_IOMMU | 648 | depends on AMD_IOMMU |
649 | select DEBUG_FS | 649 | select DEBUG_FS |
650 | ---help--- | 650 | ---help--- |
651 | This option enables code in the AMD IOMMU driver to collect various | 651 | This option enables code in the AMD IOMMU driver to collect various |
652 | statistics about whats happening in the driver and exports that | 652 | statistics about whats happening in the driver and exports that |
653 | information to userspace via debugfs. | 653 | information to userspace via debugfs. |
654 | If unsure, say N. | 654 | If unsure, say N. |
655 | 655 | ||
656 | # need this always selected by IOMMU for the VIA workaround | 656 | # need this always selected by IOMMU for the VIA workaround |
657 | config SWIOTLB | 657 | config SWIOTLB |
658 | def_bool y if X86_64 | 658 | def_bool y if X86_64 |
659 | ---help--- | 659 | ---help--- |
660 | Support for software bounce buffers used on x86-64 systems | 660 | Support for software bounce buffers used on x86-64 systems |
661 | which don't have a hardware IOMMU (e.g. the current generation | 661 | which don't have a hardware IOMMU (e.g. the current generation |
662 | of Intel's x86-64 CPUs). Using this PCI devices which can only | 662 | of Intel's x86-64 CPUs). Using this PCI devices which can only |
663 | access 32-bits of memory can be used on systems with more than | 663 | access 32-bits of memory can be used on systems with more than |
664 | 3 GB of memory. If unsure, say Y. | 664 | 3 GB of memory. If unsure, say Y. |
665 | 665 | ||
666 | config IOMMU_HELPER | 666 | config IOMMU_HELPER |
667 | def_bool (CALGARY_IOMMU || GART_IOMMU || SWIOTLB || AMD_IOMMU) | 667 | def_bool (CALGARY_IOMMU || GART_IOMMU || SWIOTLB || AMD_IOMMU) |
668 | 668 | ||
669 | config IOMMU_API | 669 | config IOMMU_API |
670 | def_bool (AMD_IOMMU || DMAR) | 670 | def_bool (AMD_IOMMU || DMAR) |
671 | 671 | ||
672 | config MAXSMP | 672 | config MAXSMP |
673 | bool "Configure Maximum number of SMP Processors and NUMA Nodes" | 673 | bool "Configure Maximum number of SMP Processors and NUMA Nodes" |
674 | depends on X86_64 && SMP && DEBUG_KERNEL && EXPERIMENTAL | 674 | depends on X86_64 && SMP && DEBUG_KERNEL && EXPERIMENTAL |
675 | select CPUMASK_OFFSTACK | 675 | select CPUMASK_OFFSTACK |
676 | default n | 676 | default n |
677 | ---help--- | 677 | ---help--- |
678 | Configure maximum number of CPUS and NUMA Nodes for this architecture. | 678 | Configure maximum number of CPUS and NUMA Nodes for this architecture. |
679 | If unsure, say N. | 679 | If unsure, say N. |
680 | 680 | ||
681 | config NR_CPUS | 681 | config NR_CPUS |
682 | int "Maximum number of CPUs" if SMP && !MAXSMP | 682 | int "Maximum number of CPUs" if SMP && !MAXSMP |
683 | range 2 8 if SMP && X86_32 && !X86_BIGSMP | 683 | range 2 8 if SMP && X86_32 && !X86_BIGSMP |
684 | range 2 512 if SMP && !MAXSMP | 684 | range 2 512 if SMP && !MAXSMP |
685 | default "1" if !SMP | 685 | default "1" if !SMP |
686 | default "4096" if MAXSMP | 686 | default "4096" if MAXSMP |
687 | default "32" if SMP && (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000) | 687 | default "32" if SMP && (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000) |
688 | default "8" if SMP | 688 | default "8" if SMP |
689 | ---help--- | 689 | ---help--- |
690 | This allows you to specify the maximum number of CPUs which this | 690 | This allows you to specify the maximum number of CPUs which this |
691 | kernel will support. The maximum supported value is 512 and the | 691 | kernel will support. The maximum supported value is 512 and the |
692 | minimum value which makes sense is 2. | 692 | minimum value which makes sense is 2. |
693 | 693 | ||
694 | This is purely to save memory - each supported CPU adds | 694 | This is purely to save memory - each supported CPU adds |
695 | approximately eight kilobytes to the kernel image. | 695 | approximately eight kilobytes to the kernel image. |
696 | 696 | ||
697 | config SCHED_SMT | 697 | config SCHED_SMT |
698 | bool "SMT (Hyperthreading) scheduler support" | 698 | bool "SMT (Hyperthreading) scheduler support" |
699 | depends on X86_HT | 699 | depends on X86_HT |
700 | ---help--- | 700 | ---help--- |
701 | SMT scheduler support improves the CPU scheduler's decision making | 701 | SMT scheduler support improves the CPU scheduler's decision making |
702 | when dealing with Intel Pentium 4 chips with HyperThreading at a | 702 | when dealing with Intel Pentium 4 chips with HyperThreading at a |
703 | cost of slightly increased overhead in some places. If unsure say | 703 | cost of slightly increased overhead in some places. If unsure say |
704 | N here. | 704 | N here. |
705 | 705 | ||
706 | config SCHED_MC | 706 | config SCHED_MC |
707 | def_bool y | 707 | def_bool y |
708 | prompt "Multi-core scheduler support" | 708 | prompt "Multi-core scheduler support" |
709 | depends on X86_HT | 709 | depends on X86_HT |
710 | ---help--- | 710 | ---help--- |
711 | Multi-core scheduler support improves the CPU scheduler's decision | 711 | Multi-core scheduler support improves the CPU scheduler's decision |
712 | making when dealing with multi-core CPU chips at a cost of slightly | 712 | making when dealing with multi-core CPU chips at a cost of slightly |
713 | increased overhead in some places. If unsure say N here. | 713 | increased overhead in some places. If unsure say N here. |
714 | 714 | ||
715 | source "kernel/Kconfig.preempt" | 715 | source "kernel/Kconfig.preempt" |
716 | 716 | ||
717 | config X86_UP_APIC | 717 | config X86_UP_APIC |
718 | bool "Local APIC support on uniprocessors" | 718 | bool "Local APIC support on uniprocessors" |
719 | depends on X86_32 && !SMP && !X86_32_NON_STANDARD | 719 | depends on X86_32 && !SMP && !X86_32_NON_STANDARD |
720 | ---help--- | 720 | ---help--- |
721 | A local APIC (Advanced Programmable Interrupt Controller) is an | 721 | A local APIC (Advanced Programmable Interrupt Controller) is an |
722 | integrated interrupt controller in the CPU. If you have a single-CPU | 722 | integrated interrupt controller in the CPU. If you have a single-CPU |
723 | system which has a processor with a local APIC, you can say Y here to | 723 | system which has a processor with a local APIC, you can say Y here to |
724 | enable and use it. If you say Y here even though your machine doesn't | 724 | enable and use it. If you say Y here even though your machine doesn't |
725 | have a local APIC, then the kernel will still run with no slowdown at | 725 | have a local APIC, then the kernel will still run with no slowdown at |
726 | all. The local APIC supports CPU-generated self-interrupts (timer, | 726 | all. The local APIC supports CPU-generated self-interrupts (timer, |
727 | performance counters), and the NMI watchdog which detects hard | 727 | performance counters), and the NMI watchdog which detects hard |
728 | lockups. | 728 | lockups. |
729 | 729 | ||
730 | config X86_UP_IOAPIC | 730 | config X86_UP_IOAPIC |
731 | bool "IO-APIC support on uniprocessors" | 731 | bool "IO-APIC support on uniprocessors" |
732 | depends on X86_UP_APIC | 732 | depends on X86_UP_APIC |
733 | ---help--- | 733 | ---help--- |
734 | An IO-APIC (I/O Advanced Programmable Interrupt Controller) is an | 734 | An IO-APIC (I/O Advanced Programmable Interrupt Controller) is an |
735 | SMP-capable replacement for PC-style interrupt controllers. Most | 735 | SMP-capable replacement for PC-style interrupt controllers. Most |
736 | SMP systems and many recent uniprocessor systems have one. | 736 | SMP systems and many recent uniprocessor systems have one. |
737 | 737 | ||
738 | If you have a single-CPU system with an IO-APIC, you can say Y here | 738 | If you have a single-CPU system with an IO-APIC, you can say Y here |
739 | to use it. If you say Y here even though your machine doesn't have | 739 | to use it. If you say Y here even though your machine doesn't have |
740 | an IO-APIC, then the kernel will still run with no slowdown at all. | 740 | an IO-APIC, then the kernel will still run with no slowdown at all. |
741 | 741 | ||
742 | config X86_LOCAL_APIC | 742 | config X86_LOCAL_APIC |
743 | def_bool y | 743 | def_bool y |
744 | depends on X86_64 || SMP || X86_32_NON_STANDARD || X86_UP_APIC | 744 | depends on X86_64 || SMP || X86_32_NON_STANDARD || X86_UP_APIC |
745 | select HAVE_PERF_COUNTERS if (!M386 && !M486) | 745 | select HAVE_PERF_COUNTERS if (!M386 && !M486) |
746 | 746 | ||
747 | config X86_IO_APIC | 747 | config X86_IO_APIC |
748 | def_bool y | 748 | def_bool y |
749 | depends on X86_64 || SMP || X86_32_NON_STANDARD || X86_UP_APIC | 749 | depends on X86_64 || SMP || X86_32_NON_STANDARD || X86_UP_APIC |
750 | 750 | ||
751 | config X86_VISWS_APIC | 751 | config X86_VISWS_APIC |
752 | def_bool y | 752 | def_bool y |
753 | depends on X86_32 && X86_VISWS | 753 | depends on X86_32 && X86_VISWS |
754 | 754 | ||
755 | config X86_REROUTE_FOR_BROKEN_BOOT_IRQS | 755 | config X86_REROUTE_FOR_BROKEN_BOOT_IRQS |
756 | bool "Reroute for broken boot IRQs" | 756 | bool "Reroute for broken boot IRQs" |
757 | default n | 757 | default n |
758 | depends on X86_IO_APIC | 758 | depends on X86_IO_APIC |
759 | ---help--- | 759 | ---help--- |
760 | This option enables a workaround that fixes a source of | 760 | This option enables a workaround that fixes a source of |
761 | spurious interrupts. This is recommended when threaded | 761 | spurious interrupts. This is recommended when threaded |
762 | interrupt handling is used on systems where the generation of | 762 | interrupt handling is used on systems where the generation of |
763 | superfluous "boot interrupts" cannot be disabled. | 763 | superfluous "boot interrupts" cannot be disabled. |
764 | 764 | ||
765 | Some chipsets generate a legacy INTx "boot IRQ" when the IRQ | 765 | Some chipsets generate a legacy INTx "boot IRQ" when the IRQ |
766 | entry in the chipset's IO-APIC is masked (as, e.g. the RT | 766 | entry in the chipset's IO-APIC is masked (as, e.g. the RT |
767 | kernel does during interrupt handling). On chipsets where this | 767 | kernel does during interrupt handling). On chipsets where this |
768 | boot IRQ generation cannot be disabled, this workaround keeps | 768 | boot IRQ generation cannot be disabled, this workaround keeps |
769 | the original IRQ line masked so that only the equivalent "boot | 769 | the original IRQ line masked so that only the equivalent "boot |
770 | IRQ" is delivered to the CPUs. The workaround also tells the | 770 | IRQ" is delivered to the CPUs. The workaround also tells the |
771 | kernel to set up the IRQ handler on the boot IRQ line. In this | 771 | kernel to set up the IRQ handler on the boot IRQ line. In this |
772 | way only one interrupt is delivered to the kernel. Otherwise | 772 | way only one interrupt is delivered to the kernel. Otherwise |
773 | the spurious second interrupt may cause the kernel to bring | 773 | the spurious second interrupt may cause the kernel to bring |
774 | down (vital) interrupt lines. | 774 | down (vital) interrupt lines. |
775 | 775 | ||
776 | Only affects "broken" chipsets. Interrupt sharing may be | 776 | Only affects "broken" chipsets. Interrupt sharing may be |
777 | increased on these systems. | 777 | increased on these systems. |
778 | 778 | ||
779 | config X86_MCE | 779 | config X86_MCE |
780 | bool "Machine Check Exception" | 780 | bool "Machine Check Exception" |
781 | ---help--- | 781 | ---help--- |
782 | Machine Check Exception support allows the processor to notify the | 782 | Machine Check Exception support allows the processor to notify the |
783 | kernel if it detects a problem (e.g. overheating, component failure). | 783 | kernel if it detects a problem (e.g. overheating, component failure). |
784 | The action the kernel takes depends on the severity of the problem, | 784 | The action the kernel takes depends on the severity of the problem, |
785 | ranging from a warning message on the console, to halting the machine. | 785 | ranging from a warning message on the console, to halting the machine. |
786 | Your processor must be a Pentium or newer to support this - check the | 786 | Your processor must be a Pentium or newer to support this - check the |
787 | flags in /proc/cpuinfo for mce. Note that some older Pentium systems | 787 | flags in /proc/cpuinfo for mce. Note that some older Pentium systems |
788 | have a design flaw which leads to false MCE events - hence MCE is | 788 | have a design flaw which leads to false MCE events - hence MCE is |
789 | disabled on all P5 processors, unless explicitly enabled with "mce" | 789 | disabled on all P5 processors, unless explicitly enabled with "mce" |
790 | as a boot argument. Similarly, if MCE is built in and creates a | 790 | as a boot argument. Similarly, if MCE is built in and creates a |
791 | problem on some new non-standard machine, you can boot with "nomce" | 791 | problem on some new non-standard machine, you can boot with "nomce" |
792 | to disable it. MCE support simply ignores non-MCE processors like | 792 | to disable it. MCE support simply ignores non-MCE processors like |
793 | the 386 and 486, so nearly everyone can say Y here. | 793 | the 386 and 486, so nearly everyone can say Y here. |
794 | 794 | ||
795 | config X86_OLD_MCE | 795 | config X86_OLD_MCE |
796 | depends on X86_32 && X86_MCE | 796 | depends on X86_32 && X86_MCE |
797 | bool "Use legacy machine check code (will go away)" | 797 | bool "Use legacy machine check code (will go away)" |
798 | default n | 798 | default n |
799 | select X86_ANCIENT_MCE | 799 | select X86_ANCIENT_MCE |
800 | ---help--- | 800 | ---help--- |
801 | Use the old i386 machine check code. This is merely intended for | 801 | Use the old i386 machine check code. This is merely intended for |
802 | testing in a transition period. Try this if you run into any machine | 802 | testing in a transition period. Try this if you run into any machine |
803 | check related software problems, but report the problem to | 803 | check related software problems, but report the problem to |
804 | linux-kernel. When in doubt say no. | 804 | linux-kernel. When in doubt say no. |
805 | 805 | ||
806 | config X86_NEW_MCE | 806 | config X86_NEW_MCE |
807 | depends on X86_MCE | 807 | depends on X86_MCE |
808 | bool | 808 | bool |
809 | default y if (!X86_OLD_MCE && X86_32) || X86_64 | 809 | default y if (!X86_OLD_MCE && X86_32) || X86_64 |
810 | 810 | ||
811 | config X86_MCE_INTEL | 811 | config X86_MCE_INTEL |
812 | def_bool y | 812 | def_bool y |
813 | prompt "Intel MCE features" | 813 | prompt "Intel MCE features" |
814 | depends on X86_NEW_MCE && X86_LOCAL_APIC | 814 | depends on X86_NEW_MCE && X86_LOCAL_APIC |
815 | ---help--- | 815 | ---help--- |
816 | Additional support for intel specific MCE features such as | 816 | Additional support for intel specific MCE features such as |
817 | the thermal monitor. | 817 | the thermal monitor. |
818 | 818 | ||
819 | config X86_MCE_AMD | 819 | config X86_MCE_AMD |
820 | def_bool y | 820 | def_bool y |
821 | prompt "AMD MCE features" | 821 | prompt "AMD MCE features" |
822 | depends on X86_NEW_MCE && X86_LOCAL_APIC | 822 | depends on X86_NEW_MCE && X86_LOCAL_APIC |
823 | ---help--- | 823 | ---help--- |
824 | Additional support for AMD specific MCE features such as | 824 | Additional support for AMD specific MCE features such as |
825 | the DRAM Error Threshold. | 825 | the DRAM Error Threshold. |
826 | 826 | ||
827 | config X86_ANCIENT_MCE | 827 | config X86_ANCIENT_MCE |
828 | def_bool n | 828 | def_bool n |
829 | depends on X86_32 | 829 | depends on X86_32 |
830 | prompt "Support for old Pentium 5 / WinChip machine checks" | 830 | prompt "Support for old Pentium 5 / WinChip machine checks" |
831 | ---help--- | 831 | ---help--- |
832 | Include support for machine check handling on old Pentium 5 or WinChip | 832 | Include support for machine check handling on old Pentium 5 or WinChip |
833 | systems. These typically need to be enabled explicitely on the command | 833 | systems. These typically need to be enabled explicitely on the command |
834 | line. | 834 | line. |
835 | 835 | ||
836 | config X86_MCE_THRESHOLD | 836 | config X86_MCE_THRESHOLD |
837 | depends on X86_MCE_AMD || X86_MCE_INTEL | 837 | depends on X86_MCE_AMD || X86_MCE_INTEL |
838 | bool | 838 | bool |
839 | default y | 839 | default y |
840 | 840 | ||
841 | config X86_MCE_INJECT | 841 | config X86_MCE_INJECT |
842 | depends on X86_NEW_MCE | 842 | depends on X86_NEW_MCE |
843 | tristate "Machine check injector support" | 843 | tristate "Machine check injector support" |
844 | ---help--- | 844 | ---help--- |
845 | Provide support for injecting machine checks for testing purposes. | 845 | Provide support for injecting machine checks for testing purposes. |
846 | If you don't know what a machine check is and you don't do kernel | 846 | If you don't know what a machine check is and you don't do kernel |
847 | QA it is safe to say n. | 847 | QA it is safe to say n. |
848 | 848 | ||
849 | config X86_MCE_NONFATAL | 849 | config X86_MCE_NONFATAL |
850 | tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4" | 850 | tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4" |
851 | depends on X86_OLD_MCE | 851 | depends on X86_OLD_MCE |
852 | ---help--- | 852 | ---help--- |
853 | Enabling this feature starts a timer that triggers every 5 seconds which | 853 | Enabling this feature starts a timer that triggers every 5 seconds which |
854 | will look at the machine check registers to see if anything happened. | 854 | will look at the machine check registers to see if anything happened. |
855 | Non-fatal problems automatically get corrected (but still logged). | 855 | Non-fatal problems automatically get corrected (but still logged). |
856 | Disable this if you don't want to see these messages. | 856 | Disable this if you don't want to see these messages. |
857 | Seeing the messages this option prints out may be indicative of dying | 857 | Seeing the messages this option prints out may be indicative of dying |
858 | or out-of-spec (ie, overclocked) hardware. | 858 | or out-of-spec (ie, overclocked) hardware. |
859 | This option only does something on certain CPUs. | 859 | This option only does something on certain CPUs. |
860 | (AMD Athlon/Duron and Intel Pentium 4) | 860 | (AMD Athlon/Duron and Intel Pentium 4) |
861 | 861 | ||
862 | config X86_MCE_P4THERMAL | 862 | config X86_MCE_P4THERMAL |
863 | bool "check for P4 thermal throttling interrupt." | 863 | bool "check for P4 thermal throttling interrupt." |
864 | depends on X86_OLD_MCE && X86_MCE && (X86_UP_APIC || SMP) | 864 | depends on X86_OLD_MCE && X86_MCE && (X86_UP_APIC || SMP) |
865 | ---help--- | 865 | ---help--- |
866 | Enabling this feature will cause a message to be printed when the P4 | 866 | Enabling this feature will cause a message to be printed when the P4 |
867 | enters thermal throttling. | 867 | enters thermal throttling. |
868 | 868 | ||
869 | config X86_THERMAL_VECTOR | 869 | config X86_THERMAL_VECTOR |
870 | def_bool y | 870 | def_bool y |
871 | depends on X86_MCE_P4THERMAL || X86_MCE_INTEL | 871 | depends on X86_MCE_P4THERMAL || X86_MCE_INTEL |
872 | 872 | ||
873 | config VM86 | 873 | config VM86 |
874 | bool "Enable VM86 support" if EMBEDDED | 874 | bool "Enable VM86 support" if EMBEDDED |
875 | default y | 875 | default y |
876 | depends on X86_32 | 876 | depends on X86_32 |
877 | ---help--- | 877 | ---help--- |
878 | This option is required by programs like DOSEMU to run 16-bit legacy | 878 | This option is required by programs like DOSEMU to run 16-bit legacy |
879 | code on X86 processors. It also may be needed by software like | 879 | code on X86 processors. It also may be needed by software like |
880 | XFree86 to initialize some video cards via BIOS. Disabling this | 880 | XFree86 to initialize some video cards via BIOS. Disabling this |
881 | option saves about 6k. | 881 | option saves about 6k. |
882 | 882 | ||
883 | config TOSHIBA | 883 | config TOSHIBA |
884 | tristate "Toshiba Laptop support" | 884 | tristate "Toshiba Laptop support" |
885 | depends on X86_32 | 885 | depends on X86_32 |
886 | ---help--- | 886 | ---help--- |
887 | This adds a driver to safely access the System Management Mode of | 887 | This adds a driver to safely access the System Management Mode of |
888 | the CPU on Toshiba portables with a genuine Toshiba BIOS. It does | 888 | the CPU on Toshiba portables with a genuine Toshiba BIOS. It does |
889 | not work on models with a Phoenix BIOS. The System Management Mode | 889 | not work on models with a Phoenix BIOS. The System Management Mode |
890 | is used to set the BIOS and power saving options on Toshiba portables. | 890 | is used to set the BIOS and power saving options on Toshiba portables. |
891 | 891 | ||
892 | For information on utilities to make use of this driver see the | 892 | For information on utilities to make use of this driver see the |
893 | Toshiba Linux utilities web site at: | 893 | Toshiba Linux utilities web site at: |
894 | <http://www.buzzard.org.uk/toshiba/>. | 894 | <http://www.buzzard.org.uk/toshiba/>. |
895 | 895 | ||
896 | Say Y if you intend to run this kernel on a Toshiba portable. | 896 | Say Y if you intend to run this kernel on a Toshiba portable. |
897 | Say N otherwise. | 897 | Say N otherwise. |
898 | 898 | ||
899 | config I8K | 899 | config I8K |
900 | tristate "Dell laptop support" | 900 | tristate "Dell laptop support" |
901 | ---help--- | 901 | ---help--- |
902 | This adds a driver to safely access the System Management Mode | 902 | This adds a driver to safely access the System Management Mode |
903 | of the CPU on the Dell Inspiron 8000. The System Management Mode | 903 | of the CPU on the Dell Inspiron 8000. The System Management Mode |
904 | is used to read cpu temperature and cooling fan status and to | 904 | is used to read cpu temperature and cooling fan status and to |
905 | control the fans on the I8K portables. | 905 | control the fans on the I8K portables. |
906 | 906 | ||
907 | This driver has been tested only on the Inspiron 8000 but it may | 907 | This driver has been tested only on the Inspiron 8000 but it may |
908 | also work with other Dell laptops. You can force loading on other | 908 | also work with other Dell laptops. You can force loading on other |
909 | models by passing the parameter `force=1' to the module. Use at | 909 | models by passing the parameter `force=1' to the module. Use at |
910 | your own risk. | 910 | your own risk. |
911 | 911 | ||
912 | For information on utilities to make use of this driver see the | 912 | For information on utilities to make use of this driver see the |
913 | I8K Linux utilities web site at: | 913 | I8K Linux utilities web site at: |
914 | <http://people.debian.org/~dz/i8k/> | 914 | <http://people.debian.org/~dz/i8k/> |
915 | 915 | ||
916 | Say Y if you intend to run this kernel on a Dell Inspiron 8000. | 916 | Say Y if you intend to run this kernel on a Dell Inspiron 8000. |
917 | Say N otherwise. | 917 | Say N otherwise. |
918 | 918 | ||
919 | config X86_REBOOTFIXUPS | 919 | config X86_REBOOTFIXUPS |
920 | bool "Enable X86 board specific fixups for reboot" | 920 | bool "Enable X86 board specific fixups for reboot" |
921 | depends on X86_32 | 921 | depends on X86_32 |
922 | ---help--- | 922 | ---help--- |
923 | This enables chipset and/or board specific fixups to be done | 923 | This enables chipset and/or board specific fixups to be done |
924 | in order to get reboot to work correctly. This is only needed on | 924 | in order to get reboot to work correctly. This is only needed on |
925 | some combinations of hardware and BIOS. The symptom, for which | 925 | some combinations of hardware and BIOS. The symptom, for which |
926 | this config is intended, is when reboot ends with a stalled/hung | 926 | this config is intended, is when reboot ends with a stalled/hung |
927 | system. | 927 | system. |
928 | 928 | ||
929 | Currently, the only fixup is for the Geode machines using | 929 | Currently, the only fixup is for the Geode machines using |
930 | CS5530A and CS5536 chipsets and the RDC R-321x SoC. | 930 | CS5530A and CS5536 chipsets and the RDC R-321x SoC. |
931 | 931 | ||
932 | Say Y if you want to enable the fixup. Currently, it's safe to | 932 | Say Y if you want to enable the fixup. Currently, it's safe to |
933 | enable this option even if you don't need it. | 933 | enable this option even if you don't need it. |
934 | Say N otherwise. | 934 | Say N otherwise. |
935 | 935 | ||
936 | config MICROCODE | 936 | config MICROCODE |
937 | tristate "/dev/cpu/microcode - microcode support" | 937 | tristate "/dev/cpu/microcode - microcode support" |
938 | select FW_LOADER | 938 | select FW_LOADER |
939 | ---help--- | 939 | ---help--- |
940 | If you say Y here, you will be able to update the microcode on | 940 | If you say Y here, you will be able to update the microcode on |
941 | certain Intel and AMD processors. The Intel support is for the | 941 | certain Intel and AMD processors. The Intel support is for the |
942 | IA32 family, e.g. Pentium Pro, Pentium II, Pentium III, | 942 | IA32 family, e.g. Pentium Pro, Pentium II, Pentium III, |
943 | Pentium 4, Xeon etc. The AMD support is for family 0x10 and | 943 | Pentium 4, Xeon etc. The AMD support is for family 0x10 and |
944 | 0x11 processors, e.g. Opteron, Phenom and Turion 64 Ultra. | 944 | 0x11 processors, e.g. Opteron, Phenom and Turion 64 Ultra. |
945 | You will obviously need the actual microcode binary data itself | 945 | You will obviously need the actual microcode binary data itself |
946 | which is not shipped with the Linux kernel. | 946 | which is not shipped with the Linux kernel. |
947 | 947 | ||
948 | This option selects the general module only, you need to select | 948 | This option selects the general module only, you need to select |
949 | at least one vendor specific module as well. | 949 | at least one vendor specific module as well. |
950 | 950 | ||
951 | To compile this driver as a module, choose M here: the | 951 | To compile this driver as a module, choose M here: the |
952 | module will be called microcode. | 952 | module will be called microcode. |
953 | 953 | ||
954 | config MICROCODE_INTEL | 954 | config MICROCODE_INTEL |
955 | bool "Intel microcode patch loading support" | 955 | bool "Intel microcode patch loading support" |
956 | depends on MICROCODE | 956 | depends on MICROCODE |
957 | default MICROCODE | 957 | default MICROCODE |
958 | select FW_LOADER | 958 | select FW_LOADER |
959 | ---help--- | 959 | ---help--- |
960 | This options enables microcode patch loading support for Intel | 960 | This options enables microcode patch loading support for Intel |
961 | processors. | 961 | processors. |
962 | 962 | ||
963 | For latest news and information on obtaining all the required | 963 | For latest news and information on obtaining all the required |
964 | Intel ingredients for this driver, check: | 964 | Intel ingredients for this driver, check: |
965 | <http://www.urbanmyth.org/microcode/>. | 965 | <http://www.urbanmyth.org/microcode/>. |
966 | 966 | ||
967 | config MICROCODE_AMD | 967 | config MICROCODE_AMD |
968 | bool "AMD microcode patch loading support" | 968 | bool "AMD microcode patch loading support" |
969 | depends on MICROCODE | 969 | depends on MICROCODE |
970 | select FW_LOADER | 970 | select FW_LOADER |
971 | ---help--- | 971 | ---help--- |
972 | If you select this option, microcode patch loading support for AMD | 972 | If you select this option, microcode patch loading support for AMD |
973 | processors will be enabled. | 973 | processors will be enabled. |
974 | 974 | ||
975 | config MICROCODE_OLD_INTERFACE | 975 | config MICROCODE_OLD_INTERFACE |
976 | def_bool y | 976 | def_bool y |
977 | depends on MICROCODE | 977 | depends on MICROCODE |
978 | 978 | ||
979 | config X86_MSR | 979 | config X86_MSR |
980 | tristate "/dev/cpu/*/msr - Model-specific register support" | 980 | tristate "/dev/cpu/*/msr - Model-specific register support" |
981 | ---help--- | 981 | ---help--- |
982 | This device gives privileged processes access to the x86 | 982 | This device gives privileged processes access to the x86 |
983 | Model-Specific Registers (MSRs). It is a character device with | 983 | Model-Specific Registers (MSRs). It is a character device with |
984 | major 202 and minors 0 to 31 for /dev/cpu/0/msr to /dev/cpu/31/msr. | 984 | major 202 and minors 0 to 31 for /dev/cpu/0/msr to /dev/cpu/31/msr. |
985 | MSR accesses are directed to a specific CPU on multi-processor | 985 | MSR accesses are directed to a specific CPU on multi-processor |
986 | systems. | 986 | systems. |
987 | 987 | ||
988 | config X86_CPUID | 988 | config X86_CPUID |
989 | tristate "/dev/cpu/*/cpuid - CPU information support" | 989 | tristate "/dev/cpu/*/cpuid - CPU information support" |
990 | ---help--- | 990 | ---help--- |
991 | This device gives processes access to the x86 CPUID instruction to | 991 | This device gives processes access to the x86 CPUID instruction to |
992 | be executed on a specific processor. It is a character device | 992 | be executed on a specific processor. It is a character device |
993 | with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to | 993 | with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to |
994 | /dev/cpu/31/cpuid. | 994 | /dev/cpu/31/cpuid. |
995 | 995 | ||
996 | config X86_CPU_DEBUG | 996 | config X86_CPU_DEBUG |
997 | tristate "/sys/kernel/debug/x86/cpu/* - CPU Debug support" | 997 | tristate "/sys/kernel/debug/x86/cpu/* - CPU Debug support" |
998 | ---help--- | 998 | ---help--- |
999 | If you select this option, this will provide various x86 CPUs | 999 | If you select this option, this will provide various x86 CPUs |
1000 | information through debugfs. | 1000 | information through debugfs. |
1001 | 1001 | ||
1002 | choice | 1002 | choice |
1003 | prompt "High Memory Support" | 1003 | prompt "High Memory Support" |
1004 | default HIGHMEM4G if !X86_NUMAQ | 1004 | default HIGHMEM4G if !X86_NUMAQ |
1005 | default HIGHMEM64G if X86_NUMAQ | 1005 | default HIGHMEM64G if X86_NUMAQ |
1006 | depends on X86_32 | 1006 | depends on X86_32 |
1007 | 1007 | ||
1008 | config NOHIGHMEM | 1008 | config NOHIGHMEM |
1009 | bool "off" | 1009 | bool "off" |
1010 | depends on !X86_NUMAQ | 1010 | depends on !X86_NUMAQ |
1011 | ---help--- | 1011 | ---help--- |
1012 | Linux can use up to 64 Gigabytes of physical memory on x86 systems. | 1012 | Linux can use up to 64 Gigabytes of physical memory on x86 systems. |
1013 | However, the address space of 32-bit x86 processors is only 4 | 1013 | However, the address space of 32-bit x86 processors is only 4 |
1014 | Gigabytes large. That means that, if you have a large amount of | 1014 | Gigabytes large. That means that, if you have a large amount of |
1015 | physical memory, not all of it can be "permanently mapped" by the | 1015 | physical memory, not all of it can be "permanently mapped" by the |
1016 | kernel. The physical memory that's not permanently mapped is called | 1016 | kernel. The physical memory that's not permanently mapped is called |
1017 | "high memory". | 1017 | "high memory". |
1018 | 1018 | ||
1019 | If you are compiling a kernel which will never run on a machine with | 1019 | If you are compiling a kernel which will never run on a machine with |
1020 | more than 1 Gigabyte total physical RAM, answer "off" here (default | 1020 | more than 1 Gigabyte total physical RAM, answer "off" here (default |
1021 | choice and suitable for most users). This will result in a "3GB/1GB" | 1021 | choice and suitable for most users). This will result in a "3GB/1GB" |
1022 | split: 3GB are mapped so that each process sees a 3GB virtual memory | 1022 | split: 3GB are mapped so that each process sees a 3GB virtual memory |
1023 | space and the remaining part of the 4GB virtual memory space is used | 1023 | space and the remaining part of the 4GB virtual memory space is used |
1024 | by the kernel to permanently map as much physical memory as | 1024 | by the kernel to permanently map as much physical memory as |
1025 | possible. | 1025 | possible. |
1026 | 1026 | ||
1027 | If the machine has between 1 and 4 Gigabytes physical RAM, then | 1027 | If the machine has between 1 and 4 Gigabytes physical RAM, then |
1028 | answer "4GB" here. | 1028 | answer "4GB" here. |
1029 | 1029 | ||
1030 | If more than 4 Gigabytes is used then answer "64GB" here. This | 1030 | If more than 4 Gigabytes is used then answer "64GB" here. This |
1031 | selection turns Intel PAE (Physical Address Extension) mode on. | 1031 | selection turns Intel PAE (Physical Address Extension) mode on. |
1032 | PAE implements 3-level paging on IA32 processors. PAE is fully | 1032 | PAE implements 3-level paging on IA32 processors. PAE is fully |
1033 | supported by Linux, PAE mode is implemented on all recent Intel | 1033 | supported by Linux, PAE mode is implemented on all recent Intel |
1034 | processors (Pentium Pro and better). NOTE: If you say "64GB" here, | 1034 | processors (Pentium Pro and better). NOTE: If you say "64GB" here, |
1035 | then the kernel will not boot on CPUs that don't support PAE! | 1035 | then the kernel will not boot on CPUs that don't support PAE! |
1036 | 1036 | ||
1037 | The actual amount of total physical memory will either be | 1037 | The actual amount of total physical memory will either be |
1038 | auto detected or can be forced by using a kernel command line option | 1038 | auto detected or can be forced by using a kernel command line option |
1039 | such as "mem=256M". (Try "man bootparam" or see the documentation of | 1039 | such as "mem=256M". (Try "man bootparam" or see the documentation of |
1040 | your boot loader (lilo or loadlin) about how to pass options to the | 1040 | your boot loader (lilo or loadlin) about how to pass options to the |
1041 | kernel at boot time.) | 1041 | kernel at boot time.) |
1042 | 1042 | ||
1043 | If unsure, say "off". | 1043 | If unsure, say "off". |
1044 | 1044 | ||
1045 | config HIGHMEM4G | 1045 | config HIGHMEM4G |
1046 | bool "4GB" | 1046 | bool "4GB" |
1047 | depends on !X86_NUMAQ | 1047 | depends on !X86_NUMAQ |
1048 | ---help--- | 1048 | ---help--- |
1049 | Select this if you have a 32-bit processor and between 1 and 4 | 1049 | Select this if you have a 32-bit processor and between 1 and 4 |
1050 | gigabytes of physical RAM. | 1050 | gigabytes of physical RAM. |
1051 | 1051 | ||
1052 | config HIGHMEM64G | 1052 | config HIGHMEM64G |
1053 | bool "64GB" | 1053 | bool "64GB" |
1054 | depends on !M386 && !M486 | 1054 | depends on !M386 && !M486 |
1055 | select X86_PAE | 1055 | select X86_PAE |
1056 | ---help--- | 1056 | ---help--- |
1057 | Select this if you have a 32-bit processor and more than 4 | 1057 | Select this if you have a 32-bit processor and more than 4 |
1058 | gigabytes of physical RAM. | 1058 | gigabytes of physical RAM. |
1059 | 1059 | ||
1060 | endchoice | 1060 | endchoice |
1061 | 1061 | ||
1062 | choice | 1062 | choice |
1063 | depends on EXPERIMENTAL | 1063 | depends on EXPERIMENTAL |
1064 | prompt "Memory split" if EMBEDDED | 1064 | prompt "Memory split" if EMBEDDED |
1065 | default VMSPLIT_3G | 1065 | default VMSPLIT_3G |
1066 | depends on X86_32 | 1066 | depends on X86_32 |
1067 | ---help--- | 1067 | ---help--- |
1068 | Select the desired split between kernel and user memory. | 1068 | Select the desired split between kernel and user memory. |
1069 | 1069 | ||
1070 | If the address range available to the kernel is less than the | 1070 | If the address range available to the kernel is less than the |
1071 | physical memory installed, the remaining memory will be available | 1071 | physical memory installed, the remaining memory will be available |
1072 | as "high memory". Accessing high memory is a little more costly | 1072 | as "high memory". Accessing high memory is a little more costly |
1073 | than low memory, as it needs to be mapped into the kernel first. | 1073 | than low memory, as it needs to be mapped into the kernel first. |
1074 | Note that increasing the kernel address space limits the range | 1074 | Note that increasing the kernel address space limits the range |
1075 | available to user programs, making the address space there | 1075 | available to user programs, making the address space there |
1076 | tighter. Selecting anything other than the default 3G/1G split | 1076 | tighter. Selecting anything other than the default 3G/1G split |
1077 | will also likely make your kernel incompatible with binary-only | 1077 | will also likely make your kernel incompatible with binary-only |
1078 | kernel modules. | 1078 | kernel modules. |
1079 | 1079 | ||
1080 | If you are not absolutely sure what you are doing, leave this | 1080 | If you are not absolutely sure what you are doing, leave this |
1081 | option alone! | 1081 | option alone! |
1082 | 1082 | ||
1083 | config VMSPLIT_3G | 1083 | config VMSPLIT_3G |
1084 | bool "3G/1G user/kernel split" | 1084 | bool "3G/1G user/kernel split" |
1085 | config VMSPLIT_3G_OPT | 1085 | config VMSPLIT_3G_OPT |
1086 | depends on !X86_PAE | 1086 | depends on !X86_PAE |
1087 | bool "3G/1G user/kernel split (for full 1G low memory)" | 1087 | bool "3G/1G user/kernel split (for full 1G low memory)" |
1088 | config VMSPLIT_2G | 1088 | config VMSPLIT_2G |
1089 | bool "2G/2G user/kernel split" | 1089 | bool "2G/2G user/kernel split" |
1090 | config VMSPLIT_2G_OPT | 1090 | config VMSPLIT_2G_OPT |
1091 | depends on !X86_PAE | 1091 | depends on !X86_PAE |
1092 | bool "2G/2G user/kernel split (for full 2G low memory)" | 1092 | bool "2G/2G user/kernel split (for full 2G low memory)" |
1093 | config VMSPLIT_1G | 1093 | config VMSPLIT_1G |
1094 | bool "1G/3G user/kernel split" | 1094 | bool "1G/3G user/kernel split" |
1095 | endchoice | 1095 | endchoice |
1096 | 1096 | ||
1097 | config PAGE_OFFSET | 1097 | config PAGE_OFFSET |
1098 | hex | 1098 | hex |
1099 | default 0xB0000000 if VMSPLIT_3G_OPT | 1099 | default 0xB0000000 if VMSPLIT_3G_OPT |
1100 | default 0x80000000 if VMSPLIT_2G | 1100 | default 0x80000000 if VMSPLIT_2G |
1101 | default 0x78000000 if VMSPLIT_2G_OPT | 1101 | default 0x78000000 if VMSPLIT_2G_OPT |
1102 | default 0x40000000 if VMSPLIT_1G | 1102 | default 0x40000000 if VMSPLIT_1G |
1103 | default 0xC0000000 | 1103 | default 0xC0000000 |
1104 | depends on X86_32 | 1104 | depends on X86_32 |
1105 | 1105 | ||
1106 | config HIGHMEM | 1106 | config HIGHMEM |
1107 | def_bool y | 1107 | def_bool y |
1108 | depends on X86_32 && (HIGHMEM64G || HIGHMEM4G) | 1108 | depends on X86_32 && (HIGHMEM64G || HIGHMEM4G) |
1109 | 1109 | ||
1110 | config X86_PAE | 1110 | config X86_PAE |
1111 | bool "PAE (Physical Address Extension) Support" | 1111 | bool "PAE (Physical Address Extension) Support" |
1112 | depends on X86_32 && !HIGHMEM4G | 1112 | depends on X86_32 && !HIGHMEM4G |
1113 | ---help--- | 1113 | ---help--- |
1114 | PAE is required for NX support, and furthermore enables | 1114 | PAE is required for NX support, and furthermore enables |
1115 | larger swapspace support for non-overcommit purposes. It | 1115 | larger swapspace support for non-overcommit purposes. It |
1116 | has the cost of more pagetable lookup overhead, and also | 1116 | has the cost of more pagetable lookup overhead, and also |
1117 | consumes more pagetable space per process. | 1117 | consumes more pagetable space per process. |
1118 | 1118 | ||
1119 | config ARCH_PHYS_ADDR_T_64BIT | 1119 | config ARCH_PHYS_ADDR_T_64BIT |
1120 | def_bool X86_64 || X86_PAE | 1120 | def_bool X86_64 || X86_PAE |
1121 | 1121 | ||
1122 | config DIRECT_GBPAGES | 1122 | config DIRECT_GBPAGES |
1123 | bool "Enable 1GB pages for kernel pagetables" if EMBEDDED | 1123 | bool "Enable 1GB pages for kernel pagetables" if EMBEDDED |
1124 | default y | 1124 | default y |
1125 | depends on X86_64 | 1125 | depends on X86_64 |
1126 | ---help--- | 1126 | ---help--- |
1127 | Allow the kernel linear mapping to use 1GB pages on CPUs that | 1127 | Allow the kernel linear mapping to use 1GB pages on CPUs that |
1128 | support it. This can improve the kernel's performance a tiny bit by | 1128 | support it. This can improve the kernel's performance a tiny bit by |
1129 | reducing TLB pressure. If in doubt, say "Y". | 1129 | reducing TLB pressure. If in doubt, say "Y". |
1130 | 1130 | ||
1131 | # Common NUMA Features | 1131 | # Common NUMA Features |
1132 | config NUMA | 1132 | config NUMA |
1133 | bool "Numa Memory Allocation and Scheduler Support" | 1133 | bool "Numa Memory Allocation and Scheduler Support" |
1134 | depends on SMP | 1134 | depends on SMP |
1135 | depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || X86_BIGSMP || X86_SUMMIT && ACPI) && EXPERIMENTAL) | 1135 | depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || X86_BIGSMP || X86_SUMMIT && ACPI) && EXPERIMENTAL) |
1136 | default y if (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP) | 1136 | default y if (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP) |
1137 | ---help--- | 1137 | ---help--- |
1138 | Enable NUMA (Non Uniform Memory Access) support. | 1138 | Enable NUMA (Non Uniform Memory Access) support. |
1139 | 1139 | ||
1140 | The kernel will try to allocate memory used by a CPU on the | 1140 | The kernel will try to allocate memory used by a CPU on the |
1141 | local memory controller of the CPU and add some more | 1141 | local memory controller of the CPU and add some more |
1142 | NUMA awareness to the kernel. | 1142 | NUMA awareness to the kernel. |
1143 | 1143 | ||
1144 | For 64-bit this is recommended if the system is Intel Core i7 | 1144 | For 64-bit this is recommended if the system is Intel Core i7 |
1145 | (or later), AMD Opteron, or EM64T NUMA. | 1145 | (or later), AMD Opteron, or EM64T NUMA. |
1146 | 1146 | ||
1147 | For 32-bit this is only needed on (rare) 32-bit-only platforms | 1147 | For 32-bit this is only needed on (rare) 32-bit-only platforms |
1148 | that support NUMA topologies, such as NUMAQ / Summit, or if you | 1148 | that support NUMA topologies, such as NUMAQ / Summit, or if you |
1149 | boot a 32-bit kernel on a 64-bit NUMA platform. | 1149 | boot a 32-bit kernel on a 64-bit NUMA platform. |
1150 | 1150 | ||
1151 | Otherwise, you should say N. | 1151 | Otherwise, you should say N. |
1152 | 1152 | ||
1153 | comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI" | 1153 | comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI" |
1154 | depends on X86_32 && X86_SUMMIT && (!HIGHMEM64G || !ACPI) | 1154 | depends on X86_32 && X86_SUMMIT && (!HIGHMEM64G || !ACPI) |
1155 | 1155 | ||
1156 | config K8_NUMA | 1156 | config K8_NUMA |
1157 | def_bool y | 1157 | def_bool y |
1158 | prompt "Old style AMD Opteron NUMA detection" | 1158 | prompt "Old style AMD Opteron NUMA detection" |
1159 | depends on X86_64 && NUMA && PCI | 1159 | depends on X86_64 && NUMA && PCI |
1160 | ---help--- | 1160 | ---help--- |
1161 | Enable K8 NUMA node topology detection. You should say Y here if | 1161 | Enable K8 NUMA node topology detection. You should say Y here if |
1162 | you have a multi processor AMD K8 system. This uses an old | 1162 | you have a multi processor AMD K8 system. This uses an old |
1163 | method to read the NUMA configuration directly from the builtin | 1163 | method to read the NUMA configuration directly from the builtin |
1164 | Northbridge of Opteron. It is recommended to use X86_64_ACPI_NUMA | 1164 | Northbridge of Opteron. It is recommended to use X86_64_ACPI_NUMA |
1165 | instead, which also takes priority if both are compiled in. | 1165 | instead, which also takes priority if both are compiled in. |
1166 | 1166 | ||
1167 | config X86_64_ACPI_NUMA | 1167 | config X86_64_ACPI_NUMA |
1168 | def_bool y | 1168 | def_bool y |
1169 | prompt "ACPI NUMA detection" | 1169 | prompt "ACPI NUMA detection" |
1170 | depends on X86_64 && NUMA && ACPI && PCI | 1170 | depends on X86_64 && NUMA && ACPI && PCI |
1171 | select ACPI_NUMA | 1171 | select ACPI_NUMA |
1172 | ---help--- | 1172 | ---help--- |
1173 | Enable ACPI SRAT based node topology detection. | 1173 | Enable ACPI SRAT based node topology detection. |
1174 | 1174 | ||
1175 | # Some NUMA nodes have memory ranges that span | 1175 | # Some NUMA nodes have memory ranges that span |
1176 | # other nodes. Even though a pfn is valid and | 1176 | # other nodes. Even though a pfn is valid and |
1177 | # between a node's start and end pfns, it may not | 1177 | # between a node's start and end pfns, it may not |
1178 | # reside on that node. See memmap_init_zone() | 1178 | # reside on that node. See memmap_init_zone() |
1179 | # for details. | 1179 | # for details. |
1180 | config NODES_SPAN_OTHER_NODES | 1180 | config NODES_SPAN_OTHER_NODES |
1181 | def_bool y | 1181 | def_bool y |
1182 | depends on X86_64_ACPI_NUMA | 1182 | depends on X86_64_ACPI_NUMA |
1183 | 1183 | ||
1184 | config NUMA_EMU | 1184 | config NUMA_EMU |
1185 | bool "NUMA emulation" | 1185 | bool "NUMA emulation" |
1186 | depends on X86_64 && NUMA | 1186 | depends on X86_64 && NUMA |
1187 | ---help--- | 1187 | ---help--- |
1188 | Enable NUMA emulation. A flat machine will be split | 1188 | Enable NUMA emulation. A flat machine will be split |
1189 | into virtual nodes when booted with "numa=fake=N", where N is the | 1189 | into virtual nodes when booted with "numa=fake=N", where N is the |
1190 | number of nodes. This is only useful for debugging. | 1190 | number of nodes. This is only useful for debugging. |
1191 | 1191 | ||
1192 | config NODES_SHIFT | 1192 | config NODES_SHIFT |
1193 | int "Maximum NUMA Nodes (as a power of 2)" if !MAXSMP | 1193 | int "Maximum NUMA Nodes (as a power of 2)" if !MAXSMP |
1194 | range 1 9 | 1194 | range 1 9 |
1195 | default "9" if MAXSMP | 1195 | default "9" if MAXSMP |
1196 | default "6" if X86_64 | 1196 | default "6" if X86_64 |
1197 | default "4" if X86_NUMAQ | 1197 | default "4" if X86_NUMAQ |
1198 | default "3" | 1198 | default "3" |
1199 | depends on NEED_MULTIPLE_NODES | 1199 | depends on NEED_MULTIPLE_NODES |
1200 | ---help--- | 1200 | ---help--- |
1201 | Specify the maximum number of NUMA Nodes available on the target | 1201 | Specify the maximum number of NUMA Nodes available on the target |
1202 | system. Increases memory reserved to accommodate various tables. | 1202 | system. Increases memory reserved to accommodate various tables. |
1203 | 1203 | ||
1204 | config HAVE_ARCH_BOOTMEM | 1204 | config HAVE_ARCH_BOOTMEM |
1205 | def_bool y | 1205 | def_bool y |
1206 | depends on X86_32 && NUMA | 1206 | depends on X86_32 && NUMA |
1207 | 1207 | ||
1208 | config ARCH_HAVE_MEMORY_PRESENT | 1208 | config ARCH_HAVE_MEMORY_PRESENT |
1209 | def_bool y | 1209 | def_bool y |
1210 | depends on X86_32 && DISCONTIGMEM | 1210 | depends on X86_32 && DISCONTIGMEM |
1211 | 1211 | ||
1212 | config NEED_NODE_MEMMAP_SIZE | 1212 | config NEED_NODE_MEMMAP_SIZE |
1213 | def_bool y | 1213 | def_bool y |
1214 | depends on X86_32 && (DISCONTIGMEM || SPARSEMEM) | 1214 | depends on X86_32 && (DISCONTIGMEM || SPARSEMEM) |
1215 | 1215 | ||
1216 | config HAVE_ARCH_ALLOC_REMAP | 1216 | config HAVE_ARCH_ALLOC_REMAP |
1217 | def_bool y | 1217 | def_bool y |
1218 | depends on X86_32 && NUMA | 1218 | depends on X86_32 && NUMA |
1219 | 1219 | ||
1220 | config ARCH_FLATMEM_ENABLE | 1220 | config ARCH_FLATMEM_ENABLE |
1221 | def_bool y | 1221 | def_bool y |
1222 | depends on X86_32 && ARCH_SELECT_MEMORY_MODEL && !NUMA | 1222 | depends on X86_32 && ARCH_SELECT_MEMORY_MODEL && !NUMA |
1223 | 1223 | ||
1224 | config ARCH_DISCONTIGMEM_ENABLE | 1224 | config ARCH_DISCONTIGMEM_ENABLE |
1225 | def_bool y | 1225 | def_bool y |
1226 | depends on NUMA && X86_32 | 1226 | depends on NUMA && X86_32 |
1227 | 1227 | ||
1228 | config ARCH_DISCONTIGMEM_DEFAULT | 1228 | config ARCH_DISCONTIGMEM_DEFAULT |
1229 | def_bool y | 1229 | def_bool y |
1230 | depends on NUMA && X86_32 | 1230 | depends on NUMA && X86_32 |
1231 | 1231 | ||
1232 | config ARCH_SPARSEMEM_DEFAULT | 1232 | config ARCH_SPARSEMEM_DEFAULT |
1233 | def_bool y | 1233 | def_bool y |
1234 | depends on X86_64 | 1234 | depends on X86_64 |
1235 | 1235 | ||
1236 | config ARCH_SPARSEMEM_ENABLE | 1236 | config ARCH_SPARSEMEM_ENABLE |
1237 | def_bool y | 1237 | def_bool y |
1238 | depends on X86_64 || NUMA || (EXPERIMENTAL && X86_32) || X86_32_NON_STANDARD | 1238 | depends on X86_64 || NUMA || (EXPERIMENTAL && X86_32) || X86_32_NON_STANDARD |
1239 | select SPARSEMEM_STATIC if X86_32 | 1239 | select SPARSEMEM_STATIC if X86_32 |
1240 | select SPARSEMEM_VMEMMAP_ENABLE if X86_64 | 1240 | select SPARSEMEM_VMEMMAP_ENABLE if X86_64 |
1241 | 1241 | ||
1242 | config ARCH_SELECT_MEMORY_MODEL | 1242 | config ARCH_SELECT_MEMORY_MODEL |
1243 | def_bool y | 1243 | def_bool y |
1244 | depends on ARCH_SPARSEMEM_ENABLE | 1244 | depends on ARCH_SPARSEMEM_ENABLE |
1245 | 1245 | ||
1246 | config ARCH_MEMORY_PROBE | 1246 | config ARCH_MEMORY_PROBE |
1247 | def_bool X86_64 | 1247 | def_bool X86_64 |
1248 | depends on MEMORY_HOTPLUG | 1248 | depends on MEMORY_HOTPLUG |
1249 | 1249 | ||
1250 | source "mm/Kconfig" | 1250 | source "mm/Kconfig" |
1251 | 1251 | ||
1252 | config HIGHPTE | 1252 | config HIGHPTE |
1253 | bool "Allocate 3rd-level pagetables from highmem" | 1253 | bool "Allocate 3rd-level pagetables from highmem" |
1254 | depends on X86_32 && (HIGHMEM4G || HIGHMEM64G) | 1254 | depends on X86_32 && (HIGHMEM4G || HIGHMEM64G) |
1255 | ---help--- | 1255 | ---help--- |
1256 | The VM uses one page table entry for each page of physical memory. | 1256 | The VM uses one page table entry for each page of physical memory. |
1257 | For systems with a lot of RAM, this can be wasteful of precious | 1257 | For systems with a lot of RAM, this can be wasteful of precious |
1258 | low memory. Setting this option will put user-space page table | 1258 | low memory. Setting this option will put user-space page table |
1259 | entries in high memory. | 1259 | entries in high memory. |
1260 | 1260 | ||
1261 | config X86_CHECK_BIOS_CORRUPTION | 1261 | config X86_CHECK_BIOS_CORRUPTION |
1262 | bool "Check for low memory corruption" | 1262 | bool "Check for low memory corruption" |
1263 | ---help--- | 1263 | ---help--- |
1264 | Periodically check for memory corruption in low memory, which | 1264 | Periodically check for memory corruption in low memory, which |
1265 | is suspected to be caused by BIOS. Even when enabled in the | 1265 | is suspected to be caused by BIOS. Even when enabled in the |
1266 | configuration, it is disabled at runtime. Enable it by | 1266 | configuration, it is disabled at runtime. Enable it by |
1267 | setting "memory_corruption_check=1" on the kernel command | 1267 | setting "memory_corruption_check=1" on the kernel command |
1268 | line. By default it scans the low 64k of memory every 60 | 1268 | line. By default it scans the low 64k of memory every 60 |
1269 | seconds; see the memory_corruption_check_size and | 1269 | seconds; see the memory_corruption_check_size and |
1270 | memory_corruption_check_period parameters in | 1270 | memory_corruption_check_period parameters in |
1271 | Documentation/kernel-parameters.txt to adjust this. | 1271 | Documentation/kernel-parameters.txt to adjust this. |
1272 | 1272 | ||
1273 | When enabled with the default parameters, this option has | 1273 | When enabled with the default parameters, this option has |
1274 | almost no overhead, as it reserves a relatively small amount | 1274 | almost no overhead, as it reserves a relatively small amount |
1275 | of memory and scans it infrequently. It both detects corruption | 1275 | of memory and scans it infrequently. It both detects corruption |
1276 | and prevents it from affecting the running system. | 1276 | and prevents it from affecting the running system. |
1277 | 1277 | ||
1278 | It is, however, intended as a diagnostic tool; if repeatable | 1278 | It is, however, intended as a diagnostic tool; if repeatable |
1279 | BIOS-originated corruption always affects the same memory, | 1279 | BIOS-originated corruption always affects the same memory, |
1280 | you can use memmap= to prevent the kernel from using that | 1280 | you can use memmap= to prevent the kernel from using that |
1281 | memory. | 1281 | memory. |
1282 | 1282 | ||
1283 | config X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK | 1283 | config X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK |
1284 | bool "Set the default setting of memory_corruption_check" | 1284 | bool "Set the default setting of memory_corruption_check" |
1285 | depends on X86_CHECK_BIOS_CORRUPTION | 1285 | depends on X86_CHECK_BIOS_CORRUPTION |
1286 | default y | 1286 | default y |
1287 | ---help--- | 1287 | ---help--- |
1288 | Set whether the default state of memory_corruption_check is | 1288 | Set whether the default state of memory_corruption_check is |
1289 | on or off. | 1289 | on or off. |
1290 | 1290 | ||
1291 | config X86_RESERVE_LOW_64K | 1291 | config X86_RESERVE_LOW_64K |
1292 | bool "Reserve low 64K of RAM on AMI/Phoenix BIOSen" | 1292 | bool "Reserve low 64K of RAM on AMI/Phoenix BIOSen" |
1293 | default y | 1293 | default y |
1294 | ---help--- | 1294 | ---help--- |
1295 | Reserve the first 64K of physical RAM on BIOSes that are known | 1295 | Reserve the first 64K of physical RAM on BIOSes that are known |
1296 | to potentially corrupt that memory range. A numbers of BIOSes are | 1296 | to potentially corrupt that memory range. A numbers of BIOSes are |
1297 | known to utilize this area during suspend/resume, so it must not | 1297 | known to utilize this area during suspend/resume, so it must not |
1298 | be used by the kernel. | 1298 | be used by the kernel. |
1299 | 1299 | ||
1300 | Set this to N if you are absolutely sure that you trust the BIOS | 1300 | Set this to N if you are absolutely sure that you trust the BIOS |
1301 | to get all its memory reservations and usages right. | 1301 | to get all its memory reservations and usages right. |
1302 | 1302 | ||
1303 | If you have doubts about the BIOS (e.g. suspend/resume does not | 1303 | If you have doubts about the BIOS (e.g. suspend/resume does not |
1304 | work or there's kernel crashes after certain hardware hotplug | 1304 | work or there's kernel crashes after certain hardware hotplug |
1305 | events) and it's not AMI or Phoenix, then you might want to enable | 1305 | events) and it's not AMI or Phoenix, then you might want to enable |
1306 | X86_CHECK_BIOS_CORRUPTION=y to allow the kernel to check typical | 1306 | X86_CHECK_BIOS_CORRUPTION=y to allow the kernel to check typical |
1307 | corruption patterns. | 1307 | corruption patterns. |
1308 | 1308 | ||
1309 | Say Y if unsure. | 1309 | Say Y if unsure. |
1310 | 1310 | ||
1311 | config MATH_EMULATION | 1311 | config MATH_EMULATION |
1312 | bool | 1312 | bool |
1313 | prompt "Math emulation" if X86_32 | 1313 | prompt "Math emulation" if X86_32 |
1314 | ---help--- | 1314 | ---help--- |
1315 | Linux can emulate a math coprocessor (used for floating point | 1315 | Linux can emulate a math coprocessor (used for floating point |
1316 | operations) if you don't have one. 486DX and Pentium processors have | 1316 | operations) if you don't have one. 486DX and Pentium processors have |
1317 | a math coprocessor built in, 486SX and 386 do not, unless you added | 1317 | a math coprocessor built in, 486SX and 386 do not, unless you added |
1318 | a 487DX or 387, respectively. (The messages during boot time can | 1318 | a 487DX or 387, respectively. (The messages during boot time can |
1319 | give you some hints here ["man dmesg"].) Everyone needs either a | 1319 | give you some hints here ["man dmesg"].) Everyone needs either a |
1320 | coprocessor or this emulation. | 1320 | coprocessor or this emulation. |
1321 | 1321 | ||
1322 | If you don't have a math coprocessor, you need to say Y here; if you | 1322 | If you don't have a math coprocessor, you need to say Y here; if you |
1323 | say Y here even though you have a coprocessor, the coprocessor will | 1323 | say Y here even though you have a coprocessor, the coprocessor will |
1324 | be used nevertheless. (This behavior can be changed with the kernel | 1324 | be used nevertheless. (This behavior can be changed with the kernel |
1325 | command line option "no387", which comes handy if your coprocessor | 1325 | command line option "no387", which comes handy if your coprocessor |
1326 | is broken. Try "man bootparam" or see the documentation of your boot | 1326 | is broken. Try "man bootparam" or see the documentation of your boot |
1327 | loader (lilo or loadlin) about how to pass options to the kernel at | 1327 | loader (lilo or loadlin) about how to pass options to the kernel at |
1328 | boot time.) This means that it is a good idea to say Y here if you | 1328 | boot time.) This means that it is a good idea to say Y here if you |
1329 | intend to use this kernel on different machines. | 1329 | intend to use this kernel on different machines. |
1330 | 1330 | ||
1331 | More information about the internals of the Linux math coprocessor | 1331 | More information about the internals of the Linux math coprocessor |
1332 | emulation can be found in <file:arch/x86/math-emu/README>. | 1332 | emulation can be found in <file:arch/x86/math-emu/README>. |
1333 | 1333 | ||
1334 | If you are not sure, say Y; apart from resulting in a 66 KB bigger | 1334 | If you are not sure, say Y; apart from resulting in a 66 KB bigger |
1335 | kernel, it won't hurt. | 1335 | kernel, it won't hurt. |
1336 | 1336 | ||
1337 | config MTRR | 1337 | config MTRR |
1338 | bool "MTRR (Memory Type Range Register) support" | 1338 | bool "MTRR (Memory Type Range Register) support" |
1339 | ---help--- | 1339 | ---help--- |
1340 | On Intel P6 family processors (Pentium Pro, Pentium II and later) | 1340 | On Intel P6 family processors (Pentium Pro, Pentium II and later) |
1341 | the Memory Type Range Registers (MTRRs) may be used to control | 1341 | the Memory Type Range Registers (MTRRs) may be used to control |
1342 | processor access to memory ranges. This is most useful if you have | 1342 | processor access to memory ranges. This is most useful if you have |
1343 | a video (VGA) card on a PCI or AGP bus. Enabling write-combining | 1343 | a video (VGA) card on a PCI or AGP bus. Enabling write-combining |
1344 | allows bus write transfers to be combined into a larger transfer | 1344 | allows bus write transfers to be combined into a larger transfer |
1345 | before bursting over the PCI/AGP bus. This can increase performance | 1345 | before bursting over the PCI/AGP bus. This can increase performance |
1346 | of image write operations 2.5 times or more. Saying Y here creates a | 1346 | of image write operations 2.5 times or more. Saying Y here creates a |
1347 | /proc/mtrr file which may be used to manipulate your processor's | 1347 | /proc/mtrr file which may be used to manipulate your processor's |
1348 | MTRRs. Typically the X server should use this. | 1348 | MTRRs. Typically the X server should use this. |
1349 | 1349 | ||
1350 | This code has a reasonably generic interface so that similar | 1350 | This code has a reasonably generic interface so that similar |
1351 | control registers on other processors can be easily supported | 1351 | control registers on other processors can be easily supported |
1352 | as well: | 1352 | as well: |
1353 | 1353 | ||
1354 | The Cyrix 6x86, 6x86MX and M II processors have Address Range | 1354 | The Cyrix 6x86, 6x86MX and M II processors have Address Range |
1355 | Registers (ARRs) which provide a similar functionality to MTRRs. For | 1355 | Registers (ARRs) which provide a similar functionality to MTRRs. For |
1356 | these, the ARRs are used to emulate the MTRRs. | 1356 | these, the ARRs are used to emulate the MTRRs. |
1357 | The AMD K6-2 (stepping 8 and above) and K6-3 processors have two | 1357 | The AMD K6-2 (stepping 8 and above) and K6-3 processors have two |
1358 | MTRRs. The Centaur C6 (WinChip) has 8 MCRs, allowing | 1358 | MTRRs. The Centaur C6 (WinChip) has 8 MCRs, allowing |
1359 | write-combining. All of these processors are supported by this code | 1359 | write-combining. All of these processors are supported by this code |
1360 | and it makes sense to say Y here if you have one of them. | 1360 | and it makes sense to say Y here if you have one of them. |
1361 | 1361 | ||
1362 | Saying Y here also fixes a problem with buggy SMP BIOSes which only | 1362 | Saying Y here also fixes a problem with buggy SMP BIOSes which only |
1363 | set the MTRRs for the boot CPU and not for the secondary CPUs. This | 1363 | set the MTRRs for the boot CPU and not for the secondary CPUs. This |
1364 | can lead to all sorts of problems, so it's good to say Y here. | 1364 | can lead to all sorts of problems, so it's good to say Y here. |
1365 | 1365 | ||
1366 | You can safely say Y even if your machine doesn't have MTRRs, you'll | 1366 | You can safely say Y even if your machine doesn't have MTRRs, you'll |
1367 | just add about 9 KB to your kernel. | 1367 | just add about 9 KB to your kernel. |
1368 | 1368 | ||
1369 | See <file:Documentation/x86/mtrr.txt> for more information. | 1369 | See <file:Documentation/x86/mtrr.txt> for more information. |
1370 | 1370 | ||
1371 | config MTRR_SANITIZER | 1371 | config MTRR_SANITIZER |
1372 | def_bool y | 1372 | def_bool y |
1373 | prompt "MTRR cleanup support" | 1373 | prompt "MTRR cleanup support" |
1374 | depends on MTRR | 1374 | depends on MTRR |
1375 | ---help--- | 1375 | ---help--- |
1376 | Convert MTRR layout from continuous to discrete, so X drivers can | 1376 | Convert MTRR layout from continuous to discrete, so X drivers can |
1377 | add writeback entries. | 1377 | add writeback entries. |
1378 | 1378 | ||
1379 | Can be disabled with disable_mtrr_cleanup on the kernel command line. | 1379 | Can be disabled with disable_mtrr_cleanup on the kernel command line. |
1380 | The largest mtrr entry size for a continuous block can be set with | 1380 | The largest mtrr entry size for a continuous block can be set with |
1381 | mtrr_chunk_size. | 1381 | mtrr_chunk_size. |
1382 | 1382 | ||
1383 | If unsure, say Y. | 1383 | If unsure, say Y. |
1384 | 1384 | ||
1385 | config MTRR_SANITIZER_ENABLE_DEFAULT | 1385 | config MTRR_SANITIZER_ENABLE_DEFAULT |
1386 | int "MTRR cleanup enable value (0-1)" | 1386 | int "MTRR cleanup enable value (0-1)" |
1387 | range 0 1 | 1387 | range 0 1 |
1388 | default "0" | 1388 | default "0" |
1389 | depends on MTRR_SANITIZER | 1389 | depends on MTRR_SANITIZER |
1390 | ---help--- | 1390 | ---help--- |
1391 | Enable mtrr cleanup default value | 1391 | Enable mtrr cleanup default value |
1392 | 1392 | ||
1393 | config MTRR_SANITIZER_SPARE_REG_NR_DEFAULT | 1393 | config MTRR_SANITIZER_SPARE_REG_NR_DEFAULT |
1394 | int "MTRR cleanup spare reg num (0-7)" | 1394 | int "MTRR cleanup spare reg num (0-7)" |
1395 | range 0 7 | 1395 | range 0 7 |
1396 | default "1" | 1396 | default "1" |
1397 | depends on MTRR_SANITIZER | 1397 | depends on MTRR_SANITIZER |
1398 | ---help--- | 1398 | ---help--- |
1399 | mtrr cleanup spare entries default, it can be changed via | 1399 | mtrr cleanup spare entries default, it can be changed via |
1400 | mtrr_spare_reg_nr=N on the kernel command line. | 1400 | mtrr_spare_reg_nr=N on the kernel command line. |
1401 | 1401 | ||
1402 | config X86_PAT | 1402 | config X86_PAT |
1403 | bool | 1403 | bool |
1404 | prompt "x86 PAT support" | 1404 | prompt "x86 PAT support" |
1405 | depends on MTRR | 1405 | depends on MTRR |
1406 | ---help--- | 1406 | ---help--- |
1407 | Use PAT attributes to setup page level cache control. | 1407 | Use PAT attributes to setup page level cache control. |
1408 | 1408 | ||
1409 | PATs are the modern equivalents of MTRRs and are much more | 1409 | PATs are the modern equivalents of MTRRs and are much more |
1410 | flexible than MTRRs. | 1410 | flexible than MTRRs. |
1411 | 1411 | ||
1412 | Say N here if you see bootup problems (boot crash, boot hang, | 1412 | Say N here if you see bootup problems (boot crash, boot hang, |
1413 | spontaneous reboots) or a non-working video driver. | 1413 | spontaneous reboots) or a non-working video driver. |
1414 | 1414 | ||
1415 | If unsure, say Y. | 1415 | If unsure, say Y. |
1416 | 1416 | ||
1417 | config EFI | 1417 | config EFI |
1418 | bool "EFI runtime service support" | 1418 | bool "EFI runtime service support" |
1419 | depends on ACPI | 1419 | depends on ACPI |
1420 | ---help--- | 1420 | ---help--- |
1421 | This enables the kernel to use EFI runtime services that are | 1421 | This enables the kernel to use EFI runtime services that are |
1422 | available (such as the EFI variable services). | 1422 | available (such as the EFI variable services). |
1423 | 1423 | ||
1424 | This option is only useful on systems that have EFI firmware. | 1424 | This option is only useful on systems that have EFI firmware. |
1425 | In addition, you should use the latest ELILO loader available | 1425 | In addition, you should use the latest ELILO loader available |
1426 | at <http://elilo.sourceforge.net> in order to take advantage | 1426 | at <http://elilo.sourceforge.net> in order to take advantage |
1427 | of EFI runtime services. However, even with this option, the | 1427 | of EFI runtime services. However, even with this option, the |
1428 | resultant kernel should continue to boot on existing non-EFI | 1428 | resultant kernel should continue to boot on existing non-EFI |
1429 | platforms. | 1429 | platforms. |
1430 | 1430 | ||
1431 | config SECCOMP | 1431 | config SECCOMP |
1432 | def_bool y | 1432 | def_bool y |
1433 | prompt "Enable seccomp to safely compute untrusted bytecode" | 1433 | prompt "Enable seccomp to safely compute untrusted bytecode" |
1434 | ---help--- | 1434 | ---help--- |
1435 | This kernel feature is useful for number crunching applications | 1435 | This kernel feature is useful for number crunching applications |
1436 | that may need to compute untrusted bytecode during their | 1436 | that may need to compute untrusted bytecode during their |
1437 | execution. By using pipes or other transports made available to | 1437 | execution. By using pipes or other transports made available to |
1438 | the process as file descriptors supporting the read/write | 1438 | the process as file descriptors supporting the read/write |
1439 | syscalls, it's possible to isolate those applications in | 1439 | syscalls, it's possible to isolate those applications in |
1440 | their own address space using seccomp. Once seccomp is | 1440 | their own address space using seccomp. Once seccomp is |
1441 | enabled via prctl(PR_SET_SECCOMP), it cannot be disabled | 1441 | enabled via prctl(PR_SET_SECCOMP), it cannot be disabled |
1442 | and the task is only allowed to execute a few safe syscalls | 1442 | and the task is only allowed to execute a few safe syscalls |
1443 | defined by each seccomp mode. | 1443 | defined by each seccomp mode. |
1444 | 1444 | ||
1445 | If unsure, say Y. Only embedded should say N here. | 1445 | If unsure, say Y. Only embedded should say N here. |
1446 | 1446 | ||
1447 | config CC_STACKPROTECTOR_ALL | 1447 | config CC_STACKPROTECTOR_ALL |
1448 | bool | 1448 | bool |
1449 | 1449 | ||
1450 | config CC_STACKPROTECTOR | 1450 | config CC_STACKPROTECTOR |
1451 | bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)" | 1451 | bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)" |
1452 | select CC_STACKPROTECTOR_ALL | 1452 | select CC_STACKPROTECTOR_ALL |
1453 | ---help--- | 1453 | ---help--- |
1454 | This option turns on the -fstack-protector GCC feature. This | 1454 | This option turns on the -fstack-protector GCC feature. This |
1455 | feature puts, at the beginning of functions, a canary value on | 1455 | feature puts, at the beginning of functions, a canary value on |
1456 | the stack just before the return address, and validates | 1456 | the stack just before the return address, and validates |
1457 | the value just before actually returning. Stack based buffer | 1457 | the value just before actually returning. Stack based buffer |
1458 | overflows (that need to overwrite this return address) now also | 1458 | overflows (that need to overwrite this return address) now also |
1459 | overwrite the canary, which gets detected and the attack is then | 1459 | overwrite the canary, which gets detected and the attack is then |
1460 | neutralized via a kernel panic. | 1460 | neutralized via a kernel panic. |
1461 | 1461 | ||
1462 | This feature requires gcc version 4.2 or above, or a distribution | 1462 | This feature requires gcc version 4.2 or above, or a distribution |
1463 | gcc with the feature backported. Older versions are automatically | 1463 | gcc with the feature backported. Older versions are automatically |
1464 | detected and for those versions, this configuration option is | 1464 | detected and for those versions, this configuration option is |
1465 | ignored. (and a warning is printed during bootup) | 1465 | ignored. (and a warning is printed during bootup) |
1466 | 1466 | ||
1467 | source kernel/Kconfig.hz | 1467 | source kernel/Kconfig.hz |
1468 | 1468 | ||
1469 | config KEXEC | 1469 | config KEXEC |
1470 | bool "kexec system call" | 1470 | bool "kexec system call" |
1471 | ---help--- | 1471 | ---help--- |
1472 | kexec is a system call that implements the ability to shutdown your | 1472 | kexec is a system call that implements the ability to shutdown your |
1473 | current kernel, and to start another kernel. It is like a reboot | 1473 | current kernel, and to start another kernel. It is like a reboot |
1474 | but it is independent of the system firmware. And like a reboot | 1474 | but it is independent of the system firmware. And like a reboot |
1475 | you can start any kernel with it, not just Linux. | 1475 | you can start any kernel with it, not just Linux. |
1476 | 1476 | ||
1477 | The name comes from the similarity to the exec system call. | 1477 | The name comes from the similarity to the exec system call. |
1478 | 1478 | ||
1479 | It is an ongoing process to be certain the hardware in a machine | 1479 | It is an ongoing process to be certain the hardware in a machine |
1480 | is properly shutdown, so do not be surprised if this code does not | 1480 | is properly shutdown, so do not be surprised if this code does not |
1481 | initially work for you. It may help to enable device hotplugging | 1481 | initially work for you. It may help to enable device hotplugging |
1482 | support. As of this writing the exact hardware interface is | 1482 | support. As of this writing the exact hardware interface is |
1483 | strongly in flux, so no good recommendation can be made. | 1483 | strongly in flux, so no good recommendation can be made. |
1484 | 1484 | ||
1485 | config CRASH_DUMP | 1485 | config CRASH_DUMP |
1486 | bool "kernel crash dumps" | 1486 | bool "kernel crash dumps" |
1487 | depends on X86_64 || (X86_32 && HIGHMEM) | 1487 | depends on X86_64 || (X86_32 && HIGHMEM) |
1488 | ---help--- | 1488 | ---help--- |
1489 | Generate crash dump after being started by kexec. | 1489 | Generate crash dump after being started by kexec. |
1490 | This should be normally only set in special crash dump kernels | 1490 | This should be normally only set in special crash dump kernels |
1491 | which are loaded in the main kernel with kexec-tools into | 1491 | which are loaded in the main kernel with kexec-tools into |
1492 | a specially reserved region and then later executed after | 1492 | a specially reserved region and then later executed after |
1493 | a crash by kdump/kexec. The crash dump kernel must be compiled | 1493 | a crash by kdump/kexec. The crash dump kernel must be compiled |
1494 | to a memory address not used by the main kernel or BIOS using | 1494 | to a memory address not used by the main kernel or BIOS using |
1495 | PHYSICAL_START, or it must be built as a relocatable image | 1495 | PHYSICAL_START, or it must be built as a relocatable image |
1496 | (CONFIG_RELOCATABLE=y). | 1496 | (CONFIG_RELOCATABLE=y). |
1497 | For more details see Documentation/kdump/kdump.txt | 1497 | For more details see Documentation/kdump/kdump.txt |
1498 | 1498 | ||
1499 | config KEXEC_JUMP | 1499 | config KEXEC_JUMP |
1500 | bool "kexec jump (EXPERIMENTAL)" | 1500 | bool "kexec jump (EXPERIMENTAL)" |
1501 | depends on EXPERIMENTAL | 1501 | depends on EXPERIMENTAL |
1502 | depends on KEXEC && HIBERNATION | 1502 | depends on KEXEC && HIBERNATION |
1503 | ---help--- | 1503 | ---help--- |
1504 | Jump between original kernel and kexeced kernel and invoke | 1504 | Jump between original kernel and kexeced kernel and invoke |
1505 | code in physical address mode via KEXEC | 1505 | code in physical address mode via KEXEC |
1506 | 1506 | ||
1507 | config PHYSICAL_START | 1507 | config PHYSICAL_START |
1508 | hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP) | 1508 | hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP) |
1509 | default "0x1000000" | 1509 | default "0x1000000" |
1510 | ---help--- | 1510 | ---help--- |
1511 | This gives the physical address where the kernel is loaded. | 1511 | This gives the physical address where the kernel is loaded. |
1512 | 1512 | ||
1513 | If kernel is a not relocatable (CONFIG_RELOCATABLE=n) then | 1513 | If kernel is a not relocatable (CONFIG_RELOCATABLE=n) then |
1514 | bzImage will decompress itself to above physical address and | 1514 | bzImage will decompress itself to above physical address and |
1515 | run from there. Otherwise, bzImage will run from the address where | 1515 | run from there. Otherwise, bzImage will run from the address where |
1516 | it has been loaded by the boot loader and will ignore above physical | 1516 | it has been loaded by the boot loader and will ignore above physical |
1517 | address. | 1517 | address. |
1518 | 1518 | ||
1519 | In normal kdump cases one does not have to set/change this option | 1519 | In normal kdump cases one does not have to set/change this option |
1520 | as now bzImage can be compiled as a completely relocatable image | 1520 | as now bzImage can be compiled as a completely relocatable image |
1521 | (CONFIG_RELOCATABLE=y) and be used to load and run from a different | 1521 | (CONFIG_RELOCATABLE=y) and be used to load and run from a different |
1522 | address. This option is mainly useful for the folks who don't want | 1522 | address. This option is mainly useful for the folks who don't want |
1523 | to use a bzImage for capturing the crash dump and want to use a | 1523 | to use a bzImage for capturing the crash dump and want to use a |
1524 | vmlinux instead. vmlinux is not relocatable hence a kernel needs | 1524 | vmlinux instead. vmlinux is not relocatable hence a kernel needs |
1525 | to be specifically compiled to run from a specific memory area | 1525 | to be specifically compiled to run from a specific memory area |
1526 | (normally a reserved region) and this option comes handy. | 1526 | (normally a reserved region) and this option comes handy. |
1527 | 1527 | ||
1528 | So if you are using bzImage for capturing the crash dump, | 1528 | So if you are using bzImage for capturing the crash dump, |
1529 | leave the value here unchanged to 0x1000000 and set | 1529 | leave the value here unchanged to 0x1000000 and set |
1530 | CONFIG_RELOCATABLE=y. Otherwise if you plan to use vmlinux | 1530 | CONFIG_RELOCATABLE=y. Otherwise if you plan to use vmlinux |
1531 | for capturing the crash dump change this value to start of | 1531 | for capturing the crash dump change this value to start of |
1532 | the reserved region. In other words, it can be set based on | 1532 | the reserved region. In other words, it can be set based on |
1533 | the "X" value as specified in the "crashkernel=YM@XM" | 1533 | the "X" value as specified in the "crashkernel=YM@XM" |
1534 | command line boot parameter passed to the panic-ed | 1534 | command line boot parameter passed to the panic-ed |
1535 | kernel. Please take a look at Documentation/kdump/kdump.txt | 1535 | kernel. Please take a look at Documentation/kdump/kdump.txt |
1536 | for more details about crash dumps. | 1536 | for more details about crash dumps. |
1537 | 1537 | ||
1538 | Usage of bzImage for capturing the crash dump is recommended as | 1538 | Usage of bzImage for capturing the crash dump is recommended as |
1539 | one does not have to build two kernels. Same kernel can be used | 1539 | one does not have to build two kernels. Same kernel can be used |
1540 | as production kernel and capture kernel. Above option should have | 1540 | as production kernel and capture kernel. Above option should have |
1541 | gone away after relocatable bzImage support is introduced. But it | 1541 | gone away after relocatable bzImage support is introduced. But it |
1542 | is present because there are users out there who continue to use | 1542 | is present because there are users out there who continue to use |
1543 | vmlinux for dump capture. This option should go away down the | 1543 | vmlinux for dump capture. This option should go away down the |
1544 | line. | 1544 | line. |
1545 | 1545 | ||
1546 | Don't change this unless you know what you are doing. | 1546 | Don't change this unless you know what you are doing. |
1547 | 1547 | ||
1548 | config RELOCATABLE | 1548 | config RELOCATABLE |
1549 | bool "Build a relocatable kernel" | 1549 | bool "Build a relocatable kernel" |
1550 | default y | 1550 | default y |
1551 | ---help--- | 1551 | ---help--- |
1552 | This builds a kernel image that retains relocation information | 1552 | This builds a kernel image that retains relocation information |
1553 | so it can be loaded someplace besides the default 1MB. | 1553 | so it can be loaded someplace besides the default 1MB. |
1554 | The relocations tend to make the kernel binary about 10% larger, | 1554 | The relocations tend to make the kernel binary about 10% larger, |
1555 | but are discarded at runtime. | 1555 | but are discarded at runtime. |
1556 | 1556 | ||
1557 | One use is for the kexec on panic case where the recovery kernel | 1557 | One use is for the kexec on panic case where the recovery kernel |
1558 | must live at a different physical address than the primary | 1558 | must live at a different physical address than the primary |
1559 | kernel. | 1559 | kernel. |
1560 | 1560 | ||
1561 | Note: If CONFIG_RELOCATABLE=y, then the kernel runs from the address | 1561 | Note: If CONFIG_RELOCATABLE=y, then the kernel runs from the address |
1562 | it has been loaded at and the compile time physical address | 1562 | it has been loaded at and the compile time physical address |
1563 | (CONFIG_PHYSICAL_START) is ignored. | 1563 | (CONFIG_PHYSICAL_START) is ignored. |
1564 | 1564 | ||
1565 | # Relocation on x86-32 needs some additional build support | 1565 | # Relocation on x86-32 needs some additional build support |
1566 | config X86_NEED_RELOCS | 1566 | config X86_NEED_RELOCS |
1567 | def_bool y | 1567 | def_bool y |
1568 | depends on X86_32 && RELOCATABLE | 1568 | depends on X86_32 && RELOCATABLE |
1569 | 1569 | ||
1570 | config PHYSICAL_ALIGN | 1570 | config PHYSICAL_ALIGN |
1571 | hex | 1571 | hex |
1572 | prompt "Alignment value to which kernel should be aligned" if X86_32 | 1572 | prompt "Alignment value to which kernel should be aligned" if X86_32 |
1573 | default "0x1000000" | 1573 | default "0x1000000" |
1574 | range 0x2000 0x1000000 | 1574 | range 0x2000 0x1000000 |
1575 | ---help--- | 1575 | ---help--- |
1576 | This value puts the alignment restrictions on physical address | 1576 | This value puts the alignment restrictions on physical address |
1577 | where kernel is loaded and run from. Kernel is compiled for an | 1577 | where kernel is loaded and run from. Kernel is compiled for an |
1578 | address which meets above alignment restriction. | 1578 | address which meets above alignment restriction. |
1579 | 1579 | ||
1580 | If bootloader loads the kernel at a non-aligned address and | 1580 | If bootloader loads the kernel at a non-aligned address and |
1581 | CONFIG_RELOCATABLE is set, kernel will move itself to nearest | 1581 | CONFIG_RELOCATABLE is set, kernel will move itself to nearest |
1582 | address aligned to above value and run from there. | 1582 | address aligned to above value and run from there. |
1583 | 1583 | ||
1584 | If bootloader loads the kernel at a non-aligned address and | 1584 | If bootloader loads the kernel at a non-aligned address and |
1585 | CONFIG_RELOCATABLE is not set, kernel will ignore the run time | 1585 | CONFIG_RELOCATABLE is not set, kernel will ignore the run time |
1586 | load address and decompress itself to the address it has been | 1586 | load address and decompress itself to the address it has been |
1587 | compiled for and run from there. The address for which kernel is | 1587 | compiled for and run from there. The address for which kernel is |
1588 | compiled already meets above alignment restrictions. Hence the | 1588 | compiled already meets above alignment restrictions. Hence the |
1589 | end result is that kernel runs from a physical address meeting | 1589 | end result is that kernel runs from a physical address meeting |
1590 | above alignment restrictions. | 1590 | above alignment restrictions. |
1591 | 1591 | ||
1592 | Don't change this unless you know what you are doing. | 1592 | Don't change this unless you know what you are doing. |
1593 | 1593 | ||
1594 | config HOTPLUG_CPU | 1594 | config HOTPLUG_CPU |
1595 | bool "Support for hot-pluggable CPUs" | 1595 | bool "Support for hot-pluggable CPUs" |
1596 | depends on SMP && HOTPLUG | 1596 | depends on SMP && HOTPLUG |
1597 | ---help--- | 1597 | ---help--- |
1598 | Say Y here to allow turning CPUs off and on. CPUs can be | 1598 | Say Y here to allow turning CPUs off and on. CPUs can be |
1599 | controlled through /sys/devices/system/cpu. | 1599 | controlled through /sys/devices/system/cpu. |
1600 | ( Note: power management support will enable this option | 1600 | ( Note: power management support will enable this option |
1601 | automatically on SMP systems. ) | 1601 | automatically on SMP systems. ) |
1602 | Say N if you want to disable CPU hotplug. | 1602 | Say N if you want to disable CPU hotplug. |
1603 | 1603 | ||
1604 | config COMPAT_VDSO | 1604 | config COMPAT_VDSO |
1605 | def_bool y | 1605 | def_bool y |
1606 | prompt "Compat VDSO support" | 1606 | prompt "Compat VDSO support" |
1607 | depends on X86_32 || IA32_EMULATION | 1607 | depends on X86_32 || IA32_EMULATION |
1608 | ---help--- | 1608 | ---help--- |
1609 | Map the 32-bit VDSO to the predictable old-style address too. | 1609 | Map the 32-bit VDSO to the predictable old-style address too. |
1610 | ---help--- | 1610 | ---help--- |
1611 | Say N here if you are running a sufficiently recent glibc | 1611 | Say N here if you are running a sufficiently recent glibc |
1612 | version (2.3.3 or later), to remove the high-mapped | 1612 | version (2.3.3 or later), to remove the high-mapped |
1613 | VDSO mapping and to exclusively use the randomized VDSO. | 1613 | VDSO mapping and to exclusively use the randomized VDSO. |
1614 | 1614 | ||
1615 | If unsure, say Y. | 1615 | If unsure, say Y. |
1616 | 1616 | ||
1617 | config CMDLINE_BOOL | 1617 | config CMDLINE_BOOL |
1618 | bool "Built-in kernel command line" | 1618 | bool "Built-in kernel command line" |
1619 | default n | 1619 | default n |
1620 | ---help--- | 1620 | ---help--- |
1621 | Allow for specifying boot arguments to the kernel at | 1621 | Allow for specifying boot arguments to the kernel at |
1622 | build time. On some systems (e.g. embedded ones), it is | 1622 | build time. On some systems (e.g. embedded ones), it is |
1623 | necessary or convenient to provide some or all of the | 1623 | necessary or convenient to provide some or all of the |
1624 | kernel boot arguments with the kernel itself (that is, | 1624 | kernel boot arguments with the kernel itself (that is, |
1625 | to not rely on the boot loader to provide them.) | 1625 | to not rely on the boot loader to provide them.) |
1626 | 1626 | ||
1627 | To compile command line arguments into the kernel, | 1627 | To compile command line arguments into the kernel, |
1628 | set this option to 'Y', then fill in the | 1628 | set this option to 'Y', then fill in the |
1629 | the boot arguments in CONFIG_CMDLINE. | 1629 | the boot arguments in CONFIG_CMDLINE. |
1630 | 1630 | ||
1631 | Systems with fully functional boot loaders (i.e. non-embedded) | 1631 | Systems with fully functional boot loaders (i.e. non-embedded) |
1632 | should leave this option set to 'N'. | 1632 | should leave this option set to 'N'. |
1633 | 1633 | ||
1634 | config CMDLINE | 1634 | config CMDLINE |
1635 | string "Built-in kernel command string" | 1635 | string "Built-in kernel command string" |
1636 | depends on CMDLINE_BOOL | 1636 | depends on CMDLINE_BOOL |
1637 | default "" | 1637 | default "" |
1638 | ---help--- | 1638 | ---help--- |
1639 | Enter arguments here that should be compiled into the kernel | 1639 | Enter arguments here that should be compiled into the kernel |
1640 | image and used at boot time. If the boot loader provides a | 1640 | image and used at boot time. If the boot loader provides a |
1641 | command line at boot time, it is appended to this string to | 1641 | command line at boot time, it is appended to this string to |
1642 | form the full kernel command line, when the system boots. | 1642 | form the full kernel command line, when the system boots. |
1643 | 1643 | ||
1644 | However, you can use the CONFIG_CMDLINE_OVERRIDE option to | 1644 | However, you can use the CONFIG_CMDLINE_OVERRIDE option to |
1645 | change this behavior. | 1645 | change this behavior. |
1646 | 1646 | ||
1647 | In most cases, the command line (whether built-in or provided | 1647 | In most cases, the command line (whether built-in or provided |
1648 | by the boot loader) should specify the device for the root | 1648 | by the boot loader) should specify the device for the root |
1649 | file system. | 1649 | file system. |
1650 | 1650 | ||
1651 | config CMDLINE_OVERRIDE | 1651 | config CMDLINE_OVERRIDE |
1652 | bool "Built-in command line overrides boot loader arguments" | 1652 | bool "Built-in command line overrides boot loader arguments" |
1653 | default n | 1653 | default n |
1654 | depends on CMDLINE_BOOL | 1654 | depends on CMDLINE_BOOL |
1655 | ---help--- | 1655 | ---help--- |
1656 | Set this option to 'Y' to have the kernel ignore the boot loader | 1656 | Set this option to 'Y' to have the kernel ignore the boot loader |
1657 | command line, and use ONLY the built-in command line. | 1657 | command line, and use ONLY the built-in command line. |
1658 | 1658 | ||
1659 | This is used to work around broken boot loaders. This should | 1659 | This is used to work around broken boot loaders. This should |
1660 | be set to 'N' under normal conditions. | 1660 | be set to 'N' under normal conditions. |
1661 | 1661 | ||
1662 | endmenu | 1662 | endmenu |
1663 | 1663 | ||
1664 | config ARCH_ENABLE_MEMORY_HOTPLUG | 1664 | config ARCH_ENABLE_MEMORY_HOTPLUG |
1665 | def_bool y | 1665 | def_bool y |
1666 | depends on X86_64 || (X86_32 && HIGHMEM) | 1666 | depends on X86_64 || (X86_32 && HIGHMEM) |
1667 | 1667 | ||
1668 | config ARCH_ENABLE_MEMORY_HOTREMOVE | 1668 | config ARCH_ENABLE_MEMORY_HOTREMOVE |
1669 | def_bool y | 1669 | def_bool y |
1670 | depends on MEMORY_HOTPLUG | 1670 | depends on MEMORY_HOTPLUG |
1671 | 1671 | ||
1672 | config HAVE_ARCH_EARLY_PFN_TO_NID | 1672 | config HAVE_ARCH_EARLY_PFN_TO_NID |
1673 | def_bool X86_64 | 1673 | def_bool X86_64 |
1674 | depends on NUMA | 1674 | depends on NUMA |
1675 | 1675 | ||
1676 | menu "Power management and ACPI options" | 1676 | menu "Power management and ACPI options" |
1677 | 1677 | ||
1678 | config ARCH_HIBERNATION_HEADER | 1678 | config ARCH_HIBERNATION_HEADER |
1679 | def_bool y | 1679 | def_bool y |
1680 | depends on X86_64 && HIBERNATION | 1680 | depends on X86_64 && HIBERNATION |
1681 | 1681 | ||
1682 | source "kernel/power/Kconfig" | 1682 | source "kernel/power/Kconfig" |
1683 | 1683 | ||
1684 | source "drivers/acpi/Kconfig" | 1684 | source "drivers/acpi/Kconfig" |
1685 | 1685 | ||
1686 | config X86_APM_BOOT | 1686 | config X86_APM_BOOT |
1687 | bool | 1687 | bool |
1688 | default y | 1688 | default y |
1689 | depends on APM || APM_MODULE | 1689 | depends on APM || APM_MODULE |
1690 | 1690 | ||
1691 | menuconfig APM | 1691 | menuconfig APM |
1692 | tristate "APM (Advanced Power Management) BIOS support" | 1692 | tristate "APM (Advanced Power Management) BIOS support" |
1693 | depends on X86_32 && PM_SLEEP | 1693 | depends on X86_32 && PM_SLEEP |
1694 | ---help--- | 1694 | ---help--- |
1695 | APM is a BIOS specification for saving power using several different | 1695 | APM is a BIOS specification for saving power using several different |
1696 | techniques. This is mostly useful for battery powered laptops with | 1696 | techniques. This is mostly useful for battery powered laptops with |
1697 | APM compliant BIOSes. If you say Y here, the system time will be | 1697 | APM compliant BIOSes. If you say Y here, the system time will be |
1698 | reset after a RESUME operation, the /proc/apm device will provide | 1698 | reset after a RESUME operation, the /proc/apm device will provide |
1699 | battery status information, and user-space programs will receive | 1699 | battery status information, and user-space programs will receive |
1700 | notification of APM "events" (e.g. battery status change). | 1700 | notification of APM "events" (e.g. battery status change). |
1701 | 1701 | ||
1702 | If you select "Y" here, you can disable actual use of the APM | 1702 | If you select "Y" here, you can disable actual use of the APM |
1703 | BIOS by passing the "apm=off" option to the kernel at boot time. | 1703 | BIOS by passing the "apm=off" option to the kernel at boot time. |
1704 | 1704 | ||
1705 | Note that the APM support is almost completely disabled for | 1705 | Note that the APM support is almost completely disabled for |
1706 | machines with more than one CPU. | 1706 | machines with more than one CPU. |
1707 | 1707 | ||
1708 | In order to use APM, you will need supporting software. For location | 1708 | In order to use APM, you will need supporting software. For location |
1709 | and more information, read <file:Documentation/power/pm.txt> and the | 1709 | and more information, read <file:Documentation/power/pm.txt> and the |
1710 | Battery Powered Linux mini-HOWTO, available from | 1710 | Battery Powered Linux mini-HOWTO, available from |
1711 | <http://www.tldp.org/docs.html#howto>. | 1711 | <http://www.tldp.org/docs.html#howto>. |
1712 | 1712 | ||
1713 | This driver does not spin down disk drives (see the hdparm(8) | 1713 | This driver does not spin down disk drives (see the hdparm(8) |
1714 | manpage ("man 8 hdparm") for that), and it doesn't turn off | 1714 | manpage ("man 8 hdparm") for that), and it doesn't turn off |
1715 | VESA-compliant "green" monitors. | 1715 | VESA-compliant "green" monitors. |
1716 | 1716 | ||
1717 | This driver does not support the TI 4000M TravelMate and the ACER | 1717 | This driver does not support the TI 4000M TravelMate and the ACER |
1718 | 486/DX4/75 because they don't have compliant BIOSes. Many "green" | 1718 | 486/DX4/75 because they don't have compliant BIOSes. Many "green" |
1719 | desktop machines also don't have compliant BIOSes, and this driver | 1719 | desktop machines also don't have compliant BIOSes, and this driver |
1720 | may cause those machines to panic during the boot phase. | 1720 | may cause those machines to panic during the boot phase. |
1721 | 1721 | ||
1722 | Generally, if you don't have a battery in your machine, there isn't | 1722 | Generally, if you don't have a battery in your machine, there isn't |
1723 | much point in using this driver and you should say N. If you get | 1723 | much point in using this driver and you should say N. If you get |
1724 | random kernel OOPSes or reboots that don't seem to be related to | 1724 | random kernel OOPSes or reboots that don't seem to be related to |
1725 | anything, try disabling/enabling this option (or disabling/enabling | 1725 | anything, try disabling/enabling this option (or disabling/enabling |
1726 | APM in your BIOS). | 1726 | APM in your BIOS). |
1727 | 1727 | ||
1728 | Some other things you should try when experiencing seemingly random, | 1728 | Some other things you should try when experiencing seemingly random, |
1729 | "weird" problems: | 1729 | "weird" problems: |
1730 | 1730 | ||
1731 | 1) make sure that you have enough swap space and that it is | 1731 | 1) make sure that you have enough swap space and that it is |
1732 | enabled. | 1732 | enabled. |
1733 | 2) pass the "no-hlt" option to the kernel | 1733 | 2) pass the "no-hlt" option to the kernel |
1734 | 3) switch on floating point emulation in the kernel and pass | 1734 | 3) switch on floating point emulation in the kernel and pass |
1735 | the "no387" option to the kernel | 1735 | the "no387" option to the kernel |
1736 | 4) pass the "floppy=nodma" option to the kernel | 1736 | 4) pass the "floppy=nodma" option to the kernel |
1737 | 5) pass the "mem=4M" option to the kernel (thereby disabling | 1737 | 5) pass the "mem=4M" option to the kernel (thereby disabling |
1738 | all but the first 4 MB of RAM) | 1738 | all but the first 4 MB of RAM) |
1739 | 6) make sure that the CPU is not over clocked. | 1739 | 6) make sure that the CPU is not over clocked. |
1740 | 7) read the sig11 FAQ at <http://www.bitwizard.nl/sig11/> | 1740 | 7) read the sig11 FAQ at <http://www.bitwizard.nl/sig11/> |
1741 | 8) disable the cache from your BIOS settings | 1741 | 8) disable the cache from your BIOS settings |
1742 | 9) install a fan for the video card or exchange video RAM | 1742 | 9) install a fan for the video card or exchange video RAM |
1743 | 10) install a better fan for the CPU | 1743 | 10) install a better fan for the CPU |
1744 | 11) exchange RAM chips | 1744 | 11) exchange RAM chips |
1745 | 12) exchange the motherboard. | 1745 | 12) exchange the motherboard. |
1746 | 1746 | ||
1747 | To compile this driver as a module, choose M here: the | 1747 | To compile this driver as a module, choose M here: the |
1748 | module will be called apm. | 1748 | module will be called apm. |
1749 | 1749 | ||
1750 | if APM | 1750 | if APM |
1751 | 1751 | ||
1752 | config APM_IGNORE_USER_SUSPEND | 1752 | config APM_IGNORE_USER_SUSPEND |
1753 | bool "Ignore USER SUSPEND" | 1753 | bool "Ignore USER SUSPEND" |
1754 | ---help--- | 1754 | ---help--- |
1755 | This option will ignore USER SUSPEND requests. On machines with a | 1755 | This option will ignore USER SUSPEND requests. On machines with a |
1756 | compliant APM BIOS, you want to say N. However, on the NEC Versa M | 1756 | compliant APM BIOS, you want to say N. However, on the NEC Versa M |
1757 | series notebooks, it is necessary to say Y because of a BIOS bug. | 1757 | series notebooks, it is necessary to say Y because of a BIOS bug. |
1758 | 1758 | ||
1759 | config APM_DO_ENABLE | 1759 | config APM_DO_ENABLE |
1760 | bool "Enable PM at boot time" | 1760 | bool "Enable PM at boot time" |
1761 | ---help--- | 1761 | ---help--- |
1762 | Enable APM features at boot time. From page 36 of the APM BIOS | 1762 | Enable APM features at boot time. From page 36 of the APM BIOS |
1763 | specification: "When disabled, the APM BIOS does not automatically | 1763 | specification: "When disabled, the APM BIOS does not automatically |
1764 | power manage devices, enter the Standby State, enter the Suspend | 1764 | power manage devices, enter the Standby State, enter the Suspend |
1765 | State, or take power saving steps in response to CPU Idle calls." | 1765 | State, or take power saving steps in response to CPU Idle calls." |
1766 | This driver will make CPU Idle calls when Linux is idle (unless this | 1766 | This driver will make CPU Idle calls when Linux is idle (unless this |
1767 | feature is turned off -- see "Do CPU IDLE calls", below). This | 1767 | feature is turned off -- see "Do CPU IDLE calls", below). This |
1768 | should always save battery power, but more complicated APM features | 1768 | should always save battery power, but more complicated APM features |
1769 | will be dependent on your BIOS implementation. You may need to turn | 1769 | will be dependent on your BIOS implementation. You may need to turn |
1770 | this option off if your computer hangs at boot time when using APM | 1770 | this option off if your computer hangs at boot time when using APM |
1771 | support, or if it beeps continuously instead of suspending. Turn | 1771 | support, or if it beeps continuously instead of suspending. Turn |
1772 | this off if you have a NEC UltraLite Versa 33/C or a Toshiba | 1772 | this off if you have a NEC UltraLite Versa 33/C or a Toshiba |
1773 | T400CDT. This is off by default since most machines do fine without | 1773 | T400CDT. This is off by default since most machines do fine without |
1774 | this feature. | 1774 | this feature. |
1775 | 1775 | ||
1776 | config APM_CPU_IDLE | 1776 | config APM_CPU_IDLE |
1777 | bool "Make CPU Idle calls when idle" | 1777 | bool "Make CPU Idle calls when idle" |
1778 | ---help--- | 1778 | ---help--- |
1779 | Enable calls to APM CPU Idle/CPU Busy inside the kernel's idle loop. | 1779 | Enable calls to APM CPU Idle/CPU Busy inside the kernel's idle loop. |
1780 | On some machines, this can activate improved power savings, such as | 1780 | On some machines, this can activate improved power savings, such as |
1781 | a slowed CPU clock rate, when the machine is idle. These idle calls | 1781 | a slowed CPU clock rate, when the machine is idle. These idle calls |
1782 | are made after the idle loop has run for some length of time (e.g., | 1782 | are made after the idle loop has run for some length of time (e.g., |
1783 | 333 mS). On some machines, this will cause a hang at boot time or | 1783 | 333 mS). On some machines, this will cause a hang at boot time or |
1784 | whenever the CPU becomes idle. (On machines with more than one CPU, | 1784 | whenever the CPU becomes idle. (On machines with more than one CPU, |
1785 | this option does nothing.) | 1785 | this option does nothing.) |
1786 | 1786 | ||
1787 | config APM_DISPLAY_BLANK | 1787 | config APM_DISPLAY_BLANK |
1788 | bool "Enable console blanking using APM" | 1788 | bool "Enable console blanking using APM" |
1789 | ---help--- | 1789 | ---help--- |
1790 | Enable console blanking using the APM. Some laptops can use this to | 1790 | Enable console blanking using the APM. Some laptops can use this to |
1791 | turn off the LCD backlight when the screen blanker of the Linux | 1791 | turn off the LCD backlight when the screen blanker of the Linux |
1792 | virtual console blanks the screen. Note that this is only used by | 1792 | virtual console blanks the screen. Note that this is only used by |
1793 | the virtual console screen blanker, and won't turn off the backlight | 1793 | the virtual console screen blanker, and won't turn off the backlight |
1794 | when using the X Window system. This also doesn't have anything to | 1794 | when using the X Window system. This also doesn't have anything to |
1795 | do with your VESA-compliant power-saving monitor. Further, this | 1795 | do with your VESA-compliant power-saving monitor. Further, this |
1796 | option doesn't work for all laptops -- it might not turn off your | 1796 | option doesn't work for all laptops -- it might not turn off your |
1797 | backlight at all, or it might print a lot of errors to the console, | 1797 | backlight at all, or it might print a lot of errors to the console, |
1798 | especially if you are using gpm. | 1798 | especially if you are using gpm. |
1799 | 1799 | ||
1800 | config APM_ALLOW_INTS | 1800 | config APM_ALLOW_INTS |
1801 | bool "Allow interrupts during APM BIOS calls" | 1801 | bool "Allow interrupts during APM BIOS calls" |
1802 | ---help--- | 1802 | ---help--- |
1803 | Normally we disable external interrupts while we are making calls to | 1803 | Normally we disable external interrupts while we are making calls to |
1804 | the APM BIOS as a measure to lessen the effects of a badly behaving | 1804 | the APM BIOS as a measure to lessen the effects of a badly behaving |
1805 | BIOS implementation. The BIOS should reenable interrupts if it | 1805 | BIOS implementation. The BIOS should reenable interrupts if it |
1806 | needs to. Unfortunately, some BIOSes do not -- especially those in | 1806 | needs to. Unfortunately, some BIOSes do not -- especially those in |
1807 | many of the newer IBM Thinkpads. If you experience hangs when you | 1807 | many of the newer IBM Thinkpads. If you experience hangs when you |
1808 | suspend, try setting this to Y. Otherwise, say N. | 1808 | suspend, try setting this to Y. Otherwise, say N. |
1809 | 1809 | ||
1810 | endif # APM | 1810 | endif # APM |
1811 | 1811 | ||
1812 | source "arch/x86/kernel/cpu/cpufreq/Kconfig" | 1812 | source "arch/x86/kernel/cpu/cpufreq/Kconfig" |
1813 | 1813 | ||
1814 | source "drivers/cpuidle/Kconfig" | 1814 | source "drivers/cpuidle/Kconfig" |
1815 | 1815 | ||
1816 | source "drivers/idle/Kconfig" | 1816 | source "drivers/idle/Kconfig" |
1817 | 1817 | ||
1818 | endmenu | 1818 | endmenu |
1819 | 1819 | ||
1820 | 1820 | ||
1821 | menu "Bus options (PCI etc.)" | 1821 | menu "Bus options (PCI etc.)" |
1822 | 1822 | ||
1823 | config PCI | 1823 | config PCI |
1824 | bool "PCI support" | 1824 | bool "PCI support" |
1825 | default y | 1825 | default y |
1826 | select ARCH_SUPPORTS_MSI if (X86_LOCAL_APIC && X86_IO_APIC) | 1826 | select ARCH_SUPPORTS_MSI if (X86_LOCAL_APIC && X86_IO_APIC) |
1827 | ---help--- | 1827 | ---help--- |
1828 | Find out whether you have a PCI motherboard. PCI is the name of a | 1828 | Find out whether you have a PCI motherboard. PCI is the name of a |
1829 | bus system, i.e. the way the CPU talks to the other stuff inside | 1829 | bus system, i.e. the way the CPU talks to the other stuff inside |
1830 | your box. Other bus systems are ISA, EISA, MicroChannel (MCA) or | 1830 | your box. Other bus systems are ISA, EISA, MicroChannel (MCA) or |
1831 | VESA. If you have PCI, say Y, otherwise N. | 1831 | VESA. If you have PCI, say Y, otherwise N. |
1832 | 1832 | ||
1833 | choice | 1833 | choice |
1834 | prompt "PCI access mode" | 1834 | prompt "PCI access mode" |
1835 | depends on X86_32 && PCI | 1835 | depends on X86_32 && PCI |
1836 | default PCI_GOANY | 1836 | default PCI_GOANY |
1837 | ---help--- | 1837 | ---help--- |
1838 | On PCI systems, the BIOS can be used to detect the PCI devices and | 1838 | On PCI systems, the BIOS can be used to detect the PCI devices and |
1839 | determine their configuration. However, some old PCI motherboards | 1839 | determine their configuration. However, some old PCI motherboards |
1840 | have BIOS bugs and may crash if this is done. Also, some embedded | 1840 | have BIOS bugs and may crash if this is done. Also, some embedded |
1841 | PCI-based systems don't have any BIOS at all. Linux can also try to | 1841 | PCI-based systems don't have any BIOS at all. Linux can also try to |
1842 | detect the PCI hardware directly without using the BIOS. | 1842 | detect the PCI hardware directly without using the BIOS. |
1843 | 1843 | ||
1844 | With this option, you can specify how Linux should detect the | 1844 | With this option, you can specify how Linux should detect the |
1845 | PCI devices. If you choose "BIOS", the BIOS will be used, | 1845 | PCI devices. If you choose "BIOS", the BIOS will be used, |
1846 | if you choose "Direct", the BIOS won't be used, and if you | 1846 | if you choose "Direct", the BIOS won't be used, and if you |
1847 | choose "MMConfig", then PCI Express MMCONFIG will be used. | 1847 | choose "MMConfig", then PCI Express MMCONFIG will be used. |
1848 | If you choose "Any", the kernel will try MMCONFIG, then the | 1848 | If you choose "Any", the kernel will try MMCONFIG, then the |
1849 | direct access method and falls back to the BIOS if that doesn't | 1849 | direct access method and falls back to the BIOS if that doesn't |
1850 | work. If unsure, go with the default, which is "Any". | 1850 | work. If unsure, go with the default, which is "Any". |
1851 | 1851 | ||
1852 | config PCI_GOBIOS | 1852 | config PCI_GOBIOS |
1853 | bool "BIOS" | 1853 | bool "BIOS" |
1854 | 1854 | ||
1855 | config PCI_GOMMCONFIG | 1855 | config PCI_GOMMCONFIG |
1856 | bool "MMConfig" | 1856 | bool "MMConfig" |
1857 | 1857 | ||
1858 | config PCI_GODIRECT | 1858 | config PCI_GODIRECT |
1859 | bool "Direct" | 1859 | bool "Direct" |
1860 | 1860 | ||
1861 | config PCI_GOOLPC | 1861 | config PCI_GOOLPC |
1862 | bool "OLPC" | 1862 | bool "OLPC" |
1863 | depends on OLPC | 1863 | depends on OLPC |
1864 | 1864 | ||
1865 | config PCI_GOANY | 1865 | config PCI_GOANY |
1866 | bool "Any" | 1866 | bool "Any" |
1867 | 1867 | ||
1868 | endchoice | 1868 | endchoice |
1869 | 1869 | ||
1870 | config PCI_BIOS | 1870 | config PCI_BIOS |
1871 | def_bool y | 1871 | def_bool y |
1872 | depends on X86_32 && PCI && (PCI_GOBIOS || PCI_GOANY) | 1872 | depends on X86_32 && PCI && (PCI_GOBIOS || PCI_GOANY) |
1873 | 1873 | ||
1874 | # x86-64 doesn't support PCI BIOS access from long mode so always go direct. | 1874 | # x86-64 doesn't support PCI BIOS access from long mode so always go direct. |
1875 | config PCI_DIRECT | 1875 | config PCI_DIRECT |
1876 | def_bool y | 1876 | def_bool y |
1877 | depends on PCI && (X86_64 || (PCI_GODIRECT || PCI_GOANY || PCI_GOOLPC)) | 1877 | depends on PCI && (X86_64 || (PCI_GODIRECT || PCI_GOANY || PCI_GOOLPC)) |
1878 | 1878 | ||
1879 | config PCI_MMCONFIG | 1879 | config PCI_MMCONFIG |
1880 | def_bool y | 1880 | def_bool y |
1881 | depends on X86_32 && PCI && ACPI && (PCI_GOMMCONFIG || PCI_GOANY) | 1881 | depends on X86_32 && PCI && ACPI && (PCI_GOMMCONFIG || PCI_GOANY) |
1882 | 1882 | ||
1883 | config PCI_OLPC | 1883 | config PCI_OLPC |
1884 | def_bool y | 1884 | def_bool y |
1885 | depends on PCI && OLPC && (PCI_GOOLPC || PCI_GOANY) | 1885 | depends on PCI && OLPC && (PCI_GOOLPC || PCI_GOANY) |
1886 | 1886 | ||
1887 | config PCI_DOMAINS | 1887 | config PCI_DOMAINS |
1888 | def_bool y | 1888 | def_bool y |
1889 | depends on PCI | 1889 | depends on PCI |
1890 | 1890 | ||
1891 | config PCI_MMCONFIG | 1891 | config PCI_MMCONFIG |
1892 | bool "Support mmconfig PCI config space access" | 1892 | bool "Support mmconfig PCI config space access" |
1893 | depends on X86_64 && PCI && ACPI | 1893 | depends on X86_64 && PCI && ACPI |
1894 | 1894 | ||
1895 | config DMAR | 1895 | config DMAR |
1896 | bool "Support for DMA Remapping Devices (EXPERIMENTAL)" | 1896 | bool "Support for DMA Remapping Devices (EXPERIMENTAL)" |
1897 | depends on PCI_MSI && ACPI && EXPERIMENTAL | 1897 | depends on PCI_MSI && ACPI && EXPERIMENTAL |
1898 | help | 1898 | help |
1899 | DMA remapping (DMAR) devices support enables independent address | 1899 | DMA remapping (DMAR) devices support enables independent address |
1900 | translations for Direct Memory Access (DMA) from devices. | 1900 | translations for Direct Memory Access (DMA) from devices. |
1901 | These DMA remapping devices are reported via ACPI tables | 1901 | These DMA remapping devices are reported via ACPI tables |
1902 | and include PCI device scope covered by these DMA | 1902 | and include PCI device scope covered by these DMA |
1903 | remapping devices. | 1903 | remapping devices. |
1904 | 1904 | ||
1905 | config DMAR_DEFAULT_ON | 1905 | config DMAR_DEFAULT_ON |
1906 | def_bool y | 1906 | def_bool y |
1907 | prompt "Enable DMA Remapping Devices by default" | 1907 | prompt "Enable DMA Remapping Devices by default" |
1908 | depends on DMAR | 1908 | depends on DMAR |
1909 | help | 1909 | help |
1910 | Selecting this option will enable a DMAR device at boot time if | 1910 | Selecting this option will enable a DMAR device at boot time if |
1911 | one is found. If this option is not selected, DMAR support can | 1911 | one is found. If this option is not selected, DMAR support can |
1912 | be enabled by passing intel_iommu=on to the kernel. It is | 1912 | be enabled by passing intel_iommu=on to the kernel. It is |
1913 | recommended you say N here while the DMAR code remains | 1913 | recommended you say N here while the DMAR code remains |
1914 | experimental. | 1914 | experimental. |
1915 | 1915 | ||
1916 | config DMAR_BROKEN_GFX_WA | 1916 | config DMAR_BROKEN_GFX_WA |
1917 | def_bool n | 1917 | def_bool n |
1918 | prompt "Workaround broken graphics drivers (going away soon)" | 1918 | prompt "Workaround broken graphics drivers (going away soon)" |
1919 | depends on DMAR | 1919 | depends on DMAR && BROKEN |
1920 | ---help--- | 1920 | ---help--- |
1921 | Current Graphics drivers tend to use physical address | 1921 | Current Graphics drivers tend to use physical address |
1922 | for DMA and avoid using DMA APIs. Setting this config | 1922 | for DMA and avoid using DMA APIs. Setting this config |
1923 | option permits the IOMMU driver to set a unity map for | 1923 | option permits the IOMMU driver to set a unity map for |
1924 | all the OS-visible memory. Hence the driver can continue | 1924 | all the OS-visible memory. Hence the driver can continue |
1925 | to use physical addresses for DMA, at least until this | 1925 | to use physical addresses for DMA, at least until this |
1926 | option is removed in the 2.6.32 kernel. | 1926 | option is removed in the 2.6.32 kernel. |
1927 | 1927 | ||
1928 | config DMAR_FLOPPY_WA | 1928 | config DMAR_FLOPPY_WA |
1929 | def_bool y | 1929 | def_bool y |
1930 | depends on DMAR | 1930 | depends on DMAR |
1931 | ---help--- | 1931 | ---help--- |
1932 | Floppy disk drivers are known to bypass DMA API calls | 1932 | Floppy disk drivers are known to bypass DMA API calls |
1933 | thereby failing to work when IOMMU is enabled. This | 1933 | thereby failing to work when IOMMU is enabled. This |
1934 | workaround will setup a 1:1 mapping for the first | 1934 | workaround will setup a 1:1 mapping for the first |
1935 | 16MiB to make floppy (an ISA device) work. | 1935 | 16MiB to make floppy (an ISA device) work. |
1936 | 1936 | ||
1937 | config INTR_REMAP | 1937 | config INTR_REMAP |
1938 | bool "Support for Interrupt Remapping (EXPERIMENTAL)" | 1938 | bool "Support for Interrupt Remapping (EXPERIMENTAL)" |
1939 | depends on X86_64 && X86_IO_APIC && PCI_MSI && ACPI && EXPERIMENTAL | 1939 | depends on X86_64 && X86_IO_APIC && PCI_MSI && ACPI && EXPERIMENTAL |
1940 | ---help--- | 1940 | ---help--- |
1941 | Supports Interrupt remapping for IO-APIC and MSI devices. | 1941 | Supports Interrupt remapping for IO-APIC and MSI devices. |
1942 | To use x2apic mode in the CPU's which support x2APIC enhancements or | 1942 | To use x2apic mode in the CPU's which support x2APIC enhancements or |
1943 | to support platforms with CPU's having > 8 bit APIC ID, say Y. | 1943 | to support platforms with CPU's having > 8 bit APIC ID, say Y. |
1944 | 1944 | ||
1945 | source "drivers/pci/pcie/Kconfig" | 1945 | source "drivers/pci/pcie/Kconfig" |
1946 | 1946 | ||
1947 | source "drivers/pci/Kconfig" | 1947 | source "drivers/pci/Kconfig" |
1948 | 1948 | ||
1949 | # x86_64 have no ISA slots, but do have ISA-style DMA. | 1949 | # x86_64 have no ISA slots, but do have ISA-style DMA. |
1950 | config ISA_DMA_API | 1950 | config ISA_DMA_API |
1951 | def_bool y | 1951 | def_bool y |
1952 | 1952 | ||
1953 | if X86_32 | 1953 | if X86_32 |
1954 | 1954 | ||
1955 | config ISA | 1955 | config ISA |
1956 | bool "ISA support" | 1956 | bool "ISA support" |
1957 | ---help--- | 1957 | ---help--- |
1958 | Find out whether you have ISA slots on your motherboard. ISA is the | 1958 | Find out whether you have ISA slots on your motherboard. ISA is the |
1959 | name of a bus system, i.e. the way the CPU talks to the other stuff | 1959 | name of a bus system, i.e. the way the CPU talks to the other stuff |
1960 | inside your box. Other bus systems are PCI, EISA, MicroChannel | 1960 | inside your box. Other bus systems are PCI, EISA, MicroChannel |
1961 | (MCA) or VESA. ISA is an older system, now being displaced by PCI; | 1961 | (MCA) or VESA. ISA is an older system, now being displaced by PCI; |
1962 | newer boards don't support it. If you have ISA, say Y, otherwise N. | 1962 | newer boards don't support it. If you have ISA, say Y, otherwise N. |
1963 | 1963 | ||
1964 | config EISA | 1964 | config EISA |
1965 | bool "EISA support" | 1965 | bool "EISA support" |
1966 | depends on ISA | 1966 | depends on ISA |
1967 | ---help--- | 1967 | ---help--- |
1968 | The Extended Industry Standard Architecture (EISA) bus was | 1968 | The Extended Industry Standard Architecture (EISA) bus was |
1969 | developed as an open alternative to the IBM MicroChannel bus. | 1969 | developed as an open alternative to the IBM MicroChannel bus. |
1970 | 1970 | ||
1971 | The EISA bus provided some of the features of the IBM MicroChannel | 1971 | The EISA bus provided some of the features of the IBM MicroChannel |
1972 | bus while maintaining backward compatibility with cards made for | 1972 | bus while maintaining backward compatibility with cards made for |
1973 | the older ISA bus. The EISA bus saw limited use between 1988 and | 1973 | the older ISA bus. The EISA bus saw limited use between 1988 and |
1974 | 1995 when it was made obsolete by the PCI bus. | 1974 | 1995 when it was made obsolete by the PCI bus. |
1975 | 1975 | ||
1976 | Say Y here if you are building a kernel for an EISA-based machine. | 1976 | Say Y here if you are building a kernel for an EISA-based machine. |
1977 | 1977 | ||
1978 | Otherwise, say N. | 1978 | Otherwise, say N. |
1979 | 1979 | ||
1980 | source "drivers/eisa/Kconfig" | 1980 | source "drivers/eisa/Kconfig" |
1981 | 1981 | ||
1982 | config MCA | 1982 | config MCA |
1983 | bool "MCA support" | 1983 | bool "MCA support" |
1984 | ---help--- | 1984 | ---help--- |
1985 | MicroChannel Architecture is found in some IBM PS/2 machines and | 1985 | MicroChannel Architecture is found in some IBM PS/2 machines and |
1986 | laptops. It is a bus system similar to PCI or ISA. See | 1986 | laptops. It is a bus system similar to PCI or ISA. See |
1987 | <file:Documentation/mca.txt> (and especially the web page given | 1987 | <file:Documentation/mca.txt> (and especially the web page given |
1988 | there) before attempting to build an MCA bus kernel. | 1988 | there) before attempting to build an MCA bus kernel. |
1989 | 1989 | ||
1990 | source "drivers/mca/Kconfig" | 1990 | source "drivers/mca/Kconfig" |
1991 | 1991 | ||
1992 | config SCx200 | 1992 | config SCx200 |
1993 | tristate "NatSemi SCx200 support" | 1993 | tristate "NatSemi SCx200 support" |
1994 | ---help--- | 1994 | ---help--- |
1995 | This provides basic support for National Semiconductor's | 1995 | This provides basic support for National Semiconductor's |
1996 | (now AMD's) Geode processors. The driver probes for the | 1996 | (now AMD's) Geode processors. The driver probes for the |
1997 | PCI-IDs of several on-chip devices, so its a good dependency | 1997 | PCI-IDs of several on-chip devices, so its a good dependency |
1998 | for other scx200_* drivers. | 1998 | for other scx200_* drivers. |
1999 | 1999 | ||
2000 | If compiled as a module, the driver is named scx200. | 2000 | If compiled as a module, the driver is named scx200. |
2001 | 2001 | ||
2002 | config SCx200HR_TIMER | 2002 | config SCx200HR_TIMER |
2003 | tristate "NatSemi SCx200 27MHz High-Resolution Timer Support" | 2003 | tristate "NatSemi SCx200 27MHz High-Resolution Timer Support" |
2004 | depends on SCx200 && GENERIC_TIME | 2004 | depends on SCx200 && GENERIC_TIME |
2005 | default y | 2005 | default y |
2006 | ---help--- | 2006 | ---help--- |
2007 | This driver provides a clocksource built upon the on-chip | 2007 | This driver provides a clocksource built upon the on-chip |
2008 | 27MHz high-resolution timer. Its also a workaround for | 2008 | 27MHz high-resolution timer. Its also a workaround for |
2009 | NSC Geode SC-1100's buggy TSC, which loses time when the | 2009 | NSC Geode SC-1100's buggy TSC, which loses time when the |
2010 | processor goes idle (as is done by the scheduler). The | 2010 | processor goes idle (as is done by the scheduler). The |
2011 | other workaround is idle=poll boot option. | 2011 | other workaround is idle=poll boot option. |
2012 | 2012 | ||
2013 | config GEODE_MFGPT_TIMER | 2013 | config GEODE_MFGPT_TIMER |
2014 | def_bool y | 2014 | def_bool y |
2015 | prompt "Geode Multi-Function General Purpose Timer (MFGPT) events" | 2015 | prompt "Geode Multi-Function General Purpose Timer (MFGPT) events" |
2016 | depends on MGEODE_LX && GENERIC_TIME && GENERIC_CLOCKEVENTS | 2016 | depends on MGEODE_LX && GENERIC_TIME && GENERIC_CLOCKEVENTS |
2017 | ---help--- | 2017 | ---help--- |
2018 | This driver provides a clock event source based on the MFGPT | 2018 | This driver provides a clock event source based on the MFGPT |
2019 | timer(s) in the CS5535 and CS5536 companion chip for the geode. | 2019 | timer(s) in the CS5535 and CS5536 companion chip for the geode. |
2020 | MFGPTs have a better resolution and max interval than the | 2020 | MFGPTs have a better resolution and max interval than the |
2021 | generic PIT, and are suitable for use as high-res timers. | 2021 | generic PIT, and are suitable for use as high-res timers. |
2022 | 2022 | ||
2023 | config OLPC | 2023 | config OLPC |
2024 | bool "One Laptop Per Child support" | 2024 | bool "One Laptop Per Child support" |
2025 | default n | 2025 | default n |
2026 | ---help--- | 2026 | ---help--- |
2027 | Add support for detecting the unique features of the OLPC | 2027 | Add support for detecting the unique features of the OLPC |
2028 | XO hardware. | 2028 | XO hardware. |
2029 | 2029 | ||
2030 | endif # X86_32 | 2030 | endif # X86_32 |
2031 | 2031 | ||
2032 | config K8_NB | 2032 | config K8_NB |
2033 | def_bool y | 2033 | def_bool y |
2034 | depends on AGP_AMD64 || (X86_64 && (GART_IOMMU || (PCI && NUMA))) | 2034 | depends on AGP_AMD64 || (X86_64 && (GART_IOMMU || (PCI && NUMA))) |
2035 | 2035 | ||
2036 | source "drivers/pcmcia/Kconfig" | 2036 | source "drivers/pcmcia/Kconfig" |
2037 | 2037 | ||
2038 | source "drivers/pci/hotplug/Kconfig" | 2038 | source "drivers/pci/hotplug/Kconfig" |
2039 | 2039 | ||
2040 | endmenu | 2040 | endmenu |
2041 | 2041 | ||
2042 | 2042 | ||
2043 | menu "Executable file formats / Emulations" | 2043 | menu "Executable file formats / Emulations" |
2044 | 2044 | ||
2045 | source "fs/Kconfig.binfmt" | 2045 | source "fs/Kconfig.binfmt" |
2046 | 2046 | ||
2047 | config IA32_EMULATION | 2047 | config IA32_EMULATION |
2048 | bool "IA32 Emulation" | 2048 | bool "IA32 Emulation" |
2049 | depends on X86_64 | 2049 | depends on X86_64 |
2050 | select COMPAT_BINFMT_ELF | 2050 | select COMPAT_BINFMT_ELF |
2051 | ---help--- | 2051 | ---help--- |
2052 | Include code to run 32-bit programs under a 64-bit kernel. You should | 2052 | Include code to run 32-bit programs under a 64-bit kernel. You should |
2053 | likely turn this on, unless you're 100% sure that you don't have any | 2053 | likely turn this on, unless you're 100% sure that you don't have any |
2054 | 32-bit programs left. | 2054 | 32-bit programs left. |
2055 | 2055 | ||
2056 | config IA32_AOUT | 2056 | config IA32_AOUT |
2057 | tristate "IA32 a.out support" | 2057 | tristate "IA32 a.out support" |
2058 | depends on IA32_EMULATION | 2058 | depends on IA32_EMULATION |
2059 | ---help--- | 2059 | ---help--- |
2060 | Support old a.out binaries in the 32bit emulation. | 2060 | Support old a.out binaries in the 32bit emulation. |
2061 | 2061 | ||
2062 | config COMPAT | 2062 | config COMPAT |
2063 | def_bool y | 2063 | def_bool y |
2064 | depends on IA32_EMULATION | 2064 | depends on IA32_EMULATION |
2065 | 2065 | ||
2066 | config COMPAT_FOR_U64_ALIGNMENT | 2066 | config COMPAT_FOR_U64_ALIGNMENT |
2067 | def_bool COMPAT | 2067 | def_bool COMPAT |
2068 | depends on X86_64 | 2068 | depends on X86_64 |
2069 | 2069 | ||
2070 | config SYSVIPC_COMPAT | 2070 | config SYSVIPC_COMPAT |
2071 | def_bool y | 2071 | def_bool y |
2072 | depends on COMPAT && SYSVIPC | 2072 | depends on COMPAT && SYSVIPC |
2073 | 2073 | ||
2074 | endmenu | 2074 | endmenu |
2075 | 2075 | ||
2076 | 2076 | ||
2077 | config HAVE_ATOMIC_IOMAP | 2077 | config HAVE_ATOMIC_IOMAP |
2078 | def_bool y | 2078 | def_bool y |
2079 | depends on X86_32 | 2079 | depends on X86_32 |
2080 | 2080 | ||
2081 | source "net/Kconfig" | 2081 | source "net/Kconfig" |
2082 | 2082 | ||
2083 | source "drivers/Kconfig" | 2083 | source "drivers/Kconfig" |
2084 | 2084 | ||
2085 | source "drivers/firmware/Kconfig" | 2085 | source "drivers/firmware/Kconfig" |
2086 | 2086 | ||
2087 | source "fs/Kconfig" | 2087 | source "fs/Kconfig" |
2088 | 2088 | ||
2089 | source "arch/x86/Kconfig.debug" | 2089 | source "arch/x86/Kconfig.debug" |
2090 | 2090 | ||
2091 | source "security/Kconfig" | 2091 | source "security/Kconfig" |
2092 | 2092 | ||
2093 | source "crypto/Kconfig" | 2093 | source "crypto/Kconfig" |
2094 | 2094 | ||
2095 | source "arch/x86/kvm/Kconfig" | 2095 | source "arch/x86/kvm/Kconfig" |
2096 | 2096 | ||
2097 | source "lib/Kconfig" | 2097 | source "lib/Kconfig" |
2098 | 2098 |