| 1 | menu "Kernel hacking" |
| 2 | |
| 3 | config TRACE_IRQFLAGS_SUPPORT |
| 4 | def_bool y |
| 5 | |
| 6 | source "lib/Kconfig.debug" |
| 7 | |
| 8 | config STRICT_DEVMEM |
| 9 | bool "Filter access to /dev/mem" |
| 10 | ---help--- |
| 11 | If this option is disabled, you allow userspace (root) access to all |
| 12 | of memory, including kernel and userspace memory. Accidental |
| 13 | access to this is obviously disastrous, but specific access can |
| 14 | be used by people debugging the kernel. Note that with PAT support |
| 15 | enabled, even in this case there are restrictions on /dev/mem |
| 16 | use due to the cache aliasing requirements. |
| 17 | |
| 18 | If this option is switched on, the /dev/mem file only allows |
| 19 | userspace access to PCI space and the BIOS code and data regions. |
| 20 | This is sufficient for dosemu and X and all common users of |
| 21 | /dev/mem. |
| 22 | |
| 23 | If in doubt, say Y. |
| 24 | |
| 25 | config X86_VERBOSE_BOOTUP |
| 26 | bool "Enable verbose x86 bootup info messages" |
| 27 | default y |
| 28 | ---help--- |
| 29 | Enables the informational output from the decompression stage |
| 30 | (e.g. bzImage) of the boot. If you disable this you will still |
| 31 | see errors. Disable this if you want silent bootup. |
| 32 | |
| 33 | config EARLY_PRINTK |
| 34 | bool "Early printk" if EXPERT |
| 35 | default y |
| 36 | ---help--- |
| 37 | Write kernel log output directly into the VGA buffer or to a serial |
| 38 | port. |
| 39 | |
| 40 | This is useful for kernel debugging when your machine crashes very |
| 41 | early before the console code is initialized. For normal operation |
| 42 | it is not recommended because it looks ugly and doesn't cooperate |
| 43 | with klogd/syslogd or the X server. You should normally N here, |
| 44 | unless you want to debug such a crash. |
| 45 | |
| 46 | config EARLY_PRINTK_INTEL_MID |
| 47 | bool "Early printk for Intel MID platform support" |
| 48 | depends on EARLY_PRINTK && X86_INTEL_MID |
| 49 | |
| 50 | config EARLY_PRINTK_DBGP |
| 51 | bool "Early printk via EHCI debug port" |
| 52 | depends on EARLY_PRINTK && PCI |
| 53 | ---help--- |
| 54 | Write kernel log output directly into the EHCI debug port. |
| 55 | |
| 56 | This is useful for kernel debugging when your machine crashes very |
| 57 | early before the console code is initialized. For normal operation |
| 58 | it is not recommended because it looks ugly and doesn't cooperate |
| 59 | with klogd/syslogd or the X server. You should normally N here, |
| 60 | unless you want to debug such a crash. You need usb debug device. |
| 61 | |
| 62 | config EARLY_PRINTK_EFI |
| 63 | bool "Early printk via the EFI framebuffer" |
| 64 | depends on EFI && EARLY_PRINTK |
| 65 | select FONT_SUPPORT |
| 66 | ---help--- |
| 67 | Write kernel log output directly into the EFI framebuffer. |
| 68 | |
| 69 | This is useful for kernel debugging when your machine crashes very |
| 70 | early before the console code is initialized. |
| 71 | |
| 72 | config X86_PTDUMP |
| 73 | bool "Export kernel pagetable layout to userspace via debugfs" |
| 74 | depends on DEBUG_KERNEL |
| 75 | select DEBUG_FS |
| 76 | ---help--- |
| 77 | Say Y here if you want to show the kernel pagetable layout in a |
| 78 | debugfs file. This information is only useful for kernel developers |
| 79 | who are working in architecture specific areas of the kernel. |
| 80 | It is probably not a good idea to enable this feature in a production |
| 81 | kernel. |
| 82 | If in doubt, say "N" |
| 83 | |
| 84 | config EFI_PGT_DUMP |
| 85 | bool "Dump the EFI pagetable" |
| 86 | depends on EFI && X86_PTDUMP |
| 87 | ---help--- |
| 88 | Enable this if you want to dump the EFI page table before |
| 89 | enabling virtual mode. This can be used to debug miscellaneous |
| 90 | issues with the mapping of the EFI runtime regions into that |
| 91 | table. |
| 92 | |
| 93 | config DEBUG_RODATA |
| 94 | bool "Write protect kernel read-only data structures" |
| 95 | default y |
| 96 | depends on DEBUG_KERNEL |
| 97 | ---help--- |
| 98 | Mark the kernel read-only data as write-protected in the pagetables, |
| 99 | in order to catch accidental (and incorrect) writes to such const |
| 100 | data. This is recommended so that we can catch kernel bugs sooner. |
| 101 | If in doubt, say "Y". |
| 102 | |
| 103 | config DEBUG_RODATA_TEST |
| 104 | bool "Testcase for the DEBUG_RODATA feature" |
| 105 | depends on DEBUG_RODATA |
| 106 | default y |
| 107 | ---help--- |
| 108 | This option enables a testcase for the DEBUG_RODATA |
| 109 | feature as well as for the change_page_attr() infrastructure. |
| 110 | If in doubt, say "N" |
| 111 | |
| 112 | config DEBUG_SET_MODULE_RONX |
| 113 | bool "Set loadable kernel module data as NX and text as RO" |
| 114 | depends on MODULES |
| 115 | ---help--- |
| 116 | This option helps catch unintended modifications to loadable |
| 117 | kernel module's text and read-only data. It also prevents execution |
| 118 | of module data. Such protection may interfere with run-time code |
| 119 | patching and dynamic kernel tracing - and they might also protect |
| 120 | against certain classes of kernel exploits. |
| 121 | If in doubt, say "N". |
| 122 | |
| 123 | config DEBUG_NX_TEST |
| 124 | tristate "Testcase for the NX non-executable stack feature" |
| 125 | depends on DEBUG_KERNEL && m |
| 126 | ---help--- |
| 127 | This option enables a testcase for the CPU NX capability |
| 128 | and the software setup of this feature. |
| 129 | If in doubt, say "N" |
| 130 | |
| 131 | config DOUBLEFAULT |
| 132 | default y |
| 133 | bool "Enable doublefault exception handler" if EXPERT |
| 134 | ---help--- |
| 135 | This option allows trapping of rare doublefault exceptions that |
| 136 | would otherwise cause a system to silently reboot. Disabling this |
| 137 | option saves about 4k and might cause you much additional grey |
| 138 | hair. |
| 139 | |
| 140 | config DEBUG_TLBFLUSH |
| 141 | bool "Set upper limit of TLB entries to flush one-by-one" |
| 142 | depends on DEBUG_KERNEL |
| 143 | ---help--- |
| 144 | |
| 145 | X86-only for now. |
| 146 | |
| 147 | This option allows the user to tune the amount of TLB entries the |
| 148 | kernel flushes one-by-one instead of doing a full TLB flush. In |
| 149 | certain situations, the former is cheaper. This is controlled by the |
| 150 | tlb_flushall_shift knob under /sys/kernel/debug/x86. If you set it |
| 151 | to -1, the code flushes the whole TLB unconditionally. Otherwise, |
| 152 | for positive values of it, the kernel will use single TLB entry |
| 153 | invalidating instructions according to the following formula: |
| 154 | |
| 155 | flush_entries <= active_tlb_entries / 2^tlb_flushall_shift |
| 156 | |
| 157 | If in doubt, say "N". |
| 158 | |
| 159 | config IOMMU_DEBUG |
| 160 | bool "Enable IOMMU debugging" |
| 161 | depends on GART_IOMMU && DEBUG_KERNEL |
| 162 | depends on X86_64 |
| 163 | ---help--- |
| 164 | Force the IOMMU to on even when you have less than 4GB of |
| 165 | memory and add debugging code. On overflow always panic. And |
| 166 | allow to enable IOMMU leak tracing. Can be disabled at boot |
| 167 | time with iommu=noforce. This will also enable scatter gather |
| 168 | list merging. Currently not recommended for production |
| 169 | code. When you use it make sure you have a big enough |
| 170 | IOMMU/AGP aperture. Most of the options enabled by this can |
| 171 | be set more finegrained using the iommu= command line |
| 172 | options. See Documentation/x86/x86_64/boot-options.txt for more |
| 173 | details. |
| 174 | |
| 175 | config IOMMU_STRESS |
| 176 | bool "Enable IOMMU stress-test mode" |
| 177 | ---help--- |
| 178 | This option disables various optimizations in IOMMU related |
| 179 | code to do real stress testing of the IOMMU code. This option |
| 180 | will cause a performance drop and should only be enabled for |
| 181 | testing. |
| 182 | |
| 183 | config IOMMU_LEAK |
| 184 | bool "IOMMU leak tracing" |
| 185 | depends on IOMMU_DEBUG && DMA_API_DEBUG |
| 186 | ---help--- |
| 187 | Add a simple leak tracer to the IOMMU code. This is useful when you |
| 188 | are debugging a buggy device driver that leaks IOMMU mappings. |
| 189 | |
| 190 | config HAVE_MMIOTRACE_SUPPORT |
| 191 | def_bool y |
| 192 | |
| 193 | config X86_DECODER_SELFTEST |
| 194 | bool "x86 instruction decoder selftest" |
| 195 | depends on DEBUG_KERNEL && KPROBES |
| 196 | depends on !COMPILE_TEST |
| 197 | ---help--- |
| 198 | Perform x86 instruction decoder selftests at build time. |
| 199 | This option is useful for checking the sanity of x86 instruction |
| 200 | decoder code. |
| 201 | If unsure, say "N". |
| 202 | |
| 203 | # |
| 204 | # IO delay types: |
| 205 | # |
| 206 | |
| 207 | config IO_DELAY_TYPE_0X80 |
| 208 | int |
| 209 | default "0" |
| 210 | |
| 211 | config IO_DELAY_TYPE_0XED |
| 212 | int |
| 213 | default "1" |
| 214 | |
| 215 | config IO_DELAY_TYPE_UDELAY |
| 216 | int |
| 217 | default "2" |
| 218 | |
| 219 | config IO_DELAY_TYPE_NONE |
| 220 | int |
| 221 | default "3" |
| 222 | |
| 223 | choice |
| 224 | prompt "IO delay type" |
| 225 | default IO_DELAY_0X80 |
| 226 | |
| 227 | config IO_DELAY_0X80 |
| 228 | bool "port 0x80 based port-IO delay [recommended]" |
| 229 | ---help--- |
| 230 | This is the traditional Linux IO delay used for in/out_p. |
| 231 | It is the most tested hence safest selection here. |
| 232 | |
| 233 | config IO_DELAY_0XED |
| 234 | bool "port 0xed based port-IO delay" |
| 235 | ---help--- |
| 236 | Use port 0xed as the IO delay. This frees up port 0x80 which is |
| 237 | often used as a hardware-debug port. |
| 238 | |
| 239 | config IO_DELAY_UDELAY |
| 240 | bool "udelay based port-IO delay" |
| 241 | ---help--- |
| 242 | Use udelay(2) as the IO delay method. This provides the delay |
| 243 | while not having any side-effect on the IO port space. |
| 244 | |
| 245 | config IO_DELAY_NONE |
| 246 | bool "no port-IO delay" |
| 247 | ---help--- |
| 248 | No port-IO delay. Will break on old boxes that require port-IO |
| 249 | delay for certain operations. Should work on most new machines. |
| 250 | |
| 251 | endchoice |
| 252 | |
| 253 | if IO_DELAY_0X80 |
| 254 | config DEFAULT_IO_DELAY_TYPE |
| 255 | int |
| 256 | default IO_DELAY_TYPE_0X80 |
| 257 | endif |
| 258 | |
| 259 | if IO_DELAY_0XED |
| 260 | config DEFAULT_IO_DELAY_TYPE |
| 261 | int |
| 262 | default IO_DELAY_TYPE_0XED |
| 263 | endif |
| 264 | |
| 265 | if IO_DELAY_UDELAY |
| 266 | config DEFAULT_IO_DELAY_TYPE |
| 267 | int |
| 268 | default IO_DELAY_TYPE_UDELAY |
| 269 | endif |
| 270 | |
| 271 | if IO_DELAY_NONE |
| 272 | config DEFAULT_IO_DELAY_TYPE |
| 273 | int |
| 274 | default IO_DELAY_TYPE_NONE |
| 275 | endif |
| 276 | |
| 277 | config DEBUG_BOOT_PARAMS |
| 278 | bool "Debug boot parameters" |
| 279 | depends on DEBUG_KERNEL |
| 280 | depends on DEBUG_FS |
| 281 | ---help--- |
| 282 | This option will cause struct boot_params to be exported via debugfs. |
| 283 | |
| 284 | config CPA_DEBUG |
| 285 | bool "CPA self-test code" |
| 286 | depends on DEBUG_KERNEL |
| 287 | ---help--- |
| 288 | Do change_page_attr() self-tests every 30 seconds. |
| 289 | |
| 290 | config OPTIMIZE_INLINING |
| 291 | bool "Allow gcc to uninline functions marked 'inline'" |
| 292 | ---help--- |
| 293 | This option determines if the kernel forces gcc to inline the functions |
| 294 | developers have marked 'inline'. Doing so takes away freedom from gcc to |
| 295 | do what it thinks is best, which is desirable for the gcc 3.x series of |
| 296 | compilers. The gcc 4.x series have a rewritten inlining algorithm and |
| 297 | enabling this option will generate a smaller kernel there. Hopefully |
| 298 | this algorithm is so good that allowing gcc 4.x and above to make the |
| 299 | decision will become the default in the future. Until then this option |
| 300 | is there to test gcc for this. |
| 301 | |
| 302 | If unsure, say N. |
| 303 | |
| 304 | config DEBUG_NMI_SELFTEST |
| 305 | bool "NMI Selftest" |
| 306 | depends on DEBUG_KERNEL && X86_LOCAL_APIC |
| 307 | ---help--- |
| 308 | Enabling this option turns on a quick NMI selftest to verify |
| 309 | that the NMI behaves correctly. |
| 310 | |
| 311 | This might help diagnose strange hangs that rely on NMI to |
| 312 | function properly. |
| 313 | |
| 314 | If unsure, say N. |
| 315 | |
| 316 | config X86_DEBUG_STATIC_CPU_HAS |
| 317 | bool "Debug alternatives" |
| 318 | depends on DEBUG_KERNEL |
| 319 | ---help--- |
| 320 | This option causes additional code to be generated which |
| 321 | fails if static_cpu_has() is used before alternatives have |
| 322 | run. |
| 323 | |
| 324 | If unsure, say N. |
| 325 | |
| 326 | endmenu |