panic: print out stacktrace if DEBUG_BUGVERBOSE
[linux-block.git] / arch / x86 / Kconfig.debug
... / ...
CommitLineData
1menu "Kernel hacking"
2
3config TRACE_IRQFLAGS_SUPPORT
4 def_bool y
5
6source "lib/Kconfig.debug"
7
8config NONPROMISC_DEVMEM
9 bool "Disable promiscuous /dev/mem"
10 help
11 The /dev/mem file by default only allows userspace access to PCI
12 space and the BIOS code and data regions. This is sufficient for
13 dosemu and X and all common users of /dev/mem. With this config
14 option, you allow userspace access to all of memory, including
15 kernel and userspace memory. Accidental access to this is
16 obviously disasterous, but specific access can be used by people
17 debugging the kernel.
18
19config EARLY_PRINTK
20 bool "Early printk" if EMBEDDED
21 default y
22 help
23 Write kernel log output directly into the VGA buffer or to a serial
24 port.
25
26 This is useful for kernel debugging when your machine crashes very
27 early before the console code is initialized. For normal operation
28 it is not recommended because it looks ugly and doesn't cooperate
29 with klogd/syslogd or the X server. You should normally N here,
30 unless you want to debug such a crash.
31
32config DEBUG_STACKOVERFLOW
33 bool "Check for stack overflows"
34 depends on DEBUG_KERNEL
35 help
36 This option will cause messages to be printed if free stack space
37 drops below a certain limit.
38
39config DEBUG_STACK_USAGE
40 bool "Stack utilization instrumentation"
41 depends on DEBUG_KERNEL
42 help
43 Enables the display of the minimum amount of free stack which each
44 task has ever had available in the sysrq-T and sysrq-P debug output.
45
46 This option will slow down process creation somewhat.
47
48config DEBUG_PAGEALLOC
49 bool "Debug page memory allocations"
50 depends on DEBUG_KERNEL
51 help
52 Unmap pages from the kernel linear mapping after free_pages().
53 This results in a large slowdown, but helps to find certain types
54 of memory corruptions.
55
56config DEBUG_PER_CPU_MAPS
57 bool "Debug access to per_cpu maps"
58 depends on DEBUG_KERNEL
59 depends on X86_64_SMP
60 default n
61 help
62 Say Y to verify that the per_cpu map being accessed has
63 been setup. Adds a fair amount of code to kernel memory
64 and decreases performance.
65
66 Say N if unsure.
67
68config X86_PTDUMP
69 bool "Export kernel pagetable layout to userspace via debugfs"
70 depends on DEBUG_KERNEL
71 select DEBUG_FS
72 help
73 Say Y here if you want to show the kernel pagetable layout in a
74 debugfs file. This information is only useful for kernel developers
75 who are working in architecture specific areas of the kernel.
76 It is probably not a good idea to enable this feature in a production
77 kernel.
78 If in doubt, say "N"
79
80config DEBUG_RODATA
81 bool "Write protect kernel read-only data structures"
82 default y
83 depends on DEBUG_KERNEL
84 help
85 Mark the kernel read-only data as write-protected in the pagetables,
86 in order to catch accidental (and incorrect) writes to such const
87 data. This is recommended so that we can catch kernel bugs sooner.
88 If in doubt, say "Y".
89
90config DIRECT_GBPAGES
91 bool "Enable gbpages-mapped kernel pagetables"
92 depends on DEBUG_KERNEL && EXPERIMENTAL && X86_64
93 help
94 Enable gigabyte pages support (if the CPU supports it). This can
95 improve the kernel's performance a tiny bit by reducing TLB
96 pressure.
97
98 This is experimental code.
99
100 If in doubt, say "N".
101
102config DEBUG_RODATA_TEST
103 bool "Testcase for the DEBUG_RODATA feature"
104 depends on DEBUG_RODATA
105 help
106 This option enables a testcase for the DEBUG_RODATA
107 feature as well as for the change_page_attr() infrastructure.
108 If in doubt, say "N"
109
110config DEBUG_NX_TEST
111 tristate "Testcase for the NX non-executable stack feature"
112 depends on DEBUG_KERNEL && m
113 help
114 This option enables a testcase for the CPU NX capability
115 and the software setup of this feature.
116 If in doubt, say "N"
117
118config 4KSTACKS
119 bool "Use 4Kb for kernel stacks instead of 8Kb"
120 depends on X86_32
121 help
122 If you say Y here the kernel will use a 4Kb stacksize for the
123 kernel stack attached to each process/thread. This facilitates
124 running more threads on a system and also reduces the pressure
125 on the VM subsystem for higher order allocations. This option
126 will also use IRQ stacks to compensate for the reduced stackspace.
127
128config X86_FIND_SMP_CONFIG
129 def_bool y
130 depends on X86_LOCAL_APIC || X86_VOYAGER
131 depends on X86_32
132
133config X86_MPPARSE
134 def_bool y
135 depends on (X86_32 && (X86_LOCAL_APIC && !X86_VISWS)) || X86_64
136
137config DOUBLEFAULT
138 default y
139 bool "Enable doublefault exception handler" if EMBEDDED
140 depends on X86_32
141 help
142 This option allows trapping of rare doublefault exceptions that
143 would otherwise cause a system to silently reboot. Disabling this
144 option saves about 4k and might cause you much additional grey
145 hair.
146
147config IOMMU_DEBUG
148 bool "Enable IOMMU debugging"
149 depends on GART_IOMMU && DEBUG_KERNEL
150 depends on X86_64
151 help
152 Force the IOMMU to on even when you have less than 4GB of
153 memory and add debugging code. On overflow always panic. And
154 allow to enable IOMMU leak tracing. Can be disabled at boot
155 time with iommu=noforce. This will also enable scatter gather
156 list merging. Currently not recommended for production
157 code. When you use it make sure you have a big enough
158 IOMMU/AGP aperture. Most of the options enabled by this can
159 be set more finegrained using the iommu= command line
160 options. See Documentation/x86_64/boot-options.txt for more
161 details.
162
163config IOMMU_LEAK
164 bool "IOMMU leak tracing"
165 depends on DEBUG_KERNEL
166 depends on IOMMU_DEBUG
167 help
168 Add a simple leak tracer to the IOMMU code. This is useful when you
169 are debugging a buggy device driver that leaks IOMMU mappings.
170
171#
172# IO delay types:
173#
174
175config IO_DELAY_TYPE_0X80
176 int
177 default "0"
178
179config IO_DELAY_TYPE_0XED
180 int
181 default "1"
182
183config IO_DELAY_TYPE_UDELAY
184 int
185 default "2"
186
187config IO_DELAY_TYPE_NONE
188 int
189 default "3"
190
191choice
192 prompt "IO delay type"
193 default IO_DELAY_0X80
194
195config IO_DELAY_0X80
196 bool "port 0x80 based port-IO delay [recommended]"
197 help
198 This is the traditional Linux IO delay used for in/out_p.
199 It is the most tested hence safest selection here.
200
201config IO_DELAY_0XED
202 bool "port 0xed based port-IO delay"
203 help
204 Use port 0xed as the IO delay. This frees up port 0x80 which is
205 often used as a hardware-debug port.
206
207config IO_DELAY_UDELAY
208 bool "udelay based port-IO delay"
209 help
210 Use udelay(2) as the IO delay method. This provides the delay
211 while not having any side-effect on the IO port space.
212
213config IO_DELAY_NONE
214 bool "no port-IO delay"
215 help
216 No port-IO delay. Will break on old boxes that require port-IO
217 delay for certain operations. Should work on most new machines.
218
219endchoice
220
221if IO_DELAY_0X80
222config DEFAULT_IO_DELAY_TYPE
223 int
224 default IO_DELAY_TYPE_0X80
225endif
226
227if IO_DELAY_0XED
228config DEFAULT_IO_DELAY_TYPE
229 int
230 default IO_DELAY_TYPE_0XED
231endif
232
233if IO_DELAY_UDELAY
234config DEFAULT_IO_DELAY_TYPE
235 int
236 default IO_DELAY_TYPE_UDELAY
237endif
238
239if IO_DELAY_NONE
240config DEFAULT_IO_DELAY_TYPE
241 int
242 default IO_DELAY_TYPE_NONE
243endif
244
245config DEBUG_BOOT_PARAMS
246 bool "Debug boot parameters"
247 depends on DEBUG_KERNEL
248 depends on DEBUG_FS
249 help
250 This option will cause struct boot_params to be exported via debugfs.
251
252config CPA_DEBUG
253 bool "CPA self-test code"
254 depends on DEBUG_KERNEL
255 help
256 Do change_page_attr() self-tests every 30 seconds.
257
258config OPTIMIZE_INLINING
259 bool "Allow gcc to uninline functions marked 'inline'"
260 depends on BROKEN
261 help
262 This option determines if the kernel forces gcc to inline the functions
263 developers have marked 'inline'. Doing so takes away freedom from gcc to
264 do what it thinks is best, which is desirable for the gcc 3.x series of
265 compilers. The gcc 4.x series have a rewritten inlining algorithm and
266 disabling this option will generate a smaller kernel there. Hopefully
267 this algorithm is so good that allowing gcc4 to make the decision can
268 become the default in the future, until then this option is there to
269 test gcc for this.
270
271endmenu
272