3 config TRACE_IRQFLAGS_SUPPORT
6 source "lib/Kconfig.debug"
8 config X86_VERBOSE_BOOTUP
9 bool "Enable verbose x86 bootup info messages"
12 Enables the informational output from the decompression stage
13 (e.g. bzImage) of the boot. If you disable this you will still
14 see errors. Disable this if you want silent bootup.
17 bool "Early printk" if EXPERT
20 Write kernel log output directly into the VGA buffer or to a serial
23 This is useful for kernel debugging when your machine crashes very
24 early before the console code is initialized. For normal operation
25 it is not recommended because it looks ugly and doesn't cooperate
26 with klogd/syslogd or the X server. You should normally N here,
27 unless you want to debug such a crash.
29 config EARLY_PRINTK_DBGP
30 bool "Early printk via EHCI debug port"
31 depends on EARLY_PRINTK && PCI
33 Write kernel log output directly into the EHCI debug port.
35 This is useful for kernel debugging when your machine crashes very
36 early before the console code is initialized. For normal operation
37 it is not recommended because it looks ugly and doesn't cooperate
38 with klogd/syslogd or the X server. You should normally N here,
39 unless you want to debug such a crash. You need usb debug device.
41 config EARLY_PRINTK_EFI
42 bool "Early printk via the EFI framebuffer"
43 depends on EFI && EARLY_PRINTK
46 Write kernel log output directly into the EFI framebuffer.
48 This is useful for kernel debugging when your machine crashes very
49 early before the console code is initialized.
51 config X86_PTDUMP_CORE
55 tristate "Export kernel pagetable layout to userspace via debugfs"
56 depends on DEBUG_KERNEL
58 select X86_PTDUMP_CORE
60 Say Y here if you want to show the kernel pagetable layout in a
61 debugfs file. This information is only useful for kernel developers
62 who are working in architecture specific areas of the kernel.
63 It is probably not a good idea to enable this feature in a production
68 bool "Dump the EFI pagetable"
70 select X86_PTDUMP_CORE
72 Enable this if you want to dump the EFI page table before
73 enabling virtual mode. This can be used to debug miscellaneous
74 issues with the mapping of the EFI runtime regions into that
77 config DEBUG_RODATA_TEST
78 bool "Testcase for the marking rodata read-only"
81 This option enables a testcase for the setting rodata read-only
82 as well as for the change_page_attr() infrastructure.
86 bool "Warn on W+X mappings at boot"
87 select X86_PTDUMP_CORE
89 Generate a warning if any W+X mappings are found at boot.
91 This is useful for discovering cases where the kernel is leaving
92 W+X mappings after applying NX, as such mappings are a security risk.
94 Look for a message in dmesg output like this:
96 x86/mm: Checked W+X mappings: passed, no W+X pages found.
98 or like this, if the check failed:
100 x86/mm: Checked W+X mappings: FAILED, <N> W+X pages found.
102 Note that even if the check fails, your kernel is possibly
103 still fine, as W+X mappings are not a security hole in
104 themselves, what they do is that they make the exploitation
105 of other unfixed kernel bugs easier.
107 There is no runtime or memory usage effect of this option
108 once the kernel has booted up - it's a one time check.
110 If in doubt, say "Y".
114 bool "Enable doublefault exception handler" if EXPERT
116 This option allows trapping of rare doublefault exceptions that
117 would otherwise cause a system to silently reboot. Disabling this
118 option saves about 4k and might cause you much additional grey
121 config DEBUG_TLBFLUSH
122 bool "Set upper limit of TLB entries to flush one-by-one"
123 depends on DEBUG_KERNEL
128 This option allows the user to tune the amount of TLB entries the
129 kernel flushes one-by-one instead of doing a full TLB flush. In
130 certain situations, the former is cheaper. This is controlled by the
131 tlb_flushall_shift knob under /sys/kernel/debug/x86. If you set it
132 to -1, the code flushes the whole TLB unconditionally. Otherwise,
133 for positive values of it, the kernel will use single TLB entry
134 invalidating instructions according to the following formula:
136 flush_entries <= active_tlb_entries / 2^tlb_flushall_shift
138 If in doubt, say "N".
141 bool "Enable IOMMU debugging"
142 depends on GART_IOMMU && DEBUG_KERNEL
145 Force the IOMMU to on even when you have less than 4GB of
146 memory and add debugging code. On overflow always panic. And
147 allow to enable IOMMU leak tracing. Can be disabled at boot
148 time with iommu=noforce. This will also enable scatter gather
149 list merging. Currently not recommended for production
150 code. When you use it make sure you have a big enough
151 IOMMU/AGP aperture. Most of the options enabled by this can
152 be set more finegrained using the iommu= command line
153 options. See Documentation/x86/x86_64/boot-options.txt for more
157 bool "Enable IOMMU stress-test mode"
159 This option disables various optimizations in IOMMU related
160 code to do real stress testing of the IOMMU code. This option
161 will cause a performance drop and should only be enabled for
165 bool "IOMMU leak tracing"
166 depends on IOMMU_DEBUG && DMA_API_DEBUG
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.
171 config HAVE_MMIOTRACE_SUPPORT
174 config X86_DECODER_SELFTEST
175 bool "x86 instruction decoder selftest"
176 depends on DEBUG_KERNEL && KPROBES
177 depends on !COMPILE_TEST
179 Perform x86 instruction decoder selftests at build time.
180 This option is useful for checking the sanity of x86 instruction
188 config IO_DELAY_TYPE_0X80
192 config IO_DELAY_TYPE_0XED
196 config IO_DELAY_TYPE_UDELAY
200 config IO_DELAY_TYPE_NONE
205 prompt "IO delay type"
206 default IO_DELAY_0X80
209 bool "port 0x80 based port-IO delay [recommended]"
211 This is the traditional Linux IO delay used for in/out_p.
212 It is the most tested hence safest selection here.
215 bool "port 0xed based port-IO delay"
217 Use port 0xed as the IO delay. This frees up port 0x80 which is
218 often used as a hardware-debug port.
220 config IO_DELAY_UDELAY
221 bool "udelay based port-IO delay"
223 Use udelay(2) as the IO delay method. This provides the delay
224 while not having any side-effect on the IO port space.
227 bool "no port-IO delay"
229 No port-IO delay. Will break on old boxes that require port-IO
230 delay for certain operations. Should work on most new machines.
235 config DEFAULT_IO_DELAY_TYPE
237 default IO_DELAY_TYPE_0X80
241 config DEFAULT_IO_DELAY_TYPE
243 default IO_DELAY_TYPE_0XED
247 config DEFAULT_IO_DELAY_TYPE
249 default IO_DELAY_TYPE_UDELAY
253 config DEFAULT_IO_DELAY_TYPE
255 default IO_DELAY_TYPE_NONE
258 config DEBUG_BOOT_PARAMS
259 bool "Debug boot parameters"
260 depends on DEBUG_KERNEL
263 This option will cause struct boot_params to be exported via debugfs.
266 bool "CPA self-test code"
267 depends on DEBUG_KERNEL
269 Do change_page_attr() self-tests every 30 seconds.
271 config OPTIMIZE_INLINING
272 bool "Allow gcc to uninline functions marked 'inline'"
274 This option determines if the kernel forces gcc to inline the functions
275 developers have marked 'inline'. Doing so takes away freedom from gcc to
276 do what it thinks is best, which is desirable for the gcc 3.x series of
277 compilers. The gcc 4.x series have a rewritten inlining algorithm and
278 enabling this option will generate a smaller kernel there. Hopefully
279 this algorithm is so good that allowing gcc 4.x and above to make the
280 decision will become the default in the future. Until then this option
281 is there to test gcc for this.
286 bool "Debug low-level entry code"
287 depends on DEBUG_KERNEL
289 This option enables sanity checks in x86's low-level entry code.
290 Some of these sanity checks may slow down kernel entries and
291 exits or otherwise impact performance.
293 This is currently used to help test NMI code.
297 config DEBUG_NMI_SELFTEST
299 depends on DEBUG_KERNEL && X86_LOCAL_APIC
301 Enabling this option turns on a quick NMI selftest to verify
302 that the NMI behaves correctly.
304 This might help diagnose strange hangs that rely on NMI to
309 config DEBUG_IMR_SELFTEST
310 bool "Isolated Memory Region self test"
314 This option enables automated sanity testing of the IMR code.
315 Some simple tests are run to verify IMR bounds checking, alignment
316 and overlapping. This option is really only useful if you are
317 debugging an IMR memory map or are modifying the IMR code and want to
320 If unsure say N here.
323 bool "Debug the x86 FPU code"
324 depends on DEBUG_KERNEL
327 If this option is enabled then there will be extra sanity
328 checks and (boot time) debug printouts added to the kernel.
329 This debugging adds some small amount of runtime overhead
334 config PUNIT_ATOM_DEBUG
335 tristate "ATOM Punit debug driver"
339 This is a debug driver, which gets the power states
340 of all Punit North Complex devices. The power states of
341 each device is exposed as part of the debugfs interface.
342 The current power state can be read from
343 /sys/kernel/debug/punit_atom/dev_power_state