1 # SPDX-License-Identifier: GPL-2.0-only
3 bool "Extend memmap on extra space for more information on page"
5 Extend memmap on extra space for more information on page. This
6 could be used for debugging features that need to insert extra
7 field for every page. This extension enables us to save memory
8 by not allocating this extra memory according to boottime
11 config DEBUG_PAGEALLOC
12 bool "Debug page memory allocations"
13 depends on DEBUG_KERNEL
14 depends on !HIBERNATION || ARCH_SUPPORTS_DEBUG_PAGEALLOC && !PPC && !SPARC
15 select PAGE_POISONING if !ARCH_SUPPORTS_DEBUG_PAGEALLOC
17 Unmap pages from the kernel linear mapping after free_pages().
18 Depending on runtime enablement, this results in a small or large
19 slowdown, but helps to find certain types of memory corruption.
21 Also, the state of page tracking structures is checked more often as
22 pages are being allocated and freed, as unexpected state changes
23 often happen for same reasons as memory corruption (e.g. double free,
24 use-after-free). The error reports for these checks can be augmented
25 with stack traces of last allocation and freeing of the page, when
26 PAGE_OWNER is also selected and enabled on boot.
28 For architectures which don't enable ARCH_SUPPORTS_DEBUG_PAGEALLOC,
29 fill the pages with poison patterns after free_pages() and verify
30 the patterns before alloc_pages(). Additionally, this option cannot
31 be enabled in combination with hibernation as that would result in
32 incorrect warnings of memory corruption after a resume because free
33 pages are not saved to the suspend image.
35 By default this option will have a small overhead, e.g. by not
36 allowing the kernel mapping to be backed by large pages on some
37 architectures. Even bigger overhead comes when the debugging is
38 enabled by DEBUG_PAGEALLOC_ENABLE_DEFAULT or the debug_pagealloc
39 command line parameter.
41 config DEBUG_PAGEALLOC_ENABLE_DEFAULT
42 bool "Enable debug page memory allocations by default?"
43 depends on DEBUG_PAGEALLOC
45 Enable debug page memory allocations by default? This value
46 can be overridden by debug_pagealloc=off|on.
49 bool "Debug slab memory allocations"
50 depends on DEBUG_KERNEL && SLAB
52 Say Y here to have the kernel do limited verification on memory
53 allocation as well as poisoning memory on free to catch use of freed
54 memory. This can make kmalloc/kfree-intensive workloads much slower.
58 bool "Enable SLUB debugging support" if EXPERT
59 depends on SLUB && SYSFS && !SLUB_TINY
60 select STACKDEPOT if STACKTRACE_SUPPORT
62 SLUB has extensive debug support features. Disabling these can
63 result in significant savings in code size. While /sys/kernel/slab
64 will still exist (with SYSFS enabled), it will not provide e.g. cache
68 bool "SLUB debugging on by default"
69 depends on SLUB && SLUB_DEBUG
70 select STACKDEPOT_ALWAYS_INIT if STACKTRACE_SUPPORT
73 Boot with debugging on by default. SLUB boots by default with
74 the runtime debug capabilities switched off. Enabling this is
75 equivalent to specifying the "slub_debug" parameter on boot.
76 There is no support for more fine grained debug control like
77 possible with slub_debug=xxx. SLUB debugging may be switched
78 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
82 bool "Track page owner"
83 depends on DEBUG_KERNEL && STACKTRACE_SUPPORT
89 This keeps track of what call chain is the owner of a page, may
90 help to find bare alloc_page(s) leaks. Even if you include this
91 feature on your build, it is disabled in default. You should pass
92 "page_owner=on" to boot parameter in order to enable it. Eats
93 a fair amount of memory if enabled. See tools/mm/page_owner_sort.c
94 for user-space helper.
98 config PAGE_TABLE_CHECK
99 bool "Check for invalid mappings in user page tables"
100 depends on ARCH_SUPPORTS_PAGE_TABLE_CHECK
101 depends on EXCLUSIVE_SYSTEM_RAM
102 select PAGE_EXTENSION
104 Check that anonymous page is not being mapped twice with read write
105 permissions. Check that anonymous and file pages are not being
106 erroneously shared. Since the checking is performed at the time
107 entries are added and removed to user page tables, leaking, corruption
108 and double mapping problems are detected synchronously.
112 config PAGE_TABLE_CHECK_ENFORCED
113 bool "Enforce the page table checking by default"
114 depends on PAGE_TABLE_CHECK
116 Always enable page table checking. By default the page table checking
117 is disabled, and can be optionally enabled via page_table_check=on
118 kernel parameter. This config enforces that page table check is always
123 config PAGE_POISONING
124 bool "Poison pages after freeing"
126 Fill the pages with poison patterns after free_pages() and verify
127 the patterns before alloc_pages. The filling of the memory helps
128 reduce the risk of information leaks from freed data. This does
129 have a potential performance impact if enabled with the
130 "page_poison=1" kernel boot option.
132 Note that "poison" here is not the same thing as the "HWPoison"
133 for CONFIG_MEMORY_FAILURE. This is software poisoning only.
135 If you are only interested in sanitization of freed pages without
136 checking the poison pattern on alloc, you can boot the kernel with
137 "init_on_free=1" instead of enabling this.
141 config DEBUG_PAGE_REF
142 bool "Enable tracepoint to track down page reference manipulation"
143 depends on DEBUG_KERNEL
144 depends on TRACEPOINTS
146 This is a feature to add tracepoint for tracking down page reference
147 manipulation. This tracking is useful to diagnose functional failure
148 due to migration failures caused by page reference mismatches. Be
149 careful when enabling this feature because it adds about 30 KB to the
150 kernel code. However the runtime performance overhead is virtually
151 nil until the tracepoints are actually enabled.
153 config DEBUG_RODATA_TEST
154 bool "Testcase for the marking rodata read-only"
155 depends on STRICT_KERNEL_RWX
157 This option enables a testcase for the setting rodata read-only.
159 config ARCH_HAS_DEBUG_WX
163 bool "Warn on W+X mappings at boot"
164 depends on ARCH_HAS_DEBUG_WX
168 Generate a warning if any W+X mappings are found at boot.
170 This is useful for discovering cases where the kernel is leaving W+X
171 mappings after applying NX, as such mappings are a security risk.
173 Look for a message in dmesg output like this:
175 <arch>/mm: Checked W+X mappings: passed, no W+X pages found.
177 or like this, if the check failed:
179 <arch>/mm: Checked W+X mappings: failed, <N> W+X pages found.
181 Note that even if the check fails, your kernel is possibly
182 still fine, as W+X mappings are not a security hole in
183 themselves, what they do is that they make the exploitation
184 of other unfixed kernel bugs easier.
186 There is no runtime or memory usage effect of this option
187 once the kernel has booted up - it's a one time check.
189 If in doubt, say "Y".
191 config GENERIC_PTDUMP
197 config PTDUMP_DEBUGFS
198 bool "Export kernel pagetable layout to userspace via debugfs"
199 depends on DEBUG_KERNEL
201 depends on GENERIC_PTDUMP
204 Say Y here if you want to show the kernel pagetable layout in a
205 debugfs file. This information is only useful for kernel developers
206 who are working in architecture specific areas of the kernel.
207 It is probably not a good idea to enable this feature in a production
212 config HAVE_DEBUG_KMEMLEAK
215 config DEBUG_KMEMLEAK
216 bool "Kernel memory leak detector"
217 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK
219 select STACKTRACE if STACKTRACE_SUPPORT
223 select STACKDEPOT_ALWAYS_INIT if !DEBUG_KMEMLEAK_DEFAULT_OFF
225 Say Y here if you want to enable the memory leak
226 detector. The memory allocation/freeing is traced in a way
227 similar to the Boehm's conservative garbage collector, the
228 difference being that the orphan objects are not freed but
229 only shown in /sys/kernel/debug/kmemleak. Enabling this
230 feature will introduce an overhead to memory
231 allocations. See Documentation/dev-tools/kmemleak.rst for more
234 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
235 of finding leaks due to the slab objects poisoning.
237 In order to access the kmemleak file, debugfs needs to be
238 mounted (usually at /sys/kernel/debug).
240 config DEBUG_KMEMLEAK_MEM_POOL_SIZE
241 int "Kmemleak memory pool size"
242 depends on DEBUG_KMEMLEAK
246 Kmemleak must track all the memory allocations to avoid
247 reporting false positives. Since memory may be allocated or
248 freed before kmemleak is fully initialised, use a static pool
249 of metadata objects to track such callbacks. After kmemleak is
250 fully initialised, this memory pool acts as an emergency one
251 if slab allocations fail.
253 config DEBUG_KMEMLEAK_DEFAULT_OFF
254 bool "Default kmemleak to off"
255 depends on DEBUG_KMEMLEAK
257 Say Y here to disable kmemleak by default. It can then be enabled
258 on the command line via kmemleak=on.
260 config DEBUG_KMEMLEAK_AUTO_SCAN
261 bool "Enable kmemleak auto scan thread on boot up"
263 depends on DEBUG_KMEMLEAK
265 Depending on the cpu, kmemleak scan may be cpu intensive and can
266 stall user tasks at times. This option enables/disables automatic
267 kmemleak scan at boot up.
269 Say N here to disable kmemleak auto scan thread to stop automatic
270 scanning. Disabling this option disables automatic reporting of
275 config PER_VMA_LOCK_STATS
276 bool "Statistics for per-vma locks"
277 depends on PER_VMA_LOCK
279 Say Y here to enable success, retry and failure counters of page
280 faults handled under protection of per-vma locks. When enabled, the
281 counters are exposed in /proc/vmstat. This information is useful for
282 kernel developers to evaluate effectiveness of per-vma locks and to
283 identify pathological cases. Counting these events introduces a small
284 overhead in the page fault path.