Thomas Gleixner | ec8f24b | 2019-05-19 13:07:45 +0100 | [diff] [blame] | 1 | # SPDX-License-Identifier: GPL-2.0-only |
Joonsoo Kim | eefa864b | 2014-12-12 16:55:46 -0800 | [diff] [blame] | 2 | config PAGE_EXTENSION |
| 3 | bool "Extend memmap on extra space for more information on page" |
Masahiro Yamada | a7f7f62 | 2020-06-14 01:50:22 +0900 | [diff] [blame] | 4 | help |
Joonsoo Kim | eefa864b | 2014-12-12 16:55:46 -0800 | [diff] [blame] | 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 |
| 9 | configuration. |
| 10 | |
Akinobu Mita | ee3b429 | 2009-04-02 16:56:30 -0700 | [diff] [blame] | 11 | config DEBUG_PAGEALLOC |
| 12 | bool "Debug page memory allocations" |
Akinobu Mita | 7bc32f6 | 2011-03-22 16:32:46 -0700 | [diff] [blame] | 13 | depends on DEBUG_KERNEL |
| 14 | depends on !HIBERNATION || ARCH_SUPPORTS_DEBUG_PAGEALLOC && !PPC && !SPARC |
Akinobu Mita | 7bc32f6 | 2011-03-22 16:32:46 -0700 | [diff] [blame] | 15 | select PAGE_POISONING if !ARCH_SUPPORTS_DEBUG_PAGEALLOC |
Masahiro Yamada | a7f7f62 | 2020-06-14 01:50:22 +0900 | [diff] [blame] | 16 | help |
Akinobu Mita | ee3b429 | 2009-04-02 16:56:30 -0700 | [diff] [blame] | 17 | Unmap pages from the kernel linear mapping after free_pages(). |
Christian Borntraeger | ea6eabb | 2016-03-15 14:55:30 -0700 | [diff] [blame] | 18 | Depending on runtime enablement, this results in a small or large |
| 19 | slowdown, but helps to find certain types of memory corruption. |
Akinobu Mita | ee3b429 | 2009-04-02 16:56:30 -0700 | [diff] [blame] | 20 | |
Vlastimil Babka | 4462b32 | 2019-07-11 20:55:09 -0700 | [diff] [blame] | 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, |
Vlastimil Babka | 8974558 | 2019-09-23 15:34:42 -0700 | [diff] [blame] | 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. |
Vlastimil Babka | 4462b32 | 2019-07-11 20:55:09 -0700 | [diff] [blame] | 27 | |
Akinobu Mita | 7bc32f6 | 2011-03-22 16:32:46 -0700 | [diff] [blame] | 28 | For architectures which don't enable ARCH_SUPPORTS_DEBUG_PAGEALLOC, |
| 29 | fill the pages with poison patterns after free_pages() and verify |
Vlastimil Babka | 4462b32 | 2019-07-11 20:55:09 -0700 | [diff] [blame] | 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. |
Akinobu Mita | 7bc32f6 | 2011-03-22 16:32:46 -0700 | [diff] [blame] | 34 | |
Christian Borntraeger | ea6eabb | 2016-03-15 14:55:30 -0700 | [diff] [blame] | 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. |
| 40 | |
| 41 | config DEBUG_PAGEALLOC_ENABLE_DEFAULT |
| 42 | bool "Enable debug page memory allocations by default?" |
Christian Borntraeger | ea6eabb | 2016-03-15 14:55:30 -0700 | [diff] [blame] | 43 | depends on DEBUG_PAGEALLOC |
Masahiro Yamada | a7f7f62 | 2020-06-14 01:50:22 +0900 | [diff] [blame] | 44 | help |
Christian Borntraeger | ea6eabb | 2016-03-15 14:55:30 -0700 | [diff] [blame] | 45 | Enable debug page memory allocations by default? This value |
| 46 | can be overridden by debug_pagealloc=off|on. |
| 47 | |
Changbin Du | 8aa4976 | 2019-03-05 15:46:19 -0800 | [diff] [blame] | 48 | config PAGE_OWNER |
| 49 | bool "Track page owner" |
| 50 | depends on DEBUG_KERNEL && STACKTRACE_SUPPORT |
| 51 | select DEBUG_FS |
| 52 | select STACKTRACE |
| 53 | select STACKDEPOT |
| 54 | select PAGE_EXTENSION |
| 55 | help |
| 56 | This keeps track of what call chain is the owner of a page, may |
| 57 | help to find bare alloc_page(s) leaks. Even if you include this |
| 58 | feature on your build, it is disabled in default. You should pass |
| 59 | "page_owner=on" to boot parameter in order to enable it. Eats |
| 60 | a fair amount of memory if enabled. See tools/vm/page_owner_sort.c |
| 61 | for user-space helper. |
| 62 | |
| 63 | If unsure, say N. |
| 64 | |
Akinobu Mita | 6a11f75 | 2009-03-31 15:23:17 -0700 | [diff] [blame] | 65 | config PAGE_POISONING |
Laura Abbott | 8823b1d | 2016-03-15 14:56:27 -0700 | [diff] [blame] | 66 | bool "Poison pages after freeing" |
Masahiro Yamada | a7f7f62 | 2020-06-14 01:50:22 +0900 | [diff] [blame] | 67 | help |
Laura Abbott | 8823b1d | 2016-03-15 14:56:27 -0700 | [diff] [blame] | 68 | Fill the pages with poison patterns after free_pages() and verify |
| 69 | the patterns before alloc_pages. The filling of the memory helps |
| 70 | reduce the risk of information leaks from freed data. This does |
Kees Cook | 8c9a134 | 2018-08-21 21:53:10 -0700 | [diff] [blame] | 71 | have a potential performance impact if enabled with the |
| 72 | "page_poison=1" kernel boot option. |
Laura Abbott | 8823b1d | 2016-03-15 14:56:27 -0700 | [diff] [blame] | 73 | |
| 74 | Note that "poison" here is not the same thing as the "HWPoison" |
| 75 | for CONFIG_MEMORY_FAILURE. This is software poisoning only. |
| 76 | |
Vlastimil Babka | 8f42475 | 2020-12-14 19:13:41 -0800 | [diff] [blame] | 77 | If you are only interested in sanitization of freed pages without |
| 78 | checking the poison pattern on alloc, you can boot the kernel with |
| 79 | "init_on_free=1" instead of enabling this. |
| 80 | |
Laura Abbott | 8823b1d | 2016-03-15 14:56:27 -0700 | [diff] [blame] | 81 | If unsure, say N |
| 82 | |
Joonsoo Kim | 95813b8 | 2016-03-17 14:19:29 -0700 | [diff] [blame] | 83 | config DEBUG_PAGE_REF |
| 84 | bool "Enable tracepoint to track down page reference manipulation" |
| 85 | depends on DEBUG_KERNEL |
| 86 | depends on TRACEPOINTS |
Masahiro Yamada | a7f7f62 | 2020-06-14 01:50:22 +0900 | [diff] [blame] | 87 | help |
Joonsoo Kim | 95813b8 | 2016-03-17 14:19:29 -0700 | [diff] [blame] | 88 | This is a feature to add tracepoint for tracking down page reference |
| 89 | manipulation. This tracking is useful to diagnose functional failure |
| 90 | due to migration failures caused by page reference mismatches. Be |
| 91 | careful when enabling this feature because it adds about 30 KB to the |
| 92 | kernel code. However the runtime performance overhead is virtually |
| 93 | nil until the tracepoints are actually enabled. |
Jinbum Park | 2959a5f | 2017-02-27 14:30:22 -0800 | [diff] [blame] | 94 | |
| 95 | config DEBUG_RODATA_TEST |
| 96 | bool "Testcase for the marking rodata read-only" |
| 97 | depends on STRICT_KERNEL_RWX |
Masahiro Yamada | a7f7f62 | 2020-06-14 01:50:22 +0900 | [diff] [blame] | 98 | help |
Jinbum Park | 2959a5f | 2017-02-27 14:30:22 -0800 | [diff] [blame] | 99 | This option enables a testcase for the setting rodata read-only. |
Steven Price | 30d621f | 2020-02-03 17:36:20 -0800 | [diff] [blame] | 100 | |
Zong Li | 375d315 | 2020-06-03 16:03:52 -0700 | [diff] [blame] | 101 | config ARCH_HAS_DEBUG_WX |
| 102 | bool |
| 103 | |
| 104 | config DEBUG_WX |
| 105 | bool "Warn on W+X mappings at boot" |
| 106 | depends on ARCH_HAS_DEBUG_WX |
| 107 | depends on MMU |
| 108 | select PTDUMP_CORE |
| 109 | help |
| 110 | Generate a warning if any W+X mappings are found at boot. |
| 111 | |
| 112 | This is useful for discovering cases where the kernel is leaving W+X |
| 113 | mappings after applying NX, as such mappings are a security risk. |
| 114 | |
| 115 | Look for a message in dmesg output like this: |
| 116 | |
| 117 | <arch>/mm: Checked W+X mappings: passed, no W+X pages found. |
| 118 | |
| 119 | or like this, if the check failed: |
| 120 | |
| 121 | <arch>/mm: Checked W+X mappings: failed, <N> W+X pages found. |
| 122 | |
| 123 | Note that even if the check fails, your kernel is possibly |
| 124 | still fine, as W+X mappings are not a security hole in |
| 125 | themselves, what they do is that they make the exploitation |
| 126 | of other unfixed kernel bugs easier. |
| 127 | |
| 128 | There is no runtime or memory usage effect of this option |
| 129 | once the kernel has booted up - it's a one time check. |
| 130 | |
| 131 | If in doubt, say "Y". |
| 132 | |
Steven Price | 30d621f | 2020-02-03 17:36:20 -0800 | [diff] [blame] | 133 | config GENERIC_PTDUMP |
| 134 | bool |
| 135 | |
| 136 | config PTDUMP_CORE |
| 137 | bool |
| 138 | |
| 139 | config PTDUMP_DEBUGFS |
| 140 | bool "Export kernel pagetable layout to userspace via debugfs" |
| 141 | depends on DEBUG_KERNEL |
| 142 | depends on DEBUG_FS |
| 143 | depends on GENERIC_PTDUMP |
| 144 | select PTDUMP_CORE |
| 145 | help |
| 146 | Say Y here if you want to show the kernel pagetable layout in a |
| 147 | debugfs file. This information is only useful for kernel developers |
| 148 | who are working in architecture specific areas of the kernel. |
| 149 | It is probably not a good idea to enable this feature in a production |
| 150 | kernel. |
| 151 | |
| 152 | If in doubt, say N. |