forked from linux-test-project/ltp
-
Notifications
You must be signed in to change notification settings - Fork 0
/
README.kernel_config
301 lines (273 loc) · 10.7 KB
/
README.kernel_config
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
---------------------------------
Enable CODE COVERAGE for your Kernel:
---------------------------------
1) Apply kernel-gcov patch(s) against the corresponding Kernel. They are available here:
http://ltp.cvs.sourceforge.net/viewvc/ltp/utils/analysis/gcov-kernel/linux-2.*.*-gcov.patch?view=log,
2) Also enable the following options in your .config file before building the kernel
CONFIG_GCOV_PROFILE=y
CONFIG_GCOV_ALL=y
CONFIG_GCOV_PROC=m
CONFIG_GCOV_HAMMER=y
---------------------------------
Enabling Kernel Configuration to test Containers/Namespaces
---------------------------------
CONFIG_NAMESPACES=y
CONFIG_UTS_NS=y
CONFIG_IPC_NS=y
CONFIG_USER_NS=y
CONFIG_PID_NS=y
CONFIG_NET_NS=y
CONFIG_VETH=y
CONFIG_MACVLAN=y
The IPC namespaces do not automatically enable IPC, so you may
also want to have:
CONFIG_SYSVIPC=y
CONFIG_SYSVIPC_SYSCTL=y
CONFIG_POSIX_MQUEUE=y
---------------------------------
Enabling Kernel Configuration to test Controllers
---------------------------------
CONFIG_CGROUPS=y
CONFIG_CGROUP_DEBUG=y
CONFIG_CGROUP_NS=y
CONFIG_GROUP_SCHED=y
CONFIG_FAIR_GROUP_SCHED=y
CONFIG_RT_GROUP_SCHED=y
CONFIG_CGROUP_SCHED=y
CONFIG_CGROUP_MEM_RES_CTLR=y
CONFIG_LOCKDEP=y
---------------------------------
Enabling Kernel Configuration to test Power Management features
---------------------------------
CONFIG_CPU_FREQ=y
CONFIG_CPU_FREQ_TABLE=y
CONFIG_CPU_FREQ_DEBUG=y
CONFIG_CPU_FREQ_STAT=y
CONFIG_CPU_FREQ_STAT_DETAILS=y
CONFIG_CPU_IDLE=y
CONFIG_CPU_IDLE_GOV_LADDER=y
CONFIG_CPU_IDLE_GOV_MENU=y
CONFIG_SCHED_MC=y
CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE=y
CONFIG_CPU_FREQ_DEFAULT_GOV_POWERSAVE=y
CONFIG_CPU_FREQ_DEFAULT_GOV_USERSPACE=y
CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_DEFAULT_GOV_CONSERVATIVE=y
CONFIG_CPU_FREQ_GOV_PERFORMANCE=y
CONFIG_CPU_FREQ_GOV_POWERSAVE=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_CONSERVATIVE=y
---------------------------------
Enabling Kernel Configuration to test filecaps security feature
---------------------------------
CONFIG_SECURITY_FILE_CAPABILITIES=y
---------------------------------
Enabling Kernel Configuration to test SELinux security feature
---------------------------------
Your Kernel should have been built with the following options to
test SELinux:
CONFIG_SECURITY=y
CONFIG_SECURITY_NETWORK=y
CONFIG_SECURITY_NETWORK_XFRM=y
CONFIG_SECURITY_DEFAULT_MMAP_MIN_ADDR=0
This has to be set to a positive value if you want to test this check.
Fedora kernels set it to 65536.
CONFIG_SECURITY_SELINUX=y
CONFIG_SECURITY_SELINUX_BOOTPARAM=y
CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=1
CONFIG_SECURITY_SELINUX_DEVELOP=y
CONFIG_SECURITY_SELINUX_AVC_STATS=y
CONFIG_SECURITY_SELINUX_CHECKREQPROT_VALUE=1
CONFIG_SECURITY_SELINUX_ENABLE_SECMARK_DEFAULT=y
CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX=y
You don't want this one unless you are running Fedora 3 or 4.
On anything newer, it will cause unnecessary policy expansion.
CONFIG_SECURITY_SMACK=y
CONFIG_SECURITY_SELINUX=y
By default, if you boot with multiple LSMs compiled into the kernel, the
kernel won't boot succesfully - there can be only one (aside from
explicit internal "stacking" e.g. as is done for combining SELinux or
Smack with capabilities). Unless you use the security= option to select
one at boot. SELinux and Smack will honor the security= option.
---------------------------------
---------------------------------
Enabling Kernel Configuration to test SMACK security feature
---------------------------------
CONFIG_NETLABEL=y
CONFIG_SECURITY=y
CONFIG_SECURITY_NETWORK=y
CONFIG_SECURITY_SMACK=y
CONFIG_SECURITY_SELINUX should not be set
For more information to build/install/run these tests, look through:
ltp/testcases/kernel/security/smack/README
---------------------------------
---------------------------------
Enablement for Enhancement to kexec/kdump for implementing the following features:
- Backup/restore memory used by the original kernel before/after kexec.
- Save/restore CPU state before/after kexec.
Now, only the i386 architecture is supported. More from the following git logs:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=3ab83521378268044a448113c6aa9a9e245f4d2f,
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=89081d17f7bb81d89fa1aa9b70f821c5cf4d39e9,
---------------------------------
CONFIG_X86_32=y
CONFIG_RELOCATABLE=y
CONFIG_KEXEC=y
CONFIG_CRASH_DUMP=y
CONFIG_PM=y
CONFIG_HIBERNATION=y
CONFIG_KEXEC_JUMP=y
---------------------------------
Enabling HOTPLUG for your Kernels
---------------------------------
CONFIG_HOTPLUG_CPU=y
CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y
CONFIG_ACPI_HOTPLUG_CPU=y
CONFIG_HOTPLUG_PCI_PCIE=y
CONFIG_HOTPLUG_PCI=y
CONFIG_HOTPLUG_PCI_FAKE=y
CONFIG_HOTPLUG_PCI_COMPAQ=y
CONFIG_HOTPLUG_PCI_COMPAQ_NVRAM=y
CONFIG_HOTPLUG_PCI_IBM=y
CONFIG_HOTPLUG_PCI_ACPI=y
CONFIG_HOTPLUG_PCI_ACPI_IBM=y
CONFIG_HOTPLUG_PCI_CPCI=y
CONFIG_HOTPLUG_PCI_CPCI_ZT5550=y
CONFIG_HOTPLUG_PCI_CPCI_GENERIC=y
CONFIG_HOTPLUG_PCI_SHPC=y
---------------------------------
Video For Linux Two API testing Requirements:
You need to have a video device (i.e. webcam, tuner card, etc.) connected to your system and available under /dev/video0. If you don't have any hardware device available, you can still test the "Virtual Video Driver". To compile this you need to compile your kernel with CONFIG_VIDEO_VIVI=m under:
-> Device Drivers
-> Multimedia devices
-> Video For Linux
-> Video capture adapters
-> Virtual Video Driver
The tests also require CUnit Framework to be installed before compiling the tests. Download & Install the same from:
http://sourceforge.net/projects/cunit/
---------------------------------
---------------------------------
Native language support (nls) testsuite requirements
----------------------------------------------------
CONFIG_NLS=m
CONFIG_BLOCK=y
---------------------------------
Enabling Controller area network (CAN) protocol support for your Kernels
---------------------------------
CONFIG_NET=y
CONFIG_CAN=m
CONFIG_CAN_RAW=m
CONFIG_CAN_BCM=m
# CAN Device Drivers
CONFIG_CAN_VCAN=m
---------------------------------
Enabling Fault Injection Support for your kernel (version 2.6.29).
Please check with the original kernel for the fault injection
types it supports. Following supports will be available:
/sys/kernel/debug/fail_io_timeout/interval
/sys/kernel/debug/fail_io_timeout/probability
/sys/kernel/debug/fail_io_timeout/reject-end
/sys/kernel/debug/fail_io_timeout/reject-start
/sys/kernel/debug/fail_io_timeout/require-end
/sys/kernel/debug/fail_io_timeout/require-start
/sys/kernel/debug/fail_io_timeout/space
/sys/kernel/debug/fail_io_timeout/stacktrace-depth
/sys/kernel/debug/fail_io_timeout/task-filter
/sys/kernel/debug/fail_io_timeout/times
/sys/kernel/debug/fail_io_timeout/verbose
/sys/kernel/debug/fail_make_request/interval
/sys/kernel/debug/fail_make_request/probability
/sys/kernel/debug/fail_make_request/reject-end
/sys/kernel/debug/fail_make_request/reject-start
/sys/kernel/debug/fail_make_request/require-end
/sys/kernel/debug/fail_make_request/require-start
/sys/kernel/debug/fail_make_request/space
/sys/kernel/debug/fail_make_request/stacktrace-depth
/sys/kernel/debug/fail_make_request/task-filter
/sys/kernel/debug/fail_make_request/times
/sys/kernel/debug/fail_make_request/verbose
/sys/kernel/debug/fail_page_alloc/ignore-gfp-highmem
/sys/kernel/debug/fail_page_alloc/ignore-gfp-wait
/sys/kernel/debug/fail_page_alloc/interval
/sys/kernel/debug/fail_page_alloc/min-order
/sys/kernel/debug/fail_page_alloc/probability
/sys/kernel/debug/fail_page_alloc/reject-end
/sys/kernel/debug/fail_page_alloc/reject-start
/sys/kernel/debug/fail_page_alloc/require-end
/sys/kernel/debug/fail_page_alloc/require-start
/sys/kernel/debug/fail_page_alloc/space
/sys/kernel/debug/fail_page_alloc/stacktrace-depth
/sys/kernel/debug/fail_page_alloc/task-filter
/sys/kernel/debug/fail_page_alloc/times
/sys/kernel/debug/fail_page_alloc/verbose
/sys/kernel/debug/failslab/ignore-gfp-wait
/sys/kernel/debug/failslab/interval
/sys/kernel/debug/failslab/probability
/sys/kernel/debug/failslab/reject-end
/sys/kernel/debug/failslab/reject-start
/sys/kernel/debug/failslab/require-end
/sys/kernel/debug/failslab/require-start
/sys/kernel/debug/failslab/space
/sys/kernel/debug/failslab/stacktrace-depth
/sys/kernel/debug/failslab/task-filter
/sys/kernel/debug/failslab/times
/sys/kernel/debug/failslab/verbose
when the below kernel config options are set:
CONFIG_FAULT_INJECTION=y
CONFIG_DEBUG_KERNEL=y
CONFIG_FAILSLAB=y (Fault-injection capability for kmalloc)
(CONFIG_SLAB=y || CONFIG_SLUB=y) if CONFIG_FAILSLAB=y
CONFIG_FAIL_PAGE_ALLOC=y (Fault-injection capabilitiy for alloc_pages())
CONFIG_FAIL_MAKE_REQUEST=y (Fault-injection capability for disk IO)
CONFIG_BLOCK=y if CONFIG_FAIL_MAKE_REQUEST=y
CONFIG_FAIL_IO_TIMEOUT=y (Faul-injection capability for faking disk interrupts)
CONFIG_BLOCK=y if CONFIG_FAIL_IO_TIMEOUT=y
CONFIG_FAULT_INJECTION_DEBUG_FS=y (Debugfs entries for fault-injection capabilities)
(CONFIG_SYSFS=y && CONFIG_DEBUG_FS=y) if CONFIG_FAULT_INJECTION_DEBUG_FS=y
CONFIG_FAULT_INJECTION_STACKTRACE_FILTER=y (stacktrace filter for fault-injection capabilities)
(CONFIG_FAULT_INJECTION_DEBUG_FS=y && CONFIG_STACKTRACE_SUPPORT=y && !CONFIG_X86_64) if
CONFIG_FAULT_INJECTION_STACKTRACE_FILTER=y
For more information on Fault injection, please refer to:
linux-2.6/Documentation/fault-injection/fault-injection.txt,
You should also have made the following entries in your /etc/fstab file
once the kernel is booted with the above CONFIG options set:
debugfs /sys/kernel/debug/ debugfs
# How the Kernel Fault Injection works for LTP ?
1) Build Kernel with all the above possible kernel CONFIG Options,
2) Create the above entry in /etc/fstab file,
3) Reboot in the new kernel,
4) Goto LTPROOT. Build and Install LTP as per ltp/INSTALL file,
5) Choose your own test(or default) to run with fault injection as follows:
./runltp -f <command_file> -F <LOOPS>,<FAULT_PROBABILITY>
The algorithm functions like:
loop (for each testcase)
begin
execute_testcase(inside_stable_kernel)
begin
insert_fault_into_kernel()
loop X Times
begin
execute_testcase(inside_fault_kernel)
end
restore_kernel_to_normal()
end
end
# External TODOs:
1) Add or modify testcases when relevant kernel functionality changes.
---------------------------------
---------------------------------
Enabling Kernel Configuration to test ext4 new features
---------------------------------
CONFIG_EXT4_FS=y
CONFIG_EXT4DEV_COMPAT=y
CONFIG_EXT4_FS_XATTR=y
CONFIG_EXT4_FS_POSIX_ACL=y
CONFIG_EXT4_FS_SECURITY=y
Beside that, the following packages are necessary.
e2fsprogs
e2fsprogs-devel
e2fsprogs-libs
And the version of packages must be 1.41.4 or above.
For more information to build/install/run these tests, look through:
ltp/testcases/kernel/fs/ext4-new-features/README
---------------------------------