aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation/RCU/torture.txt
diff options
context:
space:
mode:
Diffstat (limited to 'Documentation/RCU/torture.txt')
-rw-r--r--Documentation/RCU/torture.txt48
1 files changed, 33 insertions, 15 deletions
diff --git a/Documentation/RCU/torture.txt b/Documentation/RCU/torture.txt
index 2967a65269d..a342b6e1cc1 100644
--- a/Documentation/RCU/torture.txt
+++ b/Documentation/RCU/torture.txt
@@ -10,23 +10,30 @@ status messages via printk(), which can be examined via the dmesg
10command (perhaps grepping for "torture"). The test is started 10command (perhaps grepping for "torture"). The test is started
11when the module is loaded, and stops when the module is unloaded. 11when the module is loaded, and stops when the module is unloaded.
12 12
13However, actually setting this config option to "y" results in the system 13CONFIG_RCU_TORTURE_TEST_RUNNABLE
14running the test immediately upon boot, and ending only when the system 14
15is taken down. Normally, one will instead want to build the system 15It is also possible to specify CONFIG_RCU_TORTURE_TEST=y, which will
16with CONFIG_RCU_TORTURE_TEST=m and to use modprobe and rmmod to control 16result in the tests being loaded into the base kernel. In this case,
17the test, perhaps using a script similar to the one shown at the end of 17the CONFIG_RCU_TORTURE_TEST_RUNNABLE config option is used to specify
18this document. Note that you will need CONFIG_MODULE_UNLOAD in order 18whether the RCU torture tests are to be started immediately during
19to be able to end the test. 19boot or whether the /proc/sys/kernel/rcutorture_runnable file is used
20to enable them. This /proc file can be used to repeatedly pause and
21restart the tests, regardless of the initial state specified by the
22CONFIG_RCU_TORTURE_TEST_RUNNABLE config option.
23
24You will normally -not- want to start the RCU torture tests during boot
25(and thus the default is CONFIG_RCU_TORTURE_TEST_RUNNABLE=n), but doing
26this can sometimes be useful in finding boot-time bugs.
20 27
21 28
22MODULE PARAMETERS 29MODULE PARAMETERS
23 30
24This module has the following parameters: 31This module has the following parameters:
25 32
26nreaders This is the number of RCU reading threads supported. 33irqreaders Says to invoke RCU readers from irq level. This is currently
27 The default is twice the number of CPUs. Why twice? 34 done via timers. Defaults to "1" for variants of RCU that
28 To properly exercise RCU implementations with preemptible 35 permit this. (Or, more accurately, variants of RCU that do
29 read-side critical sections. 36 -not- permit this know to ignore this variable.)
30 37
31nfakewriters This is the number of RCU fake writer threads to run. Fake 38nfakewriters This is the number of RCU fake writer threads to run. Fake
32 writer threads repeatedly use the synchronous "wait for 39 writer threads repeatedly use the synchronous "wait for
@@ -37,6 +44,16 @@ nfakewriters This is the number of RCU fake writer threads to run. Fake
37 to trigger special cases caused by multiple writers, such as 44 to trigger special cases caused by multiple writers, such as
38 the synchronize_srcu() early return optimization. 45 the synchronize_srcu() early return optimization.
39 46
47nreaders This is the number of RCU reading threads supported.
48 The default is twice the number of CPUs. Why twice?
49 To properly exercise RCU implementations with preemptible
50 read-side critical sections.
51
52shuffle_interval
53 The number of seconds to keep the test threads affinitied
54 to a particular subset of the CPUs, defaults to 3 seconds.
55 Used in conjunction with test_no_idle_hz.
56
40stat_interval The number of seconds between output of torture 57stat_interval The number of seconds between output of torture
41 statistics (via printk()). Regardless of the interval, 58 statistics (via printk()). Regardless of the interval,
42 statistics are printed when the module is unloaded. 59 statistics are printed when the module is unloaded.
@@ -44,10 +61,11 @@ stat_interval The number of seconds between output of torture
44 be printed -only- when the module is unloaded, and this 61 be printed -only- when the module is unloaded, and this
45 is the default. 62 is the default.
46 63
47shuffle_interval 64stutter The length of time to run the test before pausing for this
48 The number of seconds to keep the test threads affinitied 65 same period of time. Defaults to "stutter=5", so as
49 to a particular subset of the CPUs, defaults to 5 seconds. 66 to run and pause for (roughly) five-second intervals.
50 Used in conjunction with test_no_idle_hz. 67 Specifying "stutter=0" causes the test to run continuously
68 without pausing, which is the old default behavior.
51 69
52test_no_idle_hz Whether or not to test the ability of RCU to operate in 70test_no_idle_hz Whether or not to test the ability of RCU to operate in
53 a kernel that disables the scheduling-clock interrupt to 71 a kernel that disables the scheduling-clock interrupt to