summaryrefslogtreecommitdiffstats
path: root/Documentation/fault-injection
diff options
context:
space:
mode:
authorDmitry Vyukov <dvyukov@google.com>2017-07-12 14:34:35 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2017-07-12 16:26:01 -0700
commite41d58185f1444368873d4d7422f7664a68be61d (patch)
tree1d7dc405b33a4aead67da40d0caec446a3da900d /Documentation/fault-injection
parent92ef6da3d06ff551a86de41ae37df9cc4b58d7a0 (diff)
downloadlinux-stable-e41d58185f1444368873d4d7422f7664a68be61d.tar.gz
linux-stable-e41d58185f1444368873d4d7422f7664a68be61d.tar.bz2
linux-stable-e41d58185f1444368873d4d7422f7664a68be61d.zip
fault-inject: support systematic fault injection
Add /proc/self/task/<current-tid>/fail-nth file that allows failing 0-th, 1-st, 2-nd and so on calls systematically. Excerpt from the added documentation: "Write to this file of integer N makes N-th call in the current task fail (N is 0-based). Read from this file returns a single char 'Y' or 'N' that says if the fault setup with a previous write to this file was injected or not, and disables the fault if it wasn't yet injected. Note that this file enables all types of faults (slab, futex, etc). This setting takes precedence over all other generic settings like probability, interval, times, etc. But per-capability settings (e.g. fail_futex/ignore-private) take precedence over it. This feature is intended for systematic testing of faults in a single system call. See an example below" Why add a new setting: 1. Existing settings are global rather than per-task. So parallel testing is not possible. 2. attr->interval is close but it depends on attr->count which is non reset to 0, so interval does not work as expected. 3. Trying to model this with existing settings requires manipulations of all of probability, interval, times, space, task-filter and unexposed count and per-task make-it-fail files. 4. Existing settings are per-failure-type, and the set of failure types is potentially expanding. 5. make-it-fail can't be changed by unprivileged user and aggressive stress testing better be done from an unprivileged user. Similarly, this would require opening the debugfs files to the unprivileged user, as he would need to reopen at least times file (not possible to pre-open before dropping privs). The proposed interface solves all of the above (see the example). We want to integrate this into syzkaller fuzzer. A prototype has found 10 bugs in kernel in first day of usage: https://groups.google.com/forum/#!searchin/syzkaller/%22FAULT_INJECTION%22%7Csort:relevance I've made the current interface work with all types of our sandboxes. For setuid the secret sauce was prctl(PR_SET_DUMPABLE, 1, 0, 0, 0) to make /proc entries non-root owned. So I am fine with the current version of the code. [akpm@linux-foundation.org: fix build] Link: http://lkml.kernel.org/r/20170328130128.101773-1-dvyukov@google.com Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Cc: Akinobu Mita <akinobu.mita@gmail.com> Cc: Michal Hocko <mhocko@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'Documentation/fault-injection')
-rw-r--r--Documentation/fault-injection/fault-injection.txt78
1 files changed, 78 insertions, 0 deletions
diff --git a/Documentation/fault-injection/fault-injection.txt b/Documentation/fault-injection/fault-injection.txt
index 415484f3d59a..192d8cbcc5f9 100644
--- a/Documentation/fault-injection/fault-injection.txt
+++ b/Documentation/fault-injection/fault-injection.txt
@@ -134,6 +134,22 @@ use the boot option:
fail_futex=
mmc_core.fail_request=<interval>,<probability>,<space>,<times>
+o proc entries
+
+- /proc/self/task/<current-tid>/fail-nth:
+
+ Write to this file of integer N makes N-th call in the current task fail
+ (N is 0-based). Read from this file returns a single char 'Y' or 'N'
+ that says if the fault setup with a previous write to this file was
+ injected or not, and disables the fault if it wasn't yet injected.
+ Note that this file enables all types of faults (slab, futex, etc).
+ This setting takes precedence over all other generic debugfs settings
+ like probability, interval, times, etc. But per-capability settings
+ (e.g. fail_futex/ignore-private) take precedence over it.
+
+ This feature is intended for systematic testing of faults in a single
+ system call. See an example below.
+
How to add new fault injection capability
-----------------------------------------
@@ -278,3 +294,65 @@ allocation failure.
# env FAILCMD_TYPE=fail_page_alloc \
./tools/testing/fault-injection/failcmd.sh --times=100 \
-- make -C tools/testing/selftests/ run_tests
+
+Systematic faults using fail-nth
+---------------------------------
+
+The following code systematically faults 0-th, 1-st, 2-nd and so on
+capabilities in the socketpair() system call.
+
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/socket.h>
+#include <sys/syscall.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <string.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <errno.h>
+
+int main()
+{
+ int i, err, res, fail_nth, fds[2];
+ char buf[128];
+
+ system("echo N > /sys/kernel/debug/failslab/ignore-gfp-wait");
+ sprintf(buf, "/proc/self/task/%ld/fail-nth", syscall(SYS_gettid));
+ fail_nth = open(buf, O_RDWR);
+ for (i = 0;; i++) {
+ sprintf(buf, "%d", i);
+ write(fail_nth, buf, strlen(buf));
+ res = socketpair(AF_LOCAL, SOCK_STREAM, 0, fds);
+ err = errno;
+ read(fail_nth, buf, 1);
+ if (res == 0) {
+ close(fds[0]);
+ close(fds[1]);
+ }
+ printf("%d-th fault %c: res=%d/%d\n", i, buf[0], res, err);
+ if (buf[0] != 'Y')
+ break;
+ }
+ return 0;
+}
+
+An example output:
+
+0-th fault Y: res=-1/23
+1-th fault Y: res=-1/23
+2-th fault Y: res=-1/23
+3-th fault Y: res=-1/12
+4-th fault Y: res=-1/12
+5-th fault Y: res=-1/23
+6-th fault Y: res=-1/23
+7-th fault Y: res=-1/23
+8-th fault Y: res=-1/12
+9-th fault Y: res=-1/12
+10-th fault Y: res=-1/12
+11-th fault Y: res=-1/12
+12-th fault Y: res=-1/12
+13-th fault Y: res=-1/12
+14-th fault Y: res=-1/12
+15-th fault Y: res=-1/12
+16-th fault N: res=0/12