From: Mian Yousaf Kaukab <yousaf.kaukab@suse.com>
To: linux-modules@vger.kernel.org, lucas.demarchi@intel.com
Cc: bjorn.andersson@linaro.org, afaerber@suse.de,
Mian Yousaf Kaukab <yousaf.kaukab@suse.com>
Subject: [PATCH v1 1/2] testsuite: depmod: add module dependency outside cyclic chain
Date: Tue, 8 Nov 2016 17:45:49 +0100 [thread overview]
Message-ID: <1478623550-18716-1-git-send-email-yousaf.kaukab@suse.com> (raw)
Check that depmod do not report modules outside cyclic chain
Two modules f and g are added which do not have any dependency.
modules a and b are made dependent on f and g.
Signed-off-by: Mian Yousaf Kaukab <yousaf.kaukab@suse.com>
---
Here is the output of loop dependency check test after adding this
patch:
ESTSUITE: ERR: wrong:
depmod: ERROR: Found 7 modules in dependency cycles!
depmod: ERROR: Cycle detected: mod_loop_d -> mod_loop_e -> mod_loop_d
depmod: ERROR: Cycle detected: mod_loop_b -> mod_loop_c -> mod_loop_a -> mod_loop_b
depmod: ERROR: Cycle detected: mod_loop_b -> mod_loop_c -> mod_loop_a -> mod_loop_g
depmod: ERROR: Cycle detected: mod_loop_b -> mod_loop_c -> mod_loop_a -> mod_loop_f
Buffer overflow occurs in the loop when last two lines are printed.
43 bytes buffer is allocated and 53 bytes are used.
testsuite/module-playground/Makefile | 6 +++++-
testsuite/module-playground/mod-loop-a.c | 2 ++
testsuite/module-playground/mod-loop-b.c | 2 ++
testsuite/module-playground/mod-loop-f.c | 24 ++++++++++++++++++++++++
testsuite/module-playground/mod-loop-g.c | 24 ++++++++++++++++++++++++
testsuite/module-playground/mod-loop.h | 2 ++
testsuite/populate-modules.sh | 2 ++
7 files changed, 61 insertions(+), 1 deletion(-)
create mode 100644 testsuite/module-playground/mod-loop-f.c
create mode 100644 testsuite/module-playground/mod-loop-g.c
diff --git a/testsuite/module-playground/Makefile b/testsuite/module-playground/Makefile
index a5f142f..bf364a9 100644
--- a/testsuite/module-playground/Makefile
+++ b/testsuite/module-playground/Makefile
@@ -12,13 +12,17 @@ obj-m += mod-foo-c.o
obj-m += mod-foo.o
# mod-loop: create loops in dependencies:
-# 1) mod-loop-a -> mod-loop-b -> mod-loop-c -> mod-loop-a
+# 1) mod-loop-a -> mod-loop-b -> mod-loop-c -> mod-loop-a
+# |-> mod-loop-f |-> mod-loop-f
+# \-> mod-loop-g \-> mod-loop-g
# 2) mod-loop-d -> mod-loop-e -> mod-loop-d
obj-m += mod-loop-a.o
obj-m += mod-loop-b.o
obj-m += mod-loop-c.o
obj-m += mod-loop-d.o
obj-m += mod-loop-e.o
+obj-m += mod-loop-f.o
+obj-m += mod-loop-g.o
# mod-fake-*: fake the respective modules in kernel with these aliases. Aliases
# list was taken from 3.5.4
diff --git a/testsuite/module-playground/mod-loop-a.c b/testsuite/module-playground/mod-loop-a.c
index e1fd0ce..e5adb49 100644
--- a/testsuite/module-playground/mod-loop-a.c
+++ b/testsuite/module-playground/mod-loop-a.c
@@ -10,6 +10,8 @@ static int __init test_module_init(void)
{
printA();
printB();
+ printF();
+ printG();
return 0;
}
diff --git a/testsuite/module-playground/mod-loop-b.c b/testsuite/module-playground/mod-loop-b.c
index f4490b7..26232ea 100644
--- a/testsuite/module-playground/mod-loop-b.c
+++ b/testsuite/module-playground/mod-loop-b.c
@@ -10,6 +10,8 @@ static int __init test_module_init(void)
{
printB();
printC();
+ printF();
+ printG();
return 0;
}
diff --git a/testsuite/module-playground/mod-loop-f.c b/testsuite/module-playground/mod-loop-f.c
new file mode 100644
index 0000000..0abb161
--- /dev/null
+++ b/testsuite/module-playground/mod-loop-f.c
@@ -0,0 +1,24 @@
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/printk.h>
+
+#include "mod-loop.h"
+
+static int __init test_module_init(void)
+{
+ printF();
+
+ return 0;
+}
+module_init(test_module_init);
+
+void printF(void)
+{
+ pr_warn("Hello, world F\n");
+}
+EXPORT_SYMBOL(printF);
+
+MODULE_AUTHOR("Lucas De Marchi <lucas.demarchi@intel.com>");
+MODULE_LICENSE("LGPL");
diff --git a/testsuite/module-playground/mod-loop-g.c b/testsuite/module-playground/mod-loop-g.c
new file mode 100644
index 0000000..0965d76
--- /dev/null
+++ b/testsuite/module-playground/mod-loop-g.c
@@ -0,0 +1,24 @@
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/printk.h>
+
+#include "mod-loop.h"
+
+static int __init test_module_init(void)
+{
+ printG();
+
+ return 0;
+}
+module_init(test_module_init);
+
+void printG(void)
+{
+ pr_warn("Hello, world G\n");
+}
+EXPORT_SYMBOL(printG);
+
+MODULE_AUTHOR("Lucas De Marchi <lucas.demarchi@intel.com>");
+MODULE_LICENSE("LGPL");
diff --git a/testsuite/module-playground/mod-loop.h b/testsuite/module-playground/mod-loop.h
index 3244ad9..4da3e67 100644
--- a/testsuite/module-playground/mod-loop.h
+++ b/testsuite/module-playground/mod-loop.h
@@ -5,3 +5,5 @@ void printB(void);
void printC(void);
void printD(void);
void printE(void);
+void printF(void);
+void printG(void);
diff --git a/testsuite/populate-modules.sh b/testsuite/populate-modules.sh
index 409a6de..ba1f842 100755
--- a/testsuite/populate-modules.sh
+++ b/testsuite/populate-modules.sh
@@ -16,6 +16,8 @@ map=(
["test-depmod/detect-loop/lib/modules/4.4.4/kernel/mod-loop-c.ko"]="mod-loop-c.ko"
["test-depmod/detect-loop/lib/modules/4.4.4/kernel/mod-loop-d.ko"]="mod-loop-d.ko"
["test-depmod/detect-loop/lib/modules/4.4.4/kernel/mod-loop-e.ko"]="mod-loop-e.ko"
+ ["test-depmod/detect-loop/lib/modules/4.4.4/kernel/mod-loop-f.ko"]="mod-loop-f.ko"
+ ["test-depmod/detect-loop/lib/modules/4.4.4/kernel/mod-loop-g.ko"]="mod-loop-g.ko"
["test-dependencies/lib/modules/4.0.20-kmod/kernel/fs/foo/"]="mod-foo-b.ko"
["test-dependencies/lib/modules/4.0.20-kmod/kernel/"]="mod-foo-c.ko"
["test-dependencies/lib/modules/4.0.20-kmod/kernel/lib/"]="mod-foo-a.ko"
--
1.8.5.6
next reply other threads:[~2016-11-08 16:46 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-08 16:45 Mian Yousaf Kaukab [this message]
2016-11-08 16:45 ` [PATCH v1 2/2] depmod: ignore related modules in depmod_report_cycles Mian Yousaf Kaukab
2016-11-09 0:40 ` Lucas De Marchi
2016-11-09 2:59 ` Yauheni Kaliuta
2016-11-09 9:17 ` Mian Yousaf Kaukab
2016-11-09 11:23 ` Yauheni Kaliuta
2016-11-11 11:43 ` [PATCH RFC 0/3] Proposal for cycles handling Yauheni Kaliuta
2016-11-11 11:43 ` [PATCH RFC 2/3] libkmod: list: export list handling functions Yauheni Kaliuta
2017-02-13 8:05 ` Lucas De Marchi
2017-02-20 14:22 ` Yauheni Kaliuta
2017-02-22 5:26 ` Lucas De Marchi
2017-02-22 9:41 ` [PATCH v3 0/2] Proposal for cycles handling Yauheni Kaliuta
2017-02-22 9:41 ` [PATCH v3 1/2] testsuite: depmod: check netsted loops reporting Yauheni Kaliuta
2017-02-22 9:41 ` [PATCH v3 2/2] depmod: handle nested loops Yauheni Kaliuta
2017-02-23 22:30 ` Lucas De Marchi
2016-11-11 11:43 ` [PATCH RFC 3/3] " Yauheni Kaliuta
2017-02-13 8:30 ` Lucas De Marchi
2017-02-20 14:16 ` Yauheni Kaliuta
2017-02-13 8:16 ` [PATCH RFC 0/3] Proposal for cycles handling Lucas De Marchi
2017-02-13 9:56 ` Yauheni Kaliuta
2017-02-13 8:32 ` Lucas De Marchi
2017-02-20 14:18 ` [PATCH RFC v2 0/2] " Yauheni Kaliuta
2017-02-20 14:18 ` [PATCH RFC v2 1/2] testsuite: depmod: check netsted loops reporting Yauheni Kaliuta
2017-02-20 14:19 ` [PATCH RFC v2 2/2] depmod: handle nested loops Yauheni Kaliuta
2016-11-09 0:29 ` [PATCH v1 1/2] testsuite: depmod: add module dependency outside cyclic chain Lucas De Marchi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1478623550-18716-1-git-send-email-yousaf.kaukab@suse.com \
--to=yousaf.kaukab@suse.com \
--cc=afaerber@suse.de \
--cc=bjorn.andersson@linaro.org \
--cc=linux-modules@vger.kernel.org \
--cc=lucas.demarchi@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).