arm64: avoid multiple evaluation of ptr in get_user/put_user()

get_user() is defined as a function macro in arm64, and trace_get_user()
calls it as followed:
     get_user(ch, ptr++);
Since the second parameter occurs twice in the definition, 'ptr++' is
unexpectedly evaluated twice and trace_get_user() will generate a bogus
string from user-provided one. As a result, some ftrace sysfs operations,
like "echo FUNCNAME > set_ftrace_filter," hit this case and eventually fail.
This patch fixes the issue both in get_user() and put_user().

Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
[catalin.marinas@arm.com: added __user type annotation and s/optr/__p/]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
This commit is contained in:
AKASHI Takahiro 2013-09-24 10:00:50 +01:00 committed by Catalin Marinas
parent 25804e6a96
commit 1f65c13efe

View file

@ -166,9 +166,10 @@ do { \
#define get_user(x, ptr) \
({ \
__typeof__(*(ptr)) __user *__p = (ptr); \
might_fault(); \
access_ok(VERIFY_READ, (ptr), sizeof(*(ptr))) ? \
__get_user((x), (ptr)) : \
access_ok(VERIFY_READ, __p, sizeof(*__p)) ? \
__get_user((x), __p) : \
((x) = 0, -EFAULT); \
})
@ -227,9 +228,10 @@ do { \
#define put_user(x, ptr) \
({ \
__typeof__(*(ptr)) __user *__p = (ptr); \
might_fault(); \
access_ok(VERIFY_WRITE, (ptr), sizeof(*(ptr))) ? \
__put_user((x), (ptr)) : \
access_ok(VERIFY_WRITE, __p, sizeof(*__p)) ? \
__put_user((x), __p) : \
-EFAULT; \
})