2012-05-30 09:52:01

by Sasikantha Babu

Subject: [PATCH] prctl: Removed redunant assignment of "error" to zero

Just setting the "error" to error number is enough on failure and
It doesn't require to set "error" variable to zero in each switch case, since
it was already initialized with zero

Signed-off-by: Sasikantha babu <[email protected]>
---
kernel/sys.c | 7 -------
1 files changed, 0 insertions(+), 7 deletions(-)

diff --git a/kernel/sys.c b/kernel/sys.c
index 6df4262..d8e9016 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1920,7 +1920,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
break;
}
me->pdeath_signal = arg2;
- error = 0;
break;
case PR_GET_PDEATHSIG:
error = put_user(me->pdeath_signal, (int __user *)arg2);
@@ -1934,7 +1933,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
break;
}
set_dumpable(me->mm, arg2);
- error = 0;
break;

case PR_SET_UNALIGN:
@@ -1961,8 +1959,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
case PR_SET_TIMING:
if (arg2 != PR_TIMING_STATISTICAL)
error = -EINVAL;
- else
- error = 0;
break;

case PR_SET_NAME:
@@ -2013,7 +2009,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
current->default_timer_slack_ns;
else
current->timer_slack_ns = arg2;
- error = 0;
break;
case PR_MCE_KILL:
if (arg4 | arg5)
@@ -2039,7 +2034,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
default:
return -EINVAL;
}
- error = 0;
break;
case PR_MCE_KILL_GET:
if (arg2 | arg3 | arg4 | arg5)
@@ -2055,7 +2049,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
break;
case PR_SET_CHILD_SUBREAPER:
me->signal->is_child_subreaper = !!arg2;
- error = 0;
break;
case PR_GET_CHILD_SUBREAPER:
error = put_user(me->signal->is_child_subreaper,
--
1.7.3.4

2012-05-30 15:07:56

by Kees Cook

Subject: Re: [PATCH] prctl: Removed redunant assignment of "error" to zero

On Wed, May 30, 2012 at 2:55 AM, Sasikantha babu
<[email protected]> wrote:
> Just setting the "error" to error number is enough on failure and
> It doesn't require to set "error" variable to zero in each switch case, since
> it was already initialized with zero

This looks fine. Can you include an additional clean up in this? It
looks like PR_SET_NAME and PR_GET_NAME both 'return 0' instead of
using 'break' like all the rest.

-Kees

--
Kees Cook
Chrome OS Security