ipcns: fix stupid !CONFIG_IPC error
PATCH 1/1 ipcns: fix stupid !CONFIG_IPC error
Sorry, I broke CONFIG_IPC=n. This is a port of the patch Andrew used to fix it in -mm
Here is a question... When CONFIG_IPC_NS=n and the user asks for a new IPC namespace, we want to return -EINVAL to let them know we couldn't oblige. But if CONFIG_IPC=n is it ok to just return 0?
If not, then we'll need a somewhat different fix, as we can't #include sched.h in ipc.h.
thanks,
-serge
From: Serge E. Hallyn <[email protected]>
Subject: [PATCH] ipcns: fix stupid !CONFIG_IPC error
Andrew Morton pointed out my patch to fix CLONE_NEWIPC behavior when CONFIG_IPC_NS=n broke compilation when CONFIG_IPC=n.
Make sure copy_ipcs() is always defined.
Signed-off-by Serge E. Hallyn <[email protected]>
---
include/linux/ipc.h | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-)
d731d029f908b76af52d2abd9ab199fa931d2c98
include/linux/ipc.h +3 -1
@@ -92,11 +92,13 @@ extern struct ipc_namespace init_ipc_ns;#ifdef CONFIG_SYSVIPC #define INIT_IPC_NS (ns) .ns = &init_ipc_ns, +extern int copy_ipcs (unsigned long flags,struct task_struct *tsk);#else #define INIT_IPC_NS (ns) +static inline int copy_ipcs (unsigned long flags,struct task_struct *tsk) +{return 0 ; }#endif -extern int copy_ipcs (unsigned long flags,struct task_struct *tsk);#ifdef CONFIG_IPC_NS extern void free_ipc_ns (struct kref *kref);extern int unshare_ipcs (unsigned long flags,struct ipc_namespace **ns);
Signature · 1 line
1.3.2