Some devices can access process address spaces directly. When creating such bond, to check that a process controlling the device is allowed to access the target address space, the device driver uses mm_access(). Since the drivers (in this case VFIO) can be built as a module, export the mm_access symbol. Cc: felix.kuehling@xxxxxxx Cc: akpm@xxxxxxxxxxxxxxxxxxxx Signed-off-by: Jean-Philippe Brucker <jean-philippe.brucker@xxxxxxx> --- This patch was already sent last year for AMD KFD. I'm resending it for VFIO, trying to address Andrew Morton's request to comment the exported function: http://lkml.iu.edu/hypermail/linux/kernel/1705.2/06774.html --- kernel/fork.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/kernel/fork.c b/kernel/fork.c index a5d21c42acfc..1062f7450e97 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1098,6 +1098,19 @@ struct mm_struct *get_task_mm(struct task_struct *task) } EXPORT_SYMBOL_GPL(get_task_mm); +/** + * mm_access - check access permission to a task and and acquire a reference to + * its mm. + * @task: target task + * @mode: selects type of access and caller credentials + * + * Return the task's mm on success, or %NULL if it cannot be accessed. + * + * Check if the caller is allowed to read or write the target task's pages. + * @mode describes the access mode and credentials using ptrace access flags. + * See ptrace_may_access() for more details. On success, a reference to the mm + * is taken. + */ struct mm_struct *mm_access(struct task_struct *task, unsigned int mode) { struct mm_struct *mm; @@ -1117,6 +1130,7 @@ struct mm_struct *mm_access(struct task_struct *task, unsigned int mode) return mm; } +EXPORT_SYMBOL_GPL(mm_access); static void complete_vfork_done(struct task_struct *tsk) { -- 2.17.0
