Home
last modified time | relevance | path

Searched refs:mm_lock_seq (Results 1 – 3 of 3) sorted by relevance

/linux/mm/
H A Dinit-mm.c44 .mm_lock_seq = SEQCNT_ZERO(init_mm.mm_lock_seq),
H A Dmmap_lock.c141 const unsigned int mm_lock_seq = __vma_raw_mm_seqnum(vma); in vma_mark_detached()
160 WRITE_ONCE(vma->vm_lock_seq, mm_lock_seq);
204 * False locked result is possible if mm_lock_seq overflows or if vma gets in vma_start_read()
221 * We can use READ_ONCE() for the mm_lock_seq here, and don't need in vma_start_read()
223 * we don't rely on for anything - the mm_lock_seq read against which we in vma_start_read()
226 if (READ_ONCE(vma->vm_lock_seq) == READ_ONCE(mm->mm_lock_seq.sequence)) { in vma_start_read()
252 * Overflow of vm_lock_seq/mm_lock_seq might produce false locked result. in lock_vma_under_rcu()
254 * vma->vm_lock_seq under vma->vm_refcnt protection and mm->mm_lock_seq in lock_vma_under_rcu()
257 * We must use ACQUIRE semantics for the mm_lock_seq so that if we are in lock_vma_under_rcu()
262 if (unlikely(vma->vm_lock_seq == raw_read_seqcount(&mm->mm_lock_seq))) { in lock_vma_under_rcu()
99 __vma_start_write(struct vm_area_struct * vma,unsigned int mm_lock_seq,int state) __vma_start_write() argument
[all...]
/linux/include/linux/
H A Dmm_types.h1222 seqcount_t mm_lock_seq; member