if (batch->nr == batch->max) {
if (!tlb_next_batch(tlb))
return 0;
+ batch = tlb->active;
}
VM_BUG_ON(batch->nr > batch->max);
int force_flush = 0;
int rss[NR_MM_COUNTERS];
spinlock_t *ptl;
+ pte_t *start_pte;
pte_t *pte;
again:
init_rss_vec(rss);
- pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
+ start_pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
+ pte = start_pte;
arch_enter_lazy_mmu_mode();
do {
pte_t ptent = *pte;
add_mm_rss_vec(mm, rss);
arch_leave_lazy_mmu_mode();
- pte_unmap_unlock(pte - 1, ptl);
+ pte_unmap_unlock(start_pte, ptl);
/*
* mmu_gather ran out of room to batch pages, we break out of
/**
* unmap_vmas - unmap a range of memory covered by a list of vma's
- * @tlbp: address of the caller's struct mmu_gather
+ * @tlb: address of the caller's struct mmu_gather
* @vma: the starting vma
* @start_addr: virtual address at which to start unmapping
* @end_addr: virtual address at which to end unmapping
}
EXPORT_SYMBOL(unmap_mapping_range);
-int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end)
-{
- struct address_space *mapping = inode->i_mapping;
-
- /*
- * If the underlying filesystem is not going to provide
- * a way to truncate a range of blocks (punch a hole) -
- * we should return failure right now.
- */
- if (!inode->i_op->truncate_range)
- return -ENOSYS;
-
- mutex_lock(&inode->i_mutex);
- down_write(&inode->i_alloc_sem);
- unmap_mapping_range(mapping, offset, (end - offset), 1);
- truncate_inode_pages_range(mapping, offset, end);
- unmap_mapping_range(mapping, offset, (end - offset), 1);
- inode->i_op->truncate_range(inode, offset, end);
- up_write(&inode->i_alloc_sem);
- mutex_unlock(&inode->i_mutex);
-
- return 0;
-}
-
/*
* We enter with non-exclusive mmap_sem (to exclude vma changes,
* but allow concurrent faults), and pte mapped but not yet locked.