share_pool: Don't do direct reclaim or compact for vmalloc_huge* funcs

ascend inclusion
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I4EUVI
CVE: NA

-------------------------------------------------

We found a hungtask problem when do direct compact in
__alloc_pages_nodemask:

vmalloc_hugepage_user -> __vmalloc_node_range -> __vmalloc_area_node ->
sp_alloc_pages -> alloc_huge_page_node -> alloc_fresh_huge_page ->
__alloc_pages_nodemask.

Set PF_MEMALLOC then direct reclaim and direct compact won't be called.
Signed-off-by: NTang Yizhou <tangyizhou@huawei.com>
Reviewed-by: NDing Tianhong <dingtianhong@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>
Reviewed-by: NWeilong Chen <chenweilong@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

mm/share_pool.c

@@ -3219,9 +3219,15 @@ void sp_group_post_exit(struct mm_struct *mm)
 struct page *sp_alloc_pages(struct vm_struct *area, gfp_t mask,
 					  unsigned int page_order, int node)
 {
-	if (area->flags & VM_HUGE_PAGES)
-		return hugetlb_alloc_hugepage(NUMA_NO_NODE, HUGETLB_ALLOC_NONE);
-	else
+	struct page *page;
+	unsigned int noreclaim_flag = 0;
+
+	if (area->flags & VM_HUGE_PAGES) {
+		noreclaim_flag = memalloc_noreclaim_save();
+		page = hugetlb_alloc_hugepage(NUMA_NO_NODE, HUGETLB_ALLOC_NONE);
+		memalloc_noreclaim_restore(noreclaim_flag);
+		return page;
+	} else
 		return alloc_pages_node(node, mask, page_order);
 }