函数逻辑报告 |
Source Code:mm\hugetlb.c |
Create Date:2022-07-27 17:01:00 |
Last Modify:2020-03-12 14:18:49 | Copyright©Brick |
首页 | 函数Tree |
注解内核,赢得工具 | 下载SCCT | English |
函数名称:hugetlb_fault
函数原型:vm_fault_t hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, unsigned int flags)
返回类型:vm_fault_t
参数:
类型 | 参数 | 名称 |
---|---|---|
struct mm_struct * | mm | |
struct vm_area_struct * | vma | |
unsigned long | address | |
unsigned int | flags |
4007 | struct page * pagecache_page = NULL |
4008 | h等于hstate_vma(vma) |
4010 | need_wait_lock等于0 |
4011 | haddr等于address按位与huge_page_mask(h) |
4013 | ptep等于huge_pte_offset(mm, haddr, huge_page_size(h)) |
4014 | 如果ptep则 |
4015 | entry等于huge_ptep_get(ptep) |
4016 | 如果此条件成立可能性小(为编译器优化)(is_hugetlb_entry_migration(entry))则 |
4017 | migration_entry_wait_huge(vma, mm, ptep) |
4018 | 返回:0 |
4022 | 否则 |
4023 | ptep等于arch callbacks |
4024 | 如果非ptep则返回:VM_FAULT_OOM |
4029 | idx等于Convert the address within this vma to the page offset within* the mapping, in pagecache page units; huge pages here. |
4036 | hash等于For uniprocesor systems we always use a single mutex, so just* return 0 and avoid the hashing overhead. |
4037 | mutex_lock( & hugetlb_fault_mutex_table[hash]) |
4039 | entry等于huge_ptep_get(ptep) |
4040 | 如果huge_pte_none(entry)则 |
4045 | ret等于0 |
4054 | 如果非pte_present(entry)则转到:out_mutex |
4065 | 如果flags按位与Fault was a write access 且非huge_pte_write(entry)则 |
4066 | 如果vma_needs_reservation(h, vma, haddr)小于0则 |
4067 | ret等于VM_FAULT_OOM |
4068 | 转到:out_mutex |
4071 | vma_end_reservation(h, vma, haddr) |
4078 | ptl等于huge_pte_lock(h, mm, ptep) |
4081 | 如果此条件成立可能性小(为编译器优化)(!pte_same(entry, huge_ptep_get(ptep)))则转到:out_ptl |
4092 | need_wait_lock等于1 |
4093 | 转到:out_ptl |
4098 | 如果flags按位与Fault was a write access 则 |
4099 | 如果非huge_pte_write(entry)则 |
4102 | 转到:out_put_page |
4104 | entry等于huge_pte_mkdirty(entry) |
4106 | entry等于pte_mkyoung(entry) |
4110 | out_put_page : |
4114 | out_ptl : |
4115 | 自旋锁解锁 |
4117 | 如果pagecache_page则 |
4121 | out_mutex : |
4130 | 如果need_wait_lock则Wait for a page to be unlocked.* This must be called with the caller "holding" the page,* ie with increased "page->count" so that the page won't* go away during the wait.. |
4132 | 返回:ret |
名称 | 描述 |
---|---|
follow_hugetlb_page | |
handle_mm_fault | By the time we get here, we already hold the mm semaphore* The mmap_sem may have been released depending on flags and our* return value. See filemap_fault() and __lock_page_or_retry(). |
源代码转换工具 开放的插件接口 | X |
---|---|
支持:c/c++/esqlc/java Oracle/Informix/Mysql 插件可实现:逻辑报告 代码生成和批量转换代码 |