v6 · 内核 Patch 系列追踪

1/22

kho: generalize radix tree APIs

将 KHO radix tree API 从页跟踪中解耦，改为直接传入 key，编码移到调用方，并重命名 kho_radix_add_key/del_key 与回调参数。

2/22

kho: make radix max key width more obvious

用 KHO_RADIX_KEY_WIDTH 取代晦涩的 KHO_ORDER_0_LOG2，明确最大 key 宽度为 PFN 52 位加 order 1 位，并优化编解码注释。

3/22

kho: disallow wide keys in radix tree

在 kho_radix_add_key() 中校验 key 宽度，超限返回错误；不为尚无需求的大 key 增加树深度或内存开销。

4/22

kho: return virtual address of mem_map from kho_get_mem_map()

kho_get_mem_map_phys() 改名为 kho_get_mem_map() 并返回虚拟地址；kho_populate() 只需判空，后续调用者需要 VA。

5/22

kho: store incoming radix tree in kho_in

将传入的 radix tree 根保存到 kho_in，并复用 kho_get_mem_map_phys() 获取 FDT 中的根节点，避免重复代码。

6/22

kho: move all memory retrieval logic to kho_mem_retrieve()

把 kho_memory_init() 中的内存检索逻辑全部移到 kho_mem_retrieve()，统一处理 scratch 初始化和错误清理。

7/22

kho: add a struct for radix callbacks

为 KHO radix tree 回调引入 struct 封装，为后续增加更多回调做准备，并相应调整现有回调调用方式。

8/22

kho: add callback for table pages

新增 table page 回调，遍历 radix tree 时返回每个表页的物理地址，用于发现无保留页内存块及外部用户保留页。

9/22

kho: add data argument to radix walk callback

From: “Pratyush Yadav (Google)” pratyush@kernel.org Add an opaque data pointer argument to kho_radix_walk_cb_t. This can be used by callers to pass extra information to the callback. Reviewed-by:

10/22

kho: allow early-boot usage of the KHO radix tree

From: “Pratyush Yadav (Google)” pratyush@kernel.org The KHO radix tree allocates memory for table pages from the buddy allocator using get_zeroed_page(). This is not available in early boot when me

11/22

kho: allow destroying KHO radix tree

From: “Pratyush Yadav (Google)” pratyush@kernel.org Add kho_radix_destroy_tree() which allows destroying the radix tree and freeing all its pages. This is will be used by the upcoming scratch exte

12/22

kho: add kho_radix_init_tree()

From: “Pratyush Yadav (Google)” pratyush@kernel.org Move the initialization logic of the radix tree into kho_radix_init_tree() instead of having users open-code it. Makes the boundaries cleaner and

13/22

kho: expose kho_scratch_overlap() to kexec_handover.h

导出 kho_scratch_overlap() 供 memblock 使用，调试检查改用 IS_ENABLED 门控，并删除已空的 kexec_handover_debug.c。

14/22

kho: initialize kho_scratch pointer earlier in boot

From: “Pratyush Yadav (Google)” pratyush@kernel.org In a future patch, mm init will use kho_scratch_overlap() for deciding the migrate type of pageblocks it initializes. The earliest user currently

15/22

kho: initialize preserved memory map radix tree earlier

From: “Pratyush Yadav (Google)” pratyush@kernel.org Currently the preserved memory radix tree is initialized from kho_memory_init(), which happens relatively late in MM init. In a coming patch, the

16/22

mm/mm_init: simplify deferred_free_pages() migratetype init

From: Muchun Song songmuchun@bytedance.com deferred_free_pages() open-codes two loops to initialize the pageblock migratetype for a range of pages. Replace them with pageblock_migratetype_init_ran

17/22

mm/mm_init: don't rely on memblock to get KHO scratch migratetype

From: “Pratyush Yadav (Google)” pratyush@kernel.org Currently struct page init via memmap_init() or deferred_init_memmap() only queries the migrate type from KHO for each discrete memory range. Tha

18/22

kho: extend scratch

From: “Pratyush Yadav (Google)” pratyush@kernel.org Motivation ========== The scratch space is allocated by the first kernel in the KHO chain, and is reused by all subsequent kernels. The size of

19/22

memblock: always include KHO headers

From: “Pratyush Yadav (Google)” pratyush@kernel.org In a coming commit, memblock will start using kho_scratch_overlap() without a compile guard. The compile guard for the function is in kexec_hando

20/22

memblock: make HugeTLB bootmem allocation work with KHO

From: “Pratyush Yadav (Google)” pratyush@kernel.org Gigantic huge page allocation is somewhat broken currently when KHO is used. Firstly, they break KHO scratch size accounting. RSRV_KERN is used

21/22

memblock: add memblock_reserved_hugetlb_size()

From: “Pratyush Yadav (Google)” pratyush@kernel.org Similar to memblock_reserved_kern_size(), but calculates only the memory reserved for hugetlb pages. This is needed in an upcoming commit that su

22/22

kho: exclude hugetlb memory from scratch size calculation

From: “Pratyush Yadav (Google)” pratyush@kernel.org HugeTLB pages can be preserved memory. So they are never allocated from scratch. Instead, they are allocated from the memory blocks with no prese

KHO 下巨型页分配修复 · v6

原始补丁链接

子 Patch（22）