6.828 Lab2 part 1

Ex1:physical page management in JOS.
Ans:這題要求完成在 JOS 中對 physical pages 的管理，第一個要完成的是 bootstrap 時一開始所需要的 boot_alloc()。首先，從 Lab1 我們可以知道進入 kernel 後，physical memory 的使用狀況如下（但在entry.S中只先將[0, 4MB）對映到[KERNBASE, KERBASE+4MB),所以boot_alloc()能直接以指標存取的記憶體會受限在這個範圍）：

boot_alloc()的記憶體會在後續持續被使用，但此時尚未完成 virtual memory 的設定，所以我們在 boot_alloc()必須以 page 的大小作為單位來使用，這樣才能在後續管理 page frame 時將 boot_alloc()過的記憶體正確標注狀態。實作如下：

static void * boot_alloc(uint32_t n) { static char *nextfree = 0; // virtual address of next byte of free memory static size_t used_npages = 0; static size_t want_npages = 0; char *result = NULL; // this is defined by linker script, kern.ld, // the very beginning virtual address of kernel image extern char btext[]; extern char end[]; // Initialize nextfree if this is the first time. // 'end' is a magic symbol automatically generated by the linker, // which points to the end of the kernel's bss segment: // the first virtual address that the linker did *not* assign // to any kernel code or global variables. if (!nextfree) { nextfree = ROUNDUP((char *) end, PGSIZE); } // Allocate a chunk large enough to hold 'n' bytes, then update // nextfree. Make sure nextfree is kept aligned // to a multiple of PGSIZE. // // LAB 2: Your code here. // calculate how much memory callers want in page unit. if (n) { want_npages = (n/PGSIZE); if (n%PGSIZE) { want_npages = (n/PGSIZE) + 1; } } else { result = nextfree; goto done; } // npage is calculated in i386_detect_memory() if (want_npages > (npages - ROUNDUP((end - btext + 1), PGSIZE) - used_npages)) { panic("out of memory in boot_alloc\n"); } else { result = nextfree; nextfree += PGSIZE*want_npages; used_npages += want_npages; } done: return result; }

接著是完成page_init()，這邊要完成對全域變數 pages[] 與 free_page_list 的設定。free_page_list會指向目前可用的free pages的頭。由於boot_alloc()已使用了部份free memory，所以在串free_page_list時要將那些記憶體跳過。如下：

實作如下：

void page_init(void) { // The example code here marks all physical pages as free. // However this is not truly the case. What memory is free? // 1) Mark physical page 0 as in use. // This way we preserve the real-mode IDT and BIOS structures // in case we ever need them. (Currently we don't, but...) // 2) The rest of base memory, [PGSIZE, npages_basemem * PGSIZE) // is free. // 3) Then comes the IO hole [IOPHYSMEM, EXTPHYSMEM), which must // never be allocated. // 4) Then extended memory [EXTPHYSMEM, ...). // Some of it is in use, some is free. Where is the kernel // in physical memory? Which pages are already in use for // page tables and other data structures? // // Change the code to reflect this. // NB: DO NOT actually touch the physical memory corresponding to // free pages! size_t next_free_pgnum = PGNUM(PADDR(boot_alloc(0))); size_t i; for (i = 0; i < (npages-1); i++) { pages[i].pp_ref = 0; pages[i].pp_link = &pages[i+1]; pages[i].pp_reserved = 0; } pages[i].pp_ref = 0; pages[i].pp_link = NULL; pages[i].pp_reserved = 0; // page 0 is for real mode IDT and BIOS structure pages[0].pp_ref = 1; pages[0].pp_link = NULL; pages[0].pp_reserved = 1; // page_free_list points to page 1 page_free_list = &pages[1]; // free pages do not include [IOPHYSMEM, nextfree) pages[PGNUM(IOPHYSMEM) - 1].pp_link = &pages[next_free_pgnum]; // pages in [IOPHYSMEM, EXTPHYSMEM) are reserved for BIOS for (i = PGNUM(IOPHYSMEM);i < next_free_pgnum;++i) { pages[i].pp_ref = 1; pages[i].pp_link = NULL; pages[i].pp_reserved = 1; } }

接下來的page_alloc()與page_free()就先以最簡單的first-fit ＆ simple free list來處裡配置策略。如下：

// // Allocates a physical page. If (alloc_flags & ALLOC_ZERO), fills the entire // returned physical page with '\0' bytes. Does NOT increment the reference // count of the page - the caller must do these if necessary (either explicitly // or via page_insert). // // Returns NULL if out of free memory. // // Hint: use page2kva and memset struct Page * page_alloc(int alloc_flags) { // Fill this function in struct Page *p = NULL; if (!page_free_list) { return NULL; } p = page_free_list; page_free_list = p->pp_link; p->pp_link = NULL; if (alloc_flags & ALLOC_ZERO) { memset ((void *)KADDR(page2pa(p)), 0, PGSIZE); } return p; } // // Return a page to the free list. // (This function should only be called when pp->pp_ref reaches 0.) // void page_free(struct Page *pp) { // Fill this function in if (!pp) { return; } if (PGNUM(page2pa(pp)) >= npages) { return; } pp->pp_link = page_free_list; page_free_list = pp; }