diff options
Diffstat (limited to 'arch/um/kernel')
-rw-r--r-- | arch/um/kernel/exec.c | 5 | ||||
-rw-r--r-- | arch/um/kernel/exitcode.c | 31 | ||||
-rw-r--r-- | arch/um/kernel/gmon_syms.c | 11 | ||||
-rw-r--r-- | arch/um/kernel/gprof_syms.c | 13 | ||||
-rw-r--r-- | arch/um/kernel/initrd.c | 30 | ||||
-rw-r--r-- | arch/um/kernel/irq.c | 6 | ||||
-rw-r--r-- | arch/um/kernel/ksyms.c | 7 | ||||
-rw-r--r-- | arch/um/kernel/mem.c | 151 | ||||
-rw-r--r-- | arch/um/kernel/physmem.c | 16 | ||||
-rw-r--r-- | arch/um/kernel/process.c | 158 | ||||
-rw-r--r-- | arch/um/kernel/reboot.c | 7 | ||||
-rw-r--r-- | arch/um/kernel/sigio.c | 18 | ||||
-rw-r--r-- | arch/um/kernel/signal.c | 16 | ||||
-rw-r--r-- | arch/um/kernel/skas/clone.c | 32 | ||||
-rw-r--r-- | arch/um/kernel/skas/mmu.c | 132 | ||||
-rw-r--r-- | arch/um/kernel/skas/process.c | 20 | ||||
-rw-r--r-- | arch/um/kernel/skas/syscall.c | 6 | ||||
-rw-r--r-- | arch/um/kernel/skas/uaccess.c | 140 | ||||
-rw-r--r-- | arch/um/kernel/smp.c | 14 | ||||
-rw-r--r-- | arch/um/kernel/syscall.c | 3 | ||||
-rw-r--r-- | arch/um/kernel/sysrq.c | 44 | ||||
-rw-r--r-- | arch/um/kernel/time.c | 14 | ||||
-rw-r--r-- | arch/um/kernel/tlb.c | 51 | ||||
-rw-r--r-- | arch/um/kernel/trap.c | 33 | ||||
-rw-r--r-- | arch/um/kernel/uaccess.c | 11 | ||||
-rw-r--r-- | arch/um/kernel/um_arch.c | 112 | ||||
-rw-r--r-- | arch/um/kernel/umid.c | 15 |
27 files changed, 562 insertions, 534 deletions
diff --git a/arch/um/kernel/exec.c b/arch/um/kernel/exec.c index 8196450451c..f5d7f4569ba 100644 --- a/arch/um/kernel/exec.c +++ b/arch/um/kernel/exec.c @@ -19,12 +19,13 @@ void flush_thread(void) { void *data = NULL; - unsigned long end = proc_mm ? task_size : STUB_START; int ret; arch_flush_thread(¤t->thread.arch); - ret = unmap(¤t->mm->context.id, 0, end, 1, &data); + ret = unmap(¤t->mm->context.id, 0, STUB_START, 0, &data); + ret = ret || unmap(¤t->mm->context.id, STUB_END, + host_task_size - STUB_END, 1, &data); if (ret) { printk(KERN_ERR "flush_thread - clearing address space failed, " "err = %d\n", ret); diff --git a/arch/um/kernel/exitcode.c b/arch/um/kernel/exitcode.c index c716b5a6db1..984f80e668c 100644 --- a/arch/um/kernel/exitcode.c +++ b/arch/um/kernel/exitcode.c @@ -1,15 +1,17 @@ /* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "linux/kernel.h" -#include "linux/init.h" -#include "linux/ctype.h" -#include "linux/proc_fs.h" -#include "asm/uaccess.h" +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/proc_fs.h> +#include <linux/types.h> +#include <asm/uaccess.h> -/* If read and write race, the read will still atomically read a valid +/* + * If read and write race, the read will still atomically read a valid * value. */ int uml_exitcode = 0; @@ -19,18 +21,19 @@ static int read_proc_exitcode(char *page, char **start, off_t off, { int len, val; - /* Save uml_exitcode in a local so that we don't need to guarantee + /* + * Save uml_exitcode in a local so that we don't need to guarantee * that sprintf accesses it atomically. */ val = uml_exitcode; len = sprintf(page, "%d\n", val); len -= off; - if(len <= off+count) + if (len <= off+count) *eof = 1; *start = page + off; - if(len > count) + if (len > count) len = count; - if(len < 0) + if (len < 0) len = 0; return len; } @@ -41,11 +44,11 @@ static int write_proc_exitcode(struct file *file, const char __user *buffer, char *end, buf[sizeof("nnnnn\0")]; int tmp; - if(copy_from_user(buf, buffer, count)) + if (copy_from_user(buf, buffer, count)) return -EFAULT; tmp = simple_strtol(buf, &end, 0); - if((*end != '\0') && !isspace(*end)) + if ((*end != '\0') && !isspace(*end)) return -EINVAL; uml_exitcode = tmp; @@ -57,7 +60,7 @@ static int make_proc_exitcode(void) struct proc_dir_entry *ent; ent = create_proc_entry("exitcode", 0600, &proc_root); - if(ent == NULL){ + if (ent == NULL) { printk(KERN_WARNING "make_proc_exitcode : Failed to register " "/proc/exitcode\n"); return 0; diff --git a/arch/um/kernel/gmon_syms.c b/arch/um/kernel/gmon_syms.c index 734f873cab1..72eccd2a411 100644 --- a/arch/um/kernel/gmon_syms.c +++ b/arch/um/kernel/gmon_syms.c @@ -1,5 +1,5 @@ -/* - * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) +/* + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ @@ -8,12 +8,13 @@ extern void __bb_init_func(void *) __attribute__((weak)); EXPORT_SYMBOL(__bb_init_func); -/* This is defined (and referred to in profiling stub code) only by some GCC +/* + * This is defined (and referred to in profiling stub code) only by some GCC * versions in libgcov. * * Since SuSE backported the fix, we cannot handle it depending on GCC version. - * So, unconditionally export it. But also give it a weak declaration, which will - * be overridden by any other one. + * So, unconditionally export it. But also give it a weak declaration, which + * will be overridden by any other one. */ extern void __gcov_init(void *) __attribute__((weak)); diff --git a/arch/um/kernel/gprof_syms.c b/arch/um/kernel/gprof_syms.c index 9244f018d44..e2f043d0de6 100644 --- a/arch/um/kernel/gprof_syms.c +++ b/arch/um/kernel/gprof_syms.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ @@ -7,14 +7,3 @@ extern void mcount(void); EXPORT_SYMBOL(mcount); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/initrd.c b/arch/um/kernel/initrd.c index 16dc43e9d94..fa015565001 100644 --- a/arch/um/kernel/initrd.c +++ b/arch/um/kernel/initrd.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ @@ -7,7 +7,6 @@ #include "linux/bootmem.h" #include "linux/initrd.h" #include "asm/types.h" -#include "kern_util.h" #include "initrd.h" #include "init.h" #include "os.h" @@ -21,18 +20,27 @@ static int __init read_initrd(void) long long size; int err; - if(initrd == NULL) + if (initrd == NULL) return 0; err = os_file_size(initrd, &size); - if(err) + if (err) return 0; + /* + * This is necessary because alloc_bootmem craps out if you + * ask for no memory. + */ + if (size == 0) { + printk(KERN_ERR "\"%\" is a zero-size initrd\n"); + return 0; + } + area = alloc_bootmem(size); - if(area == NULL) + if (area == NULL) return 0; - if(load_initrd(initrd, area, size) == -1) + if (load_initrd(initrd, area, size) == -1) return 0; initrd_start = (unsigned long) area; @@ -59,13 +67,15 @@ int load_initrd(char *filename, void *buf, int size) int fd, n; fd = os_open_file(filename, of_read(OPENFLAGS()), 0); - if(fd < 0){ - printk("Opening '%s' failed - err = %d\n", filename, -fd); + if (fd < 0) { + printk(KERN_ERR "Opening '%s' failed - err = %d\n", filename, + -fd); return -1; } n = os_read_file(fd, buf, size); - if(n != size){ - printk("Read of %d bytes from '%s' failed, err = %d\n", size, + if (n != size) { + printk(KERN_ERR "Read of %d bytes from '%s' failed, " + "err = %d\n", size, filename, -n); return -1; } diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c index ba11ccd6a8a..91587f8db34 100644 --- a/arch/um/kernel/irq.c +++ b/arch/um/kernel/irq.c @@ -107,10 +107,9 @@ int activate_fd(int irq, int fd, int type, void *dev_id) struct pollfd *tmp_pfd; struct irq_fd *new_fd, *irq_fd; unsigned long flags; - int pid, events, err, n; + int events, err, n; - pid = os_getpid(); - err = os_set_fd_async(fd, pid); + err = os_set_fd_async(fd); if (err < 0) goto out; @@ -127,7 +126,6 @@ int activate_fd(int irq, int fd, int type, void *dev_id) .fd = fd, .type = type, .irq = irq, - .pid = pid, .events = events, .current_events = 0 } ); diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c index 7c7142ba3bd..66e2a305a8d 100644 --- a/arch/um/kernel/ksyms.c +++ b/arch/um/kernel/ksyms.c @@ -5,7 +5,6 @@ #include "linux/module.h" #include "linux/syscalls.h" -#include "asm/a.out.h" #include "asm/tlbflush.h" #include "asm/uaccess.h" #include "as-layout.h" @@ -18,15 +17,11 @@ EXPORT_SYMBOL(set_signals); EXPORT_SYMBOL(get_signals); EXPORT_SYMBOL(kernel_thread); EXPORT_SYMBOL(sys_waitpid); -EXPORT_SYMBOL(task_size); EXPORT_SYMBOL(flush_tlb_range); -EXPORT_SYMBOL(host_task_size); EXPORT_SYMBOL(arch_validate); -EXPORT_SYMBOL(get_kmem_end); EXPORT_SYMBOL(high_physmem); EXPORT_SYMBOL(empty_zero_page); -EXPORT_SYMBOL(um_virt_to_phys); EXPORT_SYMBOL(handle_page_fault); EXPORT_SYMBOL(find_iomem); @@ -40,7 +35,6 @@ EXPORT_SYMBOL(uml_strdup); EXPORT_SYMBOL(os_stat_fd); EXPORT_SYMBOL(os_stat_file); EXPORT_SYMBOL(os_access); -EXPORT_SYMBOL(os_get_exec_close); EXPORT_SYMBOL(os_set_exec_close); EXPORT_SYMBOL(os_getpid); EXPORT_SYMBOL(os_open_file); @@ -65,7 +59,6 @@ EXPORT_SYMBOL(os_accept_connection); EXPORT_SYMBOL(os_rcv_fd); EXPORT_SYMBOL(run_helper); EXPORT_SYMBOL(start_thread); -EXPORT_SYMBOL(dump_thread); #ifdef CONFIG_SMP diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index 59822dee438..2627ce82e91 100644 --- a/arch/um/kernel/mem.c +++ b/arch/um/kernel/mem.c @@ -1,49 +1,41 @@ /* - * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "linux/stddef.h" -#include "linux/kernel.h" -#include "linux/mm.h" -#include "linux/bootmem.h" -#include "linux/swap.h" -#include "linux/highmem.h" -#include "linux/gfp.h" -#include "asm/page.h" -#include "asm/fixmap.h" -#include "asm/pgalloc.h" -#include "kern_util.h" +#include <linux/stddef.h> +#include <linux/bootmem.h> +#include <linux/gfp.h> +#include <linux/highmem.h> +#include <linux/mm.h> +#include <linux/swap.h> +#include <asm/fixmap.h> +#include <asm/page.h> #include "as-layout.h" +#include "init.h" #include "kern.h" +#include "kern_util.h" #include "mem_user.h" -#include "um_uaccess.h" #include "os.h" -#include "linux/types.h" -#include "linux/string.h" -#include "init.h" -#include "kern_constants.h" /* allocated in paging_init, zeroed in mem_init, and unchanged thereafter */ unsigned long *empty_zero_page = NULL; /* allocated in paging_init and unchanged thereafter */ unsigned long *empty_bad_page = NULL; + +/* + * Initialized during boot, and readonly for initializing page tables + * afterwards + */ pgd_t swapper_pg_dir[PTRS_PER_PGD]; + +/* Initialized at boot time, and readonly after that */ unsigned long long highmem; int kmalloc_ok = 0; +/* Used during early boot */ static unsigned long brk_end; -void unmap_physmem(void) -{ - os_unmap_memory((void *) brk_end, uml_reserved - brk_end); -} - -static void map_cb(void *unused) -{ - map_memory(brk_end, __pa(brk_end), uml_reserved - brk_end, 1, 1, 0); -} - #ifdef CONFIG_HIGHMEM static void setup_highmem(unsigned long highmem_start, unsigned long highmem_len) @@ -53,7 +45,7 @@ static void setup_highmem(unsigned long highmem_start, int i; highmem_pfn = __pa(highmem_start) >> PAGE_SHIFT; - for(i = 0; i < highmem_len >> PAGE_SHIFT; i++){ + for (i = 0; i < highmem_len >> PAGE_SHIFT; i++) { page = &mem_map[highmem_pfn + i]; ClearPageReserved(page); init_page_count(page); @@ -65,14 +57,13 @@ static void setup_highmem(unsigned long highmem_start, void __init mem_init(void) { /* clear the zero-page */ - memset((void *) empty_zero_page, 0, PAGE_SIZE); + memset(empty_zero_page, 0, PAGE_SIZE); /* Map in the area just after the brk now that kmalloc is about * to be turned on. */ brk_end = (unsigned long) UML_ROUND_UP(sbrk(0)); - map_cb(NULL); - initial_thread_cb(map_cb, NULL); + map_memory(brk_end, __pa(brk_end), uml_reserved - brk_end, 1, 1, 0); free_bootmem(__pa(brk_end), uml_reserved - brk_end); uml_reserved = brk_end; @@ -85,7 +76,7 @@ void __init mem_init(void) #endif num_physpages = totalram_pages; max_pfn = totalram_pages; - printk(KERN_INFO "Memory: %luk available\n", + printk(KERN_INFO "Memory: %luk available\n", (unsigned long) nr_free_pages() << (PAGE_SHIFT-10)); kmalloc_ok = 1; @@ -119,7 +110,7 @@ static void __init one_md_table_init(pud_t *pud) #endif } -static void __init fixrange_init(unsigned long start, unsigned long end, +static void __init fixrange_init(unsigned long start, unsigned long end, pgd_t *pgd_base) { pgd_t *pgd; @@ -138,7 +129,7 @@ static void __init fixrange_init(unsigned long start, unsigned long end, if (pud_none(*pud)) one_md_table_init(pud); pmd = pmd_offset(pud, vaddr); - for (; (j < PTRS_PER_PMD) && (vaddr != end); pmd++, j++) { + for (; (j < PTRS_PER_PMD) && (vaddr < end); pmd++, j++) { one_page_table_init(pmd); vaddr += PMD_SIZE; } @@ -152,7 +143,7 @@ pgprot_t kmap_prot; #define kmap_get_fixmap_pte(vaddr) \ pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), (vaddr)),\ - (vaddr)), (vaddr)) + (vaddr)), (vaddr)) static void __init kmap_init(void) { @@ -197,21 +188,23 @@ static void __init fixaddr_user_init( void) pud_t *pud; pmd_t *pmd; pte_t *pte; - unsigned long paddr, vaddr = FIXADDR_USER_START; + phys_t p; + unsigned long v, vaddr = FIXADDR_USER_START; - if ( ! size ) + if (!size) return; fixrange_init( FIXADDR_USER_START, FIXADDR_USER_END, swapper_pg_dir); - paddr = (unsigned long)alloc_bootmem_low_pages( size); - memcpy( (void *)paddr, (void *)FIXADDR_USER_START, size); - paddr = __pa(paddr); - for ( ; size > 0; size-=PAGE_SIZE, vaddr+=PAGE_SIZE, paddr+=PAGE_SIZE){ + v = (unsigned long) alloc_bootmem_low_pages(size); + memcpy((void *) v , (void *) FIXADDR_USER_START, size); + p = __pa(v); + for ( ; size > 0; size -= PAGE_SIZE, vaddr += PAGE_SIZE, + p += PAGE_SIZE) { pgd = swapper_pg_dir + pgd_index(vaddr); pud = pud_offset(pgd, vaddr); pmd = pmd_offset(pud, vaddr); pte = pte_offset_kernel(pmd, vaddr); - pte_set_val( (*pte), paddr, PAGE_READONLY); + pte_set_val(*pte, p, PAGE_READONLY); } #endif } @@ -223,7 +216,7 @@ void __init paging_init(void) empty_zero_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE); empty_bad_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE); - for(i = 0; i < ARRAY_SIZE(zones_size); i++) + for (i = 0; i < ARRAY_SIZE(zones_size); i++) zones_size[i] = 0; zones_size[ZONE_NORMAL] = (end_iomem >> PAGE_SHIFT) - @@ -253,32 +246,33 @@ struct page *arch_validate(struct page *page, gfp_t mask, int order) int i; again: - if(page == NULL) + if (page == NULL) return page; - if(PageHighMem(page)) + if (PageHighMem(page)) return page; addr = (unsigned long) page_address(page); - for(i = 0; i < (1 << order); i++){ + for (i = 0; i < (1 << order); i++) { current->thread.fault_addr = (void *) addr; - if(__do_copy_to_user((void __user *) addr, &zero, + if (__do_copy_to_user((void __user *) addr, &zero, sizeof(zero), ¤t->thread.fault_addr, - ¤t->thread.fault_catcher)){ - if(!(mask & __GFP_WAIT)) + ¤t->thread.fault_catcher)) { + if (!(mask & __GFP_WAIT)) return NULL; else break; } addr += PAGE_SIZE; } - if(i == (1 << order)) + if (i == (1 << order)) return page; page = alloc_pages(mask, order); goto again; } -/* This can't do anything because nothing in the kernel image can be freed +/* + * This can't do anything because nothing in the kernel image can be freed * since it's not in kernel physical memory. */ @@ -290,8 +284,8 @@ void free_initmem(void) void free_initrd_mem(unsigned long start, unsigned long end) { if (start < end) - printk ("Freeing initrd memory: %ldk freed\n", - (end - start) >> 10); + printk(KERN_INFO "Freeing initrd memory: %ldk freed\n", + (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); init_page_count(virt_to_page(start)); @@ -308,32 +302,31 @@ void show_mem(void) int highmem = 0; struct page *page; - printk("Mem-info:\n"); + printk(KERN_INFO "Mem-info:\n"); show_free_areas(); - printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); + printk(KERN_INFO "Free swap: %6ldkB\n", + nr_swap_pages<<(PAGE_SHIFT-10)); pfn = max_mapnr; - while(pfn-- > 0) { + while (pfn-- > 0) { page = pfn_to_page(pfn); total++; - if(PageHighMem(page)) + if (PageHighMem(page)) highmem++; - if(PageReserved(page)) + if (PageReserved(page)) reserved++; - else if(PageSwapCache(page)) + else if (PageSwapCache(page)) cached++; - else if(page_count(page)) + else if (page_count(page)) shared += page_count(page) - 1; } - printk("%d pages of RAM\n", total); - printk("%d pages of HIGHMEM\n", highmem); - printk("%d reserved pages\n", reserved); - printk("%d pages shared\n", shared); - printk("%d pages swap cached\n", cached); + printk(KERN_INFO "%d pages of RAM\n", total); + printk(KERN_INFO "%d pages of HIGHMEM\n", highmem); + printk(KERN_INFO "%d reserved pages\n", reserved); + printk(KERN_INFO "%d pages shared\n", shared); + printk(KERN_INFO "%d pages swap cached\n", cached); } -/* - * Allocate and free page tables. - */ +/* Allocate and free page tables. */ pgd_t *pgd_alloc(struct mm_struct *mm) { @@ -341,14 +334,14 @@ pgd_t *pgd_alloc(struct mm_struct *mm) if (pgd) { memset(pgd, 0, USER_PTRS_PER_PGD * sizeof(pgd_t)); - memcpy(pgd + USER_PTRS_PER_PGD, - swapper_pg_dir + USER_PTRS_PER_PGD, + memcpy(pgd + USER_PTRS_PER_PGD, + swapper_pg_dir + USER_PTRS_PER_PGD, (PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t)); } return pgd; } -void pgd_free(pgd_t *pgd) +void pgd_free(struct mm_struct *mm, pgd_t *pgd) { free_page((unsigned long) pgd); } @@ -361,10 +354,24 @@ pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) return pte; } -struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address) +pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address) { struct page *pte; pte = alloc_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO); + if (pte) + pgtable_page_ctor(pte); return pte; } + +#ifdef CONFIG_3_LEVEL_PGTABLES +pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long address) +{ + pmd_t *pmd = (pmd_t *) __get_free_page(GFP_KERNEL); + + if (pmd) + memset(pmd, 0, PAGE_SIZE); + + return pmd; +} +#endif diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c index e66432f4248..9757085a022 100644 --- a/arch/um/kernel/physmem.c +++ b/arch/um/kernel/physmem.c @@ -55,16 +55,6 @@ int __init init_maps(unsigned long physmem, unsigned long iomem, return 0; } -/* Changed during early boot */ -static unsigned long kmem_top = 0; - -unsigned long get_kmem_end(void) -{ - if (kmem_top == 0) - kmem_top = host_task_size - 1024 * 1024; - return kmem_top; -} - void map_memory(unsigned long virt, unsigned long phys, unsigned long len, int r, int w, int x) { @@ -174,10 +164,10 @@ __uml_setup("iomem=", parse_iomem, * setup_iomem, both of which run during early boot. Afterwards, it's * unchanged. */ -struct iomem_region *iomem_regions = NULL; +struct iomem_region *iomem_regions; -/* Initialized in parse_iomem */ -int iomem_size = 0; +/* Initialized in parse_iomem and unchanged thereafter */ +int iomem_size; unsigned long find_iomem(char *driver, unsigned long *len_out) { diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index 0eae00b3e58..fc50d2f959d 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -4,19 +4,21 @@ * Licensed under the GPL */ -#include "linux/stddef.h" -#include "linux/err.h" -#include "linux/hardirq.h" -#include "linux/mm.h" -#include "linux/personality.h" -#include "linux/proc_fs.h" -#include "linux/ptrace.h" -#include "linux/random.h" -#include "linux/sched.h" -#include "linux/tick.h" -#include "linux/threads.h" -#include "asm/pgtable.h" -#include "asm/uaccess.h" +#include <linux/stddef.h> +#include <linux/err.h> +#include <linux/hardirq.h> +#include <linux/gfp.h> +#include <linux/mm.h> +#include <linux/personality.h> +#include <linux/proc_fs.h> +#include <linux/ptrace.h> +#include <linux/random.h> +#include <linux/sched.h> +#include <linux/tick.h> +#include <linux/threads.h> +#include <asm/current.h> +#include <asm/pgtable.h> +#include <asm/uaccess.h> #include "as-layout.h" #include "kern_util.h" #include "os.h" @@ -30,7 +32,7 @@ */ struct cpu_task cpu_tasks[NR_CPUS] = { [0 ... NR_CPUS - 1] = { -1, NULL } }; -static inline int external_pid(struct task_struct *task) +static inline int external_pid(void) { /* FIXME: Need to look up userspace_pid by cpu */ return userspace_pid[0]; @@ -40,7 +42,7 @@ int pid_to_processor_id(int pid) { int i; - for(i = 0; i < ncpus; i++) { + for (i = 0; i < ncpus; i++) { if (cpu_tasks[i].pid == pid) return i; } @@ -60,8 +62,6 @@ unsigned long alloc_stack(int order, int atomic) if (atomic) flags = GFP_ATOMIC; page = __get_free_pages(flags, order); - if (page == 0) - return 0; return page; } @@ -80,15 +80,15 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) static inline void set_current(struct task_struct *task) { cpu_tasks[task_thread_info(task)->cpu] = ((struct cpu_task) - { external_pid(task), task }); + { external_pid(), task }); } -extern void arch_switch_to(struct task_struct *from, struct task_struct *to); +extern void arch_switch_to(struct task_struct *to); void *_switch_to(void *prev, void *next, void *last) { struct task_struct *from = prev; - struct task_struct *to= next; + struct task_struct *to = next; to->thread.prev_sched = from; set_current(to); @@ -99,13 +99,13 @@ void *_switch_to(void *prev, void *next, void *last) switch_threads(&from->thread.switch_buf, &to->thread.switch_buf); - arch_switch_to(current->thread.prev_sched, current); + arch_switch_to(current); if (current->thread.saved_task) show_regs(&(current->thread.regs)); - next= current->thread.saved_task; - prev= current; - } while(current->thread.saved_task); + to = current->thread.saved_task; + from = current; + } while (current->thread.saved_task); return current->thread.prev_sched; @@ -163,8 +163,6 @@ void new_thread_handler(void) void fork_handler(void) { force_flush_all(); - if (current->thread.prev_sched == NULL) - panic("blech"); schedule_tail(current->thread.prev_sched); @@ -173,7 +171,7 @@ void fork_handler(void) * arch_switch_to isn't needed. We could want to apply this to * improve performance. -bb */ - arch_switch_to(current->thread.prev_sched, current); + arch_switch_to(current); current->thread.prev_sched = NULL; @@ -204,7 +202,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, arch_copy_thread(¤t->thread.arch, &p->thread.arch); } else { - init_thread_registers(&p->thread.regs.regs); + get_safe_registers(p->thread.regs.regs.gp); p->thread.request.u.thread = current->thread.request.u.thread; handler = new_thread_handler; } @@ -237,7 +235,7 @@ void default_idle(void) { unsigned long long nsecs; - while(1) { + while (1) { /* endless idle loop with no priority at all */ /* @@ -256,57 +254,10 @@ void default_idle(void) void cpu_idle(void) { - cpu_tasks[current_thread->cpu].pid = os_getpid(); + cpu_tasks[current_thread_info()->cpu].pid = os_getpid(); default_idle(); } -void *um_virt_to_phys(struct task_struct *task, unsigned long addr, - pte_t *pte_out) -{ - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; - pte_t *pte; - pte_t ptent; - - if (task->mm == NULL) - return ERR_PTR(-EINVAL); - pgd = pgd_offset(task->mm, addr); - if (!pgd_present(*pgd)) - return ERR_PTR(-EINVAL); - - pud = pud_offset(pgd, addr); - if (!pud_present(*pud)) - return ERR_PTR(-EINVAL); - - pmd = pmd_offset(pud, addr); - if (!pmd_present(*pmd)) - return ERR_PTR(-EINVAL); - - pte = pte_offset_kernel(pmd, addr); - ptent = *pte; - if (!pte_present(ptent)) - return ERR_PTR(-EINVAL); - - if (pte_out != NULL) - *pte_out = ptent; - return (void *) (pte_val(ptent) & PAGE_MASK) + (addr & ~PAGE_MASK); -} - -char *current_cmd(void) -{ -#if defined(CONFIG_SMP) || defined(CONFIG_HIGHMEM) - return "(Unknown)"; -#else - void *addr = um_virt_to_phys(current, current->mm->arg_start, NULL); - return IS_ERR(addr) ? "(Unknown)": __va((unsigned long) addr); -#endif -} - -void dump_thread(struct pt_regs *regs, struct user *u) -{ -} - int __cant_sleep(void) { return in_atomic() || irqs_disabled() || in_interrupt(); /* Is in_interrupt() really needed? */ @@ -317,7 +268,7 @@ int user_context(unsigned long sp) unsigned long stack; stack = sp & (PAGE_MASK << CONFIG_KERNEL_STACK_ORDER); - return stack != (unsigned long) current_thread; + return stack != (unsigned long) current_thread_info(); } extern exitcall_t __uml_exitcall_begin, __uml_exitcall_end; @@ -331,7 +282,7 @@ void do_uml_exitcalls(void) (*call)(); } -char *uml_strdup(char *string) +char *uml_strdup(const char *string) { return kstrdup(string, GFP_KERNEL); } @@ -359,7 +310,7 @@ int strlen_user_proc(char __user *str) int smp_sigio_handler(void) { #ifdef CONFIG_SMP - int cpu = current_thread->cpu; + int cpu = current_thread_info()->cpu; IPI_handler(cpu); if (cpu != 0) return 1; @@ -369,7 +320,7 @@ int smp_sigio_handler(void) int cpu(void) { - return current_thread->cpu; + return current_thread_info()->cpu; } static atomic_t using_sysemu = ATOMIC_INIT(0); @@ -435,7 +386,7 @@ int singlestepping(void * t) { struct task_struct *task = t ? t : current; - if ( ! (task->ptrace & PT_DTRACE) ) + if (!(task->ptrace & PT_DTRACE)) return 0; if (task->thread.singlestep_syscall) @@ -459,3 +410,46 @@ unsigned long arch_align_stack(unsigned long sp) return sp & ~0xf; } #endif + +unsigned long get_wchan(struct task_struct *p) +{ + unsigned long stack_page, sp, ip; + bool seen_sched = 0; + + if ((p == NULL) || (p == current) || (p->state == TASK_RUNNING)) + return 0; + + stack_page = (unsigned long) task_stack_page(p); + /* Bail if the process has no kernel stack for some reason */ + if (stack_page == 0) + return 0; + + sp = p->thread.switch_buf->JB_SP; + /* + * Bail if the stack pointer is below the bottom of the kernel + * stack for some reason + */ + if (sp < stack_page) + return 0; + + while (sp < stack_page + THREAD_SIZE) { + ip = *((unsigned long *) sp); + if (in_sched_functions(ip)) + /* Ignore everything until we're above the scheduler */ + seen_sched = 1; + else if (kernel_text_address(ip) && seen_sched) + return ip; + + sp += sizeof(unsigned long); + } + + return 0; +} + +int elf_core_copy_fpregs(struct task_struct *t, elf_fpregset_t *fpu) +{ + int cpu = current_thread_info()->cpu; + + return save_fp_registers(userspace_pid[cpu], (unsigned long *) fpu); +} + diff --git a/arch/um/kernel/reboot.c b/arch/um/kernel/reboot.c index 04cebcf0679..00197d3d21e 100644 --- a/arch/um/kernel/reboot.c +++ b/arch/um/kernel/reboot.c @@ -4,6 +4,7 @@ */ #include "linux/sched.h" +#include "kern_util.h" #include "os.h" #include "skas.h" @@ -11,7 +12,7 @@ void (*pm_power_off)(void); static void kill_off_processes(void) { - if(proc_mm) + if (proc_mm) /* * FIXME: need to loop over userspace_pids */ @@ -21,8 +22,8 @@ static void kill_off_processes(void) int pid, me; me = os_getpid(); - for_each_process(p){ - if(p->mm == NULL) + for_each_process(p) { + if (p->mm == NULL) continue; pid = p->mm->context.id.u.pid; diff --git a/arch/um/kernel/sigio.c b/arch/um/kernel/sigio.c index 89f9866a135..2b272b63b51 100644 --- a/arch/um/kernel/sigio.c +++ b/arch/um/kernel/sigio.c @@ -1,18 +1,12 @@ /* - * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com) * Licensed under the GPL */ -#include "linux/kernel.h" -#include "linux/list.h" -#include "linux/slab.h" -#include "linux/signal.h" -#include "linux/interrupt.h" -#include "init.h" -#include "sigio.h" -#include "irq_user.h" +#include <linux/interrupt.h> #include "irq_kern.h" #include "os.h" +#include "sigio.h" /* Protected by sigio_lock() called from write_sigio_workaround */ static int sigio_irq_fd = -1; @@ -33,9 +27,9 @@ int write_sigio_irq(int fd) err = um_request_irq(SIGIO_WRITE_IRQ, fd, IRQ_READ, sigio_interrupt, IRQF_DISABLED|IRQF_SAMPLE_RANDOM, "write sigio", NULL); - if(err){ - printk("write_sigio_irq : um_request_irq failed, err = %d\n", - err); + if (err) { + printk(KERN_ERR "write_sigio_irq : um_request_irq failed, " + "err = %d\n", err); return -1; } sigio_irq_fd = fd; diff --git a/arch/um/kernel/signal.c b/arch/um/kernel/signal.c index 19cb9773393..b0fce720c4d 100644 --- a/arch/um/kernel/signal.c +++ b/arch/um/kernel/signal.c @@ -3,12 +3,12 @@ * Licensed under the GPL */ -#include "linux/module.h" -#include "linux/ptrace.h" -#include "linux/sched.h" -#include "asm/siginfo.h" -#include "asm/signal.h" -#include "asm/unistd.h" +#include <linux/module.h> +#include <linux/ptrace.h> +#include <linux/sched.h> +#include <asm/siginfo.h> +#include <asm/signal.h> +#include <asm/unistd.h> #include "frame_kern.h" #include "kern_util.h" #include "sigcontext.h" @@ -36,7 +36,7 @@ static int handle_signal(struct pt_regs *regs, unsigned long signr, /* Did we come from a system call? */ if (PT_REGS_SYSCALL_NR(regs) >= 0) { /* If so, check system call restarting.. */ - switch(PT_REGS_SYSCALL_RET(regs)) { + switch (PT_REGS_SYSCALL_RET(regs)) { case -ERESTART_RESTARTBLOCK: case -ERESTARTNOHAND: PT_REGS_SYSCALL_RET(regs) = -EINTR; @@ -116,7 +116,7 @@ static int kern_do_signal(struct pt_regs *regs) /* Did we come from a system call? */ if (!handled_sig && (PT_REGS_SYSCALL_NR(regs) >= 0)) { /* Restart the system call - no handlers present */ - switch(PT_REGS_SYSCALL_RET(regs)) { + switch (PT_REGS_SYSCALL_RET(regs)) { case -ERESTARTNOHAND: case -ERESTARTSYS: case -ERESTARTNOINTR: diff --git a/arch/um/kernel/skas/clone.c b/arch/um/kernel/skas/clone.c index 8d07a7acb90..2c8583c1a34 100644 --- a/arch/um/kernel/skas/clone.c +++ b/arch/um/kernel/skas/clone.c @@ -1,17 +1,20 @@ -#include <sched.h> +/* + * Copyright (C) 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) + * Licensed under the GPL + */ + #include <signal.h> -#include <sys/mman.h> -#include <sys/time.h> +#include <sched.h> #include <asm/unistd.h> +#include <sys/time.h> #include "as-layout.h" +#include "kern_constants.h" #include "ptrace_user.h" -#include "skas.h" #include "stub-data.h" -#include "uml-config.h" #include "sysdep/stub.h" -#include "kern_constants.h" -/* This is in a separate file because it needs to be compiled with any +/* + * This is in a separate file because it needs to be compiled with any * extraneous gcc flags (-pg, -fprofile-arcs, -ftest-coverage) disabled * * Use UM_KERN_PAGE_SIZE instead of PAGE_SIZE because that calls getpagesize @@ -26,25 +29,26 @@ stub_clone_handler(void) err = stub_syscall2(__NR_clone, CLONE_PARENT | CLONE_FILES | SIGCHLD, STUB_DATA + UM_KERN_PAGE_SIZE / 2 - sizeof(void *)); - if(err != 0) + if (err != 0) goto out; err = stub_syscall4(__NR_ptrace, PTRACE_TRACEME, 0, 0, 0); - if(err) + if (err) goto out; - err = stub_syscall3(__NR_setitimer, ITIMER_VIRTUAL, + err = stub_syscall3(__NR_setitimer, ITIMER_VIRTUAL, (long) &data->timer, 0); - if(err) + if (err) goto out; remap_stack(data->fd, data->offset); goto done; out: - /* save current result. - * Parent: pid; - * child: retcode of mmap already saved and it jumps around this + /* + * save current result. + * Parent: pid; + * child: retcode of mmap already saved and it jumps around this * assignment */ data->err = err; diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c index f859ec306cd..0cd9a7a05e7 100644 --- a/arch/um/kernel/skas/mmu.c +++ b/arch/um/kernel/skas/mmu.c @@ -34,33 +34,14 @@ static int init_stub_pte(struct mm_struct *mm, unsigned long proc, if (!pte) goto out_pte; - /* - * There's an interaction between the skas0 stub pages, stack - * randomization, and the BUG at the end of exit_mmap. exit_mmap - * checks that the number of page tables freed is the same as had - * been allocated. If the stack is on the last page table page, - * then the stack pte page will be freed, and if not, it won't. To - * avoid having to know where the stack is, or if the process mapped - * something at the top of its address space for some other reason, - * we set TASK_SIZE to end at the start of the last page table. - * This keeps exit_mmap off the last page, but introduces a leak - * of that page. So, we hang onto it here and free it in - * destroy_context_skas. - */ - - mm->context.last_page_table = pmd_page_vaddr(*pmd); -#ifdef CONFIG_3_LEVEL_PGTABLES - mm->context.last_pmd = (unsigned long) __va(pud_val(*pud)); -#endif - *pte = mk_pte(virt_to_page(kernel), __pgprot(_PAGE_PRESENT)); *pte = pte_mkread(*pte); return 0; out_pmd: - pud_free(pud); + pud_free(mm, pud); out_pte: - pmd_free(pmd); + pmd_free(mm, pmd); out: return -ENOMEM; } @@ -76,24 +57,6 @@ int init_new_context(struct task_struct *task, struct mm_struct *mm) stack = get_zeroed_page(GFP_KERNEL); if (stack == 0) goto out; - - /* - * This zeros the entry that pgd_alloc didn't, needed since - * we are about to reinitialize it, and want mm.nr_ptes to - * be accurate. - */ - mm->pgd[USER_PTRS_PER_PGD] = __pgd(0); - - ret = init_stub_pte(mm, STUB_CODE, - (unsigned long) &__syscall_stub_start); - if (ret) - goto out_free; - - ret = init_stub_pte(mm, STUB_DATA, stack); - if (ret) - goto out_free; - - mm->nr_ptes--; } to_mm->id.stack = stack; @@ -114,6 +77,11 @@ int init_new_context(struct task_struct *task, struct mm_struct *mm) to_mm->id.u.pid = copy_context_skas0(stack, from_mm->id.u.pid); else to_mm->id.u.pid = start_userspace(stack); + + if (to_mm->id.u.pid < 0) { + ret = to_mm->id.u.pid; + goto out_free; + } } ret = init_new_ldt(to_mm, from_mm); @@ -123,6 +91,8 @@ int init_new_context(struct task_struct *task, struct mm_struct *mm) goto out_free; } + to_mm->stub_pages = NULL; + return 0; out_free: @@ -132,24 +102,90 @@ int init_new_context(struct task_struct *task, struct mm_struct *mm) return ret; } +void arch_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm) +{ + struct page **pages; + int err, ret; + + if (!skas_needs_stub) + return; + + ret = init_stub_pte(mm, STUB_CODE, + (unsigned long) &__syscall_stub_start); + if (ret) + goto out; + + ret = init_stub_pte(mm, STUB_DATA, mm->context.id.stack); + if (ret) + goto out; + + pages = kmalloc(2 * sizeof(struct page *), GFP_KERNEL); + if (pages == NULL) { + printk(KERN_ERR "arch_dup_mmap failed to allocate 2 page " + "pointers\n"); + goto out; + } + + pages[0] = virt_to_page(&__syscall_stub_start); + pages[1] = virt_to_page(mm->context.id.stack); + mm->context.stub_pages = pages; + + /* dup_mmap already holds mmap_sem */ + err = install_special_mapping(mm, STUB_START, STUB_END - STUB_START, + VM_READ | VM_MAYREAD | VM_EXEC | + VM_MAYEXEC | VM_DONTCOPY, pages); + if (err) { + printk(KERN_ERR "install_special_mapping returned %d\n", err); + goto out_free; + } + return; + +out_free: + kfree(pages); +out: + force_sigsegv(SIGSEGV, current); +} + +void arch_exit_mmap(struct mm_struct *mm) +{ + pte_t *pte; + + if (mm->context.stub_pages != NULL) + kfree(mm->context.stub_pages); + pte = virt_to_pte(mm, STUB_CODE); + if (pte != NULL) + pte_clear(mm, STUB_CODE, pte); + + pte = virt_to_pte(mm, STUB_DATA); + if (pte == NULL) + return; + + pte_clear(mm, STUB_DATA, pte); +} + void destroy_context(struct mm_struct *mm) { struct mm_context *mmu = &mm->context; if (proc_mm) os_close_file(mmu->id.u.mm_fd); - else + else { + /* + * If init_new_context wasn't called, this will be + * zero, resulting in a kill(0), which will result in the + * whole UML suddenly dying. Also, cover negative and + * 1 cases, since they shouldn't happen either. + */ + if (mmu->id.u.pid < 2) { + printk(KERN_ERR "corrupt mm_context - pid = %d\n", + mmu->id.u.pid); + return; + } os_kill_ptraced_process(mmu->id.u.pid, 1); + } - if (!proc_mm || !ptrace_faultinfo) { + if (skas_needs_stub) free_page(mmu->id.stack); - pte_lock_deinit(virt_to_page(mmu->last_page_table)); - pte_free_kernel((pte_t *) mmu->last_page_table); - dec_zone_page_state(virt_to_page(mmu->last_page_table), NR_PAGETABLE); -#ifdef CONFIG_3_LEVEL_PGTABLES - pmd_free((pmd_t *) mmu->last_pmd); -#endif - } free_ldt(mmu); } diff --git a/arch/um/kernel/skas/process.c b/arch/um/kernel/skas/process.c index fce389c2342..2e9852c0d48 100644 --- a/arch/um/kernel/skas/process.c +++ b/arch/um/kernel/skas/process.c @@ -6,19 +6,25 @@ #include "linux/init.h" #include "linux/sched.h" #include "as-layout.h" +#include "kern.h" #include "os.h" #include "skas.h" int new_mm(unsigned long stack) { - int fd; + int fd, err; fd = os_open_file("/proc/mm", of_cloexec(of_write(OPENFLAGS())), 0); if (fd < 0) return fd; - if (skas_needs_stub) - map_stub_pages(fd, STUB_CODE, STUB_DATA, stack); + if (skas_needs_stub) { + err = map_stub_pages(fd, STUB_CODE, STUB_DATA, stack); + if (err) { + os_close_file(fd); + return err; + } + } return fd; } @@ -49,8 +55,14 @@ int __init start_uml(void) { stack_protections((unsigned long) &cpu0_irqstack); set_sigstack(cpu0_irqstack, THREAD_SIZE); - if (proc_mm) + if (proc_mm) { userspace_pid[0] = start_userspace(0); + if (userspace_pid[0] < 0) { + printf("start_uml - start_userspace returned %d\n", + userspace_pid[0]); + exit(1); + } + } init_new_thread_signals(); diff --git a/arch/um/kernel/skas/syscall.c b/arch/um/kernel/skas/syscall.c index 50b476f2b38..4e3b820bd2b 100644 --- a/arch/um/kernel/skas/syscall.c +++ b/arch/um/kernel/skas/syscall.c @@ -9,6 +9,9 @@ #include "sysdep/ptrace.h" #include "sysdep/syscalls.h" +extern int syscall_table_size; +#define NR_syscalls (syscall_table_size / sizeof(void *)) + void handle_syscall(struct uml_pt_regs *r) { struct pt_regs *regs = container_of(r, struct pt_regs, regs); @@ -17,9 +20,6 @@ void handle_syscall(struct uml_pt_regs *r) syscall_trace(r, 0); - current->thread.nsyscalls++; - nsyscalls++; - /* * This should go in the declaration of syscall, but when I do that, * strace -f -c bash -c 'ls ; ls' breaks, sometimes not tracing diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c index 1d8b119f2d0..e22c96993db 100644 --- a/arch/um/kernel/skas/uaccess.c +++ b/arch/um/kernel/skas/uaccess.c @@ -3,128 +3,130 @@ * Licensed under the GPL */ -#include "linux/err.h" -#include "linux/highmem.h" -#include "linux/mm.h" -#include "asm/current.h" -#include "asm/page.h" -#include "asm/pgtable.h" +#include <linux/err.h> +#include <linux/highmem.h> +#include <linux/mm.h> +#include <linux/sched.h> +#include <asm/current.h> +#include <asm/page.h> +#include <asm/pgtable.h> #include "kern_util.h" #include "os.h" -extern void *um_virt_to_phys(struct task_struct *task, unsigned long addr, - pte_t *pte_out); - -static unsigned long maybe_map(unsigned long virt, int is_write) +pte_t *virt_to_pte(struct mm_struct *mm, unsigned long addr) { - pte_t pte; - int err; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + + if (mm == NULL) + return NULL; + + pgd = pgd_offset(mm, addr); + if (!pgd_present(*pgd)) + return NULL; + + pud = pud_offset(pgd, addr); + if (!pud_present(*pud)) + return NULL; - void *phys = um_virt_to_phys(current, virt, &pte); - int dummy_code; + pmd = pmd_offset(pud, addr); + if (!pmd_present(*pmd)) + return NULL; + + return pte_offset_kernel(pmd, addr); +} + +static pte_t *maybe_map(unsigned long virt, int is_write) +{ + pte_t *pte = virt_to_pte(current->mm, virt); + int err, dummy_code; - if (IS_ERR(phys) || (is_write && !pte_write(pte))) { + if ((pte == NULL) || !pte_present(*pte) || + (is_write && !pte_write(*pte))) { err = handle_page_fault(virt, 0, is_write, 1, &dummy_code); if (err) - return -1UL; - phys = um_virt_to_phys(current, virt, NULL); + return NULL; + pte = virt_to_pte(current->mm, virt); } - if (IS_ERR(phys)) - phys = (void *) -1; + if (!pte_present(*pte)) + pte = NULL; - return (unsigned long) phys; + return pte; } static int do_op_one_page(unsigned long addr, int len, int is_write, int (*op)(unsigned long addr, int len, void *arg), void *arg) { + jmp_buf buf; struct page *page; - int n; + pte_t *pte; + int n, faulted; - addr = maybe_map(addr, is_write); - if (addr == -1UL) + pte = maybe_map(addr, is_write); + if (pte == NULL) return -1; - page = phys_to_page(addr); + page = pte_page(*pte); addr = (unsigned long) kmap_atomic(page, KM_UML_USERCOPY) + (addr & ~PAGE_MASK); - n = (*op)(addr, len, arg); + current->thread.fault_catcher = &buf; + + faulted = UML_SETJMP(&buf); + if (faulted == 0) + n = (*op)(addr, len, arg); + else + n = -1; + + current->thread.fault_catcher = NULL; kunmap_atomic(page, KM_UML_USERCOPY); return n; } -static void do_buffer_op(void *jmpbuf, void *arg_ptr) +static int buffer_op(unsigned long addr, int len, int is_write, + int (*op)(unsigned long, int, void *), void *arg) { - va_list args; - unsigned long addr; - int len, is_write, size, remain, n; - int (*op)(unsigned long, int, void *); - void *arg; - int *res; - - va_copy(args, *(va_list *)arg_ptr); - addr = va_arg(args, unsigned long); - len = va_arg(args, int); - is_write = va_arg(args, int); - op = va_arg(args, void *); - arg = va_arg(args, void *); - res = va_arg(args, int *); - va_end(args); + int size, remain, n; + size = min(PAGE_ALIGN(addr) - addr, (unsigned long) len); remain = len; - current->thread.fault_catcher = jmpbuf; n = do_op_one_page(addr, size, is_write, op, arg); if (n != 0) { - *res = (n < 0 ? remain : 0); + remain = (n < 0 ? remain : 0); goto out; } addr += size; remain -= size; - if (remain == 0) { - *res = 0; + if (remain == 0) goto out; - } - while(addr < ((addr + remain) & PAGE_MASK)) { + while (addr < ((addr + remain) & PAGE_MASK)) { n = do_op_one_page(addr, PAGE_SIZE, is_write, op, arg); if (n != 0) { - *res = (n < 0 ? remain : 0); + remain = (n < 0 ? remain : 0); goto out; } addr += PAGE_SIZE; remain -= PAGE_SIZE; } - if (remain == 0) { - *res = 0; + if (remain == 0) goto out; - } n = do_op_one_page(addr, remain, is_write, op, arg); - if (n != 0) - *res = (n < 0 ? remain : 0); - else *res = 0; - out: - current->thread.fault_catcher = NULL; -} - -static int buffer_op(unsigned long addr, int len, int is_write, - int (*op)(unsigned long addr, int len, void *arg), - void *arg) -{ - int faulted, res; - - faulted = setjmp_wrapper(do_buffer_op, addr, len, is_write, op, arg, - &res); - if (!faulted) - return res; + if (n != 0) { + remain = (n < 0 ? remain : 0); + goto out; + } - return addr + len - (unsigned long) current->thread.fault_addr; + return 0; + out: + return remain; } static int copy_chunk_from_user(unsigned long from, int len, void *arg) diff --git a/arch/um/kernel/smp.c b/arch/um/kernel/smp.c index 36d89cf8d20..e1062ec36d4 100644 --- a/arch/um/kernel/smp.c +++ b/arch/um/kernel/smp.c @@ -21,7 +21,6 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); #include "asm/smp.h" #include "asm/processor.h" #include "asm/spinlock.h" -#include "kern_util.h" #include "kern.h" #include "irq_user.h" #include "os.h" @@ -61,7 +60,7 @@ void smp_send_stop(void) continue; os_write_file(cpu_data[i].ipi_pipe[1], "S", 1); } - printk(KERN_INFO "done\n"); + printk(KERN_CONT "done\n"); } static cpumask_t smp_commenced_mask = CPU_MASK_NONE; @@ -75,8 +74,7 @@ static int idle_proc(void *cpup) if (err < 0) panic("CPU#%d failed to create IPI pipe, err = %d", cpu, -err); - os_set_fd_async(cpu_data[cpu].ipi_pipe[0], - current->thread.mode.tt.extern_pid); + os_set_fd_async(cpu_data[cpu].ipi_pipe[0]); wmb(); if (cpu_test_and_set(cpu, cpu_callin_map)) { @@ -129,8 +127,7 @@ void smp_prepare_cpus(unsigned int maxcpus) if (err < 0) panic("CPU#0 failed to create IPI pipe, errno = %d", -err); - os_set_fd_async(cpu_data[me].ipi_pipe[0], - current->thread.mode.tt.extern_pid); + os_set_fd_async(cpu_data[me].ipi_pipe[0]); for (cpu = 1; cpu < ncpus; cpu++) { printk(KERN_INFO "Booting processor %d...\n", cpu); @@ -143,9 +140,8 @@ void smp_prepare_cpus(unsigned int maxcpus) while (waittime-- && !cpu_isset(cpu, cpu_callin_map)) cpu_relax(); - if (cpu_isset(cpu, cpu_callin_map)) - printk(KERN_INFO "done\n"); - else printk(KERN_INFO "failed\n"); + printk(KERN_INFO "%s\n", + cpu_isset(cpu, cpu_calling_map) ? "done" : "failed"); } } diff --git a/arch/um/kernel/syscall.c b/arch/um/kernel/syscall.c index b9d92b2089a..9cffc628a37 100644 --- a/arch/um/kernel/syscall.c +++ b/arch/um/kernel/syscall.c @@ -13,9 +13,6 @@ #include "asm/uaccess.h" #include "asm/unistd.h" -/* Unlocked, I don't care if this is a bit off */ -int nsyscalls = 0; - long sys_fork(void) { long ret; diff --git a/arch/um/kernel/sysrq.c b/arch/um/kernel/sysrq.c index 93263571d81..56d43d0a396 100644 --- a/arch/um/kernel/sysrq.c +++ b/arch/um/kernel/sysrq.c @@ -1,38 +1,37 @@ -/* - * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) +/* + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "linux/sched.h" -#include "linux/kernel.h" -#include "linux/module.h" -#include "linux/kallsyms.h" -#include "asm/page.h" -#include "asm/processor.h" +#include <linux/kallsyms.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/sched.h> #include "sysrq.h" /* Catch non-i386 SUBARCH's. */ #if !defined(CONFIG_UML_X86) || defined(CONFIG_64BIT) void show_trace(struct task_struct *task, unsigned long * stack) { - unsigned long addr; + unsigned long addr; - if (!stack) { + if (!stack) { stack = (unsigned long*) &stack; WARN_ON(1); } - printk("Call Trace: \n"); - while (((long) stack & (THREAD_SIZE-1)) != 0) { - addr = *stack; + printk(KERN_INFO "Call Trace: \n"); + while (((long) stack & (THREAD_SIZE-1)) != 0) { + addr = *stack; if (__kernel_text_address(addr)) { - printk("%08lx: [<%08lx>]", (unsigned long) stack, addr); - print_symbol(" %s", addr); - printk("\n"); - } - stack++; - } - printk("\n"); + printk(KERN_INFO "%08lx: [<%08lx>]", + (unsigned long) stack, addr); + print_symbol(KERN_CONT " %s", addr); + printk(KERN_CONT "\n"); + } + stack++; + } + printk(KERN_INFO "\n"); } #endif @@ -67,14 +66,13 @@ void show_stack(struct task_struct *task, unsigned long *esp) } stack = esp; - for(i = 0; i < kstack_depth_to_print; i++) { + for (i = 0; i < kstack_depth_to_print; i++) { if (kstack_end(stack)) break; if (i && ((i % 8) == 0)) - printk("\n "); + printk("\n" KERN_INFO " "); printk("%08lx ", *stack++); } - printk("Call Trace: \n"); show_trace(task, esp); } diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c index 1ac746a9eae..e066e84493b 100644 --- a/arch/um/kernel/time.c +++ b/arch/um/kernel/time.c @@ -3,12 +3,12 @@ * Licensed under the GPL */ -#include "linux/clockchips.h" -#include "linux/interrupt.h" -#include "linux/jiffies.h" -#include "linux/threads.h" -#include "asm/irq.h" -#include "asm/param.h" +#include <linux/clockchips.h> +#include <linux/interrupt.h> +#include <linux/jiffies.h> +#include <linux/threads.h> +#include <asm/irq.h> +#include <asm/param.h> #include "kern_util.h" #include "os.h" @@ -32,7 +32,7 @@ void timer_handler(int sig, struct uml_pt_regs *regs) static void itimer_set_mode(enum clock_event_mode mode, struct clock_event_device *evt) { - switch(mode) { + switch (mode) { case CLOCK_EVT_MODE_PERIODIC: set_interval(); break; diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c index f4a0e407eee..d175d0566af 100644 --- a/arch/um/kernel/tlb.c +++ b/arch/um/kernel/tlb.c @@ -3,9 +3,10 @@ * Licensed under the GPL */ -#include "linux/mm.h" -#include "asm/pgtable.h" -#include "asm/tlbflush.h" +#include <linux/mm.h> +#include <linux/sched.h> +#include <asm/pgtable.h> +#include <asm/tlbflush.h> #include "as-layout.h" #include "mem_user.h" #include "os.h" @@ -56,7 +57,7 @@ static int do_ops(struct host_vm_change *hvc, int end, for (i = 0; i < end && !ret; i++) { op = &hvc->ops[i]; - switch(op->type) { + switch (op->type) { case MMAP: ret = map(hvc->id, op->u.mmap.addr, op->u.mmap.len, op->u.mmap.prot, op->u.mmap.fd, @@ -183,27 +184,30 @@ static inline int update_pte_range(pmd_t *pmd, unsigned long addr, pte = pte_offset_kernel(pmd, addr); do { + if ((addr >= STUB_START) && (addr < STUB_END)) + continue; + r = pte_read(*pte); w = pte_write(*pte); x = pte_exec(*pte); if (!pte_young(*pte)) { r = 0; w = 0; - } else if (!pte_dirty(*pte)) { + } else if (!pte_dirty(*pte)) w = 0; - } + prot = ((r ? UM_PROT_READ : 0) | (w ? UM_PROT_WRITE : 0) | (x ? UM_PROT_EXEC : 0)); if (hvc->force || pte_newpage(*pte)) { if (pte_present(*pte)) ret = add_mmap(addr, pte_val(*pte) & PAGE_MASK, PAGE_SIZE, prot, hvc); - else ret = add_munmap(addr, PAGE_SIZE, hvc); - } - else if (pte_newprot(*pte)) + else + ret = add_munmap(addr, PAGE_SIZE, hvc); + } else if (pte_newprot(*pte)) ret = add_mprotect(addr, PAGE_SIZE, prot, hvc); *pte = pte_mkuptodate(*pte); - } while (pte++, addr += PAGE_SIZE, ((addr != end) && !ret)); + } while (pte++, addr += PAGE_SIZE, ((addr < end) && !ret)); return ret; } @@ -225,7 +229,7 @@ static inline int update_pmd_range(pud_t *pud, unsigned long addr, } } else ret = update_pte_range(pmd, addr, next, hvc); - } while (pmd++, addr = next, ((addr != end) && !ret)); + } while (pmd++, addr = next, ((addr < end) && !ret)); return ret; } @@ -247,7 +251,7 @@ static inline int update_pud_range(pgd_t *pgd, unsigned long addr, } } else ret = update_pmd_range(pud, addr, next, hvc); - } while (pud++, addr = next, ((addr != end) && !ret)); + } while (pud++, addr = next, ((addr < end) && !ret)); return ret; } @@ -270,7 +274,7 @@ void fix_range_common(struct mm_struct *mm, unsigned long start_addr, } } else ret = update_pud_range(pgd, addr, next, &hvc); - } while (pgd++, addr = next, ((addr != end_addr) && !ret)); + } while (pgd++, addr = next, ((addr < end_addr) && !ret)); if (!ret) ret = do_ops(&hvc, hvc.index, 1); @@ -485,9 +489,6 @@ void __flush_tlb_one(unsigned long addr) static void fix_range(struct mm_struct *mm, unsigned long start_addr, unsigned long end_addr, int force) { - if (!proc_mm && (end_addr > STUB_START)) - end_addr = STUB_START; - fix_range_common(mm, start_addr, end_addr, force); } @@ -499,10 +500,9 @@ void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, else fix_range(vma->vm_mm, start, end, 0); } -void flush_tlb_mm(struct mm_struct *mm) +void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, + unsigned long end) { - unsigned long end; - /* * Don't bother flushing if this address space is about to be * destroyed. @@ -510,8 +510,17 @@ void flush_tlb_mm(struct mm_struct *mm) if (atomic_read(&mm->mm_users) == 0) return; - end = proc_mm ? task_size : STUB_START; - fix_range(mm, 0, end, 0); + fix_range(mm, start, end, 0); +} + +void flush_tlb_mm(struct mm_struct *mm) +{ + struct vm_area_struct *vma = mm->mmap; + + while (vma != NULL) { + fix_range(mm, vma->vm_start, vma->vm_end, 0); + vma = vma->vm_next; + } } void force_flush_all(void) diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c index cb3321f8e0a..44e49041949 100644 --- a/arch/um/kernel/trap.c +++ b/arch/um/kernel/trap.c @@ -13,6 +13,7 @@ #include "as-layout.h" #include "kern_util.h" #include "os.h" +#include "skas.h" #include "sysdep/sigcontext.h" /* @@ -128,7 +129,19 @@ static void bad_segv(struct faultinfo fi, unsigned long ip) force_sig_info(SIGSEGV, &si, current); } -static void segv_handler(int sig, struct uml_pt_regs *regs) +void fatal_sigsegv(void) +{ + force_sigsegv(SIGSEGV, current); + do_signal(); + /* + * This is to tell gcc that we're not returning - do_signal + * can, in general, return, but in this case, it's not, since + * we just got a fatal SIGSEGV queued. + */ + os_dump_core(); +} + +void segv_handler(int sig, struct uml_pt_regs *regs) { struct faultinfo * fi = UPT_FAULTINFO(regs); @@ -216,9 +229,6 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, void relay_signal(int sig, struct uml_pt_regs *regs) { - if (arch_handle_signal(sig, regs)) - return; - if (!UPT_IS_USER(regs)) { if (sig == SIGBUS) printk(KERN_ERR "Bus error - the host /dev/shm or /tmp " @@ -226,31 +236,24 @@ void relay_signal(int sig, struct uml_pt_regs *regs) panic("Kernel mode signal %d", sig); } + arch_examine_signal(sig, regs); + current->thread.arch.faultinfo = *UPT_FAULTINFO(regs); force_sig(sig, current); } -static void bus_handler(int sig, struct uml_pt_regs *regs) +void bus_handler(int sig, struct uml_pt_regs *regs) { if (current->thread.fault_catcher != NULL) UML_LONGJMP(current->thread.fault_catcher, 1); else relay_signal(sig, regs); } -static void winch(int sig, struct uml_pt_regs *regs) +void winch(int sig, struct uml_pt_regs *regs) { do_IRQ(WINCH_IRQ, regs); } -const struct kern_handlers handlinfo_kern = { - .relay_signal = relay_signal, - .winch = winch, - .bus_handler = bus_handler, - .page_fault = segv_handler, - .sigio_handler = sigio_handler, - .timer_handler = timer_handler -}; - void trap_init(void) { } diff --git a/arch/um/kernel/uaccess.c b/arch/um/kernel/uaccess.c index d7436aacd26..f0f4b040d7c 100644 --- a/arch/um/kernel/uaccess.c +++ b/arch/um/kernel/uaccess.c @@ -1,10 +1,11 @@ /* * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk) - * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -/* These are here rather than tt/uaccess.c because skas mode needs them in +/* + * These are here rather than tt/uaccess.c because skas mode needs them in * order to do SIGBUS recovery when a tmpfs mount runs out of room. */ @@ -25,6 +26,8 @@ int __do_copy_to_user(void *to, const void *from, int n, fault = __do_user_copy(to, from, n, fault_addr, fault_catcher, __do_copy, &faulted); - if(!faulted) return(0); - else return(n - (fault - (unsigned long) to)); + if (!faulted) + return 0; + else + return n - (fault - (unsigned long) to); } diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index f1c71393f57..a6c1dd1cf5a 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -3,22 +3,23 @@ * Licensed under the GPL */ -#include "linux/delay.h" -#include "linux/mm.h" -#include "linux/module.h" -#include "linux/seq_file.h" -#include "linux/string.h" -#include "linux/utsname.h" -#include "asm/pgtable.h" -#include "asm/processor.h" -#include "asm/setup.h" -#include "arch.h" +#include <linux/delay.h> +#include <linux/init.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/seq_file.h> +#include <linux/string.h> +#include <linux/utsname.h> +#include <asm/pgtable.h> +#include <asm/processor.h> +#include <asm/setup.h> #include "as-layout.h" +#include "arch.h" #include "init.h" #include "kern.h" +#include "kern_util.h" #include "mem_user.h" #include "os.h" -#include "skas.h" #define DEFAULT_COMMAND_LINE "root=98:0" @@ -100,8 +101,6 @@ const struct seq_operations cpuinfo_op = { }; /* Set in linux_main */ -unsigned long host_task_size; -unsigned long task_size; unsigned long uml_physmem; unsigned long uml_reserved; /* Also modified in mem_init */ unsigned long start_vm; @@ -197,20 +196,19 @@ __uml_setup("--help", Usage, " Prints this message.\n\n" ); -static int __init uml_checksetup(char *line, int *add) +static void __init uml_checksetup(char *line, int *add) { struct uml_param *p; p = &__uml_setup_start; - while(p < &__uml_setup_end) { + while (p < &__uml_setup_end) { int n; n = strlen(p->str); if (!strncmp(line, p->str, n) && p->setup_func(line + n, add)) - return 1; + return; p++; } - return 0; } static void __init uml_postsetup(void) @@ -218,14 +216,36 @@ static void __init uml_postsetup(void) initcall_t *p; p = &__uml_postsetup_start; - while(p < &__uml_postsetup_end) { + while (p < &__uml_postsetup_end) { (*p)(); p++; } return; } +static int panic_exit(struct notifier_block *self, unsigned long unused1, + void *unused2) +{ + bust_spinlocks(1); + show_regs(&(current->thread.regs)); + bust_spinlocks(0); + uml_exitcode = 1; + os_dump_core(); + return 0; +} + +static struct notifier_block panic_exit_notifier = { + .notifier_call = panic_exit, + .next = NULL, + .priority = 0 +}; + /* Set during early boot */ +unsigned long task_size; +EXPORT_SYMBOL(task_size); + +unsigned long host_task_size; + unsigned long brk_start; unsigned long end_iomem; EXPORT_SYMBOL(end_iomem); @@ -234,20 +254,6 @@ EXPORT_SYMBOL(end_iomem); extern char __binary_start; -static unsigned long set_task_sizes_skas(unsigned long *task_size_out) -{ - /* Round up to the nearest 4M */ - unsigned long host_task_size = ROUND_4M((unsigned long) - &host_task_size); - - if (!skas_needs_stub) - *task_size_out = host_task_size; - else - *task_size_out = STUB_START & PGDIR_MASK; - - return host_task_size; -} - int __init linux_main(int argc, char **argv) { unsigned long avail, diff; @@ -266,6 +272,13 @@ int __init linux_main(int argc, char **argv) if (have_root == 0) add_arg(DEFAULT_COMMAND_LINE); + host_task_size = os_get_task_size(); + /* + * TASK_SIZE needs to be PGDIR_SIZE aligned or else exit_mmap craps + * out + */ + task_size = host_task_size & PGDIR_MASK; + /* OS sanity checks that need to happen before the kernel runs */ os_early_checks(); @@ -278,13 +291,6 @@ int __init linux_main(int argc, char **argv) printf("UML running in %s mode\n", mode); - host_task_size = set_task_sizes_skas(&task_size); - - /* - * Setting up handlers to 'sig_info' struct - */ - os_fill_handlinfo(handlinfo_kern); - brk_start = (unsigned long) sbrk(0); /* @@ -309,7 +315,7 @@ int __init linux_main(int argc, char **argv) highmem = 0; iomem_size = (iomem_size + PAGE_SIZE - 1) & PAGE_MASK; - max_physmem = get_kmem_end() - uml_physmem - iomem_size - MIN_VMALLOC; + max_physmem = TASK_SIZE - uml_physmem - iomem_size - MIN_VMALLOC; /* * Zones have to begin on a 1 << MAX_ORDER page boundary, @@ -341,7 +347,7 @@ int __init linux_main(int argc, char **argv) } virtmem_size = physmem_size; - avail = get_kmem_end() - start_vm; + avail = TASK_SIZE - start_vm; if (physmem_size > avail) virtmem_size = avail; end_vm = start_vm + virtmem_size; @@ -350,6 +356,9 @@ int __init linux_main(int argc, char **argv) printf("Kernel virtual memory size shrunk to %lu bytes\n", virtmem_size); + atomic_notifier_chain_register(&panic_notifier_list, + &panic_exit_notifier); + uml_postsetup(); stack_protections((unsigned long) &init_thread_info); @@ -358,29 +367,8 @@ int __init linux_main(int argc, char **argv) return start_uml(); } -extern int uml_exitcode; - -static int panic_exit(struct notifier_block *self, unsigned long unused1, - void *unused2) -{ - bust_spinlocks(1); - show_regs(&(current->thread.regs)); - bust_spinlocks(0); - uml_exitcode = 1; - os_dump_core(); - return 0; -} - -static struct notifier_block panic_exit_notifier = { - .notifier_call = panic_exit, - .next = NULL, - .priority = 0 -}; - void __init setup_arch(char **cmdline_p) { - atomic_notifier_chain_register(&panic_notifier_list, - &panic_exit_notifier); paging_init(); strlcpy(boot_command_line, command_line, COMMAND_LINE_SIZE); *cmdline_p = command_line; diff --git a/arch/um/kernel/umid.c b/arch/um/kernel/umid.c index 039e16efcd5..81e07e2be3a 100644 --- a/arch/um/kernel/umid.c +++ b/arch/um/kernel/umid.c @@ -1,13 +1,12 @@ -/* - * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) +/* + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "asm/errno.h" +#include <asm/errno.h> #include "init.h" -#include "os.h" #include "kern.h" -#include "linux/kernel.h" +#include "os.h" /* Changed by set_umid_arg */ static int umid_inited = 0; @@ -16,16 +15,16 @@ static int __init set_umid_arg(char *name, int *add) { int err; - if(umid_inited){ + if (umid_inited) { printf("umid already set\n"); return 0; } *add = 0; err = set_umid(name); - if(err == -EEXIST) + if (err == -EEXIST) printf("umid '%s' already in use\n", name); - else if(!err) + else if (!err) umid_inited = 1; return 0; |