diff options
| -rw-r--r-- | arch/x86/kernel/ldt.c | 9 | ||||
| -rw-r--r-- | arch/x86/kernel/paravirt.c | 4 | ||||
| -rw-r--r-- | arch/x86/xen/enlighten.c | 73 | ||||
| -rw-r--r-- | drivers/xen/balloon.c | 174 | ||||
| -rw-r--r-- | include/asm-x86/desc.h | 15 | ||||
| -rw-r--r-- | include/asm-x86/paravirt.h | 13 | ||||
| -rw-r--r-- | include/linux/kernel.h | 2 | ||||
| -rw-r--r-- | lib/cmdline.c | 2 |
8 files changed, 126 insertions, 166 deletions
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c index b68e21f..6e38841 100644 --- a/arch/x86/kernel/ldt.c +++ b/arch/x86/kernel/ldt.c @@ -51,6 +51,8 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload) memset(newldt + oldsize * LDT_ENTRY_SIZE, 0, (mincount - oldsize) * LDT_ENTRY_SIZE); + paravirt_alloc_ldt(newldt, mincount); + #ifdef CONFIG_X86_64 /* CHECKME: Do we really need this ? */ wmb(); @@ -73,6 +75,7 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload) #endif } if (oldsize) { + paravirt_free_ldt(oldldt, oldsize); if (oldsize * LDT_ENTRY_SIZE > PAGE_SIZE) vfree(oldldt); else @@ -84,10 +87,13 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload) static inline int copy_ldt(mm_context_t *new, mm_context_t *old) { int err = alloc_ldt(new, old->size, 0); + int i; if (err < 0) return err; - memcpy(new->ldt, old->ldt, old->size * LDT_ENTRY_SIZE); + + for(i = 0; i < old->size; i++) + write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE); return 0; } @@ -124,6 +130,7 @@ void destroy_context(struct mm_struct *mm) if (mm == current->active_mm) clear_LDT(); #endif + paravirt_free_ldt(mm->context.ldt, mm->context.size); if (mm->context.size * LDT_ENTRY_SIZE > PAGE_SIZE) vfree(mm->context.ldt); else diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 94da4d52..d8f2277 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -348,6 +348,10 @@ struct pv_cpu_ops pv_cpu_ops = { .write_ldt_entry = native_write_ldt_entry, .write_gdt_entry = native_write_gdt_entry, .write_idt_entry = native_write_idt_entry, + + .alloc_ldt = paravirt_nop, + .free_ldt = paravirt_nop, + .load_sp0 = native_load_sp0, #if defined(CONFIG_X86_32) || defined(CONFIG_IA32_EMULATION) diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 9ff6e3c..b795470 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -325,6 +325,57 @@ static unsigned long xen_store_tr(void) return 0; } +/* + * If 'v' is a vmalloc mapping, then find the linear mapping of the + * page (if any) and also set its protections to match: + */ +static void set_aliased_prot(void *v, pgprot_t prot) +{ + int level; + pte_t *ptep; + pte_t pte; + unsigned long pfn; + struct page *page; + + ptep = lookup_address((unsigned long)v, &level); + BUG_ON(ptep == NULL); + + pfn = pte_pfn(*ptep); + page = pfn_to_page(pfn); + + pte = pfn_pte(pfn, prot); + + if (HYPERVISOR_update_va_mapping((unsigned long)v, pte, 0)) + BUG(); + + if (!PageHighMem(page)) { + void *av = __va(PFN_PHYS(pfn)); + + if (av != v) + if (HYPERVISOR_update_va_mapping((unsigned long)av, pte, 0)) + BUG(); + } else + kmap_flush_unused(); +} + +static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries) +{ + const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE; + int i; + + for(i = 0; i < entries; i += entries_per_page) + set_aliased_prot(ldt + i, PAGE_KERNEL_RO); +} + +static void xen_free_ldt(struct desc_struct *ldt, unsigned entries) +{ + const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE; + int i; + + for(i = 0; i < entries; i += entries_per_page) + set_aliased_prot(ldt + i, PAGE_KERNEL); +} + static void xen_set_ldt(const void *addr, unsigned entries) { struct mmuext_op *op; @@ -426,7 +477,7 @@ static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum, const void *ptr) { unsigned long lp = (unsigned long)&dt[entrynum]; - xmaddr_t mach_lp = virt_to_machine(lp); + xmaddr_t mach_lp = arbitrary_virt_to_machine(lp); u64 entry = *(u64 *)ptr; preempt_disable(); @@ -559,7 +610,7 @@ static void xen_write_gdt_entry(struct desc_struct *dt, int entry, } static void xen_load_sp0(struct tss_struct *tss, - struct thread_struct *thread) + struct thread_struct *thread) { struct multicall_space mcs = xen_mc_entry(0); MULTI_stack_switch(mcs.mc, __KERNEL_DS, thread->sp0); @@ -803,6 +854,19 @@ static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high) ret = -EFAULT; break; #endif + + case MSR_STAR: + case MSR_CSTAR: + case MSR_LSTAR: + case MSR_SYSCALL_MASK: + case MSR_IA32_SYSENTER_CS: + case MSR_IA32_SYSENTER_ESP: + case MSR_IA32_SYSENTER_EIP: + /* Fast syscall setup is all done in hypercalls, so + these are all ignored. Stub them out here to stop + Xen console noise. */ + break; + default: ret = native_write_msr_safe(msr, low, high); } @@ -1220,6 +1284,9 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = { .load_gs_index = xen_load_gs_index, #endif + .alloc_ldt = xen_alloc_ldt, + .free_ldt = xen_free_ldt, + .store_gdt = native_store_gdt, .store_idt = native_store_idt, .store_tr = xen_store_tr, @@ -1324,7 +1391,7 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = { .ptep_modify_prot_commit = __ptep_modify_prot_commit, .pte_val = xen_pte_val, - .pte_flags = native_pte_val, + .pte_flags = native_pte_flags, .pgd_val = xen_pgd_val, .make_pte = xen_make_pte, diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c index d4427cb..fff987b 100644 --- a/drivers/xen/balloon.c +++ b/drivers/xen/balloon.c @@ -60,7 +60,7 @@ #define PAGES2KB(_p) ((_p)<<(PAGE_SHIFT-10)) -#define BALLOON_CLASS_NAME "memory" +#define BALLOON_CLASS_NAME "xen_memory" struct balloon_stats { /* We aim for 'current allocation' == 'target allocation'. */ @@ -226,9 +226,8 @@ static int increase_reservation(unsigned long nr_pages) } set_xen_guest_handle(reservation.extent_start, frame_list); - reservation.nr_extents = nr_pages; - rc = HYPERVISOR_memory_op( - XENMEM_populate_physmap, &reservation); + reservation.nr_extents = nr_pages; + rc = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation); if (rc < nr_pages) { if (rc > 0) { int ret; @@ -236,7 +235,7 @@ static int increase_reservation(unsigned long nr_pages) /* We hit the Xen hard limit: reprobe. */ reservation.nr_extents = rc; ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, - &reservation); + &reservation); BUG_ON(ret != rc); } if (rc >= 0) @@ -464,136 +463,13 @@ static void balloon_exit(void) module_exit(balloon_exit); -static void balloon_update_driver_allowance(long delta) -{ - unsigned long flags; - - spin_lock_irqsave(&balloon_lock, flags); - balloon_stats.driver_pages += delta; - spin_unlock_irqrestore(&balloon_lock, flags); -} - -static int dealloc_pte_fn( - pte_t *pte, struct page *pmd_page, unsigned long addr, void *data) -{ - unsigned long mfn = pte_mfn(*pte); - int ret; - struct xen_memory_reservation reservation = { - .nr_extents = 1, - .extent_order = 0, - .domid = DOMID_SELF - }; - set_xen_guest_handle(reservation.extent_start, &mfn); - set_pte_at(&init_mm, addr, pte, __pte_ma(0ull)); - set_phys_to_machine(__pa(addr) >> PAGE_SHIFT, INVALID_P2M_ENTRY); - ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation); - BUG_ON(ret != 1); - return 0; -} - -static struct page **alloc_empty_pages_and_pagevec(int nr_pages) -{ - unsigned long vaddr, flags; - struct page *page, **pagevec; - int i, ret; - - pagevec = kmalloc(sizeof(page) * nr_pages, GFP_KERNEL); - if (pagevec == NULL) - return NULL; - - for (i = 0; i < nr_pages; i++) { - page = pagevec[i] = alloc_page(GFP_KERNEL); - if (page == NULL) - goto err; - - vaddr = (unsigned long)page_address(page); - - scrub_page(page); - - spin_lock_irqsave(&balloon_lock, flags); - - if (xen_feature(XENFEAT_auto_translated_physmap)) { - unsigned long gmfn = page_to_pfn(page); - struct xen_memory_reservation reservation = { - .nr_extents = 1, - .extent_order = 0, - .domid = DOMID_SELF - }; - set_xen_guest_handle(reservation.extent_start, &gmfn); - ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, - &reservation); - if (ret == 1) - ret = 0; /* success */ - } else { - ret = apply_to_page_range(&init_mm, vaddr, PAGE_SIZE, - dealloc_pte_fn, NULL); - } - - if (ret != 0) { - spin_unlock_irqrestore(&balloon_lock, flags); - __free_page(page); - goto err; - } - - totalram_pages = --balloon_stats.current_pages; - - spin_unlock_irqrestore(&balloon_lock, flags); - } - - out: - schedule_work(&balloon_worker); - flush_tlb_all(); - return pagevec; - - err: - spin_lock_irqsave(&balloon_lock, flags); - while (--i >= 0) - balloon_append(pagevec[i]); - spin_unlock_irqrestore(&balloon_lock, flags); - kfree(pagevec); - pagevec = NULL; - goto out; -} - -static void free_empty_pages_and_pagevec(struct page **pagevec, int nr_pages) -{ - unsigned long flags; - int i; - - if (pagevec == NULL) - return; - - spin_lock_irqsave(&balloon_lock, flags); - for (i = 0; i < nr_pages; i++) { - BUG_ON(page_count(pagevec[i]) != 1); - balloon_append(pagevec[i]); - } - spin_unlock_irqrestore(&balloon_lock, flags); - - kfree(pagevec); - - schedule_work(&balloon_worker); -} - -static void balloon_release_driver_page(struct page *page) -{ - unsigned long flags; - - spin_lock_irqsave(&balloon_lock, flags); - balloon_append(page); - balloon_stats.driver_pages--; - spin_unlock_irqrestore(&balloon_lock, flags); - - schedule_work(&balloon_worker); -} - - -#define BALLOON_SHOW(name, format, args...) \ - static ssize_t show_##name(struct sys_device *dev, \ - char *buf) \ - { \ - return sprintf(buf, format, ##args); \ - } \ +#define BALLOON_SHOW(name, format, args...) \ + static ssize_t show_##name(struct sys_device *dev, \ + struct sysdev_attribute *attr, \ + char *buf) \ + { \ + return sprintf(buf, format, ##args); \ + } \ static SYSDEV_ATTR(name, S_IRUGO, show_##name, NULL) BALLOON_SHOW(current_kb, "%lu\n", PAGES2KB(balloon_stats.current_pages)); @@ -604,7 +480,8 @@ BALLOON_SHOW(hard_limit_kb, (balloon_stats.hard_limit!=~0UL) ? PAGES2KB(balloon_stats.hard_limit) : 0); BALLOON_SHOW(driver_kb, "%lu\n", PAGES2KB(balloon_stats.driver_pages)); -static ssize_t show_target_kb(struct sys_device *dev, char *buf) +static ssize_t show_target_kb(struct sys_device *dev, struct sysdev_attribute *attr, + char *buf) { return sprintf(buf, "%lu\n", PAGES2KB(balloon_stats.target_pages)); } @@ -614,19 +491,14 @@ static ssize_t store_target_kb(struct sys_device *dev, const char *buf, size_t count) { - char memstring[64], *endchar; + char *endchar; unsigned long long target_bytes; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (count <= 1) - return -EBADMSG; /* runt */ - if (count > sizeof(memstring)) - return -EFBIG; /* too long */ - strcpy(memstring, buf); + target_bytes = memparse(buf, &endchar); - target_bytes = memparse(memstring, &endchar); balloon_set_new_target(target_bytes >> PAGE_SHIFT); return count; @@ -694,20 +566,4 @@ static int register_balloon(struct sys_device *sysdev) return error; } -static void unregister_balloon(struct sys_device *sysdev) -{ - int i; - - sysfs_remove_group(&sysdev->kobj, &balloon_info_group); - for (i = 0; i < ARRAY_SIZE(balloon_attrs); i++) - sysdev_remove_file(sysdev, balloon_attrs[i]); - sysdev_unregister(sysdev); - sysdev_class_unregister(&balloon_sysdev_class); -} - -static void balloon_sysfs_exit(void) -{ - unregister_balloon(&balloon_sysdev); -} - MODULE_LICENSE("GPL"); diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h index a44c4dc..06f786f 100644 --- a/include/asm-x86/desc.h +++ b/include/asm-x86/desc.h @@ -24,6 +24,11 @@ static inline void fill_ldt(struct desc_struct *desc, desc->d = info->seg_32bit; desc->g = info->limit_in_pages; desc->base2 = (info->base_addr & 0xff000000) >> 24; + /* + * Don't allow setting of the lm bit. It is useless anyway + * because 64bit system calls require __USER_CS: + */ + desc->l = 0; } extern struct desc_ptr idt_descr; @@ -97,7 +102,15 @@ static inline int desc_empty(const void *ptr) native_write_gdt_entry(dt, entry, desc, type) #define write_idt_entry(dt, entry, g) \ native_write_idt_entry(dt, entry, g) -#endif + +static inline void paravirt_alloc_ldt(struct desc_struct *ldt, unsigned entries) +{ +} + +static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries) +{ +} +#endif /* CONFIG_PARAVIRT */ static inline void native_write_idt_entry(gate_desc *idt, int entry, const gate_desc *gate) diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index fbbde93..db9b064 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -124,6 +124,9 @@ struct pv_cpu_ops { int entrynum, const void *desc, int size); void (*write_idt_entry)(gate_desc *, int entrynum, const gate_desc *gate); + void (*alloc_ldt)(struct desc_struct *ldt, unsigned entries); + void (*free_ldt)(struct desc_struct *ldt, unsigned entries); + void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t); void (*set_iopl_mask)(unsigned mask); @@ -824,6 +827,16 @@ do { \ (aux) = __aux; \ } while (0) +static inline void paravirt_alloc_ldt(struct desc_struct *ldt, unsigned entries) +{ + PVOP_VCALL2(pv_cpu_ops.alloc_ldt, ldt, entries); +} + +static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries) +{ + PVOP_VCALL2(pv_cpu_ops.free_ldt, ldt, entries); +} + static inline void load_TR_desc(void) { PVOP_VCALL0(pv_cpu_ops.load_tr_desc); diff --git a/include/linux/kernel.h b/include/linux/kernel.h index fdbbf72..7889c2f 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -176,7 +176,7 @@ extern int vsscanf(const char *, const char *, va_list) extern int get_option(char **str, int *pint); extern char *get_options(const char *str, int nints, int *ints); -extern unsigned long long memparse(char *ptr, char **retptr); +extern unsigned long long memparse(const char *ptr, char **retptr); extern int core_kernel_text(unsigned long addr); extern int __kernel_text_address(unsigned long addr); diff --git a/lib/cmdline.c b/lib/cmdline.c index 5ba8a94..f5f3ad8 100644 --- a/lib/cmdline.c +++ b/lib/cmdline.c @@ -126,7 +126,7 @@ char *get_options(const char *str, int nints, int *ints) * megabyte, or one gigabyte, respectively. */ -unsigned long long memparse(char *ptr, char **retptr) +unsigned long long memparse(const char *ptr, char **retptr) { char *endptr; /* local pointer to end of parsed string */ |
