From 52441fa8f2f1ccc9fa97607c6ccf8b46b9fd15ae Mon Sep 17 00:00:00 2001 From: Sasha Levin Date: Thu, 3 Jan 2013 11:09:53 +1030 Subject: module: prevent warning when finit_module a 0 sized file If we try to finit_module on a file sized 0 bytes vmalloc will scream and spit out a warning. Since modules have to be bigger than 0 bytes anyways we can just check that beforehand and avoid the warning. Signed-off-by: Sasha Levin Signed-off-by: Rusty Russell --- kernel/module.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'kernel/module.c') diff --git a/kernel/module.c b/kernel/module.c index 250092c1d57..41bc1189b06 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2527,6 +2527,13 @@ static int copy_module_from_fd(int fd, struct load_info *info) err = -EFBIG; goto out; } + + /* Don't hand 0 to vmalloc, it whines. */ + if (stat.size == 0) { + err = -EINVAL; + goto out; + } + info->hdr = vmalloc(stat.size); if (!info->hdr) { err = -ENOMEM; -- cgit v1.2.3-70-g09d2 From 0d21b0e3477395e7ff2acc269f15df6e6a8d356d Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 12 Jan 2013 11:38:44 +1030 Subject: module: add new state MODULE_STATE_UNFORMED. You should never look at such a module, so it's excised from all paths which traverse the modules list. We add the state at the end, to avoid gratuitous ABI break (ksplice). Signed-off-by: Rusty Russell --- include/linux/module.h | 10 ++++---- kernel/debug/kdb/kdb_main.c | 2 ++ kernel/module.c | 57 +++++++++++++++++++++++++++++++++++++++++---- 3 files changed, 59 insertions(+), 10 deletions(-) (limited to 'kernel/module.c') diff --git a/include/linux/module.h b/include/linux/module.h index 7760c6d344a..1375ee3f03a 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -199,11 +199,11 @@ struct module_use { struct module *source, *target; }; -enum module_state -{ - MODULE_STATE_LIVE, - MODULE_STATE_COMING, - MODULE_STATE_GOING, +enum module_state { + MODULE_STATE_LIVE, /* Normal state. */ + MODULE_STATE_COMING, /* Full formed, running module_init. */ + MODULE_STATE_GOING, /* Going away. */ + MODULE_STATE_UNFORMED, /* Still setting it up. */ }; /** diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c index 4d5f8d5612f..8875254120b 100644 --- a/kernel/debug/kdb/kdb_main.c +++ b/kernel/debug/kdb/kdb_main.c @@ -1970,6 +1970,8 @@ static int kdb_lsmod(int argc, const char **argv) kdb_printf("Module Size modstruct Used by\n"); list_for_each_entry(mod, kdb_modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; kdb_printf("%-20s%8u 0x%p ", mod->name, mod->core_size, (void *)mod); diff --git a/kernel/module.c b/kernel/module.c index 41bc1189b06..c3a2ee8e367 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -188,6 +188,7 @@ struct load_info { ongoing or failed initialization etc. */ static inline int strong_try_module_get(struct module *mod) { + BUG_ON(mod && mod->state == MODULE_STATE_UNFORMED); if (mod && mod->state == MODULE_STATE_COMING) return -EBUSY; if (try_module_get(mod)) @@ -343,6 +344,9 @@ bool each_symbol_section(bool (*fn)(const struct symsearch *arr, #endif }; + if (mod->state == MODULE_STATE_UNFORMED) + continue; + if (each_symbol_in_section(arr, ARRAY_SIZE(arr), mod, fn, data)) return true; } @@ -450,16 +454,24 @@ const struct kernel_symbol *find_symbol(const char *name, EXPORT_SYMBOL_GPL(find_symbol); /* Search for module by name: must hold module_mutex. */ -struct module *find_module(const char *name) +static struct module *find_module_all(const char *name, + bool even_unformed) { struct module *mod; list_for_each_entry(mod, &modules, list) { + if (!even_unformed && mod->state == MODULE_STATE_UNFORMED) + continue; if (strcmp(mod->name, name) == 0) return mod; } return NULL; } + +struct module *find_module(const char *name) +{ + return find_module_all(name, false); +} EXPORT_SYMBOL_GPL(find_module); #ifdef CONFIG_SMP @@ -525,6 +537,8 @@ bool is_module_percpu_address(unsigned long addr) preempt_disable(); list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if (!mod->percpu_size) continue; for_each_possible_cpu(cpu) { @@ -1048,6 +1062,8 @@ static ssize_t show_initstate(struct module_attribute *mattr, case MODULE_STATE_GOING: state = "going"; break; + default: + BUG(); } return sprintf(buffer, "%s\n", state); } @@ -1786,6 +1802,8 @@ void set_all_modules_text_rw(void) mutex_lock(&module_mutex); list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if ((mod->module_core) && (mod->core_text_size)) { set_page_attributes(mod->module_core, mod->module_core + mod->core_text_size, @@ -1807,6 +1825,8 @@ void set_all_modules_text_ro(void) mutex_lock(&module_mutex); list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if ((mod->module_core) && (mod->core_text_size)) { set_page_attributes(mod->module_core, mod->module_core + mod->core_text_size, @@ -2998,7 +3018,8 @@ static bool finished_loading(const char *name) mutex_lock(&module_mutex); mod = find_module(name); - ret = !mod || mod->state != MODULE_STATE_COMING; + ret = !mod || mod->state == MODULE_STATE_LIVE + || mod->state == MODULE_STATE_GOING; mutex_unlock(&module_mutex); return ret; @@ -3361,6 +3382,8 @@ const char *module_address_lookup(unsigned long addr, preempt_disable(); list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if (within_module_init(addr, mod) || within_module_core(addr, mod)) { if (modname) @@ -3384,6 +3407,8 @@ int lookup_module_symbol_name(unsigned long addr, char *symname) preempt_disable(); list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if (within_module_init(addr, mod) || within_module_core(addr, mod)) { const char *sym; @@ -3408,6 +3433,8 @@ int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size, preempt_disable(); list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if (within_module_init(addr, mod) || within_module_core(addr, mod)) { const char *sym; @@ -3435,6 +3462,8 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type, preempt_disable(); list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if (symnum < mod->num_symtab) { *value = mod->symtab[symnum].st_value; *type = mod->symtab[symnum].st_info; @@ -3477,9 +3506,12 @@ unsigned long module_kallsyms_lookup_name(const char *name) ret = mod_find_symname(mod, colon+1); *colon = ':'; } else { - list_for_each_entry_rcu(mod, &modules, list) + list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if ((ret = mod_find_symname(mod, name)) != 0) break; + } } preempt_enable(); return ret; @@ -3494,6 +3526,8 @@ int module_kallsyms_on_each_symbol(int (*fn)(void *, const char *, int ret; list_for_each_entry(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; for (i = 0; i < mod->num_symtab; i++) { ret = fn(data, mod->strtab + mod->symtab[i].st_name, mod, mod->symtab[i].st_value); @@ -3509,6 +3543,7 @@ static char *module_flags(struct module *mod, char *buf) { int bx = 0; + BUG_ON(mod->state == MODULE_STATE_UNFORMED); if (mod->taints || mod->state == MODULE_STATE_GOING || mod->state == MODULE_STATE_COMING) { @@ -3550,6 +3585,10 @@ static int m_show(struct seq_file *m, void *p) struct module *mod = list_entry(p, struct module, list); char buf[8]; + /* We always ignore unformed modules. */ + if (mod->state == MODULE_STATE_UNFORMED) + return 0; + seq_printf(m, "%s %u", mod->name, mod->init_size + mod->core_size); print_unload_info(m, mod); @@ -3610,6 +3649,8 @@ const struct exception_table_entry *search_module_extables(unsigned long addr) preempt_disable(); list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if (mod->num_exentries == 0) continue; @@ -3658,10 +3699,13 @@ struct module *__module_address(unsigned long addr) if (addr < module_addr_min || addr > module_addr_max) return NULL; - list_for_each_entry_rcu(mod, &modules, list) + list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; if (within_module_core(addr, mod) || within_module_init(addr, mod)) return mod; + } return NULL; } EXPORT_SYMBOL_GPL(__module_address); @@ -3714,8 +3758,11 @@ void print_modules(void) printk(KERN_DEFAULT "Modules linked in:"); /* Most callers should already have preempt disabled, but make sure */ preempt_disable(); - list_for_each_entry_rcu(mod, &modules, list) + list_for_each_entry_rcu(mod, &modules, list) { + if (mod->state == MODULE_STATE_UNFORMED) + continue; printk(" %s%s", mod->name, module_flags(mod, buf)); + } preempt_enable(); if (last_unloaded_module[0]) printk(" [last unloaded: %s]", last_unloaded_module); -- cgit v1.2.3-70-g09d2 From 1fb9341ac34825aa40354e74d9a2c69df7d2c304 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 12 Jan 2013 13:27:34 +1030 Subject: module: put modules in list much earlier. Prarit's excellent bug report: > In recent Fedora releases (F17 & F18) some users have reported seeing > messages similar to > > [ 15.478160] kvm: Could not allocate 304 bytes percpu data > [ 15.478174] PERCPU: allocation failed, size=304 align=32, alloc from > reserved chunk failed > > during system boot. In some cases, users have also reported seeing this > message along with a failed load of other modules. > > What is happening is systemd is loading an instance of the kvm module for > each cpu found (see commit e9bda3b). When the module load occurs the kernel > currently allocates the modules percpu data area prior to checking to see > if the module is already loaded or is in the process of being loaded. If > the module is already loaded, or finishes load, the module loading code > releases the current instance's module's percpu data. Now we have a new state MODULE_STATE_UNFORMED, we can insert the module into the list (and thus guarantee its uniqueness) before we allocate the per-cpu region. Reported-by: Prarit Bhargava Signed-off-by: Rusty Russell Tested-by: Prarit Bhargava --- kernel/module.c | 90 +++++++++++++++++++++++++++++++-------------------------- lib/bug.c | 1 + 2 files changed, 50 insertions(+), 41 deletions(-) (limited to 'kernel/module.c') diff --git a/kernel/module.c b/kernel/module.c index c3a2ee8e367..ec535aa63c9 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -3017,7 +3017,7 @@ static bool finished_loading(const char *name) bool ret; mutex_lock(&module_mutex); - mod = find_module(name); + mod = find_module_all(name, true); ret = !mod || mod->state == MODULE_STATE_LIVE || mod->state == MODULE_STATE_GOING; mutex_unlock(&module_mutex); @@ -3141,6 +3141,32 @@ static int load_module(struct load_info *info, const char __user *uargs, goto free_copy; } + /* + * We try to place it in the list now to make sure it's unique + * before we dedicate too many resources. In particular, + * temporary percpu memory exhaustion. + */ + mod->state = MODULE_STATE_UNFORMED; +again: + mutex_lock(&module_mutex); + if ((old = find_module_all(mod->name, true)) != NULL) { + if (old->state == MODULE_STATE_COMING + || old->state == MODULE_STATE_UNFORMED) { + /* Wait in case it fails to load. */ + mutex_unlock(&module_mutex); + err = wait_event_interruptible(module_wq, + finished_loading(mod->name)); + if (err) + goto free_module; + goto again; + } + err = -EEXIST; + mutex_unlock(&module_mutex); + goto free_module; + } + list_add_rcu(&mod->list, &modules); + mutex_unlock(&module_mutex); + #ifdef CONFIG_MODULE_SIG mod->sig_ok = info->sig_ok; if (!mod->sig_ok) @@ -3150,7 +3176,7 @@ static int load_module(struct load_info *info, const char __user *uargs, /* Now module is in final location, initialize linked lists, etc. */ err = module_unload_init(mod); if (err) - goto free_module; + goto unlink_mod; /* Now we've got everything in the final locations, we can * find optional sections. */ @@ -3185,54 +3211,33 @@ static int load_module(struct load_info *info, const char __user *uargs, goto free_arch_cleanup; } - /* Mark state as coming so strong_try_module_get() ignores us. */ - mod->state = MODULE_STATE_COMING; - - /* Now sew it into the lists so we can get lockdep and oops - * info during argument parsing. No one should access us, since - * strong_try_module_get() will fail. - * lockdep/oops can run asynchronous, so use the RCU list insertion - * function to insert in a way safe to concurrent readers. - * The mutex protects against concurrent writers. - */ -again: - mutex_lock(&module_mutex); - if ((old = find_module(mod->name)) != NULL) { - if (old->state == MODULE_STATE_COMING) { - /* Wait in case it fails to load. */ - mutex_unlock(&module_mutex); - err = wait_event_interruptible(module_wq, - finished_loading(mod->name)); - if (err) - goto free_arch_cleanup; - goto again; - } - err = -EEXIST; - goto unlock; - } - - /* This has to be done once we're sure module name is unique. */ dynamic_debug_setup(info->debug, info->num_debug); - /* Find duplicate symbols */ + mutex_lock(&module_mutex); + /* Find duplicate symbols (must be called under lock). */ err = verify_export_symbols(mod); if (err < 0) - goto ddebug; + goto ddebug_cleanup; + /* This relies on module_mutex for list integrity. */ module_bug_finalize(info->hdr, info->sechdrs, mod); - list_add_rcu(&mod->list, &modules); + + /* Mark state as coming so strong_try_module_get() ignores us, + * but kallsyms etc. can see us. */ + mod->state = MODULE_STATE_COMING; + mutex_unlock(&module_mutex); /* Module is ready to execute: parsing args may do that. */ err = parse_args(mod->name, mod->args, mod->kp, mod->num_kp, -32768, 32767, &ddebug_dyndbg_module_param_cb); if (err < 0) - goto unlink; + goto bug_cleanup; /* Link in to syfs. */ err = mod_sysfs_setup(mod, info, mod->kp, mod->num_kp); if (err < 0) - goto unlink; + goto bug_cleanup; /* Get rid of temporary copy. */ free_copy(info); @@ -3242,16 +3247,13 @@ again: return do_init_module(mod); - unlink: + bug_cleanup: + /* module_bug_cleanup needs module_mutex protection */ mutex_lock(&module_mutex); - /* Unlink carefully: kallsyms could be walking list. */ - list_del_rcu(&mod->list); module_bug_cleanup(mod); - wake_up_all(&module_wq); - ddebug: - dynamic_debug_remove(info->debug); - unlock: mutex_unlock(&module_mutex); + ddebug_cleanup: + dynamic_debug_remove(info->debug); synchronize_sched(); kfree(mod->args); free_arch_cleanup: @@ -3260,6 +3262,12 @@ again: free_modinfo(mod); free_unload: module_unload_free(mod); + unlink_mod: + mutex_lock(&module_mutex); + /* Unlink carefully: kallsyms could be walking list. */ + list_del_rcu(&mod->list); + wake_up_all(&module_wq); + mutex_unlock(&module_mutex); free_module: module_deallocate(mod, info); free_copy: diff --git a/lib/bug.c b/lib/bug.c index a28c1415357..d0cdf14c651 100644 --- a/lib/bug.c +++ b/lib/bug.c @@ -55,6 +55,7 @@ static inline unsigned long bug_addr(const struct bug_entry *bug) } #ifdef CONFIG_MODULES +/* Updates are protected by module mutex */ static LIST_HEAD(module_bug_list); static const struct bug_entry *module_find_bug(unsigned long bugaddr) -- cgit v1.2.3-70-g09d2 From 774a1221e862b343388347bac9b318767336b20b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 15 Jan 2013 18:52:51 -0800 Subject: module, async: async_synchronize_full() on module init iff async is used If the default iosched is built as module, the kernel may deadlock while trying to load the iosched module on device probe if the probing was running off async. This is because async_synchronize_full() at the end of module init ends up waiting for the async job which initiated the module loading. async A modprobe 1. finds a device 2. registers the block device 3. request_module(default iosched) 4. modprobe in userland 5. load and init module 6. async_synchronize_full() Async A waits for modprobe to finish in request_module() and modprobe waits for async A to finish in async_synchronize_full(). Because there's no easy to track dependency once control goes out to userland, implementing properly nested flushing is difficult. For now, make module init perform async_synchronize_full() iff module init has queued async jobs as suggested by Linus. This avoids the described deadlock because iosched module doesn't use async and thus wouldn't invoke async_synchronize_full(). This is hacky and incomplete. It will deadlock if async module loading nests; however, this works around the known problem case and seems to be the best of bad options. For more details, please refer to the following thread. http://thread.gmane.org/gmane.linux.kernel/1420814 Signed-off-by: Tejun Heo Reported-by: Alex Riesen Tested-by: Ming Lei Tested-by: Alex Riesen Cc: Arjan van de Ven Cc: Jens Axboe Signed-off-by: Linus Torvalds --- include/linux/sched.h | 1 + kernel/async.c | 3 +++ kernel/module.c | 27 +++++++++++++++++++++++++-- 3 files changed, 29 insertions(+), 2 deletions(-) (limited to 'kernel/module.c') diff --git a/include/linux/sched.h b/include/linux/sched.h index 206bb089c06..6fc8f45de4e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1810,6 +1810,7 @@ extern void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, #define PF_MEMALLOC 0x00000800 /* Allocating memory */ #define PF_NPROC_EXCEEDED 0x00001000 /* set_user noticed that RLIMIT_NPROC was exceeded */ #define PF_USED_MATH 0x00002000 /* if unset the fpu must be initialized before use */ +#define PF_USED_ASYNC 0x00004000 /* used async_schedule*(), used by module init */ #define PF_NOFREEZE 0x00008000 /* this thread should not be frozen */ #define PF_FROZEN 0x00010000 /* frozen for system suspend */ #define PF_FSTRANS 0x00020000 /* inside a filesystem transaction */ diff --git a/kernel/async.c b/kernel/async.c index 9d311838485..a1d585c351d 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -196,6 +196,9 @@ static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct a atomic_inc(&entry_count); spin_unlock_irqrestore(&async_lock, flags); + /* mark that this task has queued an async job, used by module init */ + current->flags |= PF_USED_ASYNC; + /* schedule for execution */ queue_work(system_unbound_wq, &entry->work); diff --git a/kernel/module.c b/kernel/module.c index 250092c1d57..b10b048367e 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -3013,6 +3013,12 @@ static int do_init_module(struct module *mod) { int ret = 0; + /* + * We want to find out whether @mod uses async during init. Clear + * PF_USED_ASYNC. async_schedule*() will set it. + */ + current->flags &= ~PF_USED_ASYNC; + blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_COMING, mod); @@ -3058,8 +3064,25 @@ static int do_init_module(struct module *mod) blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_LIVE, mod); - /* We need to finish all async code before the module init sequence is done */ - async_synchronize_full(); + /* + * We need to finish all async code before the module init sequence + * is done. This has potential to deadlock. For example, a newly + * detected block device can trigger request_module() of the + * default iosched from async probing task. Once userland helper + * reaches here, async_synchronize_full() will wait on the async + * task waiting on request_module() and deadlock. + * + * This deadlock is avoided by perfomring async_synchronize_full() + * iff module init queued any async jobs. This isn't a full + * solution as it will deadlock the same if module loading from + * async jobs nests more than once; however, due to the various + * constraints, this hack seems to be the best option for now. + * Please refer to the following thread for details. + * + * http://thread.gmane.org/gmane.linux.kernel/1420814 + */ + if (current->flags & PF_USED_ASYNC) + async_synchronize_full(); mutex_lock(&module_mutex); /* Drop initial reference. */ -- cgit v1.2.3-70-g09d2 From ee61abb3223e28a1a14a8429c0319755d20d3e40 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 20 Jan 2013 20:22:58 -0800 Subject: module: fix missing module_mutex unlock Commit 1fb9341ac348 ("module: put modules in list much earlier") moved some of the module initialization code around, and in the process changed the exit paths too. But for the duplicate export symbol error case the change made the ddebug_cleanup path jump to after the module mutex unlock, even though it happens with the mutex held. Rusty has some patches to split this function up into some helper functions, hopefully the mess of complex goto targets will go away eventually. Reported-by: Dan Carpenter Cc: Rusty Russell Signed-off-by: Linus Torvalds --- kernel/module.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/module.c') diff --git a/kernel/module.c b/kernel/module.c index d25e359279a..eab08274ec9 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -3274,8 +3274,8 @@ again: /* module_bug_cleanup needs module_mutex protection */ mutex_lock(&module_mutex); module_bug_cleanup(mod); - mutex_unlock(&module_mutex); ddebug_cleanup: + mutex_unlock(&module_mutex); dynamic_debug_remove(info->debug); synchronize_sched(); kfree(mod->args); -- cgit v1.2.3-70-g09d2