From 42b7772812d15b86543a23b82bd6070eef9a08b1 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Wed, 23 Jul 2008 21:27:10 -0700 Subject: mm: remove double indirection on tlb parameter to free_pgd_range() & Co The double indirection here is not needed anywhere and hence (at least) confusing. Signed-off-by: Jan Beulich Cc: Hugh Dickins Cc: Nick Piggin Cc: Christoph Lameter Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: "Luck, Tony" Cc: Paul Mundt Cc: "David S. Miller" Acked-by: Jeremy Fitzhardinge Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index fd9234379e8..190ed1f9277 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -541,7 +541,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) /* * when the old and new regions overlap clear from new_end. */ - free_pgd_range(&tlb, new_end, old_end, new_end, + free_pgd_range(tlb, new_end, old_end, new_end, vma->vm_next ? vma->vm_next->vm_start : 0); } else { /* @@ -550,7 +550,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) * have constraints on va-space that make this illegal (IA64) - * for the others its just a little faster. */ - free_pgd_range(&tlb, old_start, old_end, new_end, + free_pgd_range(tlb, old_start, old_end, new_end, vma->vm_next ? vma->vm_next->vm_start : 0); } tlb_finish_mmu(tlb, new_end, old_end); -- cgit v1.2.3-70-g09d2 From ba92a43dbaee339cf5915ef766d3d3ffbaaf103c Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Fri, 25 Jul 2008 01:45:43 -0700 Subject: exec: remove some includes fs/exec.c used to need mman.h pagemap.h swap.h and rmap.h when it did mm-ish stuff in install_arg_page(); but no need for them after 2.6.22. [akpm@linux-foundation.org: unbreak arm] Signed-off-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index 190ed1f9277..e41aef0fb35 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -25,19 +25,18 @@ #include #include #include -#include +#include #include #include #include +#include #include #include -#include #include #include #include #include #include -#include #include #include #include @@ -47,7 +46,6 @@ #include #include #include -#include #include #include #include -- cgit v1.2.3-70-g09d2 From e4901f92a8dbe843e76651a50f7a2a6dd3d53474 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:31 -0700 Subject: coredump: zap_threads: comments && use while_each_thread() No changes in fs/exec.o The for_each_process() loop in zap_threads() is very subtle, it is not clear why we don't race with fork/exit/exec. Add the fat comment. Also, change the code to use while_each_thread(). Signed-off-by: Oleg Nesterov Acked-by: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 39 ++++++++++++++++++++++++++++++++------- 1 file changed, 32 insertions(+), 7 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index e41aef0fb35..af249af4cca 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1517,7 +1517,7 @@ static void zap_process(struct task_struct *start) sigaddset(&t->pending.signal, SIGKILL); signal_wake_up(t, 1); } - } while ((t = next_thread(t)) != start); + } while_each_thread(start, t); } static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, @@ -1539,7 +1539,36 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, if (atomic_read(&mm->mm_users) == mm->core_waiters + 1) goto done; - + /* + * We should find and kill all tasks which use this mm, and we should + * count them correctly into mm->core_waiters. We don't take tasklist + * lock, but this is safe wrt: + * + * fork: + * None of sub-threads can fork after zap_process(leader). All + * processes which were created before this point should be + * visible to zap_threads() because copy_process() adds the new + * process to the tail of init_task.tasks list, and lock/unlock + * of ->siglock provides a memory barrier. + * + * do_exit: + * The caller holds mm->mmap_sem. This means that the task which + * uses this mm can't pass exit_mm(), so it can't exit or clear + * its ->mm. + * + * de_thread: + * It does list_replace_rcu(&leader->tasks, ¤t->tasks), + * we must see either old or new leader, this does not matter. + * However, it can change p->sighand, so lock_task_sighand(p) + * must be used. Since p->mm != NULL and we hold ->mmap_sem + * it can't fail. + * + * Note also that "g" can be the old leader with ->mm == NULL + * and already unhashed and thus removed from ->thread_group. + * This is OK, __unhash_process()->list_del_rcu() does not + * clear the ->next pointer, we will find the new leader via + * next_thread(). + */ rcu_read_lock(); for_each_process(g) { if (g == tsk->group_leader) @@ -1549,17 +1578,13 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, do { if (p->mm) { if (p->mm == mm) { - /* - * p->sighand can't disappear, but - * may be changed by de_thread() - */ lock_task_sighand(p, &flags); zap_process(p); unlock_task_sighand(p, &flags); } break; } - } while ((p = next_thread(p)) != g); + } while_each_thread(g, p); } rcu_read_unlock(); done: -- cgit v1.2.3-70-g09d2 From 7b34e4283c685f5cc6ba6d30e939906eee0d4bcf Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:37 -0700 Subject: introduce PF_KTHREAD flag Introduce the new PF_KTHREAD flag to mark the kernel threads. It is set by INIT_TASK() and copied to the forked childs (we could set it in kthreadd() along with PF_NOFREEZE instead). daemonize() was changed as well. In that case testing of PF_KTHREAD is racy, but daemonize() is hopeless anyway. This flag is cleared in do_execve(), before search_binary_handler(). Probably not the best place, we can do this in exec_mmap() or in start_thread(), or clear it along with PF_FORKNOEXEC. But I think this doesn't matter in practice, and if do_execve() fails kthread should die soon. Signed-off-by: Oleg Nesterov Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 1 + include/linux/init_task.h | 2 +- include/linux/sched.h | 1 + kernel/exit.c | 2 +- 4 files changed, 4 insertions(+), 2 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index af249af4cca..cd2e8c9b124 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1326,6 +1326,7 @@ int do_execve(char * filename, if (retval < 0) goto out; + current->flags &= ~PF_KTHREAD; retval = search_binary_handler(bprm,regs); if (retval >= 0) { /* execve success */ diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 93c45acf249..021d8e720c7 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -122,7 +122,7 @@ extern struct group_info init_groups; .state = 0, \ .stack = &init_thread_info, \ .usage = ATOMIC_INIT(2), \ - .flags = 0, \ + .flags = PF_KTHREAD, \ .lock_depth = -1, \ .prio = MAX_PRIO-20, \ .static_prio = MAX_PRIO-20, \ diff --git a/include/linux/sched.h b/include/linux/sched.h index 79e749dbf81..eec64a4adb9 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1483,6 +1483,7 @@ static inline void put_task_struct(struct task_struct *t) #define PF_EXITING 0x00000004 /* getting shut down */ #define PF_EXITPIDONE 0x00000008 /* pi exit done on shut down */ #define PF_VCPU 0x00000010 /* I'm a virtual CPU */ +#define PF_KTHREAD 0x00000020 /* I am a kernel thread */ #define PF_FORKNOEXEC 0x00000040 /* forked but didn't exec */ #define PF_SUPERPRIV 0x00000100 /* used super-user privileges */ #define PF_DUMPCORE 0x00000200 /* dumped core */ diff --git a/kernel/exit.c b/kernel/exit.c index a7799d8a640..28a44a2612d 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -430,7 +430,7 @@ void daemonize(const char *name, ...) * We don't want to have TIF_FREEZE set if the system-wide hibernation * or suspend transition begins right now. */ - current->flags |= PF_NOFREEZE; + current->flags |= (PF_NOFREEZE | PF_KTHREAD); if (current->nsproxy != &init_nsproxy) { get_nsproxy(&init_nsproxy); -- cgit v1.2.3-70-g09d2 From 15b9f360c0316c06d37c09b02d85565edbaf9dd3 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:39 -0700 Subject: coredump: zap_threads() must skip kernel threads The main loop in zap_threads() must skip kthreads which may use the same mm. Otherwise we "kill" this thread erroneously (for example, it can not fork or exec after that), and the coredumping task stucks in the TASK_UNINTERRUPTIBLE state forever because of the wrong ->core_waiters count. Signed-off-by: Oleg Nesterov Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index cd2e8c9b124..e347e6ed161 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1574,11 +1574,12 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, for_each_process(g) { if (g == tsk->group_leader) continue; - + if (g->flags & PF_KTHREAD) + continue; p = g; do { if (p->mm) { - if (p->mm == mm) { + if (unlikely(p->mm == mm)) { lock_task_sighand(p, &flags); zap_process(p); unlock_task_sighand(p, &flags); -- cgit v1.2.3-70-g09d2 From 32ecb1f26dd50eeaac4e3f4dea4541c97848e459 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:41 -0700 Subject: coredump: turn mm->core_startup_done into the pointer to struct core_state mm->core_startup_done points to "struct completion startup_done" allocated on the coredump_wait()'s stack. Introduce the new structure, core_state, which holds this "struct completion". This way we can add more info visible to the threads participating in coredump without enlarging mm_struct. No changes in affected .o files. Signed-off-by: Oleg Nesterov Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 8 ++++---- include/linux/mm_types.h | 7 ++++++- kernel/exit.c | 2 +- 3 files changed, 11 insertions(+), 6 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index e347e6ed161..71734568f01 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1597,13 +1597,13 @@ static int coredump_wait(int exit_code) { struct task_struct *tsk = current; struct mm_struct *mm = tsk->mm; - struct completion startup_done; + struct core_state core_state; struct completion *vfork_done; int core_waiters; init_completion(&mm->core_done); - init_completion(&startup_done); - mm->core_startup_done = &startup_done; + init_completion(&core_state.startup); + mm->core_state = &core_state; core_waiters = zap_threads(tsk, mm, exit_code); up_write(&mm->mmap_sem); @@ -1622,7 +1622,7 @@ static int coredump_wait(int exit_code) } if (core_waiters) - wait_for_completion(&startup_done); + wait_for_completion(&core_state.startup); fail: BUG_ON(mm->core_waiters); return core_waiters; diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 02a27ae7853..97819efd233 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -159,6 +159,10 @@ struct vm_area_struct { #endif }; +struct core_state { + struct completion startup; +}; + struct mm_struct { struct vm_area_struct * mmap; /* list of VMAs */ struct rb_root mm_rb; @@ -220,7 +224,8 @@ struct mm_struct { unsigned long flags; /* Must use atomic bitops to access the bits */ /* coredumping support */ - struct completion *core_startup_done, core_done; + struct core_state *core_state; + struct completion core_done; /* aio bits */ rwlock_t ioctx_list_lock; /* aio lock */ diff --git a/kernel/exit.c b/kernel/exit.c index 28a44a2612d..f7fa21dbced 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -680,7 +680,7 @@ static void exit_mm(struct task_struct * tsk) up_read(&mm->mmap_sem); down_write(&mm->mmap_sem); if (!--mm->core_waiters) - complete(mm->core_startup_done); + complete(&mm->core_state->startup); up_write(&mm->mmap_sem); wait_for_completion(&mm->core_done); -- cgit v1.2.3-70-g09d2 From 999d9fc1670bc082928b93b11d1f2e0e417d973c Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:41 -0700 Subject: coredump: move mm->core_waiters into struct core_state Move mm->core_waiters into "struct core_state" allocated on stack. This shrinks mm_struct a little bit and allows further changes. This patch mostly does s/core_waiters/core_state. The only essential change is that coredump_wait() must clear mm->core_state before return. The coredump_wait()'s path is uglified and .text grows by 30 bytes, this is fixed by the next patch. Signed-off-by: Oleg Nesterov Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 21 +++++++++++---------- include/linux/mm_types.h | 2 +- kernel/exit.c | 8 ++++---- kernel/fork.c | 2 +- kernel/signal.c | 4 ++-- 5 files changed, 19 insertions(+), 18 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index 71734568f01..50de3aaff4d 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -722,12 +722,10 @@ static int exec_mmap(struct mm_struct *mm) * Make sure that if there is a core dump in progress * for the old mm, we get out and die instead of going * through with the exec. We must hold mmap_sem around - * checking core_waiters and changing tsk->mm. The - * core-inducing thread will increment core_waiters for - * each thread whose ->mm == old_mm. + * checking core_state and changing tsk->mm. */ down_read(&old_mm->mmap_sem); - if (unlikely(old_mm->core_waiters)) { + if (unlikely(old_mm->core_state)) { up_read(&old_mm->mmap_sem); return -EINTR; } @@ -1514,7 +1512,7 @@ static void zap_process(struct task_struct *start) t = start; do { if (t != current && t->mm) { - t->mm->core_waiters++; + t->mm->core_state->nr_threads++; sigaddset(&t->pending.signal, SIGKILL); signal_wake_up(t, 1); } @@ -1538,11 +1536,11 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, if (err) return err; - if (atomic_read(&mm->mm_users) == mm->core_waiters + 1) + if (atomic_read(&mm->mm_users) == mm->core_state->nr_threads + 1) goto done; /* * We should find and kill all tasks which use this mm, and we should - * count them correctly into mm->core_waiters. We don't take tasklist + * count them correctly into ->nr_threads. We don't take tasklist * lock, but this is safe wrt: * * fork: @@ -1590,7 +1588,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, } rcu_read_unlock(); done: - return mm->core_waiters; + return mm->core_state->nr_threads; } static int coredump_wait(int exit_code) @@ -1603,9 +1601,12 @@ static int coredump_wait(int exit_code) init_completion(&mm->core_done); init_completion(&core_state.startup); + core_state.nr_threads = 0; mm->core_state = &core_state; core_waiters = zap_threads(tsk, mm, exit_code); + if (core_waiters < 0) + mm->core_state = NULL; up_write(&mm->mmap_sem); if (unlikely(core_waiters < 0)) @@ -1623,8 +1624,8 @@ static int coredump_wait(int exit_code) if (core_waiters) wait_for_completion(&core_state.startup); + mm->core_state = NULL; fail: - BUG_ON(mm->core_waiters); return core_waiters; } @@ -1702,7 +1703,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) /* * If another thread got here first, or we are not dumpable, bail out. */ - if (mm->core_waiters || !get_dumpable(mm)) { + if (mm->core_state || !get_dumpable(mm)) { up_write(&mm->mmap_sem); goto fail; } diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 97819efd233..c0b1747b61a 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -160,6 +160,7 @@ struct vm_area_struct { }; struct core_state { + int nr_threads; struct completion startup; }; @@ -179,7 +180,6 @@ struct mm_struct { atomic_t mm_users; /* How many users with user space? */ atomic_t mm_count; /* How many references to "struct mm_struct" (users count as 1) */ int map_count; /* number of VMAs */ - int core_waiters; struct rw_semaphore mmap_sem; spinlock_t page_table_lock; /* Protects page tables and some counters */ diff --git a/kernel/exit.c b/kernel/exit.c index f7fa21dbced..988e232254e 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -670,16 +670,16 @@ static void exit_mm(struct task_struct * tsk) return; /* * Serialize with any possible pending coredump. - * We must hold mmap_sem around checking core_waiters + * We must hold mmap_sem around checking core_state * and clearing tsk->mm. The core-inducing thread - * will increment core_waiters for each thread in the + * will increment ->nr_threads for each thread in the * group with ->mm != NULL. */ down_read(&mm->mmap_sem); - if (mm->core_waiters) { + if (mm->core_state) { up_read(&mm->mmap_sem); down_write(&mm->mmap_sem); - if (!--mm->core_waiters) + if (!--mm->core_state->nr_threads) complete(&mm->core_state->startup); up_write(&mm->mmap_sem); diff --git a/kernel/fork.c b/kernel/fork.c index eeaec6893b0..813d5c89b9d 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -400,7 +400,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p) INIT_LIST_HEAD(&mm->mmlist); mm->flags = (current->mm) ? current->mm->flags : MMF_DUMP_FILTER_DEFAULT; - mm->core_waiters = 0; + mm->core_state = NULL; mm->nr_ptes = 0; set_mm_counter(mm, file_rss, 0); set_mm_counter(mm, anon_rss, 0); diff --git a/kernel/signal.c b/kernel/signal.c index 39c1706edf0..5c7b7eaa0dc 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1480,10 +1480,10 @@ static inline int may_ptrace_stop(void) * is a deadlock situation, and pointless because our tracer * is dead so don't allow us to stop. * If SIGKILL was already sent before the caller unlocked - * ->siglock we must see ->core_waiters != 0. Otherwise it + * ->siglock we must see ->core_state != NULL. Otherwise it * is safe to enter schedule(). */ - if (unlikely(current->mm->core_waiters) && + if (unlikely(current->mm->core_state) && unlikely(current->mm == current->parent->mm)) return 0; -- cgit v1.2.3-70-g09d2 From 8cd9c249128a59e8e833d454a784b0cbd338d468 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:42 -0700 Subject: coredump: simplify core_state->nr_threads calculation Change zap_process() to return int instead of incrementing mm->core_state->nr_threads directly. Change zap_threads() to set mm->core_state only on success. This patch restores the original size of .text, and more importantly now ->nr_threads is used in two places only. Signed-off-by: Oleg Nesterov Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 33 ++++++++++++++++----------------- 1 file changed, 16 insertions(+), 17 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index 50de3aaff4d..c74bb34eeef 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1502,9 +1502,10 @@ out: return ispipe; } -static void zap_process(struct task_struct *start) +static int zap_process(struct task_struct *start) { struct task_struct *t; + int nr = 0; start->signal->flags = SIGNAL_GROUP_EXIT; start->signal->group_stop_count = 0; @@ -1512,31 +1513,33 @@ static void zap_process(struct task_struct *start) t = start; do { if (t != current && t->mm) { - t->mm->core_state->nr_threads++; sigaddset(&t->pending.signal, SIGKILL); signal_wake_up(t, 1); + nr++; } } while_each_thread(start, t); + + return nr; } static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, - int exit_code) + struct core_state *core_state, int exit_code) { struct task_struct *g, *p; unsigned long flags; - int err = -EAGAIN; + int nr = -EAGAIN; spin_lock_irq(&tsk->sighand->siglock); if (!signal_group_exit(tsk->signal)) { + mm->core_state = core_state; tsk->signal->group_exit_code = exit_code; - zap_process(tsk); - err = 0; + nr = zap_process(tsk); } spin_unlock_irq(&tsk->sighand->siglock); - if (err) - return err; + if (unlikely(nr < 0)) + return nr; - if (atomic_read(&mm->mm_users) == mm->core_state->nr_threads + 1) + if (atomic_read(&mm->mm_users) == nr + 1) goto done; /* * We should find and kill all tasks which use this mm, and we should @@ -1579,7 +1582,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, if (p->mm) { if (unlikely(p->mm == mm)) { lock_task_sighand(p, &flags); - zap_process(p); + nr += zap_process(p); unlock_task_sighand(p, &flags); } break; @@ -1588,7 +1591,8 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, } rcu_read_unlock(); done: - return mm->core_state->nr_threads; + core_state->nr_threads = nr; + return nr; } static int coredump_wait(int exit_code) @@ -1601,12 +1605,7 @@ static int coredump_wait(int exit_code) init_completion(&mm->core_done); init_completion(&core_state.startup); - core_state.nr_threads = 0; - mm->core_state = &core_state; - - core_waiters = zap_threads(tsk, mm, exit_code); - if (core_waiters < 0) - mm->core_state = NULL; + core_waiters = zap_threads(tsk, mm, &core_state, exit_code); up_write(&mm->mmap_sem); if (unlikely(core_waiters < 0)) -- cgit v1.2.3-70-g09d2 From c5f1cc8c1828486a61ab3e575da6e2c62b34d399 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:42 -0700 Subject: coredump: turn core_state->nr_threads into atomic_t Turn core_state->nr_threads into atomic_t and kill now unneeded down_write(&mm->mmap_sem) in exit_mm(). Signed-off-by: Oleg Nesterov Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 2 +- include/linux/mm_types.h | 2 +- kernel/exit.c | 5 ++--- 3 files changed, 4 insertions(+), 5 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index c74bb34eeef..15d493fe8aa 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1591,7 +1591,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, } rcu_read_unlock(); done: - core_state->nr_threads = nr; + atomic_set(&core_state->nr_threads, nr); return nr; } diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index c0b1747b61a..ae99a28ba6a 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -160,7 +160,7 @@ struct vm_area_struct { }; struct core_state { - int nr_threads; + atomic_t nr_threads; struct completion startup; }; diff --git a/kernel/exit.c b/kernel/exit.c index 988e232254e..63d82957baa 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -678,10 +678,9 @@ static void exit_mm(struct task_struct * tsk) down_read(&mm->mmap_sem); if (mm->core_state) { up_read(&mm->mmap_sem); - down_write(&mm->mmap_sem); - if (!--mm->core_state->nr_threads) + + if (atomic_dec_and_test(&mm->core_state->nr_threads)) complete(&mm->core_state->startup); - up_write(&mm->mmap_sem); wait_for_completion(&mm->core_done); down_read(&mm->mmap_sem); -- cgit v1.2.3-70-g09d2 From 9d5b327bf198d2720666de958dcc2ae219d86952 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:43 -0700 Subject: coredump: make mm->core_state visible to ->core_dump() Move the "struct core_state core_state" from coredump_wait() to do_coredump(), this makes mm->core_state visible to binfmt->core_dump(). Signed-off-by: Oleg Nesterov Acked-by: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index 15d493fe8aa..b8ee842d93c 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1595,17 +1595,16 @@ done: return nr; } -static int coredump_wait(int exit_code) +static int coredump_wait(int exit_code, struct core_state *core_state) { struct task_struct *tsk = current; struct mm_struct *mm = tsk->mm; - struct core_state core_state; struct completion *vfork_done; int core_waiters; init_completion(&mm->core_done); - init_completion(&core_state.startup); - core_waiters = zap_threads(tsk, mm, &core_state, exit_code); + init_completion(&core_state->startup); + core_waiters = zap_threads(tsk, mm, core_state, exit_code); up_write(&mm->mmap_sem); if (unlikely(core_waiters < 0)) @@ -1622,8 +1621,7 @@ static int coredump_wait(int exit_code) } if (core_waiters) - wait_for_completion(&core_state.startup); - mm->core_state = NULL; + wait_for_completion(&core_state->startup); fail: return core_waiters; } @@ -1679,6 +1677,7 @@ int get_dumpable(struct mm_struct *mm) int do_coredump(long signr, int exit_code, struct pt_regs * regs) { + struct core_state core_state; char corename[CORENAME_MAX_SIZE + 1]; struct mm_struct *mm = current->mm; struct linux_binfmt * binfmt; @@ -1717,7 +1716,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) current->fsuid = 0; /* Dump root private */ } - retval = coredump_wait(exit_code); + retval = coredump_wait(exit_code, &core_state); if (retval < 0) goto fail; @@ -1812,6 +1811,7 @@ fail_unlock: current->fsuid = fsuid; complete_all(&mm->core_done); + mm->core_state = NULL; fail: return retval; } -- cgit v1.2.3-70-g09d2 From b564daf806d492dd4f7afe9b6c83b8d35d137669 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:44 -0700 Subject: coredump: construct the list of coredumping threads at startup time binfmt->core_dump() has to iterate over the all threads in system in order to find the coredumping threads and construct the list using the GFP_ATOMIC allocations. With this patch each thread allocates the list node on exit_mm()'s stack and adds itself to the list. This allows us to do further changes: - simplify ->core_dump() - change exit_mm() to clear ->mm first, then wait for ->core_done. this makes the coredumping process visible to oom_kill - kill mm->core_done Signed-off-by: Oleg Nesterov Acked-by: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 2 ++ include/linux/mm_types.h | 6 ++++++ kernel/exit.c | 15 ++++++++++++--- 3 files changed, 20 insertions(+), 3 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index b8ee842d93c..fe2873b8037 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1604,6 +1604,8 @@ static int coredump_wait(int exit_code, struct core_state *core_state) init_completion(&mm->core_done); init_completion(&core_state->startup); + core_state->dumper.task = tsk; + core_state->dumper.next = NULL; core_waiters = zap_threads(tsk, mm, core_state, exit_code); up_write(&mm->mmap_sem); diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index ae99a28ba6a..4d0d0abc79f 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -159,8 +159,14 @@ struct vm_area_struct { #endif }; +struct core_thread { + struct task_struct *task; + struct core_thread *next; +}; + struct core_state { atomic_t nr_threads; + struct core_thread dumper; struct completion startup; }; diff --git a/kernel/exit.c b/kernel/exit.c index 63d82957baa..b66f0d55c79 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -664,6 +664,7 @@ assign_new_owner: static void exit_mm(struct task_struct * tsk) { struct mm_struct *mm = tsk->mm; + struct core_state *core_state; mm_release(tsk, mm); if (!mm) @@ -676,11 +677,19 @@ static void exit_mm(struct task_struct * tsk) * group with ->mm != NULL. */ down_read(&mm->mmap_sem); - if (mm->core_state) { + core_state = mm->core_state; + if (core_state) { + struct core_thread self; up_read(&mm->mmap_sem); - if (atomic_dec_and_test(&mm->core_state->nr_threads)) - complete(&mm->core_state->startup); + self.task = tsk; + self.next = xchg(&core_state->dumper.next, &self); + /* + * Implies mb(), the result of xchg() must be visible + * to core_state->dumper. + */ + if (atomic_dec_and_test(&core_state->nr_threads)) + complete(&core_state->startup); wait_for_completion(&mm->core_done); down_read(&mm->mmap_sem); -- cgit v1.2.3-70-g09d2 From a94e2d408eaedbd85aae259621d46fafc10479a2 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:46 -0700 Subject: coredump: kill mm->core_done Now that we have core_state->dumper list we can use it to wake up the sub-threads waiting for the coredump completion. This uglifies the code and .text grows by 47 bytes, but otoh mm_struct lessens by sizeof(struct completion). Also, with this change we can decouple exit_mm() from the coredumping code. Signed-off-by: Oleg Nesterov Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 25 ++++++++++++++++++++++--- include/linux/mm_types.h | 4 +--- kernel/exit.c | 8 +++++++- 3 files changed, 30 insertions(+), 7 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index fe2873b8037..bff43aeb235 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1602,7 +1602,6 @@ static int coredump_wait(int exit_code, struct core_state *core_state) struct completion *vfork_done; int core_waiters; - init_completion(&mm->core_done); init_completion(&core_state->startup); core_state->dumper.task = tsk; core_state->dumper.next = NULL; @@ -1628,6 +1627,27 @@ fail: return core_waiters; } +static void coredump_finish(struct mm_struct *mm) +{ + struct core_thread *curr, *next; + struct task_struct *task; + + next = mm->core_state->dumper.next; + while ((curr = next) != NULL) { + next = curr->next; + task = curr->task; + /* + * see exit_mm(), curr->task must not see + * ->task == NULL before we read ->next. + */ + smp_mb(); + curr->task = NULL; + wake_up_process(task); + } + + mm->core_state = NULL; +} + /* * set_dumpable converts traditional three-value dumpable to two flags and * stores them into mm->flags. It modifies lower two bits of mm->flags, but @@ -1812,8 +1832,7 @@ fail_unlock: argv_free(helper_argv); current->fsuid = fsuid; - complete_all(&mm->core_done); - mm->core_state = NULL; + coredump_finish(mm); fail: return retval; } diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 4d0d0abc79f..746f975b58e 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -229,9 +229,7 @@ struct mm_struct { unsigned long flags; /* Must use atomic bitops to access the bits */ - /* coredumping support */ - struct core_state *core_state; - struct completion core_done; + struct core_state *core_state; /* coredumping support */ /* aio bits */ rwlock_t ioctx_list_lock; /* aio lock */ diff --git a/kernel/exit.c b/kernel/exit.c index b66f0d55c79..8a4d4d12e29 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -691,7 +691,13 @@ static void exit_mm(struct task_struct * tsk) if (atomic_dec_and_test(&core_state->nr_threads)) complete(&core_state->startup); - wait_for_completion(&mm->core_done); + for (;;) { + set_task_state(tsk, TASK_UNINTERRUPTIBLE); + if (!self.task) /* see coredump_finish() */ + break; + schedule(); + } + __set_task_state(tsk, TASK_RUNNING); down_read(&mm->mmap_sem); } atomic_inc(&mm->mm_count); -- cgit v1.2.3-70-g09d2 From 565b9b14e7f48131bca58840aa404bbef058fa89 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:47 -0700 Subject: coredump: format_corename: fix the "core_uses_pid" logic I don't understand why the multi-thread coredump implies the core_uses_pid behaviour, but we shouldn't use mm->mm_users for that. This counter can be incremented by get_task_mm(). Use the valued returned by coredump_wait() instead. Also, remove the "const char *pattern" argument, format_corename() can use core_pattern directly. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Oleg Nesterov Cc: Roland McGrath Cc: Alan Cox Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index bff43aeb235..5e559013e30 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1379,17 +1379,14 @@ EXPORT_SYMBOL(set_binfmt); * name into corename, which must have space for at least * CORENAME_MAX_SIZE bytes plus one byte for the zero terminator. */ -static int format_corename(char *corename, const char *pattern, long signr) +static int format_corename(char *corename, int nr_threads, long signr) { - const char *pat_ptr = pattern; + const char *pat_ptr = core_pattern; + int ispipe = (*pat_ptr == '|'); char *out_ptr = corename; char *const out_end = corename + CORENAME_MAX_SIZE; int rc; int pid_in_pattern = 0; - int ispipe = 0; - - if (*pattern == '|') - ispipe = 1; /* Repeat as long as we have more pattern to process and more output space */ @@ -1490,7 +1487,7 @@ static int format_corename(char *corename, const char *pattern, long signr) * and core_uses_pid is set, then .%pid will be appended to * the filename. Do not do this for piped commands. */ if (!ispipe && !pid_in_pattern - && (core_uses_pid || atomic_read(¤t->mm->mm_users) != 1)) { + && (core_uses_pid || nr_threads)) { rc = snprintf(out_ptr, out_end - out_ptr, ".%d", task_tgid_vnr(current)); if (rc > out_end - out_ptr) @@ -1753,7 +1750,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) * uses lock_kernel() */ lock_kernel(); - ispipe = format_corename(corename, core_pattern, signr); + ispipe = format_corename(corename, retval, signr); unlock_kernel(); /* * Don't bother to check the RLIMIT_CORE value if core_pattern points -- cgit v1.2.3-70-g09d2 From 6341c393fcc37d58727865f1ee2f65e632e9d4f0 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 25 Jul 2008 19:45:44 -0700 Subject: tracehook: exec This moves all the ptrace hooks related to exec into tracehook.h inlines. This also lifts the calls for tracing out of the binfmt load_binary hooks into search_binary_handler() after it calls into the binfmt module. This change has no effect, since all the binfmt modules' load_binary functions did the call at the end on success, and now search_binary_handler() does it immediately after return if successful. We consolidate the repeated code, and binfmt modules no longer need to import ptrace_notify(). Signed-off-by: Roland McGrath Cc: Oleg Nesterov Reviewed-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/ia32/ia32_aout.c | 6 ------ fs/binfmt_aout.c | 6 ------ fs/binfmt_elf.c | 6 ------ fs/binfmt_elf_fdpic.c | 7 ------- fs/binfmt_flat.c | 3 --- fs/binfmt_som.c | 2 -- fs/exec.c | 12 ++++-------- include/linux/tracehook.h | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 8 files changed, 50 insertions(+), 38 deletions(-) (limited to 'fs/exec.c') diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c index 58cccb6483b..a0e1dbe67dc 100644 --- a/arch/x86/ia32/ia32_aout.c +++ b/arch/x86/ia32/ia32_aout.c @@ -441,12 +441,6 @@ beyond_if: regs->r8 = regs->r9 = regs->r10 = regs->r11 = regs->r12 = regs->r13 = regs->r14 = regs->r15 = 0; set_fs(USER_DS); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } return 0; } diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index ba4cddb92f1..204cfd1d767 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -444,12 +444,6 @@ beyond_if: regs->gp = ex.a_gpvalue; #endif start_thread(regs, ex.a_entry, current->mm->start_stack); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify ((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } return 0; } diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 3b6ff854d98..655ed8d30a8 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1003,12 +1003,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) #endif start_thread(regs, elf_entry, bprm->p); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify ((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } retval = 0; out: kfree(loc); diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 1b59b1edf26..fdeadab2f18 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -433,13 +433,6 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, entryaddr = interp_params.entry_addr ?: exec_params.entry_addr; start_thread(regs, entryaddr, current->mm->start_stack); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } - retval = 0; error: diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 2cb1acda3a8..56372ecf169 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -920,9 +920,6 @@ static int load_flat_binary(struct linux_binprm * bprm, struct pt_regs * regs) start_thread(regs, start_addr, current->mm->start_stack); - if (current->ptrace & PT_PTRACED) - send_sig(SIGTRAP, current, 0); - return 0; } diff --git a/fs/binfmt_som.c b/fs/binfmt_som.c index fdc36bfd6a7..68be580ba28 100644 --- a/fs/binfmt_som.c +++ b/fs/binfmt_som.c @@ -274,8 +274,6 @@ load_som_binary(struct linux_binprm * bprm, struct pt_regs * regs) map_hpux_gateway_page(current,current->mm); start_thread_som(regs, som_entry, bprm->p); - if (current->ptrace & PT_PTRACED) - send_sig(SIGTRAP, current, 0); return 0; /* error cleanup */ diff --git a/fs/exec.c b/fs/exec.c index 5e559013e30..b8792a13153 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -42,13 +42,13 @@ #include #include #include -#include #include #include #include #include #include #include +#include #include #include @@ -1071,13 +1071,8 @@ EXPORT_SYMBOL(prepare_binprm); static int unsafe_exec(struct task_struct *p) { - int unsafe = 0; - if (p->ptrace & PT_PTRACED) { - if (p->ptrace & PT_PTRACE_CAP) - unsafe |= LSM_UNSAFE_PTRACE_CAP; - else - unsafe |= LSM_UNSAFE_PTRACE; - } + int unsafe = tracehook_unsafe_exec(p); + if (atomic_read(&p->fs->count) > 1 || atomic_read(&p->files->count) > 1 || atomic_read(&p->sighand->count) > 1) @@ -1214,6 +1209,7 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs) read_unlock(&binfmt_lock); retval = fn(bprm, regs); if (retval >= 0) { + tracehook_report_exec(fmt, bprm, regs); put_binfmt(fmt); allow_write_access(bprm->file); if (bprm->file) diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h index bea0f3eeff5..6276353709c 100644 --- a/include/linux/tracehook.h +++ b/include/linux/tracehook.h @@ -48,5 +48,51 @@ #include #include +#include +struct linux_binprm; + +/** + * tracehook_unsafe_exec - check for exec declared unsafe due to tracing + * @task: current task doing exec + * + * Return %LSM_UNSAFE_* bits applied to an exec because of tracing. + * + * Called with task_lock() held on @task. + */ +static inline int tracehook_unsafe_exec(struct task_struct *task) +{ + int unsafe = 0; + int ptrace = task_ptrace(task); + if (ptrace & PT_PTRACED) { + if (ptrace & PT_PTRACE_CAP) + unsafe |= LSM_UNSAFE_PTRACE_CAP; + else + unsafe |= LSM_UNSAFE_PTRACE; + } + return unsafe; +} + +/** + * tracehook_report_exec - a successful exec was completed + * @fmt: &struct linux_binfmt that performed the exec + * @bprm: &struct linux_binprm containing exec details + * @regs: user-mode register state + * + * An exec just completed, we are shortly going to return to user mode. + * The freshly initialized register state can be seen and changed in @regs. + * The name, file and other pointers in @bprm are still on hand to be + * inspected, but will be freed as soon as this returns. + * + * Called with no locks, but with some kernel resources held live + * and a reference on @fmt->module. + */ +static inline void tracehook_report_exec(struct linux_binfmt *fmt, + struct linux_binprm *bprm, + struct pt_regs *regs) +{ + if (!ptrace_event(PT_TRACE_EXEC, PTRACE_EVENT_EXEC, 0) && + unlikely(task_ptrace(current) & PT_PTRACED)) + send_sig(SIGTRAP, current, 0); +} #endif /* */ -- cgit v1.2.3-70-g09d2 From b77b0646ef4efe31a7449bb3d9360fd00f95433d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 17 Jul 2008 09:37:02 -0400 Subject: [PATCH] pass MAY_OPEN to vfs_permission() explicitly ... and get rid of the last "let's deduce mask from nameidata->flags" bit. Signed-off-by: Al Viro --- fs/exec.c | 4 ++-- fs/namei.c | 13 ++++--------- include/linux/security.h | 7 +++---- security/capability.c | 3 +-- security/security.c | 4 ++-- security/selinux/hooks.c | 5 ++--- security/smack/smack_lsm.c | 3 +-- 7 files changed, 15 insertions(+), 24 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index b8792a13153..0ba5d355c5a 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -118,7 +118,7 @@ asmlinkage long sys_uselib(const char __user * library) if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) goto exit; - error = vfs_permission(&nd, MAY_READ | MAY_EXEC); + error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN); if (error) goto exit; @@ -666,7 +666,7 @@ struct file *open_exec(const char *name) struct inode *inode = nd.path.dentry->d_inode; file = ERR_PTR(-EACCES); if (S_ISREG(inode->i_mode)) { - int err = vfs_permission(&nd, MAY_EXEC); + int err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN); file = ERR_PTR(err); if (!err) { file = nameidata_to_filp(&nd, diff --git a/fs/namei.c b/fs/namei.c index 33dcaf025c4..6b0e8e5e079 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -263,12 +263,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) /* Ordinary permission routines do not understand MAY_APPEND. */ if (inode->i_op && inode->i_op->permission) { - int extra = 0; - if (nd) { - if (nd->flags & LOOKUP_OPEN) - extra |= MAY_OPEN; - } - retval = inode->i_op->permission(inode, mask | extra); + retval = inode->i_op->permission(inode, mask); if (!retval) { /* * Exec permission on a regular file is denied if none @@ -292,7 +287,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) return retval; return security_inode_permission(inode, - mask & (MAY_READ|MAY_WRITE|MAY_EXEC), nd); + mask & (MAY_READ|MAY_WRITE|MAY_EXEC)); } /** @@ -492,7 +487,7 @@ static int exec_permission_lite(struct inode *inode, return -EACCES; ok: - return security_inode_permission(inode, MAY_EXEC, nd); + return security_inode_permission(inode, MAY_EXEC); } /* @@ -1692,7 +1687,7 @@ struct file *do_filp_open(int dfd, const char *pathname, int will_write; int flag = open_to_namei_flags(open_flag); - acc_mode = ACC_MODE(flag); + acc_mode = MAY_OPEN | ACC_MODE(flag); /* O_TRUNC implies we need access checks for write permissions */ if (flag & O_TRUNC) diff --git a/include/linux/security.h b/include/linux/security.h index f0e9adb22ac..fd96e7f8a6f 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -1362,7 +1362,7 @@ struct security_operations { struct inode *new_dir, struct dentry *new_dentry); int (*inode_readlink) (struct dentry *dentry); int (*inode_follow_link) (struct dentry *dentry, struct nameidata *nd); - int (*inode_permission) (struct inode *inode, int mask, struct nameidata *nd); + int (*inode_permission) (struct inode *inode, int mask); int (*inode_setattr) (struct dentry *dentry, struct iattr *attr); int (*inode_getattr) (struct vfsmount *mnt, struct dentry *dentry); void (*inode_delete) (struct inode *inode); @@ -1628,7 +1628,7 @@ int security_inode_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry); int security_inode_readlink(struct dentry *dentry); int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd); -int security_inode_permission(struct inode *inode, int mask, struct nameidata *nd); +int security_inode_permission(struct inode *inode, int mask); int security_inode_setattr(struct dentry *dentry, struct iattr *attr); int security_inode_getattr(struct vfsmount *mnt, struct dentry *dentry); void security_inode_delete(struct inode *inode); @@ -2021,8 +2021,7 @@ static inline int security_inode_follow_link(struct dentry *dentry, return 0; } -static inline int security_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static inline int security_inode_permission(struct inode *inode, int mask) { return 0; } diff --git a/security/capability.c b/security/capability.c index 5b01c0b0242..63d10da515a 100644 --- a/security/capability.c +++ b/security/capability.c @@ -211,8 +211,7 @@ static int cap_inode_follow_link(struct dentry *dentry, return 0; } -static int cap_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int cap_inode_permission(struct inode *inode, int mask) { return 0; } diff --git a/security/security.c b/security/security.c index 59f23b5918b..78ed3ffde24 100644 --- a/security/security.c +++ b/security/security.c @@ -429,11 +429,11 @@ int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd) return security_ops->inode_follow_link(dentry, nd); } -int security_inode_permission(struct inode *inode, int mask, struct nameidata *nd) +int security_inode_permission(struct inode *inode, int mask) { if (unlikely(IS_PRIVATE(inode))) return 0; - return security_ops->inode_permission(inode, mask, nd); + return security_ops->inode_permission(inode, mask); } int security_inode_setattr(struct dentry *dentry, struct iattr *attr) diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 3481cde5bf1..5ba13908b5b 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -2624,12 +2624,11 @@ static int selinux_inode_follow_link(struct dentry *dentry, struct nameidata *na return dentry_has_perm(current, NULL, dentry, FILE__READ); } -static int selinux_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int selinux_inode_permission(struct inode *inode, int mask) { int rc; - rc = secondary_ops->inode_permission(inode, mask, nd); + rc = secondary_ops->inode_permission(inode, mask); if (rc) return rc; diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index ee5a51cbc5e..1b40e558f98 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -522,8 +522,7 @@ static int smack_inode_rename(struct inode *old_inode, * * Returns 0 if access is permitted, -EACCES otherwise */ -static int smack_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int smack_inode_permission(struct inode *inode, int mask) { /* * No permission to check. Existence test. Yup, it's there. -- cgit v1.2.3-70-g09d2 From e56b6a5dda1a36ffaa532df6f975ea324298fa4d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 19 May 2008 07:53:34 +0200 Subject: Re: [PATCH 3/6] vfs: open_exec cleanup On Mon, May 19, 2008 at 12:01:49AM +0200, Marcin Slusarz wrote: > open_exec is needlessly indented, calls ERR_PTR with 0 argument > (which is not valid errno) and jumps into middle of function > just to return value. > So clean it up a bit. Still looks rather messy. See below for a better version. Signed-off-by: Christoph Hellwig Signed-off-by: Al Viro --- fs/exec.c | 58 ++++++++++++++++++++++++++++++---------------------------- 1 file changed, 30 insertions(+), 28 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index 0ba5d355c5a..346e3f69c6e 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -656,38 +656,40 @@ EXPORT_SYMBOL(setup_arg_pages); struct file *open_exec(const char *name) { struct nameidata nd; - int err; struct file *file; + int err; - err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd, FMODE_READ|FMODE_EXEC); - file = ERR_PTR(err); - - if (!err) { - struct inode *inode = nd.path.dentry->d_inode; - file = ERR_PTR(-EACCES); - if (S_ISREG(inode->i_mode)) { - int err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN); - file = ERR_PTR(err); - if (!err) { - file = nameidata_to_filp(&nd, - O_RDONLY|O_LARGEFILE); - if (!IS_ERR(file)) { - err = deny_write_access(file); - if (err) { - fput(file); - file = ERR_PTR(err); - } - } -out: - return file; - } - } - release_open_intent(&nd); - path_put(&nd.path); + err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd, + FMODE_READ|FMODE_EXEC); + if (err) + goto out; + + err = -EACCES; + if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) + goto out_path_put; + + err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN); + if (err) + goto out_path_put; + + file = nameidata_to_filp(&nd, O_RDONLY|O_LARGEFILE); + if (IS_ERR(file)) + return file; + + err = deny_write_access(file); + if (err) { + fput(file); + goto out; } - goto out; -} + return file; + + out_path_put: + release_open_intent(&nd); + path_put(&nd.path); + out: + return ERR_PTR(err); +} EXPORT_SYMBOL(open_exec); int kernel_read(struct file *file, unsigned long offset, -- cgit v1.2.3-70-g09d2 From 30524472c2f728c20d6bf35191042a5d455c0a64 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 22 Jul 2008 00:02:33 -0400 Subject: [PATCH] take noexec checks to very few callers that care Signed-off-by: Al Viro --- fs/exec.c | 7 +++++++ fs/namei.c | 9 --------- fs/open.c | 10 ++++++++++ 3 files changed, 17 insertions(+), 9 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index 346e3f69c6e..eca58c29ede 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -118,6 +118,10 @@ asmlinkage long sys_uselib(const char __user * library) if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) goto exit; + error = -EACCES; + if (nd.path.mnt->mnt_flags & MNT_NOEXEC) + goto exit; + error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN); if (error) goto exit; @@ -668,6 +672,9 @@ struct file *open_exec(const char *name) if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) goto out_path_put; + if (nd.path.mnt->mnt_flags & MNT_NOEXEC) + goto out_path_put; + err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN); if (err) goto out_path_put; diff --git a/fs/namei.c b/fs/namei.c index 6d75430358a..396cb3e5c36 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -252,15 +252,6 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) return -EACCES; } - if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) { - /* - * MAY_EXEC on regular files is denied if the fs is mounted - * with the "noexec" flag. - */ - if (mnt && (mnt->mnt_flags & MNT_NOEXEC)) - return -EACCES; - } - /* Ordinary permission routines do not understand MAY_APPEND. */ if (inode->i_op && inode->i_op->permission) { retval = inode->i_op->permission(inode, mask); diff --git a/fs/open.c b/fs/open.c index 3b3c43674be..d5e421ad0cf 100644 --- a/fs/open.c +++ b/fs/open.c @@ -461,6 +461,16 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) if (res) goto out; + if ((mode & MAY_EXEC) && S_ISREG(nd.path.dentry->d_inode->i_mode)) { + /* + * MAY_EXEC on regular files is denied if the fs is mounted + * with the "noexec" flag. + */ + res = -EACCES; + if (nd.path.mnt->mnt_flags & MNT_NOEXEC) + goto out_path_release; + } + res = vfs_permission(&nd, mode | MAY_ACCESS); /* SuS v2 requires we report a read only fs too */ if(res || !(mode & S_IWOTH) || -- cgit v1.2.3-70-g09d2 From 964bd183624c03680796b63b4ab97ee3905a806a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 26 Jul 2008 03:33:14 -0400 Subject: [PATCH] get rid of __user_path_lookup_open Signed-off-by: Al Viro --- fs/exec.c | 14 ++++++++++---- fs/namei.c | 13 ------------- include/linux/namei.h | 1 - 3 files changed, 10 insertions(+), 18 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index eca58c29ede..9696bbf0f0b 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -106,11 +106,17 @@ static inline void put_binfmt(struct linux_binfmt * fmt) */ asmlinkage long sys_uselib(const char __user * library) { - struct file * file; + struct file *file; struct nameidata nd; - int error; - - error = __user_path_lookup_open(library, LOOKUP_FOLLOW, &nd, FMODE_READ|FMODE_EXEC); + char *tmp = getname(library); + int error = PTR_ERR(tmp); + + if (!IS_ERR(tmp)) { + error = path_lookup_open(AT_FDCWD, tmp, + LOOKUP_FOLLOW, &nd, + FMODE_READ|FMODE_EXEC); + putname(tmp); + } if (error) goto out; diff --git a/fs/namei.c b/fs/namei.c index 38ceb6e06eb..a7b0a0b8012 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1193,19 +1193,6 @@ static int path_lookup_create(int dfd, const char *name, nd, open_flags, create_mode); } -int __user_path_lookup_open(const char __user *name, unsigned int lookup_flags, - struct nameidata *nd, int open_flags) -{ - char *tmp = getname(name); - int err = PTR_ERR(tmp); - - if (!IS_ERR(tmp)) { - err = __path_lookup_intent_open(AT_FDCWD, tmp, lookup_flags, nd, open_flags, 0); - putname(tmp); - } - return err; -} - static struct dentry *__lookup_hash(struct qstr *name, struct dentry *base, struct nameidata *nd) { diff --git a/include/linux/namei.h b/include/linux/namei.h index 00888ff6950..68f8c3203c8 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -65,7 +65,6 @@ extern int path_lookup(const char *, unsigned, struct nameidata *); extern int vfs_path_lookup(struct dentry *, struct vfsmount *, const char *, unsigned int, struct nameidata *); -extern int __user_path_lookup_open(const char __user *, unsigned lookup_flags, struct nameidata *nd, int open_flags); extern int path_lookup_open(int dfd, const char *name, unsigned lookup_flags, struct nameidata *, int open_flags); extern struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry, int (*open)(struct inode *, struct file *)); -- cgit v1.2.3-70-g09d2 From ca5b172bd2b2fe489e7ba11cedd46ddf772d132f Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 28 Jul 2008 15:46:18 -0700 Subject: exec: include pagemap.h again to fix build Fix compilation errors on avr32 and without CONFIG_SWAP, introduced by ba92a43dbaee339cf5915ef766d3d3ffbaaf103c ("exec: remove some includes") In file included from include/asm/tlb.h:24, from fs/exec.c:55: include/asm-generic/tlb.h: In function 'tlb_flush_mmu': include/asm-generic/tlb.h:76: error: implicit declaration of function 'release_pages' include/asm-generic/tlb.h: In function 'tlb_remove_page': include/asm-generic/tlb.h:105: error: implicit declaration of function 'page_cache_release' make[1]: *** [fs/exec.o] Error 1 This straightforward part-revert is nobody's favourite patch to address the underlying tlb.h needs swap.h needs pagemap.h (but sparc won't like that) mess; but appropriate to fix the build now before any overhaul. Reported-by: Yoichi Yuasa Reported-by: Haavard Skinnemoen Signed-off-by: Hugh Dickins Tested-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index 9696bbf0f0b..32993beecbe 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3-70-g09d2