From 3859a271a003aba01e45b85c9d8b355eb7bf25f9 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 28 Oct 2016 01:22:25 -0700 Subject: randstruct: Mark various structs for randomization This marks many critical kernel structures for randomization. These are structures that have been targeted in the past in security exploits, or contain functions pointers, pointers to function pointer tables, lists, workqueues, ref-counters, credentials, permissions, or are otherwise sensitive. This initial list was extracted from Brad Spengler/PaX Team's code in the last public patch of grsecurity/PaX based on my understanding of the code. Changes or omissions from the original code are mine and don't reflect the original grsecurity/PaX code. Left out of this list is task_struct, which requires special handling and will be covered in a subsequent patch. Signed-off-by: Kees Cook --- kernel/futex.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index 357348a6cf6b..5616511abf39 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -212,7 +212,7 @@ struct futex_pi_state { atomic_t refcount; union futex_key key; -}; +} __randomize_layout; /** * struct futex_q - The hashed futex queue entry, one per waiting task @@ -246,7 +246,7 @@ struct futex_q { struct rt_mutex_waiter *rt_waiter; union futex_key *requeue_pi_key; u32 bitset; -}; +} __randomize_layout; static const struct futex_q futex_q_init = { /* list gets initialized in queue_me()*/ -- cgit v1.2.3 From 4cc7c1864bbd4cf80f6bdc8ba3217de5aa5f4688 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 5 Jul 2017 16:24:49 +0100 Subject: bpf: Implement show_options Implement the show_options superblock op for bpf as part of a bid to get rid of s_options and generic_show_options() to make it easier to implement a context-based mount where the mount options can be passed individually over a file descriptor. Signed-off-by: David Howells cc: Alexei Starovoitov cc: Daniel Borkmann cc: netdev@vger.kernel.org Signed-off-by: Al Viro --- kernel/bpf/inode.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c index 9bbd33497d3d..e833ed914358 100644 --- a/kernel/bpf/inode.c +++ b/kernel/bpf/inode.c @@ -377,10 +377,22 @@ static void bpf_evict_inode(struct inode *inode) bpf_any_put(inode->i_private, type); } +/* + * Display the mount options in /proc/mounts. + */ +static int bpf_show_options(struct seq_file *m, struct dentry *root) +{ + umode_t mode = d_inode(root)->i_mode & S_IALLUGO & ~S_ISVTX; + + if (mode != S_IRWXUGO) + seq_printf(m, ",mode=%o", mode); + return 0; +} + static const struct super_operations bpf_super_ops = { .statfs = simple_statfs, .drop_inode = generic_delete_inode, - .show_options = generic_show_options, + .show_options = bpf_show_options, .evict_inode = bpf_evict_inode, }; @@ -434,8 +446,6 @@ static int bpf_fill_super(struct super_block *sb, void *data, int silent) struct inode *inode; int ret; - save_mount_options(sb, data); - ret = bpf_parse_options(data, &opts); if (ret) return ret; -- cgit v1.2.3 From 99c621d704cf1c4eb74c3c42e674edf3df64f92d Mon Sep 17 00:00:00 2001 From: Michael Sartain Date: Wed, 5 Jul 2017 22:07:15 -0600 Subject: tracing: Add saved_tgids file to show cached pid to tgid mappings Export the cached pid / tgid mappings in debugfs tracing saved_tgids file. This allows user apps to translate the pids from a trace to their respective thread group. Example saved_tgids file with pid / tgid values separated by ' ': # cat saved_tgids 1048 1048 1047 1047 7 7 1049 1047 1054 1047 1053 1047 Link: http://lkml.kernel.org/r/20170630004023.064965233@goodmis.org Link: http://lkml.kernel.org/r/20170706040713.unwkumbta5menygi@mikesart-cos Reviewed-by: Joel Fernandes Signed-off-by: Michael Sartain Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 73 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 00e2e4169b1e..f079a8ca1117 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -4688,6 +4688,76 @@ static const struct file_operations tracing_readme_fops = { .llseek = generic_file_llseek, }; +static void *saved_tgids_next(struct seq_file *m, void *v, loff_t *pos) +{ + int *ptr = v; + + if (*pos || m->count) + ptr++; + + (*pos)++; + + for (; ptr <= &tgid_map[PID_MAX_DEFAULT]; ptr++) { + if (trace_find_tgid(*ptr)) + return ptr; + } + + return NULL; +} + +static void *saved_tgids_start(struct seq_file *m, loff_t *pos) +{ + void *v; + loff_t l = 0; + + if (!tgid_map) + return NULL; + + v = &tgid_map[0]; + while (l <= *pos) { + v = saved_tgids_next(m, v, &l); + if (!v) + return NULL; + } + + return v; +} + +static void saved_tgids_stop(struct seq_file *m, void *v) +{ +} + +static int saved_tgids_show(struct seq_file *m, void *v) +{ + int pid = (int *)v - tgid_map; + + seq_printf(m, "%d %d\n", pid, trace_find_tgid(pid)); + return 0; +} + +static const struct seq_operations tracing_saved_tgids_seq_ops = { + .start = saved_tgids_start, + .stop = saved_tgids_stop, + .next = saved_tgids_next, + .show = saved_tgids_show, +}; + +static int tracing_saved_tgids_open(struct inode *inode, struct file *filp) +{ + if (tracing_disabled) + return -ENODEV; + + return seq_open(filp, &tracing_saved_tgids_seq_ops); +} + + +static const struct file_operations tracing_saved_tgids_fops = { + .open = tracing_saved_tgids_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + static void *saved_cmdlines_next(struct seq_file *m, void *v, loff_t *pos) { unsigned int *ptr = v; @@ -7920,6 +7990,9 @@ static __init int tracer_init_tracefs(void) trace_create_file("saved_cmdlines_size", 0644, d_tracer, NULL, &tracing_saved_cmdlines_size_fops); + trace_create_file("saved_tgids", 0444, d_tracer, + NULL, &tracing_saved_tgids_fops); + trace_eval_init(); trace_create_eval_file(d_tracer); -- cgit v1.2.3 From eaf260ac04d9b4cf9f458d5c97555bfff2da526e Mon Sep 17 00:00:00 2001 From: Joel Fernandes Date: Thu, 6 Jul 2017 16:00:21 -0700 Subject: tracing: Treat recording comm for idle task as a success Currently we stop recording comm for non-idle tasks when switching from/to idle task since we treat that as a record failure. Fix that by treat recording of comm for idle task as a success. Link: http://lkml.kernel.org/r/20170706230023.17942-1-joelaf@google.com Cc: kernel-team@android.com Cc: Ingo Molnar Reported-by: Michael Sartain Signed-off-by: Joel Fernandes Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index f079a8ca1117..6722d86f2af5 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1916,7 +1916,11 @@ static int trace_save_cmdline(struct task_struct *tsk) { unsigned pid, idx; - if (!tsk->pid || unlikely(tsk->pid > PID_MAX_DEFAULT)) + /* treat recording of idle task as a success */ + if (!tsk->pid) + return 1; + + if (unlikely(tsk->pid > PID_MAX_DEFAULT)) return 0; /* -- cgit v1.2.3 From bd45d34d25720a820021c8ea45de5cd607eace64 Mon Sep 17 00:00:00 2001 From: Joel Fernandes Date: Thu, 6 Jul 2017 16:00:22 -0700 Subject: tracing: Treat recording tgid for idle task as a success Currently we stop recording tgid for non-idle tasks when switching from/to idle task since we treat that as a record failure. Fix that by treat recording of tgid for idle task as a success. Link: http://lkml.kernel.org/r/20170706230023.17942-2-joelaf@google.com Cc: kernel-team@android.com Cc: Ingo Molnar Reported-by: Michael Sartain Signed-off-by: Joel Fernandes Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 6722d86f2af5..aee11e3a394f 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2006,7 +2006,11 @@ int trace_find_tgid(int pid) static int trace_save_tgid(struct task_struct *tsk) { - if (unlikely(!tgid_map || !tsk->pid || tsk->pid > PID_MAX_DEFAULT)) + /* treat recording of idle task as a success */ + if (!tsk->pid) + return 1; + + if (unlikely(!tgid_map || tsk->pid > PID_MAX_DEFAULT)) return 0; tgid_map[tsk->pid] = tsk->tgid; -- cgit v1.2.3 From 29b1a8ad7df4528b862a79e3d5fb0936f4d199c7 Mon Sep 17 00:00:00 2001 From: Joel Fernandes Date: Thu, 6 Jul 2017 16:00:23 -0700 Subject: tracing: Attempt to record other information even if some fail In recent patches where we record comm and tgid at the same time, we skip continuing to record if any fail. Fix that by trying to record as many things as we can even if some couldn't be recorded. If any information isn't recorded, then we don't set trace_taskinfo_save as before. Link: http://lkml.kernel.org/r/20170706230023.17942-3-joelaf@google.com Cc: kernel-team@android.com Cc: Ingo Molnar Signed-off-by: Joel Fernandes Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 32 ++++++++++++++++++++++++-------- 1 file changed, 24 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index aee11e3a394f..92af8fd1429b 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2037,11 +2037,20 @@ static bool tracing_record_taskinfo_skip(int flags) */ void tracing_record_taskinfo(struct task_struct *task, int flags) { + bool done; + if (tracing_record_taskinfo_skip(flags)) return; - if ((flags & TRACE_RECORD_CMDLINE) && !trace_save_cmdline(task)) - return; - if ((flags & TRACE_RECORD_TGID) && !trace_save_tgid(task)) + + /* + * Record as much task information as possible. If some fail, continue + * to try to record the others. + */ + done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(task); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(task); + + /* If recording any information failed, retry again soon. */ + if (!done) return; __this_cpu_write(trace_taskinfo_save, false); @@ -2058,15 +2067,22 @@ void tracing_record_taskinfo(struct task_struct *task, int flags) void tracing_record_taskinfo_sched_switch(struct task_struct *prev, struct task_struct *next, int flags) { + bool done; + if (tracing_record_taskinfo_skip(flags)) return; - if ((flags & TRACE_RECORD_CMDLINE) && - (!trace_save_cmdline(prev) || !trace_save_cmdline(next))) - return; + /* + * Record as much task information as possible. If some fail, continue + * to try to record the others. + */ + done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(prev); + done &= !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(next); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(prev); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(next); - if ((flags & TRACE_RECORD_TGID) && - (!trace_save_tgid(prev) || !trace_save_tgid(next))) + /* If recording any information failed, retry again soon. */ + if (!done) return; __this_cpu_write(trace_taskinfo_save, false); -- cgit v1.2.3 From fca18a47cf3eb8425ec19c2dfc374f3d04f5219f Mon Sep 17 00:00:00 2001 From: "Naveen N. Rao" Date: Sat, 8 Jul 2017 00:27:30 +0530 Subject: trace/kprobes: Sanitize derived event names When we derive event names, convert some expected symbols (such as ':' used to specify module:name and '.' present in some symbols) into underscores so that the event name is not rejected. Before this patch: # echo 'p kobject_example:foo_store' > kprobe_events trace_kprobe: Failed to allocate trace_probe.(-22) -sh: write error: Invalid argument After this patch: # echo 'p kobject_example:foo_store' > kprobe_events # cat kprobe_events p:kprobes/p_kobject_example_foo_store_0 kobject_example:foo_store Link: http://lkml.kernel.org/r/66c189e09e71361aba91dd4a5bd146a1b62a7a51.1499453040.git.naveen.n.rao@linux.vnet.ibm.com Acked-by: Masami Hiramatsu Signed-off-by: Naveen N. Rao Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_kprobe.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index c129fca6ec99..44fd819aa33d 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -598,6 +598,14 @@ static struct notifier_block trace_kprobe_module_nb = { .priority = 1 /* Invoked after kprobe module callback */ }; +/* Convert certain expected symbols into '_' when generating event names */ +static inline void sanitize_event_name(char *name) +{ + while (*name++ != '\0') + if (*name == ':' || *name == '.') + *name = '_'; +} + static int create_trace_kprobe(int argc, char **argv) { /* @@ -740,6 +748,7 @@ static int create_trace_kprobe(int argc, char **argv) else snprintf(buf, MAX_EVENT_NAME_LEN, "%c_0x%p", is_return ? 'r' : 'p', addr); + sanitize_event_name(buf); event = buf; } tk = alloc_trace_kprobe(group, event, addr, symbol, offset, maxactive, -- cgit v1.2.3 From dea1d0f5f1284e3defee4b8484d9fc230686cd42 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 11 Jul 2017 22:06:24 +0200 Subject: smp/hotplug: Replace BUG_ON and react useful The move of the unpark functions to the control thread moved the BUG_ON() there as well. While it made some sense in the idle thread of the upcoming CPU, it's bogus to crash the control thread on the already online CPU, especially as the function has a return value and the callsite is prepared to handle an error return. Replace it with a WARN_ON_ONCE() and return a proper error code. Fixes: 9cd4f1a4e7a8 ("smp/hotplug: Move unparking of percpu threads to the control CPU") Rightfully-ranted-at-by: Linux Torvalds Signed-off-by: Thomas Gleixner --- kernel/cpu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/cpu.c b/kernel/cpu.c index ab860453841d..eee033134262 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -279,7 +279,8 @@ static int bringup_wait_for_ap(unsigned int cpu) /* Wait for the CPU to reach CPUHP_AP_ONLINE_IDLE */ wait_for_completion(&st->done); - BUG_ON(!cpu_online(cpu)); + if (WARN_ON_ONCE((!cpu_online(cpu)))) + return -ECANCELED; /* Unpark the stopper thread and the hotplug thread of the target cpu */ stop_machine_unpark(cpu); -- cgit v1.2.3 From b11fb73743fc406204e0749ead18560aeda8b136 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Tue, 11 Jul 2017 15:43:24 -0400 Subject: tracing: Fixup trace file header alignment The addition of TGID to the tracing header added a check to see if TGID shoudl be displayed or not, and updated the header accordingly. Unfortunately, it broke the default header. Also add constant strings to use for spacing. This does remove the visibility of the header a bit, but cuts it down from the extended lines much greater than 80 characters. Before this change: # tracer: function # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU#|||| TIMESTAMP FUNCTION # | | | |||| | | swapper/0-1 [000] .... 0.277830: migration_init <-do_one_initcall swapper/0-1 [002] d... 13.861967: Unknown type 1201 swapper/0-1 [002] d..1 13.861970: Unknown type 1202 After this change: # tracer: function # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | swapper/0-1 [000] .... 0.278245: migration_init <-do_one_initcall swapper/0-1 [003] d... 13.861189: Unknown type 1201 swapper/0-1 [003] d..1 13.861192: Unknown type 1202 Cc: Joel Fernandes Fixes: 441dae8f2f29 ("tracing: Add support for display of tgid in trace output") Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 25 +++++++++++++++++-------- 1 file changed, 17 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 92af8fd1429b..dabd810a10cd 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3358,14 +3358,23 @@ static void print_func_help_header_irq(struct trace_buffer *buf, struct seq_file unsigned int flags) { bool tgid = flags & TRACE_ITER_RECORD_TGID; - - seq_printf(m, "# %s _-----=> irqs-off\n", tgid ? " " : ""); - seq_printf(m, "# %s / _----=> need-resched\n", tgid ? " " : ""); - seq_printf(m, "# %s| / _---=> hardirq/softirq\n", tgid ? " " : ""); - seq_printf(m, "# %s|| / _--=> preempt-depth\n", tgid ? " " : ""); - seq_printf(m, "# %s||| / delay\n", tgid ? " " : ""); - seq_printf(m, "# TASK-PID CPU#%s|||| TIMESTAMP FUNCTION\n", tgid ? " TGID " : ""); - seq_printf(m, "# | | | %s|||| | |\n", tgid ? " | " : ""); + const char tgid_space[] = " "; + const char space[] = " "; + + seq_printf(m, "# %s _-----=> irqs-off\n", + tgid ? tgid_space : space); + seq_printf(m, "# %s / _----=> need-resched\n", + tgid ? tgid_space : space); + seq_printf(m, "# %s| / _---=> hardirq/softirq\n", + tgid ? tgid_space : space); + seq_printf(m, "# %s|| / _--=> preempt-depth\n", + tgid ? tgid_space : space); + seq_printf(m, "# %s||| / delay\n", + tgid ? tgid_space : space); + seq_printf(m, "# TASK-PID CPU#%s|||| TIMESTAMP FUNCTION\n", + tgid ? " TGID " : space); + seq_printf(m, "# | | | %s|||| | |\n", + tgid ? " | " : space); } void -- cgit v1.2.3 From bbd1d27d863d5c0acee65ecd0c2e34035e1df5ea Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Tue, 11 Jul 2017 19:21:04 -0400 Subject: tracing: Do note expose stack_trace_filter without DYNAMIC_FTRACE The "stack_trace_filter" file only makes sense if DYNAMIC_FTRACE is configured in. If it is not, then the user can not filter any functions. Not only that, the open function causes warnings when DYNAMIC_FTRACE is not set. Link: http://lkml.kernel.org/r/20170710110521.600806-1-arnd@arndb.de Reported-by: Arnd Bergmann Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_stack.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index b4a751e8f9d6..a4df67cbc711 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -406,6 +406,8 @@ static const struct file_operations stack_trace_fops = { .release = seq_release, }; +#ifdef CONFIG_DYNAMIC_FTRACE + static int stack_trace_filter_open(struct inode *inode, struct file *file) { @@ -423,6 +425,8 @@ static const struct file_operations stack_trace_filter_fops = { .release = ftrace_regex_release, }; +#endif /* CONFIG_DYNAMIC_FTRACE */ + int stack_trace_sysctl(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, @@ -477,8 +481,10 @@ static __init int stack_trace_init(void) trace_create_file("stack_trace", 0444, d_tracer, NULL, &stack_trace_fops); +#ifdef CONFIG_DYNAMIC_FTRACE trace_create_file("stack_trace_filter", 0444, d_tracer, &trace_ops, &stack_trace_filter_fops); +#endif if (stack_trace_filter_buf[0]) ftrace_set_early_filter(&trace_ops, stack_trace_filter_buf, 1); -- cgit v1.2.3 From 69449bbd65687e8e5fb968a5a0c46089f6af6001 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 10 Jul 2017 10:44:03 +0200 Subject: ftrace: Hide cached module code for !CONFIG_MODULES When modules are disabled, we get a harmless build warning: kernel/trace/ftrace.c:4051:13: error: 'process_cached_mods' defined but not used [-Werror=unused-function] This adds the same #ifdef around the new code that exists around its caller. Link: http://lkml.kernel.org/r/20170710084413.1820568-1-arnd@arndb.de Fixes: d7fbf8df7ca0 ("ftrace: Implement cached modules tracing on module load") Signed-off-by: Arnd Bergmann Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 2953d558bbee..4706f0ed193e 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -3978,6 +3978,7 @@ static int ftrace_set_regex(struct ftrace_ops *ops, unsigned char *buf, int len, int reset, int enable); +#ifdef CONFIG_MODULES static void process_mod_list(struct list_head *head, struct ftrace_ops *ops, char *mod, bool enable) { @@ -4068,6 +4069,7 @@ static void process_cached_mods(const char *mod_name) kfree(mod); } +#endif /* * We register the module command as a template to show others how -- cgit v1.2.3 From 19d39a3810e7032f311ef83effdac40339b9d022 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 11 Jul 2017 23:41:52 +0200 Subject: genirq: Keep chip buslock across irq_request/release_resources() Moving the irq_request/release_resources() callbacks out of the spinlocked, irq disabled and bus locked region, unearthed an interesting abuse of the irq_bus_lock/irq_bus_sync_unlock() callbacks. The OMAP GPIO driver does merily power management inside of them. The irq_request_resources() callback of this GPIO irqchip calls a function which reads a GPIO register. That read aborts now because the clock of the GPIO block is not magically enabled via the irq_bus_lock() callback. Move the callbacks under the bus lock again to prevent this. In the free_irq() path this requires to drop the bus_lock before calling synchronize_irq() and reaquiring it before calling the irq_release_resources() callback. The bus lock can't be held because: 1) The data which has been changed between bus_lock/un_lock is cached in the irq chip driver private data and needs to go out to the irq chip via the slow bus (usually SPI or I2C) before calling synchronize_irq(). That's the reason why this bus_lock/unlock magic exists in the first place, as you cannot do SPI/I2C transactions while holding desc->lock with interrupts disabled. 2) synchronize_irq() will actually deadlock, if there is a handler on flight. These chips use threaded handlers for obvious reasons, as they allow to do SPI/I2C communication. When the threaded handler returns then bus_lock needs to be taken in irq_finalize_oneshot() as we need to talk to the actual irq chip once more. After that the threaded handler is marked done, which makes synchronize_irq() return. So if we hold bus_lock accross the synchronize_irq() call, the handler cannot mark itself done because it blocks on the bus lock. That in turn makes synchronize_irq() wait forever on the threaded handler to complete.... Add the missing unlock of desc->request_mutex in the error path of __free_irq() and add a bunch of comments to explain the locking and protection rules. Fixes: 46e48e257360 ("genirq: Move irq resource handling out of spinlocked region") Reported-and-tested-by: Sebastian Reichel Reported-and-tested-by: Tony Lindgren Reported-by: Pavel Machek Signed-off-by: Thomas Gleixner Not-longer-ranted-at-by: Linus Torvalds Cc: Linus Walleij Cc: Grygorii Strashko Cc: Marc Zyngier --- kernel/irq/manage.c | 63 ++++++++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 53 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 5624b2dd6b58..1d1a5b945ab4 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -1090,6 +1090,16 @@ setup_irq_thread(struct irqaction *new, unsigned int irq, bool secondary) /* * Internal function to register an irqaction - typically used to * allocate special interrupts that are part of the architecture. + * + * Locking rules: + * + * desc->request_mutex Provides serialization against a concurrent free_irq() + * chip_bus_lock Provides serialization for slow bus operations + * desc->lock Provides serialization against hard interrupts + * + * chip_bus_lock and desc->lock are sufficient for all other management and + * interrupt related functions. desc->request_mutex solely serializes + * request/free_irq(). */ static int __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) @@ -1167,20 +1177,35 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) if (desc->irq_data.chip->flags & IRQCHIP_ONESHOT_SAFE) new->flags &= ~IRQF_ONESHOT; + /* + * Protects against a concurrent __free_irq() call which might wait + * for synchronize_irq() to complete without holding the optional + * chip bus lock and desc->lock. + */ mutex_lock(&desc->request_mutex); + + /* + * Acquire bus lock as the irq_request_resources() callback below + * might rely on the serialization or the magic power management + * functions which are abusing the irq_bus_lock() callback, + */ + chip_bus_lock(desc); + + /* First installed action requests resources. */ if (!desc->action) { ret = irq_request_resources(desc); if (ret) { pr_err("Failed to request resources for %s (irq %d) on irqchip %s\n", new->name, irq, desc->irq_data.chip->name); - goto out_mutex; + goto out_bus_unlock; } } - chip_bus_lock(desc); - /* * The following block of code has to be executed atomically + * protected against a concurrent interrupt and any of the other + * management calls which are not serialized via + * desc->request_mutex or the optional bus lock. */ raw_spin_lock_irqsave(&desc->lock, flags); old_ptr = &desc->action; @@ -1286,10 +1311,8 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) ret = __irq_set_trigger(desc, new->flags & IRQF_TRIGGER_MASK); - if (ret) { - irq_release_resources(desc); + if (ret) goto out_unlock; - } } desc->istate &= ~(IRQS_AUTODETECT | IRQS_SPURIOUS_DISABLED | \ @@ -1385,12 +1408,10 @@ mismatch: out_unlock: raw_spin_unlock_irqrestore(&desc->lock, flags); - chip_bus_sync_unlock(desc); - if (!desc->action) irq_release_resources(desc); - -out_mutex: +out_bus_unlock: + chip_bus_sync_unlock(desc); mutex_unlock(&desc->request_mutex); out_thread: @@ -1472,6 +1493,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) WARN(1, "Trying to free already-free IRQ %d\n", irq); raw_spin_unlock_irqrestore(&desc->lock, flags); chip_bus_sync_unlock(desc); + mutex_unlock(&desc->request_mutex); return NULL; } @@ -1498,6 +1520,20 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) #endif raw_spin_unlock_irqrestore(&desc->lock, flags); + /* + * Drop bus_lock here so the changes which were done in the chip + * callbacks above are synced out to the irq chips which hang + * behind a slow bus (I2C, SPI) before calling synchronize_irq(). + * + * Aside of that the bus_lock can also be taken from the threaded + * handler in irq_finalize_oneshot() which results in a deadlock + * because synchronize_irq() would wait forever for the thread to + * complete, which is blocked on the bus lock. + * + * The still held desc->request_mutex() protects against a + * concurrent request_irq() of this irq so the release of resources + * and timing data is properly serialized. + */ chip_bus_sync_unlock(desc); unregister_handler_proc(irq, action); @@ -1530,8 +1566,15 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) } } + /* Last action releases resources */ if (!desc->action) { + /* + * Reaquire bus lock as irq_release_resources() might + * require it to deallocate resources over the slow bus. + */ + chip_bus_lock(desc); irq_release_resources(desc); + chip_bus_sync_unlock(desc); irq_remove_timings(desc); } -- cgit v1.2.3 From ab2f7cf141aa6734c4ca7525132d8cc236efee77 Mon Sep 17 00:00:00 2001 From: Vikram Mulukutla Date: Thu, 6 Jul 2017 10:53:20 -0700 Subject: cpufreq: schedutil: Fix sugov_start() versus sugov_update_shared() race With a shared policy in place, when one of the CPUs in the policy is hotplugged out and then brought back online, sugov_stop() and sugov_start() are called in order. sugov_stop() removes utilization hooks for each CPU in the policy and does nothing else in the for_each_cpu() loop. sugov_start() on the other hand iterates through the CPUs in the policy and re-initializes the per-cpu structure _and_ adds the utilization hook. This implies that the scheduler is allowed to invoke a CPU's utilization update hook when the rest of the per-cpu structures have yet to be re-inited. Apart from some strange values in tracepoints this doesn't cause a problem, but if we do end up accessing a pointer from the per-cpu sugov_cpu structure somewhere in the sugov_update_shared() path, we will likely see crashes since the memset for another CPU in the policy is free to race with sugov_update_shared from the CPU that is ready to go. So let's fix this now to first init all per-cpu structures, and then add the per-cpu utilization update hooks all at once. Signed-off-by: Vikram Mulukutla Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki --- kernel/sched/cpufreq_schedutil.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel') diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index 076a2e31951c..29a397067ffa 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -610,6 +610,11 @@ static int sugov_start(struct cpufreq_policy *policy) sg_cpu->sg_policy = sg_policy; sg_cpu->flags = SCHED_CPUFREQ_RT; sg_cpu->iowait_boost_max = policy->cpuinfo.max_freq; + } + + for_each_cpu(cpu, policy->cpus) { + struct sugov_cpu *sg_cpu = &per_cpu(sugov_cpu, cpu); + cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util, policy_is_shared(policy) ? sugov_update_shared : -- cgit v1.2.3 From 44925dfff05fd1a897992d278b15a6b6b55e79a7 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 12 Jul 2017 10:33:40 +0300 Subject: ftrace: Remove an unneeded NULL check "func" can't be NULL and it doesn't make sense to check because we've already derefenced it. Link: http://lkml.kernel.org/r/20170712073340.4enzeojeoupuds5a@mwanda Signed-off-by: Dan Carpenter Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 4706f0ed193e..5fb5b40b3ae8 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -3950,7 +3950,7 @@ static int cache_mod(struct trace_array *tr, continue; /* no func matches all */ - if (!func || strcmp(func, "*") == 0 || + if (strcmp(func, "*") == 0 || (ftrace_mod->func && strcmp(ftrace_mod->func, func) == 0)) { ret = 0; -- cgit v1.2.3 From 2e028c4fe12907f226b8221815f16c2486ad3aa7 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 12 Jul 2017 10:35:57 +0300 Subject: ftrace: Fix uninitialized variable in match_records() My static checker complains that if "func" is NULL then "clear_filter" is uninitialized. This seems like it could be true, although it's possible something subtle is happening that I haven't seen. kernel/trace/ftrace.c:3844 match_records() error: uninitialized symbol 'clear_filter'. Link: http://lkml.kernel.org/r/20170712073556.h6tkpjcdzjaozozs@mwanda Cc: stable@vger.kernel.org Fixes: f0a3b154bd7 ("ftrace: Clarify code for mod command") Signed-off-by: Dan Carpenter Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 5fb5b40b3ae8..53f6b6401cf0 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -3816,7 +3816,7 @@ match_records(struct ftrace_hash *hash, char *func, int len, char *mod) int exclude_mod = 0; int found = 0; int ret; - int clear_filter; + int clear_filter = 0; if (func) { func_g.type = filter_parse_regex(func, len, &func_g.search, -- cgit v1.2.3 From 112166f88cf83dd11486cf1818672d42b540865b Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 12 Jul 2017 14:33:11 -0700 Subject: kernel/fork.c: virtually mapped stacks: do not disable interrupts The reason to disable interrupts seems to be to avoid switching to a different processor while handling per cpu data using individual loads and stores. If we use per cpu RMV primitives we will not have to disable interrupts. Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1705171055130.5898@east.gentwo.org Signed-off-by: Christoph Lameter Cc: Andy Lutomirski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 16 +++++----------- 1 file changed, 5 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index 0f69a3e5281e..d2b9d7c31eaf 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -205,19 +205,17 @@ static unsigned long *alloc_thread_stack_node(struct task_struct *tsk, int node) void *stack; int i; - local_irq_disable(); for (i = 0; i < NR_CACHED_STACKS; i++) { - struct vm_struct *s = this_cpu_read(cached_stacks[i]); + struct vm_struct *s; + + s = this_cpu_xchg(cached_stacks[i], NULL); if (!s) continue; - this_cpu_write(cached_stacks[i], NULL); tsk->stack_vm_area = s; - local_irq_enable(); return s->addr; } - local_irq_enable(); stack = __vmalloc_node_range(THREAD_SIZE, THREAD_SIZE, VMALLOC_START, VMALLOC_END, @@ -245,19 +243,15 @@ static inline void free_thread_stack(struct task_struct *tsk) { #ifdef CONFIG_VMAP_STACK if (task_stack_vm_area(tsk)) { - unsigned long flags; int i; - local_irq_save(flags); for (i = 0; i < NR_CACHED_STACKS; i++) { - if (this_cpu_read(cached_stacks[i])) + if (this_cpu_cmpxchg(cached_stacks[i], + NULL, tsk->stack_vm_area) != NULL) continue; - this_cpu_write(cached_stacks[i], tsk->stack_vm_area); - local_irq_restore(flags); return; } - local_irq_restore(flags); vfree_atomic(tsk->stack); return; -- cgit v1.2.3 From 203e9e41219b4e7357104e525e91ac609fba2c6c Mon Sep 17 00:00:00 2001 From: Xunlei Pang Date: Wed, 12 Jul 2017 14:33:14 -0700 Subject: kexec: move vmcoreinfo out of the kernel's .bss section As Eric said, "what we need to do is move the variable vmcoreinfo_note out of the kernel's .bss section. And modify the code to regenerate and keep this information in something like the control page. Definitely something like this needs a page all to itself, and ideally far away from any other kernel data structures. I clearly was not watching closely the data someone decided to keep this silly thing in the kernel's .bss section." This patch allocates extra pages for these vmcoreinfo_XXX variables, one advantage is that it enhances some safety of vmcoreinfo, because vmcoreinfo now is kept far away from other kernel data structures. Link: http://lkml.kernel.org/r/1493281021-20737-1-git-send-email-xlpang@redhat.com Signed-off-by: Xunlei Pang Tested-by: Michael Holzheu Reviewed-by: Juergen Gross Suggested-by: Eric Biederman Cc: Benjamin Herrenschmidt Cc: Dave Young Cc: Hari Bathini Cc: Mahesh Salgaonkar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/machine_kexec.c | 5 ----- arch/s390/kernel/machine_kexec.c | 1 + arch/s390/kernel/setup.c | 6 ------ arch/x86/kernel/crash.c | 2 +- arch/x86/xen/mmu_pv.c | 4 ++-- include/linux/crash_core.h | 4 ++-- kernel/crash_core.c | 26 ++++++++++++++++++++++---- kernel/ksysfs.c | 2 +- 8 files changed, 29 insertions(+), 21 deletions(-) (limited to 'kernel') diff --git a/arch/ia64/kernel/machine_kexec.c b/arch/ia64/kernel/machine_kexec.c index 599507bcec91..c14815dca747 100644 --- a/arch/ia64/kernel/machine_kexec.c +++ b/arch/ia64/kernel/machine_kexec.c @@ -163,8 +163,3 @@ void arch_crash_save_vmcoreinfo(void) #endif } -phys_addr_t paddr_vmcoreinfo_note(void) -{ - return ia64_tpa((unsigned long)(char *)&vmcoreinfo_note); -} - diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c index 49a6bd45957b..3d0b14afa232 100644 --- a/arch/s390/kernel/machine_kexec.c +++ b/arch/s390/kernel/machine_kexec.c @@ -246,6 +246,7 @@ void arch_crash_save_vmcoreinfo(void) VMCOREINFO_SYMBOL(lowcore_ptr); VMCOREINFO_SYMBOL(high_memory); VMCOREINFO_LENGTH(lowcore_ptr, NR_CPUS); + mem_assign_absolute(S390_lowcore.vmcore_info, paddr_vmcoreinfo_note()); } void machine_shutdown(void) diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 3ae756c0db3d..3d1d808ea8a9 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -496,11 +496,6 @@ static void __init setup_memory_end(void) pr_notice("The maximum memory size is %luMB\n", memory_end >> 20); } -static void __init setup_vmcoreinfo(void) -{ - mem_assign_absolute(S390_lowcore.vmcore_info, paddr_vmcoreinfo_note()); -} - #ifdef CONFIG_CRASH_DUMP /* @@ -939,7 +934,6 @@ void __init setup_arch(char **cmdline_p) #endif setup_resources(); - setup_vmcoreinfo(); setup_lowcore(); smp_fill_possible_mask(); cpu_detect_mhz_feature(); diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index 22217ece26c8..44404e2307bb 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -457,7 +457,7 @@ static int prepare_elf64_headers(struct crash_elf_data *ced, bufp += sizeof(Elf64_Phdr); phdr->p_type = PT_NOTE; phdr->p_offset = phdr->p_paddr = paddr_vmcoreinfo_note(); - phdr->p_filesz = phdr->p_memsz = sizeof(vmcoreinfo_note); + phdr->p_filesz = phdr->p_memsz = VMCOREINFO_NOTE_SIZE; (ehdr->e_phnum)++; #ifdef CONFIG_X86_64 diff --git a/arch/x86/xen/mmu_pv.c b/arch/x86/xen/mmu_pv.c index 1d7a7213a310..cab28cf2cffb 100644 --- a/arch/x86/xen/mmu_pv.c +++ b/arch/x86/xen/mmu_pv.c @@ -2693,8 +2693,8 @@ EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region); phys_addr_t paddr_vmcoreinfo_note(void) { if (xen_pv_domain()) - return virt_to_machine(&vmcoreinfo_note).maddr; + return virt_to_machine(vmcoreinfo_note).maddr; else - return __pa_symbol(&vmcoreinfo_note); + return __pa(vmcoreinfo_note); } #endif /* CONFIG_KEXEC_CORE */ diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index 4090a42578a8..87506a02e914 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -19,7 +19,7 @@ CRASH_CORE_NOTE_NAME_BYTES + \ CRASH_CORE_NOTE_DESC_BYTES) -#define VMCOREINFO_BYTES (4096) +#define VMCOREINFO_BYTES PAGE_SIZE #define VMCOREINFO_NOTE_NAME "VMCOREINFO" #define VMCOREINFO_NOTE_NAME_BYTES ALIGN(sizeof(VMCOREINFO_NOTE_NAME), 4) #define VMCOREINFO_NOTE_SIZE ((CRASH_CORE_NOTE_HEAD_BYTES * 2) + \ @@ -56,7 +56,7 @@ phys_addr_t paddr_vmcoreinfo_note(void); #define VMCOREINFO_CONFIG(name) \ vmcoreinfo_append_str("CONFIG_%s=y\n", #name) -extern u32 vmcoreinfo_note[VMCOREINFO_NOTE_SIZE/4]; +extern u32 *vmcoreinfo_note; extern size_t vmcoreinfo_size; extern size_t vmcoreinfo_max_size; diff --git a/kernel/crash_core.c b/kernel/crash_core.c index fcbd568f1e95..2837d6164db8 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -14,10 +14,10 @@ #include /* vmcoreinfo stuff */ -static unsigned char vmcoreinfo_data[VMCOREINFO_BYTES]; -u32 vmcoreinfo_note[VMCOREINFO_NOTE_SIZE/4]; +static unsigned char *vmcoreinfo_data; size_t vmcoreinfo_size; -size_t vmcoreinfo_max_size = sizeof(vmcoreinfo_data); +size_t vmcoreinfo_max_size = VMCOREINFO_BYTES; +u32 *vmcoreinfo_note; /* * parsing the "crashkernel" commandline @@ -326,6 +326,9 @@ static void update_vmcoreinfo_note(void) void crash_save_vmcoreinfo(void) { + if (!vmcoreinfo_note) + return; + vmcoreinfo_append_str("CRASHTIME=%ld\n", get_seconds()); update_vmcoreinfo_note(); } @@ -356,11 +359,26 @@ void __weak arch_crash_save_vmcoreinfo(void) phys_addr_t __weak paddr_vmcoreinfo_note(void) { - return __pa_symbol((unsigned long)(char *)&vmcoreinfo_note); + return __pa(vmcoreinfo_note); } static int __init crash_save_vmcoreinfo_init(void) { + vmcoreinfo_data = (unsigned char *)get_zeroed_page(GFP_KERNEL); + if (!vmcoreinfo_data) { + pr_warn("Memory allocation for vmcoreinfo_data failed\n"); + return -ENOMEM; + } + + vmcoreinfo_note = alloc_pages_exact(VMCOREINFO_NOTE_SIZE, + GFP_KERNEL | __GFP_ZERO); + if (!vmcoreinfo_note) { + free_page((unsigned long)vmcoreinfo_data); + vmcoreinfo_data = NULL; + pr_warn("Memory allocation for vmcoreinfo_note failed\n"); + return -ENOMEM; + } + VMCOREINFO_OSRELEASE(init_uts_ns.name.release); VMCOREINFO_PAGESIZE(PAGE_SIZE); diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c index df1a9aa602a0..46ba853656f6 100644 --- a/kernel/ksysfs.c +++ b/kernel/ksysfs.c @@ -134,7 +134,7 @@ static ssize_t vmcoreinfo_show(struct kobject *kobj, { phys_addr_t vmcore_base = paddr_vmcoreinfo_note(); return sprintf(buf, "%pa %x\n", &vmcore_base, - (unsigned int)sizeof(vmcoreinfo_note)); + (unsigned int)VMCOREINFO_NOTE_SIZE); } KERNEL_ATTR_RO(vmcoreinfo); -- cgit v1.2.3 From 5203f4995d9a87952a83c2ce7866adbbe8f97bb5 Mon Sep 17 00:00:00 2001 From: Xunlei Pang Date: Wed, 12 Jul 2017 14:33:17 -0700 Subject: powerpc/fadump: use the correct VMCOREINFO_NOTE_SIZE for phdr vmcoreinfo_max_size stands for the vmcoreinfo_data, the correct one we should use is vmcoreinfo_note whose total size is VMCOREINFO_NOTE_SIZE. Like explained in commit 77019967f06b ("kdump: fix exported size of vmcoreinfo note"), it should not affect the actual function, but we better fix it, also this change should be safe and backward compatible. After this, we can get rid of variable vmcoreinfo_max_size, let's use the corresponding macros directly, fewer variables means more safety for vmcoreinfo operation. [xlpang@redhat.com: fix build warning] Link: http://lkml.kernel.org/r/1494830606-27736-1-git-send-email-xlpang@redhat.com Link: http://lkml.kernel.org/r/1493281021-20737-2-git-send-email-xlpang@redhat.com Signed-off-by: Xunlei Pang Reviewed-by: Mahesh Salgaonkar Reviewed-by: Dave Young Cc: Hari Bathini Cc: Benjamin Herrenschmidt Cc: Eric Biederman Cc: Juergen Gross Cc: Michael Holzheu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/kernel/fadump.c | 3 +-- include/linux/crash_core.h | 1 - kernel/crash_core.c | 3 +-- 3 files changed, 2 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/arch/powerpc/kernel/fadump.c b/arch/powerpc/kernel/fadump.c index 3079518f2245..dc0c49cfd90a 100644 --- a/arch/powerpc/kernel/fadump.c +++ b/arch/powerpc/kernel/fadump.c @@ -999,8 +999,7 @@ static int fadump_create_elfcore_headers(char *bufp) phdr->p_paddr = fadump_relocate(paddr_vmcoreinfo_note()); phdr->p_offset = phdr->p_paddr; - phdr->p_memsz = vmcoreinfo_max_size; - phdr->p_filesz = vmcoreinfo_max_size; + phdr->p_memsz = phdr->p_filesz = VMCOREINFO_NOTE_SIZE; /* Increment number of program headers. */ (elf->e_phnum)++; diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index 87506a02e914..e5df1b3cf072 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -58,7 +58,6 @@ phys_addr_t paddr_vmcoreinfo_note(void); extern u32 *vmcoreinfo_note; extern size_t vmcoreinfo_size; -extern size_t vmcoreinfo_max_size; Elf_Word *append_elf_note(Elf_Word *buf, char *name, unsigned int type, void *data, size_t data_len); diff --git a/kernel/crash_core.c b/kernel/crash_core.c index 2837d6164db8..315adbf9cb68 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -16,7 +16,6 @@ /* vmcoreinfo stuff */ static unsigned char *vmcoreinfo_data; size_t vmcoreinfo_size; -size_t vmcoreinfo_max_size = VMCOREINFO_BYTES; u32 *vmcoreinfo_note; /* @@ -343,7 +342,7 @@ void vmcoreinfo_append_str(const char *fmt, ...) r = vscnprintf(buf, sizeof(buf), fmt, args); va_end(args); - r = min(r, vmcoreinfo_max_size - vmcoreinfo_size); + r = min(r, (size_t)VMCOREINFO_BYTES - vmcoreinfo_size); memcpy(&vmcoreinfo_data[vmcoreinfo_size], buf, r); -- cgit v1.2.3 From 1229384f5b856d83698c38f9dedfd836e26711cb Mon Sep 17 00:00:00 2001 From: Xunlei Pang Date: Wed, 12 Jul 2017 14:33:21 -0700 Subject: kdump: protect vmcoreinfo data under the crash memory Currently vmcoreinfo data is updated at boot time subsys_initcall(), it has the risk of being modified by some wrong code during system is running. As a result, vmcore dumped may contain the wrong vmcoreinfo. Later on, when using "crash", "makedumpfile", etc utility to parse this vmcore, we probably will get "Segmentation fault" or other unexpected errors. E.g. 1) wrong code overwrites vmcoreinfo_data; 2) further crashes the system; 3) trigger kdump, then we obviously will fail to recognize the crash context correctly due to the corrupted vmcoreinfo. Now except for vmcoreinfo, all the crash data is well protected(including the cpu note which is fully updated in the crash path, thus its correctness is guaranteed). Given that vmcoreinfo data is a large chunk prepared for kdump, we better protect it as well. To solve this, we relocate and copy vmcoreinfo_data to the crash memory when kdump is loading via kexec syscalls. Because the whole crash memory will be protected by existing arch_kexec_protect_crashkres() mechanism, we naturally protect vmcoreinfo_data from write(even read) access under kernel direct mapping after kdump is loaded. Since kdump is usually loaded at the very early stage after boot, we can trust the correctness of the vmcoreinfo data copied. On the other hand, we still need to operate the vmcoreinfo safe copy when crash happens to generate vmcoreinfo_note again, we rely on vmap() to map out a new kernel virtual address and update to use this new one instead in the following crash_save_vmcoreinfo(). BTW, we do not touch vmcoreinfo_note, because it will be fully updated using the protected vmcoreinfo_data after crash which is surely correct just like the cpu crash note. Link: http://lkml.kernel.org/r/1493281021-20737-3-git-send-email-xlpang@redhat.com Signed-off-by: Xunlei Pang Tested-by: Michael Holzheu Cc: Benjamin Herrenschmidt Cc: Dave Young Cc: Eric Biederman Cc: Hari Bathini Cc: Juergen Gross Cc: Mahesh Salgaonkar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/crash_core.h | 2 +- include/linux/kexec.h | 2 ++ kernel/crash_core.c | 17 ++++++++++++++++- kernel/kexec.c | 8 ++++++++ kernel/kexec_core.c | 39 +++++++++++++++++++++++++++++++++++++++ kernel/kexec_file.c | 8 ++++++++ 6 files changed, 74 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index e5df1b3cf072..2df2118fbe13 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -28,6 +28,7 @@ typedef u32 note_buf_t[CRASH_CORE_NOTE_BYTES/4]; +void crash_update_vmcoreinfo_safecopy(void *ptr); void crash_save_vmcoreinfo(void); void arch_crash_save_vmcoreinfo(void); __printf(1, 2) @@ -57,7 +58,6 @@ phys_addr_t paddr_vmcoreinfo_note(void); vmcoreinfo_append_str("CONFIG_%s=y\n", #name) extern u32 *vmcoreinfo_note; -extern size_t vmcoreinfo_size; Elf_Word *append_elf_note(Elf_Word *buf, char *name, unsigned int type, void *data, size_t data_len); diff --git a/include/linux/kexec.h b/include/linux/kexec.h index 65888418fb69..dd056fab9e35 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -172,6 +172,7 @@ struct kimage { unsigned long start; struct page *control_code_page; struct page *swap_page; + void *vmcoreinfo_data_copy; /* locates in the crash memory */ unsigned long nr_segments; struct kexec_segment segment[KEXEC_SEGMENT_MAX]; @@ -241,6 +242,7 @@ extern void crash_kexec(struct pt_regs *); int kexec_should_crash(struct task_struct *); int kexec_crash_loaded(void); void crash_save_cpu(struct pt_regs *regs, int cpu); +extern int kimage_crash_copy_vmcoreinfo(struct kimage *image); extern struct kimage *kexec_image; extern struct kimage *kexec_crash_image; diff --git a/kernel/crash_core.c b/kernel/crash_core.c index 315adbf9cb68..6db80fc0810b 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -15,9 +15,12 @@ /* vmcoreinfo stuff */ static unsigned char *vmcoreinfo_data; -size_t vmcoreinfo_size; +static size_t vmcoreinfo_size; u32 *vmcoreinfo_note; +/* trusted vmcoreinfo, e.g. we can make a copy in the crash memory */ +static unsigned char *vmcoreinfo_data_safecopy; + /* * parsing the "crashkernel" commandline * @@ -323,11 +326,23 @@ static void update_vmcoreinfo_note(void) final_note(buf); } +void crash_update_vmcoreinfo_safecopy(void *ptr) +{ + if (ptr) + memcpy(ptr, vmcoreinfo_data, vmcoreinfo_size); + + vmcoreinfo_data_safecopy = ptr; +} + void crash_save_vmcoreinfo(void) { if (!vmcoreinfo_note) return; + /* Use the safe copy to generate vmcoreinfo note if have */ + if (vmcoreinfo_data_safecopy) + vmcoreinfo_data = vmcoreinfo_data_safecopy; + vmcoreinfo_append_str("CRASHTIME=%ld\n", get_seconds()); update_vmcoreinfo_note(); } diff --git a/kernel/kexec.c b/kernel/kexec.c index 980936a90ee6..e62ec4dc6620 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -144,6 +144,14 @@ static int do_kexec_load(unsigned long entry, unsigned long nr_segments, if (ret) goto out; + /* + * Some architecture(like S390) may touch the crash memory before + * machine_kexec_prepare(), we must copy vmcoreinfo data after it. + */ + ret = kimage_crash_copy_vmcoreinfo(image); + if (ret) + goto out; + for (i = 0; i < nr_segments; i++) { ret = kimage_load_segment(image, &image->segment[i]); if (ret) diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c index 154ffb489b93..1ae7c41c33c1 100644 --- a/kernel/kexec_core.c +++ b/kernel/kexec_core.c @@ -482,6 +482,40 @@ struct page *kimage_alloc_control_pages(struct kimage *image, return pages; } +int kimage_crash_copy_vmcoreinfo(struct kimage *image) +{ + struct page *vmcoreinfo_page; + void *safecopy; + + if (image->type != KEXEC_TYPE_CRASH) + return 0; + + /* + * For kdump, allocate one vmcoreinfo safe copy from the + * crash memory. as we have arch_kexec_protect_crashkres() + * after kexec syscall, we naturally protect it from write + * (even read) access under kernel direct mapping. But on + * the other hand, we still need to operate it when crash + * happens to generate vmcoreinfo note, hereby we rely on + * vmap for this purpose. + */ + vmcoreinfo_page = kimage_alloc_control_pages(image, 0); + if (!vmcoreinfo_page) { + pr_warn("Could not allocate vmcoreinfo buffer\n"); + return -ENOMEM; + } + safecopy = vmap(&vmcoreinfo_page, 1, VM_MAP, PAGE_KERNEL); + if (!safecopy) { + pr_warn("Could not vmap vmcoreinfo buffer\n"); + return -ENOMEM; + } + + image->vmcoreinfo_data_copy = safecopy; + crash_update_vmcoreinfo_safecopy(safecopy); + + return 0; +} + static int kimage_add_entry(struct kimage *image, kimage_entry_t entry) { if (*image->entry != 0) @@ -569,6 +603,11 @@ void kimage_free(struct kimage *image) if (!image) return; + if (image->vmcoreinfo_data_copy) { + crash_update_vmcoreinfo_safecopy(NULL); + vunmap(image->vmcoreinfo_data_copy); + } + kimage_free_extra_pages(image); for_each_kimage_entry(image, ptr, entry) { if (entry & IND_INDIRECTION) { diff --git a/kernel/kexec_file.c b/kernel/kexec_file.c index 766e7e4d3ad9..c8f7f77e9fa9 100644 --- a/kernel/kexec_file.c +++ b/kernel/kexec_file.c @@ -298,6 +298,14 @@ SYSCALL_DEFINE5(kexec_file_load, int, kernel_fd, int, initrd_fd, if (ret) goto out; + /* + * Some architecture(like S390) may touch the crash memory before + * machine_kexec_prepare(), we must copy vmcoreinfo data after it. + */ + ret = kimage_crash_copy_vmcoreinfo(image); + if (ret) + goto out; + ret = kexec_calculate_store_digests(image); if (ret) goto out; -- cgit v1.2.3 From a19ac3374995382a994653ff372b98ea7cbad548 Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Wed, 12 Jul 2017 14:33:30 -0700 Subject: sysctl: kdoc'ify sysctl_writes_strict Document the different sysctl_writes_strict modes in code. Link: http://lkml.kernel.org/r/20170519033554.18592-3-mcgrof@kernel.org Signed-off-by: Luis R. Rodriguez Cc: Al Viro Cc: "Eric W. Biederman" Cc: Alexey Dobriyan Cc: Kees Cook Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 29 +++++++++++++++++++++++++---- 1 file changed, 25 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 4dfba1a76cc3..02725178694a 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -174,11 +174,32 @@ extern int no_unaligned_warning; #ifdef CONFIG_PROC_SYSCTL -#define SYSCTL_WRITES_LEGACY -1 -#define SYSCTL_WRITES_WARN 0 -#define SYSCTL_WRITES_STRICT 1 +/** + * enum sysctl_writes_mode - supported sysctl write modes + * + * @SYSCTL_WRITES_LEGACY: each write syscall must fully contain the sysctl value + * to be written, and multiple writes on the same sysctl file descriptor + * will rewrite the sysctl value, regardless of file position. No warning + * is issued when the initial position is not 0. + * @SYSCTL_WRITES_WARN: same as above but warn when the initial file position is + * not 0. + * @SYSCTL_WRITES_STRICT: writes to numeric sysctl entries must always be at + * file position 0 and the value must be fully contained in the buffer + * sent to the write syscall. If dealing with strings respect the file + * position, but restrict this to the max length of the buffer, anything + * passed the max lenght will be ignored. Multiple writes will append + * to the buffer. + * + * These write modes control how current file position affects the behavior of + * updating sysctl values through the proc interface on each write. + */ +enum sysctl_writes_mode { + SYSCTL_WRITES_LEGACY = -1, + SYSCTL_WRITES_WARN = 0, + SYSCTL_WRITES_STRICT = 1, +}; -static int sysctl_writes_strict = SYSCTL_WRITES_STRICT; +static enum sysctl_writes_mode sysctl_writes_strict = SYSCTL_WRITES_STRICT; static int proc_do_cad_pid(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos); -- cgit v1.2.3 From d383d48470819e86fe30eb72f0e9494e1ee0e2af Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Wed, 12 Jul 2017 14:33:33 -0700 Subject: sysctl: fold sysctl_writes_strict checks into helper The mode sysctl_writes_strict positional checks keep being copy and pasted as we add new proc handlers. Just add a helper to avoid code duplication. Link: http://lkml.kernel.org/r/20170519033554.18592-4-mcgrof@kernel.org Signed-off-by: Luis R. Rodriguez Suggested-by: Kees Cook Cc: Al Viro Cc: "Eric W. Biederman" Cc: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 56 ++++++++++++++++++++++++++++++++------------------------ 1 file changed, 32 insertions(+), 24 deletions(-) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 02725178694a..6f3bb1f099fa 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1970,6 +1970,32 @@ static void warn_sysctl_write(struct ctl_table *table) current->comm, table->procname); } +/** + * proc_first_pos_non_zero_ignore - check if firs position is allowed + * @ppos: file position + * @table: the sysctl table + * + * Returns true if the first position is non-zero and the sysctl_writes_strict + * mode indicates this is not allowed for numeric input types. String proc + * hadlers can ignore the return value. + */ +static bool proc_first_pos_non_zero_ignore(loff_t *ppos, + struct ctl_table *table) +{ + if (!*ppos) + return false; + + switch (sysctl_writes_strict) { + case SYSCTL_WRITES_STRICT: + return true; + case SYSCTL_WRITES_WARN: + warn_sysctl_write(table); + return false; + default: + return false; + } +} + /** * proc_dostring - read a string sysctl * @table: the sysctl table @@ -1990,8 +2016,8 @@ static void warn_sysctl_write(struct ctl_table *table) int proc_dostring(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - if (write && *ppos && sysctl_writes_strict == SYSCTL_WRITES_WARN) - warn_sysctl_write(table); + if (write) + proc_first_pos_non_zero_ignore(ppos, table); return _proc_do_string((char *)(table->data), table->maxlen, write, (char __user *)buffer, lenp, ppos); @@ -2193,17 +2219,8 @@ static int __do_proc_dointvec(void *tbl_data, struct ctl_table *table, conv = do_proc_dointvec_conv; if (write) { - if (*ppos) { - switch (sysctl_writes_strict) { - case SYSCTL_WRITES_STRICT: - goto out; - case SYSCTL_WRITES_WARN: - warn_sysctl_write(table); - break; - default: - break; - } - } + if (proc_first_pos_non_zero_ignore(ppos, table)) + goto out; if (left > PAGE_SIZE - 1) left = PAGE_SIZE - 1; @@ -2468,17 +2485,8 @@ static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, int left = *lenp; if (write) { - if (*ppos) { - switch (sysctl_writes_strict) { - case SYSCTL_WRITES_STRICT: - goto out; - case SYSCTL_WRITES_WARN: - warn_sysctl_write(table); - break; - default: - break; - } - } + if (proc_first_pos_non_zero_ignore(ppos, table)) + goto out; if (left > PAGE_SIZE - 1) left = PAGE_SIZE - 1; -- cgit v1.2.3 From 4f2fec00afa60aa8e5d1b7f2a8e0526900f55623 Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Wed, 12 Jul 2017 14:33:36 -0700 Subject: sysctl: simplify unsigned int support Commit e7d316a02f68 ("sysctl: handle error writing UINT_MAX to u32 fields") added proc_douintvec() to start help adding support for unsigned int, this however was only half the work needed. Two fixes have come in since then for the following issues: o Printing the values shows a negative value, this happens since do_proc_dointvec() and this uses proc_put_long() This was fixed by commit 5380e5644afbba9 ("sysctl: don't print negative flag for proc_douintvec"). o We can easily wrap around the int values: UINT_MAX is 4294967295, if we echo in 4294967295 + 1 we end up with 0, using 4294967295 + 2 we end up with 1. o We echo negative values in and they are accepted This was fixed by commit 425fffd886ba ("sysctl: report EINVAL if value is larger than UINT_MAX for proc_douintvec"). It still also failed to be added to sysctl_check_table()... instead of adding it with the current implementation just provide a proper and simplified unsigned int support without any array unsigned int support with no negative support at all. Historically sysctl proc helpers have supported arrays, due to the complexity this adds though we've taken a step back to evaluate array users to determine if its worth upkeeping for unsigned int. An evaluation using Coccinelle has been done to perform a grammatical search to ask ourselves: o How many sysctl proc_dointvec() (int) users exist which likely should be moved over to proc_douintvec() (unsigned int) ? Answer: about 8 - Of these how many are array users ? Answer: Probably only 1 o How many sysctl array users exist ? Answer: about 12 This last question gives us an idea just how popular arrays: they are not. Array support should probably just be kept for strings. The identified uint ports are: drivers/infiniband/core/ucma.c - max_backlog drivers/infiniband/core/iwcm.c - default_backlog net/core/sysctl_net_core.c - rps_sock_flow_sysctl() net/netfilter/nf_conntrack_timestamp.c - nf_conntrack_timestamp -- bool net/netfilter/nf_conntrack_acct.c nf_conntrack_acct -- bool net/netfilter/nf_conntrack_ecache.c - nf_conntrack_events -- bool net/netfilter/nf_conntrack_helper.c - nf_conntrack_helper -- bool net/phonet/sysctl.c proc_local_port_range() The only possible array users is proc_local_port_range() but it does not seem worth it to add array support just for this given the range support works just as well. Unsigned int support should be desirable more for when you *need* more than INT_MAX or using int min/max support then does not suffice for your ranges. If you forget and by mistake happen to register an unsigned int proc entry with an array, the driver will fail and you will get something as follows: sysctl table check failed: debug/test_sysctl//uint_0002 array now allowed CPU: 2 PID: 1342 Comm: modprobe Tainted: G W E Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Call Trace: dump_stack+0x63/0x81 __register_sysctl_table+0x350/0x650 ? kmem_cache_alloc_trace+0x107/0x240 __register_sysctl_paths+0x1b3/0x1e0 ? 0xffffffffc005f000 register_sysctl_table+0x1f/0x30 test_sysctl_init+0x10/0x1000 [test_sysctl] do_one_initcall+0x52/0x1a0 ? kmem_cache_alloc_trace+0x107/0x240 do_init_module+0x5f/0x200 load_module+0x1867/0x1bd0 ? __symbol_put+0x60/0x60 SYSC_finit_module+0xdf/0x110 SyS_finit_module+0xe/0x10 entry_SYSCALL_64_fastpath+0x1e/0xad RIP: 0033:0x7f042b22d119 Fixes: e7d316a02f68 ("sysctl: handle error writing UINT_MAX to u32 fields") Link: http://lkml.kernel.org/r/20170519033554.18592-5-mcgrof@kernel.org Signed-off-by: Luis R. Rodriguez Suggested-by: Alexey Dobriyan Cc: Subash Abhinov Kasiviswanathan Cc: Liping Zhang Cc: Alexey Dobriyan Cc: Heinrich Schuchardt Cc: Kees Cook Cc: "David S. Miller" Cc: Ingo Molnar Cc: Al Viro Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/proc_sysctl.c | 14 +++++ kernel/sysctl.c | 153 +++++++++++++++++++++++++++++++++++++++++++++++--- 2 files changed, 160 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 32c9c5630507..ee6feba8b6c0 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -1061,6 +1061,18 @@ static int sysctl_err(const char *path, struct ctl_table *table, char *fmt, ...) return -EINVAL; } +static int sysctl_check_table_array(const char *path, struct ctl_table *table) +{ + int err = 0; + + if (table->proc_handler == proc_douintvec) { + if (table->maxlen != sizeof(unsigned int)) + err |= sysctl_err(path, table, "array now allowed"); + } + + return err; +} + static int sysctl_check_table(const char *path, struct ctl_table *table) { int err = 0; @@ -1081,6 +1093,8 @@ static int sysctl_check_table(const char *path, struct ctl_table *table) err |= sysctl_err(path, table, "No data"); if (!table->maxlen) err |= sysctl_err(path, table, "No maxlen"); + else + err |= sysctl_check_table_array(path, table); } if (!table->proc_handler) err |= sysctl_err(path, table, "No proc_handler"); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 6f3bb1f099fa..d12078fc215f 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2175,19 +2175,18 @@ static int do_proc_dointvec_conv(bool *negp, unsigned long *lvalp, return 0; } -static int do_proc_douintvec_conv(bool *negp, unsigned long *lvalp, - int *valp, - int write, void *data) +static int do_proc_douintvec_conv(unsigned long *lvalp, + unsigned int *valp, + int write, void *data) { if (write) { - if (*negp) + if (*lvalp > UINT_MAX) return -EINVAL; if (*lvalp > UINT_MAX) return -EINVAL; *valp = *lvalp; } else { unsigned int val = *valp; - *negp = false; *lvalp = (unsigned long)val; } return 0; @@ -2287,6 +2286,146 @@ static int do_proc_dointvec(struct ctl_table *table, int write, buffer, lenp, ppos, conv, data); } +static int do_proc_douintvec_w(unsigned int *tbl_data, + struct ctl_table *table, + void __user *buffer, + size_t *lenp, loff_t *ppos, + int (*conv)(unsigned long *lvalp, + unsigned int *valp, + int write, void *data), + void *data) +{ + unsigned long lval; + int err = 0; + size_t left; + bool neg; + char *kbuf = NULL, *p; + + left = *lenp; + + if (proc_first_pos_non_zero_ignore(ppos, table)) + goto bail_early; + + if (left > PAGE_SIZE - 1) + left = PAGE_SIZE - 1; + + p = kbuf = memdup_user_nul(buffer, left); + if (IS_ERR(kbuf)) + return -EINVAL; + + left -= proc_skip_spaces(&p); + if (!left) { + err = -EINVAL; + goto out_free; + } + + err = proc_get_long(&p, &left, &lval, &neg, + proc_wspace_sep, + sizeof(proc_wspace_sep), NULL); + if (err || neg) { + err = -EINVAL; + goto out_free; + } + + if (conv(&lval, tbl_data, 1, data)) { + err = -EINVAL; + goto out_free; + } + + if (!err && left) + left -= proc_skip_spaces(&p); + +out_free: + kfree(kbuf); + if (err) + return -EINVAL; + + return 0; + + /* This is in keeping with old __do_proc_dointvec() */ +bail_early: + *ppos += *lenp; + return err; +} + +static int do_proc_douintvec_r(unsigned int *tbl_data, void __user *buffer, + size_t *lenp, loff_t *ppos, + int (*conv)(unsigned long *lvalp, + unsigned int *valp, + int write, void *data), + void *data) +{ + unsigned long lval; + int err = 0; + size_t left; + + left = *lenp; + + if (conv(&lval, tbl_data, 0, data)) { + err = -EINVAL; + goto out; + } + + err = proc_put_long(&buffer, &left, lval, false); + if (err || !left) + goto out; + + err = proc_put_char(&buffer, &left, '\n'); + +out: + *lenp -= left; + *ppos += *lenp; + + return err; +} + +static int __do_proc_douintvec(void *tbl_data, struct ctl_table *table, + int write, void __user *buffer, + size_t *lenp, loff_t *ppos, + int (*conv)(unsigned long *lvalp, + unsigned int *valp, + int write, void *data), + void *data) +{ + unsigned int *i, vleft; + + if (!tbl_data || !table->maxlen || !*lenp || (*ppos && !write)) { + *lenp = 0; + return 0; + } + + i = (unsigned int *) tbl_data; + vleft = table->maxlen / sizeof(*i); + + /* + * Arrays are not supported, keep this simple. *Do not* add + * support for them. + */ + if (vleft != 1) { + *lenp = 0; + return -EINVAL; + } + + if (!conv) + conv = do_proc_douintvec_conv; + + if (write) + return do_proc_douintvec_w(i, table, buffer, lenp, ppos, + conv, data); + return do_proc_douintvec_r(i, buffer, lenp, ppos, conv, data); +} + +static int do_proc_douintvec(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos, + int (*conv)(unsigned long *lvalp, + unsigned int *valp, + int write, void *data), + void *data) +{ + return __do_proc_douintvec(table->data, table, write, + buffer, lenp, ppos, conv, data); +} + /** * proc_dointvec - read a vector of integers * @table: the sysctl table @@ -2322,8 +2461,8 @@ int proc_dointvec(struct ctl_table *table, int write, int proc_douintvec(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - return do_proc_dointvec(table, write, buffer, lenp, ppos, - do_proc_douintvec_conv, NULL); + return do_proc_douintvec(table, write, buffer, lenp, ppos, + do_proc_douintvec_conv, NULL); } /* -- cgit v1.2.3 From 61d9b56a89208d8cccd0b4cfec7e6959717e16e3 Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Wed, 12 Jul 2017 14:33:40 -0700 Subject: sysctl: add unsigned int range support To keep parity with regular int interfaces provide the an unsigned int proc_douintvec_minmax() which allows you to specify a range of allowed valid numbers. Adding proc_douintvec_minmax_sysadmin() is easy but we can wait for an actual user for that. Link: http://lkml.kernel.org/r/20170519033554.18592-6-mcgrof@kernel.org Signed-off-by: Luis R. Rodriguez Acked-by: Kees Cook Cc: Subash Abhinov Kasiviswanathan Cc: Heinrich Schuchardt Cc: Kees Cook Cc: "David S. Miller" Cc: Ingo Molnar Cc: Al Viro Cc: "Eric W. Biederman" Cc: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/proc_sysctl.c | 4 ++- include/linux/sysctl.h | 3 +++ kernel/sysctl.c | 66 ++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 72 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index ee6feba8b6c0..8f9d564d0969 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -1065,7 +1065,8 @@ static int sysctl_check_table_array(const char *path, struct ctl_table *table) { int err = 0; - if (table->proc_handler == proc_douintvec) { + if ((table->proc_handler == proc_douintvec) || + (table->proc_handler == proc_douintvec_minmax)) { if (table->maxlen != sizeof(unsigned int)) err |= sysctl_err(path, table, "array now allowed"); } @@ -1083,6 +1084,7 @@ static int sysctl_check_table(const char *path, struct ctl_table *table) if ((table->proc_handler == proc_dostring) || (table->proc_handler == proc_dointvec) || (table->proc_handler == proc_douintvec) || + (table->proc_handler == proc_douintvec_minmax) || (table->proc_handler == proc_dointvec_minmax) || (table->proc_handler == proc_dointvec_jiffies) || (table->proc_handler == proc_dointvec_userhz_jiffies) || diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 80d07816def0..225001d437ae 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -47,6 +47,9 @@ extern int proc_douintvec(struct ctl_table *, int, void __user *, size_t *, loff_t *); extern int proc_dointvec_minmax(struct ctl_table *, int, void __user *, size_t *, loff_t *); +extern int proc_douintvec_minmax(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, + loff_t *ppos); extern int proc_dointvec_jiffies(struct ctl_table *, int, void __user *, size_t *, loff_t *); extern int proc_dointvec_userhz_jiffies(struct ctl_table *, int, diff --git a/kernel/sysctl.c b/kernel/sysctl.c index d12078fc215f..df9f2a367882 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2567,6 +2567,65 @@ int proc_dointvec_minmax(struct ctl_table *table, int write, do_proc_dointvec_minmax_conv, ¶m); } +struct do_proc_douintvec_minmax_conv_param { + unsigned int *min; + unsigned int *max; +}; + +static int do_proc_douintvec_minmax_conv(unsigned long *lvalp, + unsigned int *valp, + int write, void *data) +{ + struct do_proc_douintvec_minmax_conv_param *param = data; + + if (write) { + unsigned int val = *lvalp; + + if ((param->min && *param->min > val) || + (param->max && *param->max < val)) + return -ERANGE; + + if (*lvalp > UINT_MAX) + return -EINVAL; + *valp = val; + } else { + unsigned int val = *valp; + *lvalp = (unsigned long) val; + } + + return 0; +} + +/** + * proc_douintvec_minmax - read a vector of unsigned ints with min/max values + * @table: the sysctl table + * @write: %TRUE if this is a write to the sysctl file + * @buffer: the user buffer + * @lenp: the size of the user buffer + * @ppos: file position + * + * Reads/writes up to table->maxlen/sizeof(unsigned int) unsigned integer + * values from/to the user buffer, treated as an ASCII string. Negative + * strings are not allowed. + * + * This routine will ensure the values are within the range specified by + * table->extra1 (min) and table->extra2 (max). There is a final sanity + * check for UINT_MAX to avoid having to support wrap around uses from + * userspace. + * + * Returns 0 on success. + */ +int proc_douintvec_minmax(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + struct do_proc_douintvec_minmax_conv_param param = { + .min = (unsigned int *) table->extra1, + .max = (unsigned int *) table->extra2, + }; + return do_proc_douintvec(table, write, buffer, lenp, ppos, + do_proc_douintvec_minmax_conv, ¶m); +} + static void validate_coredump_safety(void) { #ifdef CONFIG_COREDUMP @@ -3066,6 +3125,12 @@ int proc_dointvec_minmax(struct ctl_table *table, int write, return -ENOSYS; } +int proc_douintvec_minmax(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + return -ENOSYS; +} + int proc_dointvec_jiffies(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { @@ -3108,6 +3173,7 @@ EXPORT_SYMBOL(proc_dointvec); EXPORT_SYMBOL(proc_douintvec); EXPORT_SYMBOL(proc_dointvec_jiffies); EXPORT_SYMBOL(proc_dointvec_minmax); +EXPORT_SYMBOL_GPL(proc_douintvec_minmax); EXPORT_SYMBOL(proc_dointvec_userhz_jiffies); EXPORT_SYMBOL(proc_dointvec_ms_jiffies); EXPORT_SYMBOL(proc_dostring); -- cgit v1.2.3 From 9380fa60b10ebd6ee7c3fcdb2cf162f4d7cf9fc5 Mon Sep 17 00:00:00 2001 From: Mateusz Jurczyk Date: Wed, 12 Jul 2017 14:34:01 -0700 Subject: kernel/sysctl_binary.c: check name array length in deprecated_sysctl_warning() Prevent use of uninitialized memory (originating from the stack frame of do_sysctl()) by verifying that the name array is filled with sufficient input data before comparing its specific entries with integer constants. Through timing measurement or analyzing the kernel debug logs, a user-mode program could potentially infer the results of comparisons against the uninitialized memory, and acquire some (very limited) information about the state of the kernel stack. The change also eliminates possible future warnings by tools such as KMSAN and other code checkers / instrumentations. Link: http://lkml.kernel.org/r/20170524122139.21333-1-mjurczyk@google.com Signed-off-by: Mateusz Jurczyk Acked-by: Kees Cook Cc: "David S. Miller" Cc: Matthew Whitehead Cc: "Eric W. Biederman" Cc: Tetsuo Handa Cc: Alexander Potapenko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl_binary.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index 939a158eab11..02e1859f2ca8 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -1346,7 +1346,7 @@ static void deprecated_sysctl_warning(const int *name, int nlen) * CTL_KERN/KERN_VERSION is used by older glibc and cannot * ever go away. */ - if (name[0] == CTL_KERN && name[1] == KERN_VERSION) + if (nlen >= 2 && name[0] == CTL_KERN && name[1] == KERN_VERSION) return; if (printk_ratelimit()) { -- cgit v1.2.3 From 0791e3644e5ef21646fe565b9061788d05ec71d4 Mon Sep 17 00:00:00 2001 From: Cyrill Gorcunov Date: Wed, 12 Jul 2017 14:34:28 -0700 Subject: kcmp: add KCMP_EPOLL_TFD mode to compare epoll target files With current epoll architecture target files are addressed with file_struct and file descriptor number, where the last is not unique. Moreover files can be transferred from another process via unix socket, added into queue and closed then so we won't find this descriptor in the task fdinfo list. Thus to checkpoint and restore such processes CRIU needs to find out where exactly the target file is present to add it into epoll queue. For this sake one can use kcmp call where some particular target file from the queue is compared with arbitrary file passed as an argument. Because epoll target files can have same file descriptor number but different file_struct a caller should explicitly specify the offset within. To test if some particular file is matching entry inside epoll one have to - fill kcmp_epoll_slot structure with epoll file descriptor, target file number and target file offset (in case if only one target is present then it should be 0) - call kcmp as kcmp(pid1, pid2, KCMP_EPOLL_TFD, fd, &kcmp_epoll_slot) - the kernel fetch file pointer matching file descriptor @fd of pid1 - lookups for file struct in epoll queue of pid2 and returns traditional 0,1,2 result for sorting purpose Link: http://lkml.kernel.org/r/20170424154423.511592110@gmail.com Signed-off-by: Cyrill Gorcunov Acked-by: Andrey Vagin Cc: Al Viro Cc: Pavel Emelyanov Cc: Michael Kerrisk Cc: Jason Baron Cc: Andy Lutomirski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/eventpoll.c | 42 ++++++++++++++++++++++++++++++++++ include/linux/eventpoll.h | 3 +++ include/uapi/linux/kcmp.h | 10 +++++++++ kernel/kcmp.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 112 insertions(+) (limited to 'kernel') diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 322904c3ebdf..e7e9901c3790 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1077,6 +1077,48 @@ static struct epitem *ep_find(struct eventpoll *ep, struct file *file, int fd) return epir; } +static struct epitem *ep_find_tfd(struct eventpoll *ep, int tfd, unsigned long toff) +{ + struct rb_node *rbp; + struct epitem *epi; + + for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) { + epi = rb_entry(rbp, struct epitem, rbn); + if (epi->ffd.fd == tfd) { + if (toff == 0) + return epi; + else + toff--; + } + cond_resched(); + } + + return NULL; +} + +struct file *get_epoll_tfile_raw_ptr(struct file *file, int tfd, + unsigned long toff) +{ + struct file *file_raw; + struct eventpoll *ep; + struct epitem *epi; + + if (!is_file_epoll(file)) + return ERR_PTR(-EINVAL); + + ep = file->private_data; + + mutex_lock(&ep->mtx); + epi = ep_find_tfd(ep, tfd, toff); + if (epi) + file_raw = epi->ffd.file; + else + file_raw = ERR_PTR(-ENOENT); + mutex_unlock(&ep->mtx); + + return file_raw; +} + /* * This is the callback that is passed to the wait queue wakeup * mechanism. It is called by the stored file descriptors when they diff --git a/include/linux/eventpoll.h b/include/linux/eventpoll.h index 6daf6d4971f6..d8625d214ea7 100644 --- a/include/linux/eventpoll.h +++ b/include/linux/eventpoll.h @@ -14,6 +14,7 @@ #define _LINUX_EVENTPOLL_H #include +#include /* Forward declarations to avoid compiler errors */ @@ -22,6 +23,8 @@ struct file; #ifdef CONFIG_EPOLL +struct file *get_epoll_tfile_raw_ptr(struct file *file, int tfd, unsigned long toff); + /* Used to initialize the epoll bits inside the "struct file" */ static inline void eventpoll_init_file(struct file *file) { diff --git a/include/uapi/linux/kcmp.h b/include/uapi/linux/kcmp.h index 84df14b37360..481e103da78e 100644 --- a/include/uapi/linux/kcmp.h +++ b/include/uapi/linux/kcmp.h @@ -1,6 +1,8 @@ #ifndef _UAPI_LINUX_KCMP_H #define _UAPI_LINUX_KCMP_H +#include + /* Comparison type */ enum kcmp_type { KCMP_FILE, @@ -10,8 +12,16 @@ enum kcmp_type { KCMP_SIGHAND, KCMP_IO, KCMP_SYSVSEM, + KCMP_EPOLL_TFD, KCMP_TYPES, }; +/* Slot for KCMP_EPOLL_TFD */ +struct kcmp_epoll_slot { + __u32 efd; /* epoll file descriptor */ + __u32 tfd; /* target file number */ + __u32 toff; /* target offset within same numbered sequence */ +}; + #endif /* _UAPI_LINUX_KCMP_H */ diff --git a/kernel/kcmp.c b/kernel/kcmp.c index 3a47fa998fe0..ea34ed8bb952 100644 --- a/kernel/kcmp.c +++ b/kernel/kcmp.c @@ -11,6 +11,10 @@ #include #include #include +#include +#include +#include +#include #include @@ -94,6 +98,56 @@ static int kcmp_lock(struct mutex *m1, struct mutex *m2) return err; } +#ifdef CONFIG_EPOLL +static int kcmp_epoll_target(struct task_struct *task1, + struct task_struct *task2, + unsigned long idx1, + struct kcmp_epoll_slot __user *uslot) +{ + struct file *filp, *filp_epoll, *filp_tgt; + struct kcmp_epoll_slot slot; + struct files_struct *files; + + if (copy_from_user(&slot, uslot, sizeof(slot))) + return -EFAULT; + + filp = get_file_raw_ptr(task1, idx1); + if (!filp) + return -EBADF; + + files = get_files_struct(task2); + if (!files) + return -EBADF; + + spin_lock(&files->file_lock); + filp_epoll = fcheck_files(files, slot.efd); + if (filp_epoll) + get_file(filp_epoll); + else + filp_tgt = ERR_PTR(-EBADF); + spin_unlock(&files->file_lock); + put_files_struct(files); + + if (filp_epoll) { + filp_tgt = get_epoll_tfile_raw_ptr(filp_epoll, slot.tfd, slot.toff); + fput(filp_epoll); + } else + + if (IS_ERR(filp_tgt)) + return PTR_ERR(filp_tgt); + + return kcmp_ptr(filp, filp_tgt, KCMP_FILE); +} +#else +static int kcmp_epoll_target(struct task_struct *task1, + struct task_struct *task2, + unsigned long idx1, + struct kcmp_epoll_slot __user *uslot) +{ + return -EOPNOTSUPP; +} +#endif + SYSCALL_DEFINE5(kcmp, pid_t, pid1, pid_t, pid2, int, type, unsigned long, idx1, unsigned long, idx2) { @@ -165,6 +219,9 @@ SYSCALL_DEFINE5(kcmp, pid_t, pid1, pid_t, pid2, int, type, ret = -EOPNOTSUPP; #endif break; + case KCMP_EPOLL_TFD: + ret = kcmp_epoll_target(task1, task2, idx1, (void *)idx2); + break; default: ret = -EINVAL; break; -- cgit v1.2.3 From e41d58185f1444368873d4d7422f7664a68be61d Mon Sep 17 00:00:00 2001 From: Dmitry Vyukov Date: Wed, 12 Jul 2017 14:34:35 -0700 Subject: fault-inject: support systematic fault injection Add /proc/self/task//fail-nth file that allows failing 0-th, 1-st, 2-nd and so on calls systematically. Excerpt from the added documentation: "Write to this file of integer N makes N-th call in the current task fail (N is 0-based). Read from this file returns a single char 'Y' or 'N' that says if the fault setup with a previous write to this file was injected or not, and disables the fault if it wasn't yet injected. Note that this file enables all types of faults (slab, futex, etc). This setting takes precedence over all other generic settings like probability, interval, times, etc. But per-capability settings (e.g. fail_futex/ignore-private) take precedence over it. This feature is intended for systematic testing of faults in a single system call. See an example below" Why add a new setting: 1. Existing settings are global rather than per-task. So parallel testing is not possible. 2. attr->interval is close but it depends on attr->count which is non reset to 0, so interval does not work as expected. 3. Trying to model this with existing settings requires manipulations of all of probability, interval, times, space, task-filter and unexposed count and per-task make-it-fail files. 4. Existing settings are per-failure-type, and the set of failure types is potentially expanding. 5. make-it-fail can't be changed by unprivileged user and aggressive stress testing better be done from an unprivileged user. Similarly, this would require opening the debugfs files to the unprivileged user, as he would need to reopen at least times file (not possible to pre-open before dropping privs). The proposed interface solves all of the above (see the example). We want to integrate this into syzkaller fuzzer. A prototype has found 10 bugs in kernel in first day of usage: https://groups.google.com/forum/#!searchin/syzkaller/%22FAULT_INJECTION%22%7Csort:relevance I've made the current interface work with all types of our sandboxes. For setuid the secret sauce was prctl(PR_SET_DUMPABLE, 1, 0, 0, 0) to make /proc entries non-root owned. So I am fine with the current version of the code. [akpm@linux-foundation.org: fix build] Link: http://lkml.kernel.org/r/20170328130128.101773-1-dvyukov@google.com Signed-off-by: Dmitry Vyukov Cc: Akinobu Mita Cc: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/fault-injection/fault-injection.txt | 78 +++++++++++++++++++++++ fs/proc/base.c | 52 +++++++++++++++ include/linux/sched.h | 1 + kernel/fork.c | 4 ++ lib/fault-inject.c | 7 ++ 5 files changed, 142 insertions(+) (limited to 'kernel') diff --git a/Documentation/fault-injection/fault-injection.txt b/Documentation/fault-injection/fault-injection.txt index 415484f3d59a..192d8cbcc5f9 100644 --- a/Documentation/fault-injection/fault-injection.txt +++ b/Documentation/fault-injection/fault-injection.txt @@ -134,6 +134,22 @@ use the boot option: fail_futex= mmc_core.fail_request=,,, +o proc entries + +- /proc/self/task//fail-nth: + + Write to this file of integer N makes N-th call in the current task fail + (N is 0-based). Read from this file returns a single char 'Y' or 'N' + that says if the fault setup with a previous write to this file was + injected or not, and disables the fault if it wasn't yet injected. + Note that this file enables all types of faults (slab, futex, etc). + This setting takes precedence over all other generic debugfs settings + like probability, interval, times, etc. But per-capability settings + (e.g. fail_futex/ignore-private) take precedence over it. + + This feature is intended for systematic testing of faults in a single + system call. See an example below. + How to add new fault injection capability ----------------------------------------- @@ -278,3 +294,65 @@ allocation failure. # env FAILCMD_TYPE=fail_page_alloc \ ./tools/testing/fault-injection/failcmd.sh --times=100 \ -- make -C tools/testing/selftests/ run_tests + +Systematic faults using fail-nth +--------------------------------- + +The following code systematically faults 0-th, 1-st, 2-nd and so on +capabilities in the socketpair() system call. + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +int main() +{ + int i, err, res, fail_nth, fds[2]; + char buf[128]; + + system("echo N > /sys/kernel/debug/failslab/ignore-gfp-wait"); + sprintf(buf, "/proc/self/task/%ld/fail-nth", syscall(SYS_gettid)); + fail_nth = open(buf, O_RDWR); + for (i = 0;; i++) { + sprintf(buf, "%d", i); + write(fail_nth, buf, strlen(buf)); + res = socketpair(AF_LOCAL, SOCK_STREAM, 0, fds); + err = errno; + read(fail_nth, buf, 1); + if (res == 0) { + close(fds[0]); + close(fds[1]); + } + printf("%d-th fault %c: res=%d/%d\n", i, buf[0], res, err); + if (buf[0] != 'Y') + break; + } + return 0; +} + +An example output: + +0-th fault Y: res=-1/23 +1-th fault Y: res=-1/23 +2-th fault Y: res=-1/23 +3-th fault Y: res=-1/12 +4-th fault Y: res=-1/12 +5-th fault Y: res=-1/23 +6-th fault Y: res=-1/23 +7-th fault Y: res=-1/23 +8-th fault Y: res=-1/12 +9-th fault Y: res=-1/12 +10-th fault Y: res=-1/12 +11-th fault Y: res=-1/12 +12-th fault Y: res=-1/12 +13-th fault Y: res=-1/12 +14-th fault Y: res=-1/12 +15-th fault Y: res=-1/12 +16-th fault N: res=0/12 diff --git a/fs/proc/base.c b/fs/proc/base.c index f1e1927ccd48..88b773f318cd 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1355,6 +1355,53 @@ static const struct file_operations proc_fault_inject_operations = { .write = proc_fault_inject_write, .llseek = generic_file_llseek, }; + +static ssize_t proc_fail_nth_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task; + int err, n; + + task = get_proc_task(file_inode(file)); + if (!task) + return -ESRCH; + put_task_struct(task); + if (task != current) + return -EPERM; + err = kstrtoint_from_user(buf, count, 10, &n); + if (err) + return err; + if (n < 0 || n == INT_MAX) + return -EINVAL; + current->fail_nth = n + 1; + return count; +} + +static ssize_t proc_fail_nth_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task; + int err; + + task = get_proc_task(file_inode(file)); + if (!task) + return -ESRCH; + put_task_struct(task); + if (task != current) + return -EPERM; + if (count < 1) + return -EINVAL; + err = put_user((char)(current->fail_nth ? 'N' : 'Y'), buf); + if (err) + return err; + current->fail_nth = 0; + return 1; +} + +static const struct file_operations proc_fail_nth_operations = { + .read = proc_fail_nth_read, + .write = proc_fail_nth_write, +}; #endif @@ -3311,6 +3358,11 @@ static const struct pid_entry tid_base_stuff[] = { #endif #ifdef CONFIG_FAULT_INJECTION REG("make-it-fail", S_IRUGO|S_IWUSR, proc_fault_inject_operations), + /* + * Operations on the file check that the task is current, + * so we create it with 0666 to support testing under unprivileged user. + */ + REG("fail-nth", 0666, proc_fail_nth_operations), #endif #ifdef CONFIG_TASK_IO_ACCOUNTING ONE("io", S_IRUSR, proc_tid_io_accounting), diff --git a/include/linux/sched.h b/include/linux/sched.h index 20814b7d7d70..3822d749fc9e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -974,6 +974,7 @@ struct task_struct { #ifdef CONFIG_FAULT_INJECTION int make_it_fail; + int fail_nth; #endif /* * When (nr_dirtied >= nr_dirtied_pause), it's time to call diff --git a/kernel/fork.c b/kernel/fork.c index d2b9d7c31eaf..ade237a96308 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -573,6 +573,10 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node) kcov_task_init(tsk); +#ifdef CONFIG_FAULT_INJECTION + tsk->fail_nth = 0; +#endif + return tsk; free_stack: diff --git a/lib/fault-inject.c b/lib/fault-inject.c index 4ff157159a0d..09ac73c177fd 100644 --- a/lib/fault-inject.c +++ b/lib/fault-inject.c @@ -107,6 +107,12 @@ static inline bool fail_stacktrace(struct fault_attr *attr) bool should_fail(struct fault_attr *attr, ssize_t size) { + if (in_task() && current->fail_nth) { + if (--current->fail_nth == 0) + goto fail; + return false; + } + /* No need to check any other properties if the probability is 0 */ if (attr->probability == 0) return false; @@ -134,6 +140,7 @@ bool should_fail(struct fault_attr *attr, ssize_t size) if (!fail_stacktrace(attr)) return false; +fail: fail_dump(attr); if (atomic_read(&attr->times) != -1) -- cgit v1.2.3 From f2e0cff85ed111a3cf24d894c3fa11697dfae628 Mon Sep 17 00:00:00 2001 From: Nicholas Piggin Date: Wed, 12 Jul 2017 14:35:43 -0700 Subject: kernel/watchdog: introduce arch_touch_nmi_watchdog() For architectures that define HAVE_NMI_WATCHDOG, instead of having them provide the complete touch_nmi_watchdog() function, just have them provide arch_touch_nmi_watchdog(). This gives the generic code more flexibility in implementing this function, and arch implementations don't miss out on touching the softlockup watchdog or other generic details. Link: http://lkml.kernel.org/r/20170616065715.18390-3-npiggin@gmail.com Signed-off-by: Nicholas Piggin Reviewed-by: Don Zickus Reviewed-by: Babu Moger Tested-by: Babu Moger [sparc] Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/blackfin/include/asm/nmi.h | 2 ++ arch/blackfin/kernel/nmi.c | 2 +- arch/mn10300/include/asm/nmi.h | 2 ++ arch/mn10300/kernel/mn10300-watchdog-low.S | 8 ++++---- arch/mn10300/kernel/mn10300-watchdog.c | 2 +- arch/sparc/include/asm/nmi.h | 1 + arch/sparc/kernel/nmi.c | 6 ++---- include/linux/nmi.h | 27 ++++++++++++++++----------- kernel/watchdog_hld.c | 5 ++--- 9 files changed, 31 insertions(+), 24 deletions(-) (limited to 'kernel') diff --git a/arch/blackfin/include/asm/nmi.h b/arch/blackfin/include/asm/nmi.h index b9caac4fcfd8..107d23705f46 100644 --- a/arch/blackfin/include/asm/nmi.h +++ b/arch/blackfin/include/asm/nmi.h @@ -9,4 +9,6 @@ #include +extern void arch_touch_nmi_watchdog(void); + #endif diff --git a/arch/blackfin/kernel/nmi.c b/arch/blackfin/kernel/nmi.c index 633c37083e87..1e714329fe8a 100644 --- a/arch/blackfin/kernel/nmi.c +++ b/arch/blackfin/kernel/nmi.c @@ -190,7 +190,7 @@ static int __init init_nmi_wdt(void) } device_initcall(init_nmi_wdt); -void touch_nmi_watchdog(void) +void arch_touch_nmi_watchdog(void) { atomic_set(&nmi_touched[smp_processor_id()], 1); } diff --git a/arch/mn10300/include/asm/nmi.h b/arch/mn10300/include/asm/nmi.h index f3671cbbc117..b05627597b1b 100644 --- a/arch/mn10300/include/asm/nmi.h +++ b/arch/mn10300/include/asm/nmi.h @@ -11,4 +11,6 @@ #ifndef _ASM_NMI_H #define _ASM_NMI_H +extern void arch_touch_nmi_watchdog(void); + #endif /* _ASM_NMI_H */ diff --git a/arch/mn10300/kernel/mn10300-watchdog-low.S b/arch/mn10300/kernel/mn10300-watchdog-low.S index f2f5c9cfaabd..34f8773de7d0 100644 --- a/arch/mn10300/kernel/mn10300-watchdog-low.S +++ b/arch/mn10300/kernel/mn10300-watchdog-low.S @@ -50,9 +50,9 @@ watchdog_handler: # we can't inline it) # ############################################################################### - .globl touch_nmi_watchdog - .type touch_nmi_watchdog,@function -touch_nmi_watchdog: + .globl arch_touch_nmi_watchdog + .type arch_touch_nmi_watchdog,@function +arch_touch_nmi_watchdog: clr d0 clr d1 mov watchdog_alert_counter, a0 @@ -63,4 +63,4 @@ touch_nmi_watchdog: lne ret [],0 - .size touch_nmi_watchdog,.-touch_nmi_watchdog + .size arch_touch_nmi_watchdog,.-arch_touch_nmi_watchdog diff --git a/arch/mn10300/kernel/mn10300-watchdog.c b/arch/mn10300/kernel/mn10300-watchdog.c index a2d8e6938d67..0d5641beadf5 100644 --- a/arch/mn10300/kernel/mn10300-watchdog.c +++ b/arch/mn10300/kernel/mn10300-watchdog.c @@ -31,7 +31,7 @@ static unsigned int watchdog; static unsigned int watchdog_hz = 1; unsigned int watchdog_alert_counter[NR_CPUS]; -EXPORT_SYMBOL(touch_nmi_watchdog); +EXPORT_SYMBOL(arch_touch_nmi_watchdog); /* * the best way to detect whether a CPU has a 'hard lockup' problem diff --git a/arch/sparc/include/asm/nmi.h b/arch/sparc/include/asm/nmi.h index 26ad2b2607c6..284eac3ffaf2 100644 --- a/arch/sparc/include/asm/nmi.h +++ b/arch/sparc/include/asm/nmi.h @@ -7,6 +7,7 @@ void nmi_adjust_hz(unsigned int new_hz); extern atomic_t nmi_active; +void arch_touch_nmi_watchdog(void); void start_nmi_watchdog(void *unused); void stop_nmi_watchdog(void *unused); diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index 95e73c63c99d..048ad783ea3f 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -51,7 +51,7 @@ static DEFINE_PER_CPU(unsigned int, last_irq_sum); static DEFINE_PER_CPU(long, alert_counter); static DEFINE_PER_CPU(int, nmi_touch); -void touch_nmi_watchdog(void) +void arch_touch_nmi_watchdog(void) { if (atomic_read(&nmi_active)) { int cpu; @@ -61,10 +61,8 @@ void touch_nmi_watchdog(void) per_cpu(nmi_touch, cpu) = 1; } } - - touch_softlockup_watchdog(); } -EXPORT_SYMBOL(touch_nmi_watchdog); +EXPORT_SYMBOL(arch_touch_nmi_watchdog); static void die_nmi(const char *str, struct pt_regs *regs, int do_panic) { diff --git a/include/linux/nmi.h b/include/linux/nmi.h index 5e2e57536d98..bd387ef8bccd 100644 --- a/include/linux/nmi.h +++ b/include/linux/nmi.h @@ -6,6 +6,9 @@ #include #include +#if defined(CONFIG_HAVE_NMI_WATCHDOG) +#include +#endif #ifdef CONFIG_LOCKUP_DETECTOR extern void touch_softlockup_watchdog_sched(void); @@ -58,6 +61,18 @@ static inline void reset_hung_task_detector(void) #define NMI_WATCHDOG_ENABLED (1 << NMI_WATCHDOG_ENABLED_BIT) #define SOFT_WATCHDOG_ENABLED (1 << SOFT_WATCHDOG_ENABLED_BIT) +#if defined(CONFIG_HARDLOCKUP_DETECTOR) +extern void hardlockup_detector_disable(void); +#else +static inline void hardlockup_detector_disable(void) {} +#endif + +#if defined(CONFIG_HARDLOCKUP_DETECTOR) || defined(CONFIG_HAVE_NMI_WATCHDOG) +extern void arch_touch_nmi_watchdog(void); +#else +static inline void arch_touch_nmi_watchdog(void) {} +#endif + /** * touch_nmi_watchdog - restart NMI watchdog timeout. * @@ -65,21 +80,11 @@ static inline void reset_hung_task_detector(void) * may be used to reset the timeout - for code which intentionally * disables interrupts for a long time. This call is stateless. */ -#if defined(CONFIG_HAVE_NMI_WATCHDOG) || defined(CONFIG_HARDLOCKUP_DETECTOR) -#include -extern void touch_nmi_watchdog(void); -#else static inline void touch_nmi_watchdog(void) { + arch_touch_nmi_watchdog(); touch_softlockup_watchdog(); } -#endif - -#if defined(CONFIG_HARDLOCKUP_DETECTOR) -extern void hardlockup_detector_disable(void); -#else -static inline void hardlockup_detector_disable(void) {} -#endif /* * Create trigger_all_cpu_backtrace() out of the arch-provided diff --git a/kernel/watchdog_hld.c b/kernel/watchdog_hld.c index 54a427d1f344..90d688df6ce1 100644 --- a/kernel/watchdog_hld.c +++ b/kernel/watchdog_hld.c @@ -56,7 +56,7 @@ static int __init hardlockup_panic_setup(char *str) } __setup("nmi_watchdog=", hardlockup_panic_setup); -void touch_nmi_watchdog(void) +void arch_touch_nmi_watchdog(void) { /* * Using __raw here because some code paths have @@ -66,9 +66,8 @@ void touch_nmi_watchdog(void) * going off. */ raw_cpu_write(watchdog_nmi_touch, true); - touch_softlockup_watchdog(); } -EXPORT_SYMBOL(touch_nmi_watchdog); +EXPORT_SYMBOL(arch_touch_nmi_watchdog); static struct perf_event_attr wd_hw_attr = { .type = PERF_TYPE_HARDWARE, -- cgit v1.2.3 From 05a4a95279311c3a4633b4277a5d21cfd616c6c7 Mon Sep 17 00:00:00 2001 From: Nicholas Piggin Date: Wed, 12 Jul 2017 14:35:46 -0700 Subject: kernel/watchdog: split up config options Split SOFTLOCKUP_DETECTOR from LOCKUP_DETECTOR, and split HARDLOCKUP_DETECTOR_PERF from HARDLOCKUP_DETECTOR. LOCKUP_DETECTOR implies the general boot, sysctl, and programming interfaces for the lockup detectors. An architecture that wants to use a hard lockup detector must define HAVE_HARDLOCKUP_DETECTOR_PERF or HAVE_HARDLOCKUP_DETECTOR_ARCH. Alternatively an arch can define HAVE_NMI_WATCHDOG, which provides the minimum arch_touch_nmi_watchdog, and it otherwise does its own thing and does not implement the LOCKUP_DETECTOR interfaces. sparc is unusual in that it has started to implement some of the interfaces, but not fully yet. It should probably be converted to a full HAVE_HARDLOCKUP_DETECTOR_ARCH. [npiggin@gmail.com: fix] Link: http://lkml.kernel.org/r/20170617223522.66c0ad88@roar.ozlabs.ibm.com Link: http://lkml.kernel.org/r/20170616065715.18390-4-npiggin@gmail.com Signed-off-by: Nicholas Piggin Reviewed-by: Don Zickus Reviewed-by: Babu Moger Tested-by: Babu Moger [sparc] Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/Kconfig | 25 ++++- arch/powerpc/Kconfig | 1 + arch/powerpc/kernel/setup_64.c | 2 +- arch/x86/Kconfig | 1 + arch/x86/kernel/apic/hw_nmi.c | 2 +- include/linux/nmi.h | 29 +++-- kernel/Makefile | 2 +- kernel/sysctl.c | 31 +++--- kernel/watchdog.c | 243 +++++++++++++++++++++++++++-------------- kernel/watchdog_hld.c | 32 ------ lib/Kconfig.debug | 45 +++++--- 11 files changed, 251 insertions(+), 162 deletions(-) (limited to 'kernel') diff --git a/arch/Kconfig b/arch/Kconfig index cae0958a2298..fb9bd7d36b05 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -198,9 +198,6 @@ config HAVE_KPROBES_ON_FTRACE config HAVE_NMI bool -config HAVE_NMI_WATCHDOG - depends on HAVE_NMI - bool # # An arch should select this if it provides all these things: # @@ -288,6 +285,28 @@ config HAVE_PERF_EVENTS_NMI subsystem. Also has support for calculating CPU cycle events to determine how many clock cycles in a given period. +config HAVE_HARDLOCKUP_DETECTOR_PERF + bool + depends on HAVE_PERF_EVENTS_NMI + help + The arch chooses to use the generic perf-NMI-based hardlockup + detector. Must define HAVE_PERF_EVENTS_NMI. + +config HAVE_NMI_WATCHDOG + depends on HAVE_NMI + bool + help + The arch provides a low level NMI watchdog. It provides + asm/nmi.h, and defines its own arch_touch_nmi_watchdog(). + +config HAVE_HARDLOCKUP_DETECTOR_ARCH + bool + select HAVE_NMI_WATCHDOG + help + The arch chooses to provide its own hardlockup detector, which is + a superset of the HAVE_NMI_WATCHDOG. It also conforms to config + interfaces and parameters provided by hardlockup detector subsystem. + config HAVE_PERF_REGS bool help diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 7177a3f4f418..63ed758e1d20 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -197,6 +197,7 @@ config PPC select HAVE_OPTPROBES if PPC64 select HAVE_PERF_EVENTS select HAVE_PERF_EVENTS_NMI if PPC64 + select HAVE_HARDLOCKUP_DETECTOR_PERF if HAVE_PERF_EVENTS_NMI select HAVE_PERF_REGS select HAVE_PERF_USER_STACK_DUMP select HAVE_RCU_TABLE_FREE if SMP diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 4640f6d64f8b..074a075a9cdb 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -752,7 +752,7 @@ struct ppc_pci_io ppc_pci_io; EXPORT_SYMBOL(ppc_pci_io); #endif -#ifdef CONFIG_HARDLOCKUP_DETECTOR +#ifdef CONFIG_HARDLOCKUP_DETECTOR_PERF u64 hw_nmi_get_sample_period(int watchdog_thresh) { return ppc_proc_freq * watchdog_thresh; diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 94a18681353d..3d2b8ce54e00 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -162,6 +162,7 @@ config X86 select HAVE_PCSPKR_PLATFORM select HAVE_PERF_EVENTS select HAVE_PERF_EVENTS_NMI + select HAVE_HARDLOCKUP_DETECTOR_PERF if HAVE_PERF_EVENTS_NMI select HAVE_PERF_REGS select HAVE_PERF_USER_STACK_DUMP select HAVE_REGS_AND_STACK_ACCESS_API diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c index c73c9fb281e1..d6f387780849 100644 --- a/arch/x86/kernel/apic/hw_nmi.c +++ b/arch/x86/kernel/apic/hw_nmi.c @@ -19,7 +19,7 @@ #include #include -#ifdef CONFIG_HARDLOCKUP_DETECTOR +#ifdef CONFIG_HARDLOCKUP_DETECTOR_PERF u64 hw_nmi_get_sample_period(int watchdog_thresh) { return (u64)(cpu_khz) * 1000 * watchdog_thresh; diff --git a/include/linux/nmi.h b/include/linux/nmi.h index bd387ef8bccd..8aa01fd859fb 100644 --- a/include/linux/nmi.h +++ b/include/linux/nmi.h @@ -11,13 +11,21 @@ #endif #ifdef CONFIG_LOCKUP_DETECTOR +void lockup_detector_init(void); +#else +static inline void lockup_detector_init(void) +{ +} +#endif + +#ifdef CONFIG_SOFTLOCKUP_DETECTOR extern void touch_softlockup_watchdog_sched(void); extern void touch_softlockup_watchdog(void); extern void touch_softlockup_watchdog_sync(void); extern void touch_all_softlockup_watchdogs(void); extern unsigned int softlockup_panic; -extern unsigned int hardlockup_panic; -void lockup_detector_init(void); +extern int soft_watchdog_enabled; +extern atomic_t watchdog_park_in_progress; #else static inline void touch_softlockup_watchdog_sched(void) { @@ -31,9 +39,6 @@ static inline void touch_softlockup_watchdog_sync(void) static inline void touch_all_softlockup_watchdogs(void) { } -static inline void lockup_detector_init(void) -{ -} #endif #ifdef CONFIG_DETECT_HUNG_TASK @@ -63,15 +68,18 @@ static inline void reset_hung_task_detector(void) #if defined(CONFIG_HARDLOCKUP_DETECTOR) extern void hardlockup_detector_disable(void); +extern unsigned int hardlockup_panic; #else static inline void hardlockup_detector_disable(void) {} #endif -#if defined(CONFIG_HARDLOCKUP_DETECTOR) || defined(CONFIG_HAVE_NMI_WATCHDOG) +#if defined(CONFIG_HARDLOCKUP_DETECTOR_PERF) extern void arch_touch_nmi_watchdog(void); #else +#if !defined(CONFIG_HAVE_NMI_WATCHDOG) static inline void arch_touch_nmi_watchdog(void) {} #endif +#endif /** * touch_nmi_watchdog - restart NMI watchdog timeout. @@ -141,15 +149,18 @@ static inline bool trigger_single_cpu_backtrace(int cpu) } #endif -#ifdef CONFIG_LOCKUP_DETECTOR +#ifdef CONFIG_HARDLOCKUP_DETECTOR_PERF u64 hw_nmi_get_sample_period(int watchdog_thresh); +#endif + +#ifdef CONFIG_LOCKUP_DETECTOR extern int nmi_watchdog_enabled; -extern int soft_watchdog_enabled; extern int watchdog_user_enabled; extern int watchdog_thresh; extern unsigned long watchdog_enabled; +extern struct cpumask watchdog_cpumask; extern unsigned long *watchdog_cpumask_bits; -extern atomic_t watchdog_park_in_progress; +extern int __read_mostly watchdog_suspended; #ifdef CONFIG_SMP extern int sysctl_softlockup_all_cpu_backtrace; extern int sysctl_hardlockup_all_cpu_backtrace; diff --git a/kernel/Makefile b/kernel/Makefile index 72aa080f91f0..4cb8e8b23c6e 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -82,7 +82,7 @@ obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_KGDB) += debug/ obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o obj-$(CONFIG_LOCKUP_DETECTOR) += watchdog.o -obj-$(CONFIG_HARDLOCKUP_DETECTOR) += watchdog_hld.o +obj-$(CONFIG_HARDLOCKUP_DETECTOR_PERF) += watchdog_hld.o obj-$(CONFIG_SECCOMP) += seccomp.o obj-$(CONFIG_RELAY) += relay.o obj-$(CONFIG_SYSCTL) += utsname_sysctl.o diff --git a/kernel/sysctl.c b/kernel/sysctl.c index df9f2a367882..6648fbbb8157 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -900,6 +900,14 @@ static struct ctl_table kern_table[] = { .extra2 = &zero, #endif }, + { + .procname = "watchdog_cpumask", + .data = &watchdog_cpumask_bits, + .maxlen = NR_CPUS, + .mode = 0644, + .proc_handler = proc_watchdog_cpumask, + }, +#ifdef CONFIG_SOFTLOCKUP_DETECTOR { .procname = "soft_watchdog", .data = &soft_watchdog_enabled, @@ -909,13 +917,6 @@ static struct ctl_table kern_table[] = { .extra1 = &zero, .extra2 = &one, }, - { - .procname = "watchdog_cpumask", - .data = &watchdog_cpumask_bits, - .maxlen = NR_CPUS, - .mode = 0644, - .proc_handler = proc_watchdog_cpumask, - }, { .procname = "softlockup_panic", .data = &softlockup_panic, @@ -925,27 +926,29 @@ static struct ctl_table kern_table[] = { .extra1 = &zero, .extra2 = &one, }, -#ifdef CONFIG_HARDLOCKUP_DETECTOR +#ifdef CONFIG_SMP { - .procname = "hardlockup_panic", - .data = &hardlockup_panic, + .procname = "softlockup_all_cpu_backtrace", + .data = &sysctl_softlockup_all_cpu_backtrace, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = &zero, .extra2 = &one, }, +#endif /* CONFIG_SMP */ #endif -#ifdef CONFIG_SMP +#ifdef CONFIG_HARDLOCKUP_DETECTOR { - .procname = "softlockup_all_cpu_backtrace", - .data = &sysctl_softlockup_all_cpu_backtrace, + .procname = "hardlockup_panic", + .data = &hardlockup_panic, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = &zero, .extra2 = &one, }, +#ifdef CONFIG_SMP { .procname = "hardlockup_all_cpu_backtrace", .data = &sysctl_hardlockup_all_cpu_backtrace, @@ -957,6 +960,8 @@ static struct ctl_table kern_table[] = { }, #endif /* CONFIG_SMP */ #endif +#endif + #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) { .procname = "unknown_nmi_panic", diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 03e0b69bb5bf..1fba9c3d66dc 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -29,15 +29,58 @@ #include #include +/* Watchdog configuration */ static DEFINE_MUTEX(watchdog_proc_mutex); -#if defined(CONFIG_HAVE_NMI_WATCHDOG) || defined(CONFIG_HARDLOCKUP_DETECTOR) -unsigned long __read_mostly watchdog_enabled = SOFT_WATCHDOG_ENABLED|NMI_WATCHDOG_ENABLED; +int __read_mostly nmi_watchdog_enabled; + +#if defined(CONFIG_HARDLOCKUP_DETECTOR) || defined(CONFIG_HAVE_NMI_WATCHDOG) +unsigned long __read_mostly watchdog_enabled = SOFT_WATCHDOG_ENABLED | + NMI_WATCHDOG_ENABLED; #else unsigned long __read_mostly watchdog_enabled = SOFT_WATCHDOG_ENABLED; #endif -int __read_mostly nmi_watchdog_enabled; + +#ifdef CONFIG_HARDLOCKUP_DETECTOR +/* boot commands */ +/* + * Should we panic when a soft-lockup or hard-lockup occurs: + */ +unsigned int __read_mostly hardlockup_panic = + CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE; +/* + * We may not want to enable hard lockup detection by default in all cases, + * for example when running the kernel as a guest on a hypervisor. In these + * cases this function can be called to disable hard lockup detection. This + * function should only be executed once by the boot processor before the + * kernel command line parameters are parsed, because otherwise it is not + * possible to override this in hardlockup_panic_setup(). + */ +void hardlockup_detector_disable(void) +{ + watchdog_enabled &= ~NMI_WATCHDOG_ENABLED; +} + +static int __init hardlockup_panic_setup(char *str) +{ + if (!strncmp(str, "panic", 5)) + hardlockup_panic = 1; + else if (!strncmp(str, "nopanic", 7)) + hardlockup_panic = 0; + else if (!strncmp(str, "0", 1)) + watchdog_enabled &= ~NMI_WATCHDOG_ENABLED; + else if (!strncmp(str, "1", 1)) + watchdog_enabled |= NMI_WATCHDOG_ENABLED; + return 1; +} +__setup("nmi_watchdog=", hardlockup_panic_setup); + +#endif + +#ifdef CONFIG_SOFTLOCKUP_DETECTOR int __read_mostly soft_watchdog_enabled; +#endif + int __read_mostly watchdog_user_enabled; int __read_mostly watchdog_thresh = 10; @@ -45,15 +88,9 @@ int __read_mostly watchdog_thresh = 10; int __read_mostly sysctl_softlockup_all_cpu_backtrace; int __read_mostly sysctl_hardlockup_all_cpu_backtrace; #endif -static struct cpumask watchdog_cpumask __read_mostly; +struct cpumask watchdog_cpumask __read_mostly; unsigned long *watchdog_cpumask_bits = cpumask_bits(&watchdog_cpumask); -/* Helper for online, unparked cpus. */ -#define for_each_watchdog_cpu(cpu) \ - for_each_cpu_and((cpu), cpu_online_mask, &watchdog_cpumask) - -atomic_t watchdog_park_in_progress = ATOMIC_INIT(0); - /* * The 'watchdog_running' variable is set to 1 when the watchdog threads * are registered/started and is set to 0 when the watchdog threads are @@ -72,7 +109,27 @@ static int __read_mostly watchdog_running; * of 'watchdog_running' cannot change while the watchdog is deactivated * temporarily (see related code in 'proc' handlers). */ -static int __read_mostly watchdog_suspended; +int __read_mostly watchdog_suspended; + +/* + * These functions can be overridden if an architecture implements its + * own hardlockup detector. + */ +int __weak watchdog_nmi_enable(unsigned int cpu) +{ + return 0; +} +void __weak watchdog_nmi_disable(unsigned int cpu) +{ +} + +#ifdef CONFIG_SOFTLOCKUP_DETECTOR + +/* Helper for online, unparked cpus. */ +#define for_each_watchdog_cpu(cpu) \ + for_each_cpu_and((cpu), cpu_online_mask, &watchdog_cpumask) + +atomic_t watchdog_park_in_progress = ATOMIC_INIT(0); static u64 __read_mostly sample_period; @@ -120,6 +177,7 @@ static int __init softlockup_all_cpu_backtrace_setup(char *str) return 1; } __setup("softlockup_all_cpu_backtrace=", softlockup_all_cpu_backtrace_setup); +#ifdef CONFIG_HARDLOCKUP_DETECTOR static int __init hardlockup_all_cpu_backtrace_setup(char *str) { sysctl_hardlockup_all_cpu_backtrace = @@ -128,6 +186,7 @@ static int __init hardlockup_all_cpu_backtrace_setup(char *str) } __setup("hardlockup_all_cpu_backtrace=", hardlockup_all_cpu_backtrace_setup); #endif +#endif /* * Hard-lockup warnings should be triggered after just a few seconds. Soft- @@ -213,18 +272,6 @@ void touch_softlockup_watchdog_sync(void) __this_cpu_write(watchdog_touch_ts, 0); } -/* watchdog detector functions */ -bool is_hardlockup(void) -{ - unsigned long hrint = __this_cpu_read(hrtimer_interrupts); - - if (__this_cpu_read(hrtimer_interrupts_saved) == hrint) - return true; - - __this_cpu_write(hrtimer_interrupts_saved, hrint); - return false; -} - static int is_softlockup(unsigned long touch_ts) { unsigned long now = get_timestamp(); @@ -237,21 +284,21 @@ static int is_softlockup(unsigned long touch_ts) return 0; } -static void watchdog_interrupt_count(void) +/* watchdog detector functions */ +bool is_hardlockup(void) { - __this_cpu_inc(hrtimer_interrupts); -} + unsigned long hrint = __this_cpu_read(hrtimer_interrupts); -/* - * These two functions are mostly architecture specific - * defining them as weak here. - */ -int __weak watchdog_nmi_enable(unsigned int cpu) -{ - return 0; + if (__this_cpu_read(hrtimer_interrupts_saved) == hrint) + return true; + + __this_cpu_write(hrtimer_interrupts_saved, hrint); + return false; } -void __weak watchdog_nmi_disable(unsigned int cpu) + +static void watchdog_interrupt_count(void) { + __this_cpu_inc(hrtimer_interrupts); } static int watchdog_enable_all_cpus(void); @@ -502,57 +549,6 @@ static void watchdog_unpark_threads(void) kthread_unpark(per_cpu(softlockup_watchdog, cpu)); } -/* - * Suspend the hard and soft lockup detector by parking the watchdog threads. - */ -int lockup_detector_suspend(void) -{ - int ret = 0; - - get_online_cpus(); - mutex_lock(&watchdog_proc_mutex); - /* - * Multiple suspend requests can be active in parallel (counted by - * the 'watchdog_suspended' variable). If the watchdog threads are - * running, the first caller takes care that they will be parked. - * The state of 'watchdog_running' cannot change while a suspend - * request is active (see related code in 'proc' handlers). - */ - if (watchdog_running && !watchdog_suspended) - ret = watchdog_park_threads(); - - if (ret == 0) - watchdog_suspended++; - else { - watchdog_disable_all_cpus(); - pr_err("Failed to suspend lockup detectors, disabled\n"); - watchdog_enabled = 0; - } - - mutex_unlock(&watchdog_proc_mutex); - - return ret; -} - -/* - * Resume the hard and soft lockup detector by unparking the watchdog threads. - */ -void lockup_detector_resume(void) -{ - mutex_lock(&watchdog_proc_mutex); - - watchdog_suspended--; - /* - * The watchdog threads are unparked if they were previously running - * and if there is no more active suspend request. - */ - if (watchdog_running && !watchdog_suspended) - watchdog_unpark_threads(); - - mutex_unlock(&watchdog_proc_mutex); - put_online_cpus(); -} - static int update_watchdog_all_cpus(void) { int ret; @@ -604,6 +600,81 @@ static void watchdog_disable_all_cpus(void) } } +#else /* SOFTLOCKUP */ +static int watchdog_park_threads(void) +{ + return 0; +} + +static void watchdog_unpark_threads(void) +{ +} + +static int watchdog_enable_all_cpus(void) +{ + return 0; +} + +static void watchdog_disable_all_cpus(void) +{ +} + +static void set_sample_period(void) +{ +} +#endif /* SOFTLOCKUP */ + +/* + * Suspend the hard and soft lockup detector by parking the watchdog threads. + */ +int lockup_detector_suspend(void) +{ + int ret = 0; + + get_online_cpus(); + mutex_lock(&watchdog_proc_mutex); + /* + * Multiple suspend requests can be active in parallel (counted by + * the 'watchdog_suspended' variable). If the watchdog threads are + * running, the first caller takes care that they will be parked. + * The state of 'watchdog_running' cannot change while a suspend + * request is active (see related code in 'proc' handlers). + */ + if (watchdog_running && !watchdog_suspended) + ret = watchdog_park_threads(); + + if (ret == 0) + watchdog_suspended++; + else { + watchdog_disable_all_cpus(); + pr_err("Failed to suspend lockup detectors, disabled\n"); + watchdog_enabled = 0; + } + + mutex_unlock(&watchdog_proc_mutex); + + return ret; +} + +/* + * Resume the hard and soft lockup detector by unparking the watchdog threads. + */ +void lockup_detector_resume(void) +{ + mutex_lock(&watchdog_proc_mutex); + + watchdog_suspended--; + /* + * The watchdog threads are unparked if they were previously running + * and if there is no more active suspend request. + */ + if (watchdog_running && !watchdog_suspended) + watchdog_unpark_threads(); + + mutex_unlock(&watchdog_proc_mutex); + put_online_cpus(); +} + #ifdef CONFIG_SYSCTL /* @@ -810,9 +881,11 @@ int proc_watchdog_cpumask(struct ctl_table *table, int write, * a temporary cpumask, so we are likely not in a * position to do much else to make things better. */ +#ifdef CONFIG_SOFTLOCKUP_DETECTOR if (smpboot_update_cpumask_percpu_thread( &watchdog_threads, &watchdog_cpumask) != 0) pr_err("cpumask update failed\n"); +#endif } } out: diff --git a/kernel/watchdog_hld.c b/kernel/watchdog_hld.c index 90d688df6ce1..295a0d84934c 100644 --- a/kernel/watchdog_hld.c +++ b/kernel/watchdog_hld.c @@ -22,39 +22,7 @@ static DEFINE_PER_CPU(bool, hard_watchdog_warn); static DEFINE_PER_CPU(bool, watchdog_nmi_touch); static DEFINE_PER_CPU(struct perf_event *, watchdog_ev); -/* boot commands */ -/* - * Should we panic when a soft-lockup or hard-lockup occurs: - */ -unsigned int __read_mostly hardlockup_panic = - CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE; static unsigned long hardlockup_allcpu_dumped; -/* - * We may not want to enable hard lockup detection by default in all cases, - * for example when running the kernel as a guest on a hypervisor. In these - * cases this function can be called to disable hard lockup detection. This - * function should only be executed once by the boot processor before the - * kernel command line parameters are parsed, because otherwise it is not - * possible to override this in hardlockup_panic_setup(). - */ -void hardlockup_detector_disable(void) -{ - watchdog_enabled &= ~NMI_WATCHDOG_ENABLED; -} - -static int __init hardlockup_panic_setup(char *str) -{ - if (!strncmp(str, "panic", 5)) - hardlockup_panic = 1; - else if (!strncmp(str, "nopanic", 7)) - hardlockup_panic = 0; - else if (!strncmp(str, "0", 1)) - watchdog_enabled &= ~NMI_WATCHDOG_ENABLED; - else if (!strncmp(str, "1", 1)) - watchdog_enabled |= NMI_WATCHDOG_ENABLED; - return 1; -} -__setup("nmi_watchdog=", hardlockup_panic_setup); void arch_touch_nmi_watchdog(void) { diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index f28f4252e54a..b0d01c6d4e03 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -778,34 +778,45 @@ config DEBUG_SHIRQ menu "Debug Lockups and Hangs" config LOCKUP_DETECTOR - bool "Detect Hard and Soft Lockups" + bool + +config SOFTLOCKUP_DETECTOR + bool "Detect Soft Lockups" depends on DEBUG_KERNEL && !S390 + select LOCKUP_DETECTOR help Say Y here to enable the kernel to act as a watchdog to detect - hard and soft lockups. + soft lockups. Softlockups are bugs that cause the kernel to loop in kernel mode for more than 20 seconds, without giving other tasks a chance to run. The current stack trace is displayed upon detection and the system will stay locked up. +config HARDLOCKUP_DETECTOR_PERF + bool + select SOFTLOCKUP_DETECTOR + +# +# arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard +# lockup detector rather than the perf based detector. +# +config HARDLOCKUP_DETECTOR + bool "Detect Hard Lockups" + depends on DEBUG_KERNEL && !S390 + depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH + select LOCKUP_DETECTOR + select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF + select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH + help + Say Y here to enable the kernel to act as a watchdog to detect + hard lockups. + Hardlockups are bugs that cause the CPU to loop in kernel mode for more than 10 seconds, without letting other interrupts have a chance to run. The current stack trace is displayed upon detection and the system will stay locked up. - The overhead should be minimal. A periodic hrtimer runs to - generate interrupts and kick the watchdog task every 4 seconds. - An NMI is generated every 10 seconds or so to check for hardlockups. - - The frequency of hrtimer and NMI events and the soft and hard lockup - thresholds can be controlled through the sysctl watchdog_thresh. - -config HARDLOCKUP_DETECTOR - def_bool y - depends on LOCKUP_DETECTOR && !HAVE_NMI_WATCHDOG - depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI - config BOOTPARAM_HARDLOCKUP_PANIC bool "Panic (Reboot) On Hard Lockups" depends on HARDLOCKUP_DETECTOR @@ -826,7 +837,7 @@ config BOOTPARAM_HARDLOCKUP_PANIC_VALUE config BOOTPARAM_SOFTLOCKUP_PANIC bool "Panic (Reboot) On Soft Lockups" - depends on LOCKUP_DETECTOR + depends on SOFTLOCKUP_DETECTOR help Say Y here to enable the kernel to panic on "soft lockups", which are bugs that cause the kernel to loop in kernel @@ -843,7 +854,7 @@ config BOOTPARAM_SOFTLOCKUP_PANIC config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE int - depends on LOCKUP_DETECTOR + depends on SOFTLOCKUP_DETECTOR range 0 1 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC default 1 if BOOTPARAM_SOFTLOCKUP_PANIC @@ -851,7 +862,7 @@ config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE config DETECT_HUNG_TASK bool "Detect Hung Tasks" depends on DEBUG_KERNEL - default LOCKUP_DETECTOR + default SOFTLOCKUP_DETECTOR help Say Y here to enable the kernel to detect "hung tasks", which are bugs that cause the task to be stuck in -- cgit v1.2.3 From a10a842ff81a7e3810817b3b04e4c432b6191e21 Mon Sep 17 00:00:00 2001 From: Nicholas Piggin Date: Wed, 12 Jul 2017 14:35:49 -0700 Subject: kernel/watchdog: provide watchdog_nmi_reconfigure() for arch watchdogs After reconfiguring watchdog sysctls etc., architecture specific watchdogs may not get all their parameters updated. watchdog_nmi_reconfigure() can be implemented to pull the new values in and set the arch NMI watchdog. [npiggin@gmail.com: add code comments] Link: http://lkml.kernel.org/r/20170617125933.774d3858@roar.ozlabs.ibm.com [arnd@arndb.de: hide unused function] Link: http://lkml.kernel.org/r/20170620204854.966601-1-arnd@arndb.de Link: http://lkml.kernel.org/r/20170616065715.18390-5-npiggin@gmail.com Signed-off-by: Nicholas Piggin Signed-off-by: Arnd Bergmann Reviewed-by: Don Zickus Tested-by: Babu Moger [sparc] Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/watchdog.c | 48 ++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 44 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 1fba9c3d66dc..cabe3e9fb620 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -114,6 +114,10 @@ int __read_mostly watchdog_suspended; /* * These functions can be overridden if an architecture implements its * own hardlockup detector. + * + * watchdog_nmi_enable/disable can be implemented to start and stop when + * softlockup watchdog threads start and stop. The arch must select the + * SOFTLOCKUP_DETECTOR Kconfig. */ int __weak watchdog_nmi_enable(unsigned int cpu) { @@ -123,6 +127,22 @@ void __weak watchdog_nmi_disable(unsigned int cpu) { } +/* + * watchdog_nmi_reconfigure can be implemented to be notified after any + * watchdog configuration change. The arch hardlockup watchdog should + * respond to the following variables: + * - nmi_watchdog_enabled + * - watchdog_thresh + * - watchdog_cpumask + * - sysctl_hardlockup_all_cpu_backtrace + * - hardlockup_panic + * - watchdog_suspended + */ +void __weak watchdog_nmi_reconfigure(void) +{ +} + + #ifdef CONFIG_SOFTLOCKUP_DETECTOR /* Helper for online, unparked cpus. */ @@ -600,6 +620,14 @@ static void watchdog_disable_all_cpus(void) } } +#ifdef CONFIG_SYSCTL +static int watchdog_update_cpus(void) +{ + return smpboot_update_cpumask_percpu_thread( + &watchdog_threads, &watchdog_cpumask); +} +#endif + #else /* SOFTLOCKUP */ static int watchdog_park_threads(void) { @@ -619,6 +647,13 @@ static void watchdog_disable_all_cpus(void) { } +#ifdef CONFIG_SYSCTL +static int watchdog_update_cpus(void) +{ + return 0; +} +#endif + static void set_sample_period(void) { } @@ -651,6 +686,8 @@ int lockup_detector_suspend(void) watchdog_enabled = 0; } + watchdog_nmi_reconfigure(); + mutex_unlock(&watchdog_proc_mutex); return ret; @@ -671,6 +708,8 @@ void lockup_detector_resume(void) if (watchdog_running && !watchdog_suspended) watchdog_unpark_threads(); + watchdog_nmi_reconfigure(); + mutex_unlock(&watchdog_proc_mutex); put_online_cpus(); } @@ -696,6 +735,8 @@ static int proc_watchdog_update(void) else watchdog_disable_all_cpus(); + watchdog_nmi_reconfigure(); + return err; } @@ -881,12 +922,11 @@ int proc_watchdog_cpumask(struct ctl_table *table, int write, * a temporary cpumask, so we are likely not in a * position to do much else to make things better. */ -#ifdef CONFIG_SOFTLOCKUP_DETECTOR - if (smpboot_update_cpumask_percpu_thread( - &watchdog_threads, &watchdog_cpumask) != 0) + if (watchdog_update_cpus() != 0) pr_err("cpumask update failed\n"); -#endif } + + watchdog_nmi_reconfigure(); } out: mutex_unlock(&watchdog_proc_mutex); -- cgit v1.2.3 From e2ae8ab4b571e2e4094a28acb60649bc2732c67f Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 12 Jul 2017 14:35:58 -0700 Subject: kexec_file: adjust declaration of kexec_purgatory Defining kexec_purgatory as a zero-length char array upsets compile time size checking. Since this is built on a per-arch basis, define it as an unsized char array (like is done for other similar things, e.g. linker sections). This silences the warning generated by the future CONFIG_FORTIFY_SOURCE, which did not like the memcmp() of a "0 byte" array. This drops the __weak and uses an extern instead, since both users define kexec_purgatory. Link: http://lkml.kernel.org/r/1497903987-21002-4-git-send-email-keescook@chromium.org Signed-off-by: Kees Cook Acked-by: "Eric W. Biederman" Cc: Daniel Micay Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kexec_file.c | 7 ------- kernel/kexec_internal.h | 2 ++ 2 files changed, 2 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/kexec_file.c b/kernel/kexec_file.c index c8f7f77e9fa9..9f48f4412297 100644 --- a/kernel/kexec_file.c +++ b/kernel/kexec_file.c @@ -26,13 +26,6 @@ #include #include "kexec_internal.h" -/* - * Declare these symbols weak so that if architecture provides a purgatory, - * these will be overridden. - */ -char __weak kexec_purgatory[0]; -size_t __weak kexec_purgatory_size = 0; - static int kexec_calculate_store_digests(struct kimage *image); /* Architectures can provide this probe function */ diff --git a/kernel/kexec_internal.h b/kernel/kexec_internal.h index 799a8a452187..50dfcb039a41 100644 --- a/kernel/kexec_internal.h +++ b/kernel/kexec_internal.h @@ -17,6 +17,8 @@ extern struct mutex kexec_mutex; #ifdef CONFIG_KEXEC_FILE #include void kimage_file_post_load_cleanup(struct kimage *image); +extern char kexec_purgatory[]; +extern size_t kexec_purgatory_size; #else /* CONFIG_KEXEC_FILE */ static inline void kimage_file_post_load_cleanup(struct kimage *image) { } #endif /* CONFIG_KEXEC_FILE */ -- cgit v1.2.3 From 7cd815bce828220deffd1654265f0ef891567774 Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Wed, 12 Jul 2017 14:36:20 -0700 Subject: fork,random: use get_random_canary() to set tsk->stack_canary Use the ascii-armor canary to prevent unterminated C string overflows from being able to successfully overwrite the canary, even if they somehow obtain the canary value. Inspired by execshield ascii-armor and Daniel Micay's linux-hardened tree. Link: http://lkml.kernel.org/r/20170524155751.424-3-riel@redhat.com Signed-off-by: Rik van Riel Acked-by: Kees Cook Cc: Daniel Micay Cc: "Theodore Ts'o" Cc: H. Peter Anvin Cc: Andy Lutomirski Cc: Ingo Molnar Cc: Catalin Marinas Cc: Yoshinori Sato Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index ade237a96308..17921b0390b4 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -554,7 +554,7 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node) set_task_stack_end_magic(tsk); #ifdef CONFIG_CC_STACKPROTECTOR - tsk->stack_canary = get_random_long(); + tsk->stack_canary = get_random_canary(); #endif /* -- cgit v1.2.3 From 5f92a7b0fcd627fbd06ceb1cee3bbe5d08d13356 Mon Sep 17 00:00:00 2001 From: Kefeng Wang Date: Fri, 14 Jul 2017 14:49:46 -0700 Subject: kernel/watchdog.c: use better pr_fmt prefix After commit 73ce0511c436 ("kernel/watchdog.c: move hardlockup detector to separate file"), 'NMI watchdog' is inappropriate in kernel/watchdog.c, using 'watchdog' only. Link: http://lkml.kernel.org/r/1499928642-48983-1-git-send-email-wangkefeng.wang@huawei.com Signed-off-by: Kefeng Wang Cc: Babu Moger Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/watchdog.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/watchdog.c b/kernel/watchdog.c index cabe3e9fb620..06d3389bca0d 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -9,7 +9,7 @@ * to those contributors as well. */ -#define pr_fmt(fmt) "NMI watchdog: " fmt +#define pr_fmt(fmt) "watchdog: " fmt #include #include -- cgit v1.2.3 From 6d7964a722afc8e4f880b947f174009063028c99 Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Fri, 14 Jul 2017 14:50:11 -0700 Subject: kmod: throttle kmod thread limit If we reach the limit of modprobe_limit threads running the next request_module() call will fail. The original reason for adding a kill was to do away with possible issues with in old circumstances which would create a recursive series of request_module() calls. We can do better than just be super aggressive and reject calls once we've reached the limit by simply making pending callers wait until the threshold has been reduced, and then throttling them in, one by one. This throttling enables requests over the kmod concurrent limit to be processed once a pending request completes. Only the first item queued up to wait is woken up. The assumption here is once a task is woken it will have no other option to also kick the queue to check if there are more pending tasks -- regardless of whether or not it was successful. By throttling and processing only max kmod concurrent tasks we ensure we avoid unexpected fatal request_module() calls, and we keep memory consumption on module loading to a minimum. With x86_64 qemu, with 4 cores, 4 GiB of RAM it takes the following run time to run both tests: time ./kmod.sh -t 0008 real 0m16.366s user 0m0.883s sys 0m8.916s time ./kmod.sh -t 0009 real 0m50.803s user 0m0.791s sys 0m9.852s Link: http://lkml.kernel.org/r/20170628223155.26472-4-mcgrof@kernel.org Signed-off-by: Luis R. Rodriguez Reviewed-by: Petr Mladek Cc: Jessica Yu Cc: Shuah Khan Cc: Rusty Russell Cc: Michal Marek Cc: Greg Kroah-Hartman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kmod.c | 16 +++++++--------- tools/testing/selftests/kmod/kmod.sh | 24 ++---------------------- 2 files changed, 9 insertions(+), 31 deletions(-) (limited to 'kernel') diff --git a/kernel/kmod.c b/kernel/kmod.c index ff68198fe83b..6d016c5d97c8 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -68,6 +68,7 @@ static DECLARE_RWSEM(umhelper_sem); */ #define MAX_KMOD_CONCURRENT 50 static atomic_t kmod_concurrent_max = ATOMIC_INIT(MAX_KMOD_CONCURRENT); +static DECLARE_WAIT_QUEUE_HEAD(kmod_wq); /* modprobe_path is set via /proc/sys. @@ -140,7 +141,6 @@ int __request_module(bool wait, const char *fmt, ...) va_list args; char module_name[MODULE_NAME_LEN]; int ret; - static int kmod_loop_msg; /* * We don't allow synchronous module loading from async. Module @@ -164,14 +164,11 @@ int __request_module(bool wait, const char *fmt, ...) return ret; if (atomic_dec_if_positive(&kmod_concurrent_max) < 0) { - /* We may be blaming an innocent here, but unlikely */ - if (kmod_loop_msg < 5) { - printk(KERN_ERR - "request_module: runaway loop modprobe %s\n", - module_name); - kmod_loop_msg++; - } - return -ENOMEM; + pr_warn_ratelimited("request_module: kmod_concurrent_max (%u) close to 0 (max_modprobes: %u), for module %s, throttling...", + atomic_read(&kmod_concurrent_max), + MAX_KMOD_CONCURRENT, module_name); + wait_event_interruptible(kmod_wq, + atomic_dec_if_positive(&kmod_concurrent_max) >= 0); } trace_module_request(module_name, wait, _RET_IP_); @@ -179,6 +176,7 @@ int __request_module(bool wait, const char *fmt, ...) ret = call_modprobe(module_name, wait ? UMH_WAIT_PROC : UMH_WAIT_EXEC); atomic_inc(&kmod_concurrent_max); + wake_up(&kmod_wq); return ret; } diff --git a/tools/testing/selftests/kmod/kmod.sh b/tools/testing/selftests/kmod/kmod.sh index 10196a62ed09..8cecae9a8bca 100644 --- a/tools/testing/selftests/kmod/kmod.sh +++ b/tools/testing/selftests/kmod/kmod.sh @@ -59,28 +59,8 @@ ALL_TESTS="$ALL_TESTS 0004:1:1" ALL_TESTS="$ALL_TESTS 0005:10:1" ALL_TESTS="$ALL_TESTS 0006:10:1" ALL_TESTS="$ALL_TESTS 0007:5:1" - -# Disabled tests: -# -# 0008 x 150 - multithreaded - push kmod_concurrent over max_modprobes for request_module()" -# Current best-effort failure interpretation: -# Enough module requests get loaded in place fast enough to reach over the -# max_modprobes limit and trigger a failure -- before we're even able to -# start processing pending requests. -ALL_TESTS="$ALL_TESTS 0008:150:0" - -# 0009 x 150 - multithreaded - push kmod_concurrent over max_modprobes for get_fs_type()" -# Current best-effort failure interpretation: -# -# get_fs_type() requests modules using aliases as such the optimization in -# place today to look for already loaded modules will not take effect and -# we end up requesting a new module to load, this bumps the kmod_concurrent, -# and in certain circumstances can lead to pushing the kmod_concurrent over -# the max_modprobe limit. -# -# This test fails much easier than test 0008 since the alias optimizations -# are not in place. -ALL_TESTS="$ALL_TESTS 0009:150:0" +ALL_TESTS="$ALL_TESTS 0008:150:1" +ALL_TESTS="$ALL_TESTS 0009:150:1" test_modprobe() { -- cgit v1.2.3 From b0659ae5e30074ede1dc08f2c6d64f0c11d64e0f Mon Sep 17 00:00:00 2001 From: Shu Wang Date: Tue, 18 Jul 2017 14:37:24 +0800 Subject: audit: fix memleak in auditd_send_unicast_skb. Found this issue by kmemleak report, auditd_send_unicast_skb did not free skb if rcu_dereference(auditd_conn) returns null. unreferenced object 0xffff88082568ce00 (size 256): comm "auditd", pid 1119, jiffies 4294708499 backtrace: [] kmemleak_alloc+0x4a/0xa0 [] kmem_cache_alloc_node+0xcc/0x210 [] __alloc_skb+0x5d/0x290 [] audit_make_reply+0x54/0xd0 [] audit_receive_msg+0x967/0xd70 ---------------- (gdb) list *audit_receive_msg+0x967 0xffffffff8113dff7 is in audit_receive_msg (kernel/audit.c:1133). 1132 skb = audit_make_reply(0, AUDIT_REPLACE, 0, 0, &pvnr, sizeof(pvnr)); --------------- [] audit_receive+0x52/0xa0 [] netlink_unicast+0x181/0x240 [] netlink_sendmsg+0x2c2/0x3b0 [] sock_sendmsg+0x38/0x50 [] SYSC_sendto+0x102/0x190 [] SyS_sendto+0xe/0x10 [] entry_SYSCALL_64_fastpath+0x1a/0xa5 [] 0xffffffffffffffff Signed-off-by: Shu Wang Signed-off-by: Paul Moore --- kernel/audit.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/audit.c b/kernel/audit.c index 7cad70214b81..07def5e49cc9 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -641,6 +641,7 @@ static int auditd_send_unicast_skb(struct sk_buff *skb) ac = rcu_dereference(auditd_conn); if (!ac) { rcu_read_unlock(); + kfree_skb(skb); rc = -ECONNREFUSED; goto err; } -- cgit v1.2.3 From 4cabc5b186b5427b9ee5a7495172542af105f02b Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 21 Jul 2017 00:00:21 +0200 Subject: bpf: fix mixed signed/unsigned derived min/max value bounds Edward reported that there's an issue in min/max value bounds tracking when signed and unsigned compares both provide hints on limits when having unknown variables. E.g. a program such as the following should have been rejected: 0: (7a) *(u64 *)(r10 -8) = 0 1: (bf) r2 = r10 2: (07) r2 += -8 3: (18) r1 = 0xffff8a94cda93400 5: (85) call bpf_map_lookup_elem#1 6: (15) if r0 == 0x0 goto pc+7 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R10=fp 7: (7a) *(u64 *)(r10 -16) = -8 8: (79) r1 = *(u64 *)(r10 -16) 9: (b7) r2 = -1 10: (2d) if r1 > r2 goto pc+3 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R1=inv,min_value=0 R2=imm-1,max_value=18446744073709551615,min_align=1 R10=fp 11: (65) if r1 s> 0x1 goto pc+2 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R1=inv,min_value=0,max_value=1 R2=imm-1,max_value=18446744073709551615,min_align=1 R10=fp 12: (0f) r0 += r1 13: (72) *(u8 *)(r0 +0) = 0 R0=map_value_adj(ks=8,vs=8,id=0),min_value=0,max_value=1 R1=inv,min_value=0,max_value=1 R2=imm-1,max_value=18446744073709551615,min_align=1 R10=fp 14: (b7) r0 = 0 15: (95) exit What happens is that in the first part ... 8: (79) r1 = *(u64 *)(r10 -16) 9: (b7) r2 = -1 10: (2d) if r1 > r2 goto pc+3 ... r1 carries an unsigned value, and is compared as unsigned against a register carrying an immediate. Verifier deduces in reg_set_min_max() that since the compare is unsigned and operation is greater than (>), that in the fall-through/false case, r1's minimum bound must be 0 and maximum bound must be r2. Latter is larger than the bound and thus max value is reset back to being 'invalid' aka BPF_REGISTER_MAX_RANGE. Thus, r1 state is now 'R1=inv,min_value=0'. The subsequent test ... 11: (65) if r1 s> 0x1 goto pc+2 ... is a signed compare of r1 with immediate value 1. Here, verifier deduces in reg_set_min_max() that since the compare is signed this time and operation is greater than (>), that in the fall-through/false case, we can deduce that r1's maximum bound must be 1, meaning with prior test, we result in r1 having the following state: R1=inv,min_value=0,max_value=1. Given that the actual value this holds is -8, the bounds are wrongly deduced. When this is being added to r0 which holds the map_value(_adj) type, then subsequent store access in above case will go through check_mem_access() which invokes check_map_access_adj(), that will then probe whether the map memory is in bounds based on the min_value and max_value as well as access size since the actual unknown value is min_value <= x <= max_value; commit fce366a9dd0d ("bpf, verifier: fix alu ops against map_value{, _adj} register types") provides some more explanation on the semantics. It's worth to note in this context that in the current code, min_value and max_value tracking are used for two things, i) dynamic map value access via check_map_access_adj() and since commit 06c1c049721a ("bpf: allow helpers access to variable memory") ii) also enforced at check_helper_mem_access() when passing a memory address (pointer to packet, map value, stack) and length pair to a helper and the length in this case is an unknown value defining an access range through min_value/max_value in that case. The min_value/max_value tracking is /not/ used in the direct packet access case to track ranges. However, the issue also affects case ii), for example, the following crafted program based on the same principle must be rejected as well: 0: (b7) r2 = 0 1: (bf) r3 = r10 2: (07) r3 += -512 3: (7a) *(u64 *)(r10 -16) = -8 4: (79) r4 = *(u64 *)(r10 -16) 5: (b7) r6 = -1 6: (2d) if r4 > r6 goto pc+5 R1=ctx R2=imm0,min_value=0,max_value=0,min_align=2147483648 R3=fp-512 R4=inv,min_value=0 R6=imm-1,max_value=18446744073709551615,min_align=1 R10=fp 7: (65) if r4 s> 0x1 goto pc+4 R1=ctx R2=imm0,min_value=0,max_value=0,min_align=2147483648 R3=fp-512 R4=inv,min_value=0,max_value=1 R6=imm-1,max_value=18446744073709551615,min_align=1 R10=fp 8: (07) r4 += 1 9: (b7) r5 = 0 10: (6a) *(u16 *)(r10 -512) = 0 11: (85) call bpf_skb_load_bytes#26 12: (b7) r0 = 0 13: (95) exit Meaning, while we initialize the max_value stack slot that the verifier thinks we access in the [1,2] range, in reality we pass -7 as length which is interpreted as u32 in the helper. Thus, this issue is relevant also for the case of helper ranges. Resetting both bounds in check_reg_overflow() in case only one of them exceeds limits is also not enough as similar test can be created that uses values which are within range, thus also here learned min value in r1 is incorrect when mixed with later signed test to create a range: 0: (7a) *(u64 *)(r10 -8) = 0 1: (bf) r2 = r10 2: (07) r2 += -8 3: (18) r1 = 0xffff880ad081fa00 5: (85) call bpf_map_lookup_elem#1 6: (15) if r0 == 0x0 goto pc+7 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R10=fp 7: (7a) *(u64 *)(r10 -16) = -8 8: (79) r1 = *(u64 *)(r10 -16) 9: (b7) r2 = 2 10: (3d) if r2 >= r1 goto pc+3 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R1=inv,min_value=3 R2=imm2,min_value=2,max_value=2,min_align=2 R10=fp 11: (65) if r1 s> 0x4 goto pc+2 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R1=inv,min_value=3,max_value=4 R2=imm2,min_value=2,max_value=2,min_align=2 R10=fp 12: (0f) r0 += r1 13: (72) *(u8 *)(r0 +0) = 0 R0=map_value_adj(ks=8,vs=8,id=0),min_value=3,max_value=4 R1=inv,min_value=3,max_value=4 R2=imm2,min_value=2,max_value=2,min_align=2 R10=fp 14: (b7) r0 = 0 15: (95) exit This leaves us with two options for fixing this: i) to invalidate all prior learned information once we switch signed context, ii) to track min/max signed and unsigned boundaries separately as done in [0]. (Given latter introduces major changes throughout the whole verifier, it's rather net-next material, thus this patch follows option i), meaning we can derive bounds either from only signed tests or only unsigned tests.) There is still the case of adjust_reg_min_max_vals(), where we adjust bounds on ALU operations, meaning programs like the following where boundaries on the reg get mixed in context later on when bounds are merged on the dst reg must get rejected, too: 0: (7a) *(u64 *)(r10 -8) = 0 1: (bf) r2 = r10 2: (07) r2 += -8 3: (18) r1 = 0xffff89b2bf87ce00 5: (85) call bpf_map_lookup_elem#1 6: (15) if r0 == 0x0 goto pc+6 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R10=fp 7: (7a) *(u64 *)(r10 -16) = -8 8: (79) r1 = *(u64 *)(r10 -16) 9: (b7) r2 = 2 10: (3d) if r2 >= r1 goto pc+2 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R1=inv,min_value=3 R2=imm2,min_value=2,max_value=2,min_align=2 R10=fp 11: (b7) r7 = 1 12: (65) if r7 s> 0x0 goto pc+2 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R1=inv,min_value=3 R2=imm2,min_value=2,max_value=2,min_align=2 R7=imm1,max_value=0 R10=fp 13: (b7) r0 = 0 14: (95) exit from 12 to 15: R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R1=inv,min_value=3 R2=imm2,min_value=2,max_value=2,min_align=2 R7=imm1,min_value=1 R10=fp 15: (0f) r7 += r1 16: (65) if r7 s> 0x4 goto pc+2 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R1=inv,min_value=3 R2=imm2,min_value=2,max_value=2,min_align=2 R7=inv,min_value=4,max_value=4 R10=fp 17: (0f) r0 += r7 18: (72) *(u8 *)(r0 +0) = 0 R0=map_value_adj(ks=8,vs=8,id=0),min_value=4,max_value=4 R1=inv,min_value=3 R2=imm2,min_value=2,max_value=2,min_align=2 R7=inv,min_value=4,max_value=4 R10=fp 19: (b7) r0 = 0 20: (95) exit Meaning, in adjust_reg_min_max_vals() we must also reset range values on the dst when src/dst registers have mixed signed/ unsigned derived min/max value bounds with one unbounded value as otherwise they can be added together deducing false boundaries. Once both boundaries are established from either ALU ops or compare operations w/o mixing signed/unsigned insns, then they can safely be added to other regs also having both boundaries established. Adding regs with one unbounded side to a map value where the bounded side has been learned w/o mixing ops is possible, but the resulting map value won't recover from that, meaning such op is considered invalid on the time of actual access. Invalid bounds are set on the dst reg in case i) src reg, or ii) in case dst reg already had them. The only way to recover would be to perform i) ALU ops but only 'add' is allowed on map value types or ii) comparisons, but these are disallowed on pointers in case they span a range. This is fine as only BPF_JEQ and BPF_JNE may be performed on PTR_TO_MAP_VALUE_OR_NULL registers which potentially turn them into PTR_TO_MAP_VALUE type depending on the branch, so only here min/max value cannot be invalidated for them. In terms of state pruning, value_from_signed is considered as well in states_equal() when dealing with adjusted map values. With regards to breaking existing programs, there is a small risk, but use-cases are rather quite narrow where this could occur and mixing compares probably unlikely. Joint work with Josef and Edward. [0] https://lists.iovisor.org/pipermail/iovisor-dev/2017-June/000822.html Fixes: 484611357c19 ("bpf: allow access into map value arrays") Reported-by: Edward Cree Signed-off-by: Daniel Borkmann Signed-off-by: Edward Cree Signed-off-by: Josef Bacik Signed-off-by: David S. Miller --- include/linux/bpf_verifier.h | 1 + kernel/bpf/verifier.c | 108 +++++++++++++++++++++++++++++++++++++------ 2 files changed, 95 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index 621076f56251..8e5d31f6faef 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -43,6 +43,7 @@ struct bpf_reg_state { u32 min_align; u32 aux_off; u32 aux_off_align; + bool value_from_signed; }; enum bpf_stack_slot_type { diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 6a86723c5b64..af9e84a4944e 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -504,6 +504,7 @@ static void reset_reg_range_values(struct bpf_reg_state *regs, u32 regno) { regs[regno].min_value = BPF_REGISTER_MIN_RANGE; regs[regno].max_value = BPF_REGISTER_MAX_RANGE; + regs[regno].value_from_signed = false; regs[regno].min_align = 0; } @@ -777,12 +778,13 @@ static int check_ctx_access(struct bpf_verifier_env *env, int insn_idx, int off, return -EACCES; } -static bool is_pointer_value(struct bpf_verifier_env *env, int regno) +static bool __is_pointer_value(bool allow_ptr_leaks, + const struct bpf_reg_state *reg) { - if (env->allow_ptr_leaks) + if (allow_ptr_leaks) return false; - switch (env->cur_state.regs[regno].type) { + switch (reg->type) { case UNKNOWN_VALUE: case CONST_IMM: return false; @@ -791,6 +793,11 @@ static bool is_pointer_value(struct bpf_verifier_env *env, int regno) } } +static bool is_pointer_value(struct bpf_verifier_env *env, int regno) +{ + return __is_pointer_value(env->allow_ptr_leaks, &env->cur_state.regs[regno]); +} + static int check_pkt_ptr_alignment(const struct bpf_reg_state *reg, int off, int size, bool strict) { @@ -1832,10 +1839,24 @@ static void adjust_reg_min_max_vals(struct bpf_verifier_env *env, dst_align = dst_reg->min_align; /* We don't know anything about what was done to this register, mark it - * as unknown. + * as unknown. Also, if both derived bounds came from signed/unsigned + * mixed compares and one side is unbounded, we cannot really do anything + * with them as boundaries cannot be trusted. Thus, arithmetic of two + * regs of such kind will get invalidated bounds on the dst side. */ - if (min_val == BPF_REGISTER_MIN_RANGE && - max_val == BPF_REGISTER_MAX_RANGE) { + if ((min_val == BPF_REGISTER_MIN_RANGE && + max_val == BPF_REGISTER_MAX_RANGE) || + (BPF_SRC(insn->code) == BPF_X && + ((min_val != BPF_REGISTER_MIN_RANGE && + max_val == BPF_REGISTER_MAX_RANGE) || + (min_val == BPF_REGISTER_MIN_RANGE && + max_val != BPF_REGISTER_MAX_RANGE) || + (dst_reg->min_value != BPF_REGISTER_MIN_RANGE && + dst_reg->max_value == BPF_REGISTER_MAX_RANGE) || + (dst_reg->min_value == BPF_REGISTER_MIN_RANGE && + dst_reg->max_value != BPF_REGISTER_MAX_RANGE)) && + regs[insn->dst_reg].value_from_signed != + regs[insn->src_reg].value_from_signed)) { reset_reg_range_values(regs, insn->dst_reg); return; } @@ -2023,6 +2044,7 @@ static int check_alu_op(struct bpf_verifier_env *env, struct bpf_insn *insn) regs[insn->dst_reg].max_value = insn->imm; regs[insn->dst_reg].min_value = insn->imm; regs[insn->dst_reg].min_align = calc_align(insn->imm); + regs[insn->dst_reg].value_from_signed = false; } } else if (opcode > BPF_END) { @@ -2198,40 +2220,63 @@ static void reg_set_min_max(struct bpf_reg_state *true_reg, struct bpf_reg_state *false_reg, u64 val, u8 opcode) { + bool value_from_signed = true; + bool is_range = true; + switch (opcode) { case BPF_JEQ: /* If this is false then we know nothing Jon Snow, but if it is * true then we know for sure. */ true_reg->max_value = true_reg->min_value = val; + is_range = false; break; case BPF_JNE: /* If this is true we know nothing Jon Snow, but if it is false * we know the value for sure; */ false_reg->max_value = false_reg->min_value = val; + is_range = false; break; case BPF_JGT: - /* Unsigned comparison, the minimum value is 0. */ - false_reg->min_value = 0; + value_from_signed = false; /* fallthrough */ case BPF_JSGT: + if (true_reg->value_from_signed != value_from_signed) + reset_reg_range_values(true_reg, 0); + if (false_reg->value_from_signed != value_from_signed) + reset_reg_range_values(false_reg, 0); + if (opcode == BPF_JGT) { + /* Unsigned comparison, the minimum value is 0. */ + false_reg->min_value = 0; + } /* If this is false then we know the maximum val is val, * otherwise we know the min val is val+1. */ false_reg->max_value = val; + false_reg->value_from_signed = value_from_signed; true_reg->min_value = val + 1; + true_reg->value_from_signed = value_from_signed; break; case BPF_JGE: - /* Unsigned comparison, the minimum value is 0. */ - false_reg->min_value = 0; + value_from_signed = false; /* fallthrough */ case BPF_JSGE: + if (true_reg->value_from_signed != value_from_signed) + reset_reg_range_values(true_reg, 0); + if (false_reg->value_from_signed != value_from_signed) + reset_reg_range_values(false_reg, 0); + if (opcode == BPF_JGE) { + /* Unsigned comparison, the minimum value is 0. */ + false_reg->min_value = 0; + } /* If this is false then we know the maximum value is val - 1, * otherwise we know the mimimum value is val. */ false_reg->max_value = val - 1; + false_reg->value_from_signed = value_from_signed; true_reg->min_value = val; + true_reg->value_from_signed = value_from_signed; break; default: break; @@ -2239,6 +2284,12 @@ static void reg_set_min_max(struct bpf_reg_state *true_reg, check_reg_overflow(false_reg); check_reg_overflow(true_reg); + if (is_range) { + if (__is_pointer_value(false, false_reg)) + reset_reg_range_values(false_reg, 0); + if (__is_pointer_value(false, true_reg)) + reset_reg_range_values(true_reg, 0); + } } /* Same as above, but for the case that dst_reg is a CONST_IMM reg and src_reg @@ -2248,41 +2299,64 @@ static void reg_set_min_max_inv(struct bpf_reg_state *true_reg, struct bpf_reg_state *false_reg, u64 val, u8 opcode) { + bool value_from_signed = true; + bool is_range = true; + switch (opcode) { case BPF_JEQ: /* If this is false then we know nothing Jon Snow, but if it is * true then we know for sure. */ true_reg->max_value = true_reg->min_value = val; + is_range = false; break; case BPF_JNE: /* If this is true we know nothing Jon Snow, but if it is false * we know the value for sure; */ false_reg->max_value = false_reg->min_value = val; + is_range = false; break; case BPF_JGT: - /* Unsigned comparison, the minimum value is 0. */ - true_reg->min_value = 0; + value_from_signed = false; /* fallthrough */ case BPF_JSGT: + if (true_reg->value_from_signed != value_from_signed) + reset_reg_range_values(true_reg, 0); + if (false_reg->value_from_signed != value_from_signed) + reset_reg_range_values(false_reg, 0); + if (opcode == BPF_JGT) { + /* Unsigned comparison, the minimum value is 0. */ + true_reg->min_value = 0; + } /* * If this is false, then the val is <= the register, if it is * true the register <= to the val. */ false_reg->min_value = val; + false_reg->value_from_signed = value_from_signed; true_reg->max_value = val - 1; + true_reg->value_from_signed = value_from_signed; break; case BPF_JGE: - /* Unsigned comparison, the minimum value is 0. */ - true_reg->min_value = 0; + value_from_signed = false; /* fallthrough */ case BPF_JSGE: + if (true_reg->value_from_signed != value_from_signed) + reset_reg_range_values(true_reg, 0); + if (false_reg->value_from_signed != value_from_signed) + reset_reg_range_values(false_reg, 0); + if (opcode == BPF_JGE) { + /* Unsigned comparison, the minimum value is 0. */ + true_reg->min_value = 0; + } /* If this is false then constant < register, if it is true then * the register < constant. */ false_reg->min_value = val + 1; + false_reg->value_from_signed = value_from_signed; true_reg->max_value = val; + true_reg->value_from_signed = value_from_signed; break; default: break; @@ -2290,6 +2364,12 @@ static void reg_set_min_max_inv(struct bpf_reg_state *true_reg, check_reg_overflow(false_reg); check_reg_overflow(true_reg); + if (is_range) { + if (__is_pointer_value(false, false_reg)) + reset_reg_range_values(false_reg, 0); + if (__is_pointer_value(false, true_reg)) + reset_reg_range_values(true_reg, 0); + } } static void mark_map_reg(struct bpf_reg_state *regs, u32 regno, u32 id, -- cgit v1.2.3