diff options
author | Otavio Salvador <otavio@ossystems.com.br> | 2016-10-05 16:23:28 -0300 |
---|---|---|
committer | Otavio Salvador <otavio@ossystems.com.br> | 2016-10-05 16:23:28 -0300 |
commit | a7f858238e7c8e8c4b9040aac0eb28bb0b6ab9dc (patch) | |
tree | f87a76eb101d6b9ea2016c404982991345d95dc8 /kernel | |
parent | 5bc73053cc0a6a0e5c5f953f8b774854316b87a9 (diff) | |
parent | 04cb720142764ebf3786eba1feb8fc4b6ef87fcf (diff) |
Merge tag 'v4.1.33' into 4.1-2.0.x-imx
Linux 4.1.33
* tag 'v4.1.33': (1760 commits)
Linux 4.1.33
Revert "ARC: mm: don't loose PTE_SPECIAL in pte_modify()"
x86/AMD: Apply erratum 665 on machines without a BIOS fix
x86/paravirt: Do not trace _paravirt_ident_*() functions
ovl: listxattr: use strnlen()
ovl: remove posix_acl_default from workdir
serial: 8250: added acces i/o products quad and octal serial cards
sysfs: correctly handle read offset on PREALLOC attrs
sysfs: correctly handle short reads on PREALLOC attrs.
kernfs: don't depend on d_find_any_alias() when generating notifications
dm crypt: fix free of bad values after tfm allocation failure
dm crypt: fix error with too large bios
dm log writes: fix check of kthread_run() return value
dm log writes: fix bug with too large bios
dm log writes: move IO accounting earlier to fix error path
NFSv4.x: Fix a refcount leak in nfs_callback_up_net
xfs: prevent dropping ioend completions during buftarg wait
xfs: fix superblock inprogress check
USB: serial: option: add WeTelecom 0x6802 and 0x6803 products
USB: avoid left shift by -1
...
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Diffstat (limited to 'kernel')
34 files changed, 788 insertions, 455 deletions
diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 9fb9d1cb83ce..b1943039aab6 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -73,6 +73,7 @@ #include <linux/compat.h> #include <linux/ctype.h> #include <linux/string.h> +#include <linux/uaccess.h> #include <uapi/linux/limits.h> #include "audit.h" @@ -82,7 +83,8 @@ #define AUDITSC_SUCCESS 1 #define AUDITSC_FAILURE 2 -/* no execve audit message should be longer than this (userspace limits) */ +/* no execve audit message should be longer than this (userspace limits), + * see the note near the top of audit_log_execve_info() about this value */ #define MAX_EXECVE_AUDIT_LEN 7500 /* max length to print of cmdline/proctitle value during audit */ @@ -987,185 +989,178 @@ static int audit_log_pid_context(struct audit_context *context, pid_t pid, return rc; } -/* - * to_send and len_sent accounting are very loose estimates. We aren't - * really worried about a hard cap to MAX_EXECVE_AUDIT_LEN so much as being - * within about 500 bytes (next page boundary) - * - * why snprintf? an int is up to 12 digits long. if we just assumed when - * logging that a[%d]= was going to be 16 characters long we would be wasting - * space in every audit message. In one 7500 byte message we can log up to - * about 1000 min size arguments. That comes down to about 50% waste of space - * if we didn't do the snprintf to find out how long arg_num_len was. - */ -static int audit_log_single_execve_arg(struct audit_context *context, - struct audit_buffer **ab, - int arg_num, - size_t *len_sent, - const char __user *p, - char *buf) +static void audit_log_execve_info(struct audit_context *context, + struct audit_buffer **ab) { - char arg_num_len_buf[12]; - const char __user *tmp_p = p; - /* how many digits are in arg_num? 5 is the length of ' a=""' */ - size_t arg_num_len = snprintf(arg_num_len_buf, 12, "%d", arg_num) + 5; - size_t len, len_left, to_send; - size_t max_execve_audit_len = MAX_EXECVE_AUDIT_LEN; - unsigned int i, has_cntl = 0, too_long = 0; - int ret; - - /* strnlen_user includes the null we don't want to send */ - len_left = len = strnlen_user(p, MAX_ARG_STRLEN) - 1; - - /* - * We just created this mm, if we can't find the strings - * we just copied into it something is _very_ wrong. Similar - * for strings that are too long, we should not have created - * any. - */ - if (unlikely((len == -1) || len > MAX_ARG_STRLEN - 1)) { - WARN_ON(1); - send_sig(SIGKILL, current, 0); - return -1; + long len_max; + long len_rem; + long len_full; + long len_buf; + long len_abuf; + long len_tmp; + bool require_data; + bool encode; + unsigned int iter; + unsigned int arg; + char *buf_head; + char *buf; + const char __user *p = (const char __user *)current->mm->arg_start; + + /* NOTE: this buffer needs to be large enough to hold all the non-arg + * data we put in the audit record for this argument (see the + * code below) ... at this point in time 96 is plenty */ + char abuf[96]; + + /* NOTE: we set MAX_EXECVE_AUDIT_LEN to a rather arbitrary limit, the + * current value of 7500 is not as important as the fact that it + * is less than 8k, a setting of 7500 gives us plenty of wiggle + * room if we go over a little bit in the logging below */ + WARN_ON_ONCE(MAX_EXECVE_AUDIT_LEN > 7500); + len_max = MAX_EXECVE_AUDIT_LEN; + + /* scratch buffer to hold the userspace args */ + buf_head = kmalloc(MAX_EXECVE_AUDIT_LEN + 1, GFP_KERNEL); + if (!buf_head) { + audit_panic("out of memory for argv string"); + return; } + buf = buf_head; - /* walk the whole argument looking for non-ascii chars */ + audit_log_format(*ab, "argc=%d", context->execve.argc); + + len_rem = len_max; + len_buf = 0; + len_full = 0; + require_data = true; + encode = false; + iter = 0; + arg = 0; do { - if (len_left > MAX_EXECVE_AUDIT_LEN) - to_send = MAX_EXECVE_AUDIT_LEN; - else - to_send = len_left; - ret = copy_from_user(buf, tmp_p, to_send); - /* - * There is no reason for this copy to be short. We just - * copied them here, and the mm hasn't been exposed to user- - * space yet. - */ - if (ret) { - WARN_ON(1); - send_sig(SIGKILL, current, 0); - return -1; - } - buf[to_send] = '\0'; - has_cntl = audit_string_contains_control(buf, to_send); - if (has_cntl) { - /* - * hex messages get logged as 2 bytes, so we can only - * send half as much in each message - */ - max_execve_audit_len = MAX_EXECVE_AUDIT_LEN / 2; - break; - } - len_left -= to_send; - tmp_p += to_send; - } while (len_left > 0); - - len_left = len; - - if (len > max_execve_audit_len) - too_long = 1; - - /* rewalk the argument actually logging the message */ - for (i = 0; len_left > 0; i++) { - int room_left; - - if (len_left > max_execve_audit_len) - to_send = max_execve_audit_len; - else - to_send = len_left; - - /* do we have space left to send this argument in this ab? */ - room_left = MAX_EXECVE_AUDIT_LEN - arg_num_len - *len_sent; - if (has_cntl) - room_left -= (to_send * 2); - else - room_left -= to_send; - if (room_left < 0) { - *len_sent = 0; - audit_log_end(*ab); - *ab = audit_log_start(context, GFP_KERNEL, AUDIT_EXECVE); - if (!*ab) - return 0; - } + /* NOTE: we don't ever want to trust this value for anything + * serious, but the audit record format insists we + * provide an argument length for really long arguments, + * e.g. > MAX_EXECVE_AUDIT_LEN, so we have no choice but + * to use strncpy_from_user() to obtain this value for + * recording in the log, although we don't use it + * anywhere here to avoid a double-fetch problem */ + if (len_full == 0) + len_full = strnlen_user(p, MAX_ARG_STRLEN) - 1; + + /* read more data from userspace */ + if (require_data) { + /* can we make more room in the buffer? */ + if (buf != buf_head) { + memmove(buf_head, buf, len_buf); + buf = buf_head; + } + + /* fetch as much as we can of the argument */ + len_tmp = strncpy_from_user(&buf_head[len_buf], p, + len_max - len_buf); + if (len_tmp == -EFAULT) { + /* unable to copy from userspace */ + send_sig(SIGKILL, current, 0); + goto out; + } else if (len_tmp == (len_max - len_buf)) { + /* buffer is not large enough */ + require_data = true; + /* NOTE: if we are going to span multiple + * buffers force the encoding so we stand + * a chance at a sane len_full value and + * consistent record encoding */ + encode = true; + len_full = len_full * 2; + p += len_tmp; + } else { + require_data = false; + if (!encode) + encode = audit_string_contains_control( + buf, len_tmp); + /* try to use a trusted value for len_full */ + if (len_full < len_max) + len_full = (encode ? + len_tmp * 2 : len_tmp); + p += len_tmp + 1; + } + len_buf += len_tmp; + buf_head[len_buf] = '\0'; - /* - * first record needs to say how long the original string was - * so we can be sure nothing was lost. - */ - if ((i == 0) && (too_long)) - audit_log_format(*ab, " a%d_len=%zu", arg_num, - has_cntl ? 2*len : len); - - /* - * normally arguments are small enough to fit and we already - * filled buf above when we checked for control characters - * so don't bother with another copy_from_user - */ - if (len >= max_execve_audit_len) - ret = copy_from_user(buf, p, to_send); - else - ret = 0; - if (ret) { - WARN_ON(1); - send_sig(SIGKILL, current, 0); - return -1; + /* length of the buffer in the audit record? */ + len_abuf = (encode ? len_buf * 2 : len_buf + 2); } - buf[to_send] = '\0'; - - /* actually log it */ - audit_log_format(*ab, " a%d", arg_num); - if (too_long) - audit_log_format(*ab, "[%d]", i); - audit_log_format(*ab, "="); - if (has_cntl) - audit_log_n_hex(*ab, buf, to_send); - else - audit_log_string(*ab, buf); - - p += to_send; - len_left -= to_send; - *len_sent += arg_num_len; - if (has_cntl) - *len_sent += to_send * 2; - else - *len_sent += to_send; - } - /* include the null we didn't log */ - return len + 1; -} -static void audit_log_execve_info(struct audit_context *context, - struct audit_buffer **ab) -{ - int i, len; - size_t len_sent = 0; - const char __user *p; - char *buf; + /* write as much as we can to the audit log */ + if (len_buf > 0) { + /* NOTE: some magic numbers here - basically if we + * can't fit a reasonable amount of data into the + * existing audit buffer, flush it and start with + * a new buffer */ + if ((sizeof(abuf) + 8) > len_rem) { + len_rem = len_max; + audit_log_end(*ab); + *ab = audit_log_start(context, + GFP_KERNEL, AUDIT_EXECVE); + if (!*ab) + goto out; + } - p = (const char __user *)current->mm->arg_start; + /* create the non-arg portion of the arg record */ + len_tmp = 0; + if (require_data || (iter > 0) || + ((len_abuf + sizeof(abuf)) > len_rem)) { + if (iter == 0) { + len_tmp += snprintf(&abuf[len_tmp], + sizeof(abuf) - len_tmp, + " a%d_len=%lu", + arg, len_full); + } + len_tmp += snprintf(&abuf[len_tmp], + sizeof(abuf) - len_tmp, + " a%d[%d]=", arg, iter++); + } else + len_tmp += snprintf(&abuf[len_tmp], + sizeof(abuf) - len_tmp, + " a%d=", arg); + WARN_ON(len_tmp >= sizeof(abuf)); + abuf[sizeof(abuf) - 1] = '\0'; + + /* log the arg in the audit record */ + audit_log_format(*ab, "%s", abuf); + len_rem -= len_tmp; + len_tmp = len_buf; + if (encode) { + if (len_abuf > len_rem) + len_tmp = len_rem / 2; /* encoding */ + audit_log_n_hex(*ab, buf, len_tmp); + len_rem -= len_tmp * 2; + len_abuf -= len_tmp * 2; + } else { + if (len_abuf > len_rem) + len_tmp = len_rem - 2; /* quotes */ + audit_log_n_string(*ab, buf, len_tmp); + len_rem -= len_tmp + 2; + /* don't subtract the "2" because we still need + * to add quotes to the remaining string */ + len_abuf -= len_tmp; + } + len_buf -= len_tmp; + buf += len_tmp; + } - audit_log_format(*ab, "argc=%d", context->execve.argc); + /* ready to move to the next argument? */ + if ((len_buf == 0) && !require_data) { + arg++; + iter = 0; + len_full = 0; + require_data = true; + encode = false; + } + } while (arg < context->execve.argc); - /* - * we need some kernel buffer to hold the userspace args. Just - * allocate one big one rather than allocating one of the right size - * for every single argument inside audit_log_single_execve_arg() - * should be <8k allocation so should be pretty safe. - */ - buf = kmalloc(MAX_EXECVE_AUDIT_LEN + 1, GFP_KERNEL); - if (!buf) { - audit_panic("out of memory for argv string"); - return; - } + /* NOTE: the caller handles the final audit_log_end() call */ - for (i = 0; i < context->execve.argc; i++) { - len = audit_log_single_execve_arg(context, ab, i, - &len_sent, p, buf); - if (len <= 0) - break; - p += len; - } - kfree(buf); +out: + kfree(buf_head); } static void show_special(struct audit_context *context, int *call_panic) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 47dcd3aa6e23..efd143dcedf1 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1019,6 +1019,16 @@ static int check_alu_op(struct reg_state *regs, struct bpf_insn *insn) return -EINVAL; } + if ((opcode == BPF_LSH || opcode == BPF_RSH || + opcode == BPF_ARSH) && BPF_SRC(insn->code) == BPF_K) { + int size = BPF_CLASS(insn->code) == BPF_ALU64 ? 64 : 32; + + if (insn->imm < 0 || insn->imm >= size) { + verbose("invalid shift %d\n", insn->imm); + return -EINVAL; + } + } + /* pattern match 'bpf_add Rx, imm' instruction */ if (opcode == BPF_ADD && BPF_CLASS(insn->code) == BPF_ALU64 && regs[insn->dst_reg].type == FRAME_PTR && @@ -1217,6 +1227,7 @@ static int check_ld_abs(struct verifier_env *env, struct bpf_insn *insn) } if (insn->dst_reg != BPF_REG_0 || insn->off != 0 || + BPF_SIZE(insn->code) == BPF_DW || (mode == BPF_ABS && insn->src_reg != BPF_REG_0)) { verbose("BPF_LD_ABS uses reserved fields\n"); return -EINVAL; @@ -1854,7 +1865,6 @@ static int replace_map_fd_with_map_ptr(struct verifier_env *env) if (IS_ERR(map)) { verbose("fd %d is not pointing to valid bpf_map\n", insn->imm); - fdput(f); return PTR_ERR(map); } @@ -1934,7 +1944,7 @@ static void adjust_branches(struct bpf_prog *prog, int pos, int delta) /* adjust offset of jmps if necessary */ if (i < pos && i + insn->off + 1 > pos) insn->off += delta; - else if (i > pos && i + insn->off + 1 < pos) + else if (i > pos + delta && i + insn->off + 1 <= pos + delta) insn->off -= delta; } } diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 4d65b66ae60d..3abce1e0f910 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -4378,14 +4378,15 @@ static void css_free_work_fn(struct work_struct *work) if (ss) { /* css free path */ + struct cgroup_subsys_state *parent = css->parent; int id = css->id; - if (css->parent) - css_put(css->parent); - ss->css_free(css); cgroup_idr_remove(&ss->css_idr, id); cgroup_put(cgrp); + + if (parent) + css_put(parent); } else { /* cgroup free path */ atomic_dec(&cgrp->root->nr_cgrps); @@ -4478,9 +4479,11 @@ static void init_and_link_css(struct cgroup_subsys_state *css, memset(css, 0, sizeof(*css)); css->cgroup = cgrp; css->ss = ss; + css->id = -1; INIT_LIST_HEAD(&css->sibling); INIT_LIST_HEAD(&css->children); css->serial_nr = css_serial_nr_next++; + atomic_set(&css->online_cnt, 0); if (cgroup_parent(cgrp)) { css->parent = cgroup_css(cgroup_parent(cgrp), ss); @@ -4503,6 +4506,10 @@ static int online_css(struct cgroup_subsys_state *css) if (!ret) { css->flags |= CSS_ONLINE; rcu_assign_pointer(css->cgroup->subsys[ss->id], css); + + atomic_inc(&css->online_cnt); + if (css->parent) + atomic_inc(&css->parent->online_cnt); } return ret; } @@ -4558,7 +4565,7 @@ static int create_css(struct cgroup *cgrp, struct cgroup_subsys *ss, err = cgroup_idr_alloc(&ss->css_idr, NULL, 2, 0, GFP_NOWAIT); if (err < 0) - goto err_free_percpu_ref; + goto err_free_css; css->id = err; if (visible) { @@ -4590,9 +4597,6 @@ err_list_del: list_del_rcu(&css->sibling); cgroup_clear_dir(css->cgroup, 1 << css->ss->id); err_free_id: - cgroup_idr_remove(&ss->css_idr, css->id); -err_free_percpu_ref: - percpu_ref_exit(&css->refcnt); err_free_css: call_rcu(&css->rcu_head, css_free_rcu_fn); return err; @@ -4740,10 +4744,15 @@ static void css_killed_work_fn(struct work_struct *work) container_of(work, struct cgroup_subsys_state, destroy_work); mutex_lock(&cgroup_mutex); - offline_css(css); - mutex_unlock(&cgroup_mutex); - css_put(css); + do { + offline_css(css); + css_put(css); + /* @css can't go away while we're holding cgroup_mutex */ + css = css->parent; + } while (css && atomic_dec_and_test(&css->online_cnt)); + + mutex_unlock(&cgroup_mutex); } /* css kill confirmation processing requires process context, bounce */ @@ -4752,8 +4761,10 @@ static void css_killed_ref_fn(struct percpu_ref *ref) struct cgroup_subsys_state *css = container_of(ref, struct cgroup_subsys_state, refcnt); - INIT_WORK(&css->destroy_work, css_killed_work_fn); - queue_work(cgroup_destroy_wq, &css->destroy_work); + if (atomic_dec_and_test(&css->online_cnt)) { + INIT_WORK(&css->destroy_work, css_killed_work_fn); + queue_work(cgroup_destroy_wq, &css->destroy_work); + } } /** diff --git a/kernel/cpuset.c b/kernel/cpuset.c index f0acff0f66c9..388fc6f78c6f 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -2042,6 +2042,20 @@ static void cpuset_bind(struct cgroup_subsys_state *root_css) mutex_unlock(&cpuset_mutex); } +/* + * Make sure the new task conform to the current state of its parent, + * which could have been changed by cpuset just after it inherits the + * state from the parent and before it sits on the cgroup's task list. + */ +void cpuset_fork(struct task_struct *task) +{ + if (task_css_is_root(task, cpuset_cgrp_id)) + return; + + set_cpus_allowed_ptr(task, ¤t->cpus_allowed); + task->mems_allowed = current->mems_allowed; +} + struct cgroup_subsys cpuset_cgrp_subsys = { .css_alloc = cpuset_css_alloc, .css_online = cpuset_css_online, @@ -2051,6 +2065,7 @@ struct cgroup_subsys cpuset_cgrp_subsys = { .cancel_attach = cpuset_cancel_attach, .attach = cpuset_attach, .bind = cpuset_bind, + .fork = cpuset_fork, .legacy_cftypes = files, .early_init = 1, }; diff --git a/kernel/events/core.c b/kernel/events/core.c index e1af58e23bee..6da64f0d0630 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1562,14 +1562,14 @@ event_sched_out(struct perf_event *event, perf_pmu_disable(event->pmu); + event->tstamp_stopped = tstamp; + event->pmu->del(event, 0); + event->oncpu = -1; event->state = PERF_EVENT_STATE_INACTIVE; if (event->pending_disable) { event->pending_disable = 0; event->state = PERF_EVENT_STATE_OFF; } - event->tstamp_stopped = tstamp; - event->pmu->del(event, 0); - event->oncpu = -1; if (!is_software_event(event)) cpuctx->active_oncpu--; @@ -3314,7 +3314,7 @@ find_lively_task_by_vpid(pid_t vpid) /* Reuse ptrace permission checks for now. */ err = -EACCES; - if (!ptrace_may_access(task, PTRACE_MODE_READ)) + if (!ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS)) goto errout; return task; @@ -7641,6 +7641,9 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, } } + /* symmetric to unaccount_event() in _free_event() */ + account_event(event); + return event; err_per_task: @@ -8004,8 +8007,6 @@ SYSCALL_DEFINE5(perf_event_open, } } - account_event(event); - /* * Special case software events and allow them to be part of * any hardware group. @@ -8221,7 +8222,12 @@ err_context: perf_unpin_context(ctx); put_ctx(ctx); err_alloc: - free_event(event); + /* + * If event_file is set, the fput() above will have called ->release() + * and that will take care of freeing the event. + */ + if (!event_file) + free_event(event); err_cpus: put_online_cpus(); err_task: @@ -8265,8 +8271,6 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu, /* Mark owner so we could distinguish it from user events. */ event->owner = EVENT_OWNER_KERNEL; - account_event(event); - ctx = find_get_context(event->pmu, task, event); if (IS_ERR(ctx)) { err = PTR_ERR(ctx); diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index 7f63ad978cb8..dba8894d25cc 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -347,6 +347,7 @@ void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size, bool truncated) { struct ring_buffer *rb = handle->rb; + bool wakeup = truncated; unsigned long aux_head; u64 flags = 0; @@ -375,9 +376,16 @@ void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size, aux_head = rb->user_page->aux_head = local_read(&rb->aux_head); if (aux_head - local_read(&rb->aux_wakeup) >= rb->aux_watermark) { - perf_output_wakeup(handle); + wakeup = true; local_add(rb->aux_watermark, &rb->aux_wakeup); } + + if (wakeup) { + if (truncated) + handle->event->pending_disable = 1; + perf_output_wakeup(handle); + } + handle->event = NULL; local_set(&rb->aux_nest, 0); diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index cb346f26a22d..a89bca964b1f 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -179,8 +179,10 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); err = -EAGAIN; ptep = page_check_address(page, mm, addr, &ptl, 0); - if (!ptep) + if (!ptep) { + mem_cgroup_cancel_charge(kpage, memcg); goto unlock; + } get_page(kpage); page_add_new_anon_rmap(kpage, vma, addr); @@ -207,7 +209,6 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, err = 0; unlock: - mem_cgroup_cancel_charge(kpage, memcg); mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); unlock_page(page); return err; diff --git a/kernel/exit.c b/kernel/exit.c index 22fcc05dec40..819f51ec4f55 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -914,17 +914,28 @@ static int eligible_pid(struct wait_opts *wo, struct task_struct *p) task_pid_type(p, wo->wo_type) == wo->wo_pid; } -static int eligible_child(struct wait_opts *wo, struct task_struct *p) +static int +eligible_child(struct wait_opts *wo, bool ptrace, struct task_struct *p) { if (!eligible_pid(wo, p)) return 0; - /* Wait for all children (clone and not) if __WALL is set; - * otherwise, wait for clone children *only* if __WCLONE is - * set; otherwise, wait for non-clone children *only*. (Note: - * A "clone" child here is one that reports to its parent - * using a signal other than SIGCHLD.) */ - if (((p->exit_signal != SIGCHLD) ^ !!(wo->wo_flags & __WCLONE)) - && !(wo->wo_flags & __WALL)) + + /* + * Wait for all children (clone and not) if __WALL is set or + * if it is traced by us. + */ + if (ptrace || (wo->wo_flags & __WALL)) + return 1; + + /* + * Otherwise, wait for clone children *only* if __WCLONE is set; + * otherwise, wait for non-clone children *only*. + * + * Note: a "clone" child here is one that reports to its parent + * using a signal other than SIGCHLD, or a non-leader thread which + * we can only see if it is traced by us. + */ + if ((p->exit_signal != SIGCHLD) ^ !!(wo->wo_flags & __WCLONE)) return 0; return 1; @@ -1297,7 +1308,7 @@ static int wait_consider_task(struct wait_opts *wo, int ptrace, if (unlikely(exit_state == EXIT_DEAD)) return 0; - ret = eligible_child(wo, p); + ret = eligible_child(wo, ptrace, p); if (!ret) return ret; diff --git a/kernel/futex.c b/kernel/futex.c index 2579e407ff67..2214b70f1910 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1157,10 +1157,20 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) */ newval = FUTEX_WAITERS | task_pid_vnr(new_owner); - if (cmpxchg_futex_value_locked(&curval, uaddr, uval, newval)) + if (cmpxchg_futex_value_locked(&curval, uaddr, uval, newval)) { ret = -EFAULT; - else if (curval != uval) - ret = -EINVAL; + } else if (curval != uval) { + /* + * If a unconditional UNLOCK_PI operation (user space did not + * try the TID->0 transition) raced with a waiter setting the + * FUTEX_WAITERS flag between get_user() and locking the hash + * bucket lock, retry the operation. + */ + if ((FUTEX_TID_MASK & curval) == uval) + ret = -EAGAIN; + else + ret = -EINVAL; + } if (ret) { raw_spin_unlock(&pi_state->pi_mutex.wait_lock); return ret; @@ -1371,8 +1381,8 @@ void requeue_futex(struct futex_q *q, struct futex_hash_bucket *hb1, if (likely(&hb1->chain != &hb2->chain)) { plist_del(&q->list, &hb1->chain); hb_waiters_dec(hb1); - plist_add(&q->list, &hb2->chain); hb_waiters_inc(hb2); + plist_add(&q->list, &hb2->chain); q->lock_ptr = &hb2->lock; } get_futex_key_refs(key2); @@ -2419,6 +2429,15 @@ retry: */ if (ret == -EFAULT) goto pi_faulted; + /* + * A unconditional UNLOCK_PI op raced against a waiter + * setting the FUTEX_WAITERS bit. Try again. + */ + if (ret == -EAGAIN) { + spin_unlock(&hb->lock); + put_futex_key(&key); + goto retry; + } goto out_unlock; } @@ -2632,6 +2651,11 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, if (q.pi_state && (q.pi_state->owner != current)) { spin_lock(q.lock_ptr); ret = fixup_pi_state_owner(uaddr2, &q, current); + /* + * Drop the reference to the pi state which + * the requeue_pi() code acquired for us. + */ + free_pi_state(q.pi_state); spin_unlock(q.lock_ptr); } } else { @@ -2758,7 +2782,7 @@ SYSCALL_DEFINE3(get_robust_list, int, pid, } ret = -EPERM; - if (!ptrace_may_access(p, PTRACE_MODE_READ)) + if (!ptrace_may_access(p, PTRACE_MODE_READ_REALCREDS)) goto err_unlock; head = p->robust_list; diff --git a/kernel/futex_compat.c b/kernel/futex_compat.c index 55c8c9349cfe..4ae3232e7a28 100644 --- a/kernel/futex_compat.c +++ b/kernel/futex_compat.c @@ -155,7 +155,7 @@ COMPAT_SYSCALL_DEFINE3(get_robust_list, int, pid, } ret = -EPERM; - if (!ptrace_may_access(p, PTRACE_MODE_READ)) + if (!ptrace_may_access(p, PTRACE_MODE_READ_REALCREDS)) goto err_unlock; head = p->compat_robust_list; diff --git a/kernel/kcmp.c b/kernel/kcmp.c index 0aa69ea1d8fd..3a47fa998fe0 100644 --- a/kernel/kcmp.c +++ b/kernel/kcmp.c @@ -122,8 +122,8 @@ SYSCALL_DEFINE5(kcmp, pid_t, pid1, pid_t, pid2, int, type, &task2->signal->cred_guard_mutex); if (ret) goto err; - if (!ptrace_may_access(task1, PTRACE_MODE_READ) || - !ptrace_may_access(task2, PTRACE_MODE_READ)) { + if (!ptrace_may_access(task1, PTRACE_MODE_READ_REALCREDS) || + !ptrace_may_access(task2, PTRACE_MODE_READ_REALCREDS)) { ret = -EPERM; goto err_unlock; } diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c index 4cccea6b8934..6f1c3e41b2a6 100644 --- a/kernel/locking/mutex.c +++ b/kernel/locking/mutex.c @@ -486,9 +486,6 @@ __ww_mutex_lock_check_stamp(struct mutex *lock, struct ww_acquire_ctx *ctx) if (!hold_ctx) return 0; - if (unlikely(ctx == hold_ctx)) - return -EALREADY; - if (ctx->stamp - hold_ctx->stamp <= LONG_MAX && (ctx->stamp != hold_ctx->stamp || ctx > hold_ctx)) { #ifdef CONFIG_DEBUG_MUTEXES @@ -514,6 +511,12 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, unsigned long flags; int ret; + if (use_ww_ctx) { + struct ww_mutex *ww = container_of(lock, struct ww_mutex, base); + if (unlikely(ww_ctx == READ_ONCE(ww->ctx))) + return -EALREADY; + } + preempt_disable(); mutex_acquire_nest(&lock->dep_map, subclass, 0, nest_lock, ip); diff --git a/kernel/module.c b/kernel/module.c index 3b9ff966edb9..6920d1080cdd 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -178,6 +178,9 @@ struct load_info { struct _ddebug *debug; unsigned int num_debug; bool sig_ok; +#ifdef CONFIG_KALLSYMS + unsigned long mod_kallsyms_init_off; +#endif struct { unsigned int sym, str, mod, vers, info, pcpu; } index; @@ -2321,10 +2324,21 @@ static void layout_symtab(struct module *mod, struct load_info *info) strsect->sh_flags |= SHF_ALLOC; strsect->sh_entsize = get_offset(mod, &mod->init_size, strsect, info->index.str) | INIT_OFFSET_MASK; - mod->init_size = debug_align(mod->init_size); pr_debug("\t%s\n", info->secstrings + strsect->sh_name); + + /* We'll tack temporary mod_kallsyms on the end. */ + mod->init_size = ALIGN(mod->init_size, + __alignof__(struct mod_kallsyms)); + info->mod_kallsyms_init_off = mod->init_size; + mod->init_size += sizeof(struct mod_kallsyms); + mod->init_size = debug_align(mod->init_size); } +/* + * We use the full symtab and strtab which layout_symtab arranged to + * be appended to the init section. Later we switch to the cut-down + * core-only ones. + */ static void add_kallsyms(struct module *mod, const struct load_info *info) { unsigned int i, ndst; @@ -2333,28 +2347,33 @@ static void add_kallsyms(struct module *mod, const struct load_info *info) char *s; Elf_Shdr *symsec = &info->sechdrs[info->index.sym]; - mod->symtab = (void *)symsec->sh_addr; - mod->num_symtab = symsec->sh_size / sizeof(Elf_Sym); + /* Set up to point into init section. */ + mod->kallsyms = mod->module_init + info->mod_kallsyms_init_off; + + mod->kallsyms->symtab = (void *)symsec->sh_addr; + mod->kallsyms->num_symtab = symsec->sh_size / sizeof(Elf_Sym); /* Make sure we get permanent strtab: don't use info->strtab. */ - mod->strtab = (void *)info->sechdrs[info->index.str].sh_addr; + mod->kallsyms->strtab = (void *)info->sechdrs[info->index.str].sh_addr; /* Set types up while we still have access to sections. */ - for (i = 0; i < mod->num_symtab; i++) - mod->symtab[i].st_info = elf_type(&mod->symtab[i], info); - - mod->core_symtab = dst = mod->module_core + info->symoffs; - mod->core_strtab = s = mod->module_core + info->stroffs; - src = mod->symtab; - for (ndst = i = 0; i < mod->num_symtab; i++) { + for (i = 0; i < mod->kallsyms->num_symtab; i++) + mod->kallsyms->symtab[i].st_info + = elf_type(&mod->kallsyms->symtab[i], info); + + /* Now populate the cut down core kallsyms for after init. */ + mod->core_kallsyms.symtab = dst = mod->module_core + info->symoffs; + mod->core_kallsyms.strtab = s = mod->module_core + info->stroffs; + src = mod->kallsyms->symtab; + for (ndst = i = 0; i < mod->kallsyms->num_symtab; i++) { if (i == 0 || is_core_symbol(src+i, info->sechdrs, info->hdr->e_shnum)) { dst[ndst] = src[i]; - dst[ndst++].st_name = s - mod->core_strtab; - s += strlcpy(s, &mod->strtab[src[i].st_name], + dst[ndst++].st_name = s - mod->core_kallsyms.strtab; + s += strlcpy(s, &mod->kallsyms->strtab[src[i].st_name], KSYM_NAME_LEN) + 1; } } - mod->core_num_syms = ndst; + mod->core_kallsyms.num_symtab = ndst; } #else static inline void layout_symtab(struct module *mod, struct load_info *info) @@ -2432,13 +2451,18 @@ static inline void kmemleak_load_module(const struct module *mod, #endif #ifdef CONFIG_MODULE_SIG -static int module_sig_check(struct load_info *info) +static int module_sig_check(struct load_info *info, int flags) { int err = -ENOKEY; const unsigned long markerlen = sizeof(MODULE_SIG_STRING) - 1; const void *mod = info->hdr; - if (info->len > markerlen && + /* + * Require flags == 0, as a module with version information + * removed is no longer the module that was signed + */ + if (flags == 0 && + info->len > markerlen && memcmp(mod + info->len - markerlen, MODULE_SIG_STRING, markerlen) == 0) { /* We truncate the module to discard the signature */ info->len -= markerlen; @@ -2457,7 +2481,7 @@ static int module_sig_check(struct load_info *info) return err; } #else /* !CONFIG_MODULE_SIG */ -static int module_sig_check(struct load_info *info) +static int module_sig_check(struct load_info *info, int flags) { return 0; } @@ -3119,9 +3143,8 @@ static noinline int do_init_module(struct module *mod) module_put(mod); trim_init_extable(mod); #ifdef CONFIG_KALLSYMS - mod->num_symtab = mod->core_num_syms; - mod->symtab = mod->core_symtab; - mod->strtab = mod->core_strtab; + /* Switch to core kallsyms now init is done: kallsyms may be walking! */ + rcu_assign_pointer(mod->kallsyms, &mod->core_kallsyms); #endif unset_module_init_ro_nx(mod); module_arch_freeing_init(mod); @@ -3259,7 +3282,7 @@ static int load_module(struct load_info *info, const char __user *uargs, long err; char *after_dashes; - err = module_sig_check(info); + err = module_sig_check(info, flags); if (err) goto free_copy; @@ -3469,6 +3492,11 @@ static inline int is_arm_mapping_symbol(const char *str) && (str[2] == '\0' || str[2] == '.'); } +static const char *symname(struct mod_kallsyms *kallsyms, unsigned int symnum) +{ + return kallsyms->strtab + kallsyms->symtab[symnum].st_name; +} + static const char *get_ksymbol(struct module *mod, unsigned long addr, unsigned long *size, @@ -3476,6 +3504,7 @@ static const char *get_ksymbol(struct module *mod, { unsigned int i, best = 0; unsigned long nextval; + struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms); /* At worse, next value is at end of module */ if (within_module_init(addr, mod)) @@ -3485,32 +3514,32 @@ static const char *get_ksymbol(struct module *mod, /* Scan for closest preceding symbol, and next symbol. (ELF starts real symbols at 1). */ - for (i = 1; i < mod->num_symtab; i++) { - if (mod->symtab[i].st_shndx == SHN_UNDEF) + for (i = 1; i < kallsyms->num_symtab; i++) { + if (kallsyms->symtab[i].st_shndx == SHN_UNDEF) continue; /* We ignore unnamed symbols: they're uninformative * and inserted at a whim. */ - if (mod->symtab[i].st_value <= addr - && mod->symtab[i].st_value > mod->symtab[best].st_value - && *(mod->strtab + mod->symtab[i].st_name) != '\0' - && !is_arm_mapping_symbol(mod->strtab + mod->symtab[i].st_name)) + if (*symname(kallsyms, i) == '\0' + || is_arm_mapping_symbol(symname(kallsyms, i))) + continue; + + if (kallsyms->symtab[i].st_value <= addr + && kallsyms->symtab[i].st_value > kallsyms->symtab[best].st_value) best = i; - if (mod->symtab[i].st_value > addr - && mod->symtab[i].st_value < nextval - && *(mod->strtab + mod->symtab[i].st_name) != '\0' - && !is_arm_mapping_symbol(mod->strtab + mod->symtab[i].st_name)) - nextval = mod->symtab[i].st_value; + if (kallsyms->symtab[i].st_value > addr + && kallsyms->symtab[i].st_value < nextval) + nextval = kallsyms->symtab[i].st_value; } if (!best) return NULL; if (size) - *size = nextval - mod->symtab[best].st_value; + *size = nextval - kallsyms->symtab[best].st_value; if (offset) - *offset = addr - mod->symtab[best].st_value; - return mod->strtab + mod->symtab[best].st_name; + *offset = addr - kallsyms->symtab[best].st_value; + return symname(kallsyms, best); } /* For kallsyms to ask for address resolution. NULL means not found. Careful @@ -3603,19 +3632,21 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type, preempt_disable(); list_for_each_entry_rcu(mod, &modules, list) { + struct mod_kallsyms *kallsyms; + if (mod->state == MODULE_STATE_UNFORMED) continue; - if (symnum < mod->num_symtab) { - *value = mod->symtab[symnum].st_value; - *type = mod->symtab[symnum].st_info; - strlcpy(name, mod->strtab + mod->symtab[symnum].st_name, - KSYM_NAME_LEN); + kallsyms = rcu_dereference_sched(mod->kallsyms); + if (symnum < kallsyms->num_symtab) { + *value = kallsyms->symtab[symnum].st_value; + *type = kallsyms->symtab[symnum].st_info; + strlcpy(name, symname(kallsyms, symnum), KSYM_NAME_LEN); strlcpy(module_name, mod->name, MODULE_NAME_LEN); *exported = is_exported(name, *value, mod); preempt_enable(); return 0; } - symnum -= mod->num_symtab; + symnum -= kallsyms->num_symtab; } preempt_enable(); return -ERANGE; @@ -3624,11 +3655,12 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type, static unsigned long mod_find_symname(struct module *mod, const char *name) { unsigned int i; + struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms); - for (i = 0; i < mod->num_symtab; i++) - if (strcmp(name, mod->strtab+mod->symtab[i].st_name) == 0 && - mod->symtab[i].st_info != 'U') - return mod->symtab[i].st_value; + for (i = 0; i < kallsyms->num_symtab; i++) + if (strcmp(name, symname(kallsyms, i)) == 0 && + kallsyms->symtab[i].st_info != 'U') + return kallsyms->symtab[i].st_value; return 0; } @@ -3665,11 +3697,14 @@ int module_kallsyms_on_each_symbol(int (*fn)(void *, const char *, int ret; list_for_each_entry(mod, &modules, list) { + /* We hold module_mutex: no need for rcu_dereference_sched */ + struct mod_kallsyms *kallsyms = mod->kallsyms; + if (mod->state == MODULE_STATE_UNFORMED) continue; - for (i = 0; i < mod->num_symtab; i++) { - ret = fn(data, mod->strtab + mod->symtab[i].st_name, - mod, mod->symtab[i].st_value); + for (i = 0; i < kallsyms->num_symtab; i++) { + ret = fn(data, symname(kallsyms, i), + mod, kallsyms->symtab[i].st_value); if (ret != 0) return ret; } diff --git a/kernel/panic.c b/kernel/panic.c index 8136ad76e5fd..a4f7820f5930 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -23,6 +23,7 @@ #include <linux/sysrq.h> #include <linux/init.h> #include <linux/nmi.h> +#include <linux/console.h> #define PANIC_TIMER_STEP 100 #define PANIC_BLINK_SPD 18 @@ -146,6 +147,17 @@ void panic(const char *fmt, ...) bust_spinlocks(0); + /* + * We may have ended up stopping the CPU holding the lock (in + * smp_send_stop()) while still having some valuable data in the console + * buffer. Try to acquire the lock then release it regardless of the + * result. The release will also print the buffers out. Locks debug + * should be disabled to avoid reporting bad unlock balance when + * panic() is not being callled from OOPS. + */ + debug_locks_off(); + console_flush_on_panic(); + if (!panic_blink) panic_blink = no_blink; diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index bff0169e1ad8..3c1aca0c3543 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -2173,13 +2173,24 @@ void console_unlock(void) static u64 seen_seq; unsigned long flags; bool wake_klogd = false; - bool retry; + bool do_cond_resched, retry; if (console_suspended) { up_console_sem(); return; } + /* + * Console drivers are called under logbuf_lock, so + * @console_may_schedule should be cleared before; however, we may + * end up dumping a lot of lines, for example, if called from + * console registration path, and should invoke cond_resched() + * between lines if allowable. Not doing so can cause a very long + * scheduling stall on a slow console leading to RCU stall and + * softlockup warnings which exacerbate the issue with more + * messages practically incapacitating the system. + */ + do_cond_resched = console_may_schedule; console_may_schedule = 0; /* flush buffered message fragment immediately to console */ @@ -2241,6 +2252,9 @@ skip: call_console_drivers(level, text, len); start_critical_timings(); local_irq_restore(flags); + + if (do_cond_resched) + cond_resched(); } console_locked = 0; @@ -2308,6 +2322,25 @@ void console_unblank(void) console_unlock(); } +/** + * console_flush_on_panic - flush console content on panic + * + * Immediately output all pending messages no matter what. + */ +void console_flush_on_panic(void) +{ + /* + * If someone else is holding the console lock, trylock will fail + * and may_schedule may be set. Ignore and proceed to unlock so + * that messages are flushed out. As this can be called from any + * context and we don't want to get preempted while flushing, + * ensure may_schedule is cleared. + */ + console_trylock(); + console_may_schedule = 0; + console_unlock(); +} + /* * Return the console tty driver structure and its associated index */ diff --git a/kernel/ptrace.c b/kernel/ptrace.c index c8e0e050a36a..261ee21e62db 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -219,6 +219,14 @@ static int ptrace_has_cap(struct user_namespace *ns, unsigned int mode) static int __ptrace_may_access(struct task_struct *task, unsigned int mode) { const struct cred *cred = current_cred(), *tcred; + int dumpable = 0; + kuid_t caller_uid; + kgid_t caller_gid; + + if (!(mode & PTRACE_MODE_FSCREDS) == !(mode & PTRACE_MODE_REALCREDS)) { + WARN(1, "denying ptrace access check without PTRACE_MODE_*CREDS\n"); + return -EPERM; + } /* May we inspect the given task? * This check is used both for attaching with ptrace @@ -228,18 +236,33 @@ static int __ptrace_may_access(struct task_struct *task, unsigned int mode) * because setting up the necessary parent/child relationship * or halting the specified task is impossible. */ - int dumpable = 0; + /* Don't let security modules deny introspection */ if (same_thread_group(task, current)) return 0; rcu_read_lock(); + if (mode & PTRACE_MODE_FSCREDS) { + caller_uid = cred->fsuid; + caller_gid = cred->fsgid; + } else { + /* + * Using the euid would make more sense here, but something + * in userland might rely on the old behavior, and this + * shouldn't be a security problem since + * PTRACE_MODE_REALCREDS implies that the caller explicitly + * used a syscall that requests access to another process + * (and not a filesystem syscall to procfs). + */ + caller_uid = cred->uid; + caller_gid = cred->gid; + } tcred = __task_cred(task); - if (uid_eq(cred->uid, tcred->euid) && - uid_eq(cred->uid, tcred->suid) && - uid_eq(cred->uid, tcred->uid) && - gid_eq(cred->gid, tcred->egid) && - gid_eq(cred->gid, tcred->sgid) && - gid_eq(cred->gid, tcred->gid)) + if (uid_eq(caller_uid, tcred->euid) && + uid_eq(caller_uid, tcred->suid) && + uid_eq(caller_uid, tcred->uid) && + gid_eq(caller_gid, tcred->egid) && + gid_eq(caller_gid, tcred->sgid) && + gid_eq(caller_gid, tcred->gid)) goto ok; if (ptrace_has_cap(tcred->user_ns, mode)) goto ok; @@ -306,7 +329,7 @@ static int ptrace_attach(struct task_struct *task, long request, goto out; task_lock(task); - retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH); + retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH_REALCREDS); task_unlock(task); if (retval) goto unlock_creds; diff --git a/kernel/resource.c b/kernel/resource.c index 90552aab5f2d..a7c27cb71fc5 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -1072,9 +1072,10 @@ struct resource * __request_region(struct resource *parent, if (!conflict) break; if (conflict != parent) { - parent = conflict; - if (!(conflict->flags & IORESOURCE_BUSY)) + if (!(conflict->flags & IORESOURCE_BUSY)) { + parent = conflict; continue; + } } if (conflict->flags & flags & IORESOURCE_MUXED) { add_wait_queue(&muxed_resource_wait, &wait); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 4d870eb6086b..98e607121d09 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -4568,14 +4568,16 @@ void show_state_filter(unsigned long state_filter) /* * reset the NMI-timeout, listing all files on a slow * console might take a lot of time: + * Also, reset softlockup watchdogs on all CPUs, because + * another CPU might be blocked waiting for us to process + * an IPI. */ touch_nmi_watchdog(); + touch_all_softlockup_watchdogs(); if (!state_filter || (p->state & state_filter)) sched_show_task(p); } - touch_all_softlockup_watchdogs(); - #ifdef CONFIG_SCHED_DEBUG sysrq_sched_debug_show(); #endif @@ -6480,7 +6482,7 @@ static void sched_init_numa(void) sched_domains_numa_masks[i][j] = mask; - for (k = 0; k < nr_node_ids; k++) { + for_each_node(k) { if (node_distance(j, k) > sched_domains_numa_distance[i]) continue; diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 8394b1ee600c..87b8576cbd50 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -259,21 +259,21 @@ static __always_inline bool steal_account_process_tick(void) #ifdef CONFIG_PARAVIRT if (static_key_false(¶virt_steal_enabled)) { u64 steal; - cputime_t steal_ct; + unsigned long steal_jiffies; steal = paravirt_steal_clock(smp_processor_id()); steal -= this_rq()->prev_steal_time; /* - * cputime_t may be less precise than nsecs (eg: if it's - * based on jiffies). Lets cast the result to cputime + * steal is in nsecs but our caller is expecting steal + * time in jiffies. Lets cast the result to jiffies * granularity and account the rest on the next rounds. */ - steal_ct = nsecs_to_cputime(steal); - this_rq()->prev_steal_time += cputime_to_nsecs(steal_ct); + steal_jiffies = nsecs_to_jiffies(steal); + this_rq()->prev_steal_time += jiffies_to_nsecs(steal_jiffies); - account_steal_time(steal_ct); - return steal_ct; + account_steal_time(jiffies_to_cputime(steal_jiffies)); + return steal_jiffies; } #endif return false; diff --git a/kernel/sched/proc.c b/kernel/sched/proc.c index 8ecd552fe4f2..09dfe51f89b5 100644 --- a/kernel/sched/proc.c +++ b/kernel/sched/proc.c @@ -97,10 +97,13 @@ long calc_load_fold_active(struct rq *this_rq) static unsigned long calc_load(unsigned long load, unsigned long exp, unsigned long active) { - load *= exp; - load += active * (FIXED_1 - exp); - load += 1UL << (FSHIFT - 1); - return load >> FSHIFT; + unsigned long newload; + + newload = load * exp + active * (FIXED_1 - exp); + if (active >= load) + newload += FIXED_1-1; + + return newload / FIXED_1; } #ifdef CONFIG_NO_HZ_COMMON diff --git a/kernel/seccomp.c b/kernel/seccomp.c index 4f44028943e6..30c682adcdeb 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -317,24 +317,24 @@ static inline void seccomp_sync_threads(void) put_seccomp_filter(thread); smp_store_release(&thread->seccomp.filter, caller->seccomp.filter); + + /* + * Don't let an unprivileged task work around + * the no_new_privs restriction by creating + * a thread that sets it up, enters seccomp, + * then dies. + */ + if (task_no_new_privs(caller)) + task_set_no_new_privs(thread); + /* * Opt the other thread into seccomp if needed. * As threads are considered to be trust-realm * equivalent (see ptrace_may_access), it is safe to * allow one thread to transition the other. */ - if (thread->seccomp.mode == SECCOMP_MODE_DISABLED) { - /* - * Don't let an unprivileged task work around - * the no_new_privs restriction by creating - * a thread that sets it up, enters seccomp, - * then dies. - */ - if (task_no_new_privs(caller)) - task_set_no_new_privs(thread); - + if (thread->seccomp.mode == SECCOMP_MODE_DISABLED) seccomp_assign_mode(thread, SECCOMP_MODE_FILTER); - } } } diff --git a/kernel/sys.c b/kernel/sys.c index a4e372b798a5..25ae8d2e65e2 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1854,11 +1854,13 @@ static int prctl_set_mm_map(int opt, const void __user *addr, unsigned long data user_auxv[AT_VECTOR_SIZE - 1] = AT_NULL; } - if (prctl_map.exe_fd != (u32)-1) + if (prctl_map.exe_fd != (u32)-1) { error = prctl_set_mm_exe_file(mm, prctl_map.exe_fd); - down_read(&mm->mmap_sem); - if (error) - goto out; + if (error) + return error; + } + + down_write(&mm->mmap_sem); /* * We don't validate if these members are pointing to @@ -1895,10 +1897,8 @@ static int prctl_set_mm_map(int opt, const void __user *addr, unsigned long data if (prctl_map.auxv_size) memcpy(mm->saved_auxv, user_auxv, sizeof(user_auxv)); - error = 0; -out: - up_read(&mm->mmap_sem); - return error; + up_write(&mm->mmap_sem); + return 0; } #endif /* CONFIG_CHECKPOINT_RESTORE */ @@ -1930,7 +1930,7 @@ static int prctl_set_mm(int opt, unsigned long addr, error = -EINVAL; - down_read(&mm->mmap_sem); + down_write(&mm->mmap_sem); vma = find_vma(mm, addr); switch (opt) { @@ -2033,7 +2033,7 @@ static int prctl_set_mm(int opt, unsigned long addr, error = 0; out: - up_read(&mm->mmap_sem); + up_write(&mm->mmap_sem); return error; } diff --git a/kernel/sysctl.c b/kernel/sysctl.c index c3eee4c6d6c1..7d4900404c94 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1695,6 +1695,20 @@ static struct ctl_table fs_table[] = { .proc_handler = &pipe_proc_fn, .extra1 = &pipe_min_size, }, + { + .procname = "pipe-user-pages-hard", + .data = &pipe_user_pages_hard, + .maxlen = sizeof(pipe_user_pages_hard), + .mode = 0644, + .proc_handler = proc_doulongvec_minmax, + }, + { + .procname = "pipe-user-pages-soft", + .data = &pipe_user_pages_soft, + .maxlen = sizeof(pipe_user_pages_soft), + .mode = 0644, + .proc_handler = proc_doulongvec_minmax, + }, { } }; diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index 7e7746a42a62..10a1d7dc9313 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -1321,7 +1321,7 @@ static ssize_t binary_sysctl(const int *name, int nlen, } mnt = task_active_pid_ns(current)->proc_mnt; - file = file_open_root(mnt->mnt_root, mnt, pathname, flags); + file = file_open_root(mnt->mnt_root, mnt, pathname, flags, 0); result = PTR_ERR(file); if (IS_ERR(file)) goto out_putname; diff --git a/kernel/time/posix-clock.c b/kernel/time/posix-clock.c index ce033c7aa2e8..9cff0ab82b63 100644 --- a/kernel/time/posix-clock.c +++ b/kernel/time/posix-clock.c @@ -69,10 +69,10 @@ static ssize_t posix_clock_read(struct file *fp, char __user *buf, static unsigned int posix_clock_poll(struct file *fp, poll_table *wait) { struct posix_clock *clk = get_posix_clock(fp); - int result = 0; + unsigned int result = 0; if (!clk) - return -ENODEV; + return POLLERR; if (clk->ops.poll) result = clk->ops.poll(clk, fp, wait); diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 0075da74abf0..57d1acb91c56 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -784,6 +784,7 @@ static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec *itp) timer->it.cpu.expires = 0; sample_to_timespec(timer->it_clock, timer->it.cpu.expires, &itp->it_value); + return; } else { cpu_timer_sample_group(timer->it_clock, p, &now); unlock_task_sighand(p, &flags); diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 414d9df94724..8fcc801fde15 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -316,8 +316,7 @@ static inline s64 timekeeping_get_ns(struct tk_read_base *tkr) delta = timekeeping_get_delta(tkr); - nsec = delta * tkr->mult + tkr->xtime_nsec; - nsec >>= tkr->shift; + nsec = (delta * tkr->mult + tkr->xtime_nsec) >> tkr->shift; /* If arch requires, add in get_arch_timeoffset() */ return nsec + arch_gettimeoffset(); @@ -420,7 +419,10 @@ static __always_inline u64 __ktime_get_fast_ns(struct tk_fast *tkf) do { seq = raw_read_seqcount(&tkf->seq); tkr = tkf->base + (seq & 0x01); - now = ktime_to_ns(tkr->base) + timekeeping_get_ns(tkr); + now = ktime_to_ns(tkr->base); + + now += clocksource_delta(tkr->read(tkr->clock), + tkr->cycle_last, tkr->mask); } while (read_seqcount_retry(&tkf->seq, seq)); return now; diff --git a/kernel/time/timekeeping_debug.c b/kernel/time/timekeeping_debug.c index f6bd65236712..107310a6f36f 100644 --- a/kernel/time/timekeeping_debug.c +++ b/kernel/time/timekeeping_debug.c @@ -23,7 +23,9 @@ #include "timekeeping_internal.h" -static unsigned int sleep_time_bin[32] = {0}; +#define NUM_BINS 32 + +static unsigned int sleep_time_bin[NUM_BINS] = {0}; static int tk_debug_show_sleep_time(struct seq_file *s, void *data) { @@ -69,6 +71,9 @@ late_initcall(tk_debug_sleep_time_init); void tk_debug_account_sleep_time(struct timespec64 *t) { - sleep_time_bin[fls(t->tv_sec)]++; + /* Cap bin index so we don't overflow the array */ + int bin = min(fls(t->tv_sec), NUM_BINS-1); + + sleep_time_bin[bin]++; } diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 0315d43176d8..fb147c7811d2 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -461,7 +461,8 @@ struct ring_buffer_per_cpu { raw_spinlock_t reader_lock; /* serialize readers */ arch_spinlock_t lock; struct lock_class_key lock_key; - unsigned int nr_pages; + unsigned long nr_pages; + unsigned int current_context; struct list_head *pages; struct buffer_page *head_page; /* read from head */ struct buffer_page *tail_page; /* write to tail */ @@ -481,7 +482,7 @@ struct ring_buffer_per_cpu { u64 write_stamp; u64 read_stamp; /* ring buffer pages to update, > 0 to add, < 0 to remove */ - int nr_pages_to_update; + long nr_pages_to_update; struct list_head new_pages; /* new pages to add */ struct work_struct update_pages_work; struct completion update_done; @@ -1160,10 +1161,10 @@ static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer) return 0; } -static int __rb_allocate_pages(int nr_pages, struct list_head *pages, int cpu) +static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int cpu) { - int i; struct buffer_page *bpage, *tmp; + long i; for (i = 0; i < nr_pages; i++) { struct page *page; @@ -1200,7 +1201,7 @@ free_pages: } static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, - unsigned nr_pages) + unsigned long nr_pages) { LIST_HEAD(pages); @@ -1225,7 +1226,7 @@ static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, } static struct ring_buffer_per_cpu * -rb_allocate_cpu_buffer(struct ring_buffer *buffer, int nr_pages, int cpu) +rb_allocate_cpu_buffer(struct ring_buffer *buffer, long nr_pages, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; struct buffer_page *bpage; @@ -1325,8 +1326,9 @@ struct ring_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, struct lock_class_key *key) { struct ring_buffer *buffer; + long nr_pages; int bsize; - int cpu, nr_pages; + int cpu; /* keep it in its own cache line */ buffer = kzalloc(ALIGN(sizeof(*buffer), cache_line_size()), @@ -1452,12 +1454,12 @@ static inline unsigned long rb_page_write(struct buffer_page *bpage) } static int -rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned int nr_pages) +rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages) { struct list_head *tail_page, *to_remove, *next_page; struct buffer_page *to_remove_page, *tmp_iter_page; struct buffer_page *last_page, *first_page; - unsigned int nr_removed; + unsigned long nr_removed; unsigned long head_bit; int page_entries; @@ -1674,7 +1676,7 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size, int cpu_id) { struct ring_buffer_per_cpu *cpu_buffer; - unsigned nr_pages; + unsigned long nr_pages; int cpu, err = 0; /* @@ -1688,14 +1690,13 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size, !cpumask_test_cpu(cpu_id, buffer->cpumask)) return size; - size = DIV_ROUND_UP(size, BUF_PAGE_SIZE); - size *= BUF_PAGE_SIZE; + nr_pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE); /* we need a minimum of two pages */ - if (size < BUF_PAGE_SIZE * 2) - size = BUF_PAGE_SIZE * 2; + if (nr_pages < 2) + nr_pages = 2; - nr_pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE); + size = nr_pages * BUF_PAGE_SIZE; /* * Don't succeed if resizing is disabled, as a reader might be @@ -2675,11 +2676,11 @@ rb_reserve_next_event(struct ring_buffer *buffer, * just so happens that it is the same bit corresponding to * the current context. */ -static DEFINE_PER_CPU(unsigned int, current_context); -static __always_inline int trace_recursive_lock(void) +static __always_inline int +trace_recursive_lock(struct ring_buffer_per_cpu *cpu_buffer) { - unsigned int val = __this_cpu_read(current_context); + unsigned int val = cpu_buffer->current_context; int bit; if (in_interrupt()) { @@ -2696,20 +2697,21 @@ static __always_inline int trace_recursive_lock(void) return 1; val |= (1 << bit); - __this_cpu_write(current_context, val); + cpu_buffer->current_context = val; return 0; } -static __always_inline void trace_recursive_unlock(void) +static __always_inline void +trace_recursive_unlock(struct ring_buffer_per_cpu *cpu_buffer) { - __this_cpu_and(current_context, __this_cpu_read(current_context) - 1); + cpu_buffer->current_context &= cpu_buffer->current_context - 1; } #else -#define trace_recursive_lock() (0) -#define trace_recursive_unlock() do { } while (0) +#define trace_recursive_lock(cpu_buffer) (0) +#define trace_recursive_unlock(cpu_buffer) do { } while (0) #endif @@ -2741,35 +2743,34 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, unsigned long length) /* If we are tracing schedule, we don't want to recurse */ preempt_disable_notrace(); - if (atomic_read(&buffer->record_disabled)) - goto out_nocheck; - - if (trace_recursive_lock()) - goto out_nocheck; + if (unlikely(atomic_read(&buffer->record_disabled))) + goto out; cpu = raw_smp_processor_id(); - if (!cpumask_test_cpu(cpu, buffer->cpumask)) + if (unlikely(!cpumask_test_cpu(cpu, buffer->cpumask))) goto out; cpu_buffer = buffer->buffers[cpu]; - if (atomic_read(&cpu_buffer->record_disabled)) + if (unlikely(atomic_read(&cpu_buffer->record_disabled))) goto out; - if (length > BUF_MAX_DATA_SIZE) + if (unlikely(length > BUF_MAX_DATA_SIZE)) + goto out; + + if (unlikely(trace_recursive_lock(cpu_buffer))) goto out; event = rb_reserve_next_event(buffer, cpu_buffer, length); if (!event) - goto out; + goto out_unlock; return event; + out_unlock: + trace_recursive_unlock(cpu_buffer); out: - trace_recursive_unlock(); - - out_nocheck: preempt_enable_notrace(); return NULL; } @@ -2859,7 +2860,7 @@ int ring_buffer_unlock_commit(struct ring_buffer *buffer, rb_wakeups(buffer, cpu_buffer); - trace_recursive_unlock(); + trace_recursive_unlock(cpu_buffer); preempt_enable_notrace(); @@ -2970,7 +2971,7 @@ void ring_buffer_discard_commit(struct ring_buffer *buffer, out: rb_end_commit(cpu_buffer); - trace_recursive_unlock(); + trace_recursive_unlock(cpu_buffer); preempt_enable_notrace(); @@ -4647,8 +4648,9 @@ static int rb_cpu_notify(struct notifier_block *self, struct ring_buffer *buffer = container_of(self, struct ring_buffer, cpu_notify); long cpu = (long)hcpu; - int cpu_i, nr_pages_same; - unsigned int nr_pages; + long nr_pages_same; + int cpu_i; + unsigned long nr_pages; switch (action) { case CPU_UP_PREPARE: diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 05330494a0df..de6ea94c41bb 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -4916,7 +4916,10 @@ static ssize_t tracing_splice_read_pipe(struct file *filp, spd.nr_pages = i; - ret = splice_to_pipe(pipe, &spd); + if (i) + ret = splice_to_pipe(pipe, &spd); + else + ret = 0; out: splice_shrink_spd(&spd); return ret; diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index c4de47fc5cca..6459f77e2c72 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -683,7 +683,8 @@ t_next(struct seq_file *m, void *v, loff_t *pos) * The ftrace subsystem is for showing formats only. * They can not be enabled or disabled via the event files. */ - if (call->class && call->class->reg) + if (call->class && call->class->reg && + !(call->flags & TRACE_EVENT_FL_IGNORE_ENABLE)) return file; } @@ -1625,8 +1626,13 @@ event_create_dir(struct dentry *parent, struct ftrace_event_file *file) trace_create_file("filter", 0644, file->dir, file, &ftrace_event_filter_fops); - trace_create_file("trigger", 0644, file->dir, file, - &event_trigger_fops); + /* + * Only event directories that can be enabled should have + * triggers. + */ + if (!(call->flags & TRACE_EVENT_FL_IGNORE_ENABLE)) + trace_create_file("trigger", 0644, file->dir, file, + &event_trigger_fops); trace_create_file("format", 0444, file->dir, call, &ftrace_event_format_fops); diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c index 36c1455b7567..fc0c74f18288 100644 --- a/kernel/trace/trace_printk.c +++ b/kernel/trace/trace_printk.c @@ -36,6 +36,10 @@ struct trace_bprintk_fmt { static inline struct trace_bprintk_fmt *lookup_format(const char *fmt) { struct trace_bprintk_fmt *pos; + + if (!fmt) + return ERR_PTR(-EINVAL); + list_for_each_entry(pos, &trace_bprintk_fmt_list, list) { if (!strcmp(pos->fmt, fmt)) return pos; @@ -57,7 +61,8 @@ void hold_module_trace_bprintk_format(const char **start, const char **end) for (iter = start; iter < end; iter++) { struct trace_bprintk_fmt *tb_fmt = lookup_format(*iter); if (tb_fmt) { - *iter = tb_fmt->fmt; + if (!IS_ERR(tb_fmt)) + *iter = tb_fmt->fmt; continue; } @@ -289,6 +294,9 @@ static int t_show(struct seq_file *m, void *v) const char *str = *fmt; int i; + if (!*fmt) + return 0; + seq_printf(m, "0x%lx : \"", *(unsigned long *)fmt); /* diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 581a68a04c64..f89ea713213f 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -812,6 +812,9 @@ static int proc_watchdog_common(int which, struct ctl_table *table, int write, * Update the run state of the lockup detectors. * Restore 'watchdog_enabled' on failure. */ + if (old == new) + goto out; + err = proc_watchdog_update(); if (err) watchdog_enabled = old; @@ -857,7 +860,7 @@ int proc_soft_watchdog(struct ctl_table *table, int write, int proc_watchdog_thresh(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - int err, old; + int err, old, new; mutex_lock(&watchdog_proc_mutex); @@ -871,6 +874,10 @@ int proc_watchdog_thresh(struct ctl_table *table, int write, * Update the sample period. * Restore 'watchdog_thresh' on failure. */ + new = ACCESS_ONCE(watchdog_thresh); + if (old == new) + goto out; + set_sample_period(); err = proc_watchdog_update(); if (err) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 5c01664c26e2..d0efe9295a0e 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -127,6 +127,11 @@ enum { * * PR: wq_pool_mutex protected for writes. Sched-RCU protected for reads. * + * PW: wq_pool_mutex and wq->mutex protected for writes. Either for reads. + * + * PWR: wq_pool_mutex and wq->mutex protected for writes. Either or + * sched-RCU for reads. + * * WQ: wq->mutex protected. * * WR: wq->mutex protected for writes. Sched-RCU protected for reads. @@ -247,8 +252,8 @@ struct workqueue_struct { int nr_drainers; /* WQ: drain in progress */ int saved_max_active; /* WQ: saved pwq max_active */ - struct workqueue_attrs *unbound_attrs; /* WQ: only for unbound wqs */ - struct pool_workqueue *dfl_pwq; /* WQ: only for unbound wqs */ + struct workqueue_attrs *unbound_attrs; /* PW: only for unbound wqs */ + struct pool_workqueue *dfl_pwq; /* PW: only for unbound wqs */ #ifdef CONFIG_SYSFS struct wq_device *wq_dev; /* I: for sysfs interface */ @@ -268,7 +273,7 @@ struct workqueue_struct { /* hot fields used during command issue, aligned to cacheline */ unsigned int flags ____cacheline_aligned; /* WQ: WQ_* flags */ struct pool_workqueue __percpu *cpu_pwqs; /* I: per-cpu pwqs */ - struct pool_workqueue __rcu *numa_pwq_tbl[]; /* FR: unbound pwqs indexed by node */ + struct pool_workqueue __rcu *numa_pwq_tbl[]; /* PWR: unbound pwqs indexed by node */ }; static struct kmem_cache *pwq_cache; @@ -347,6 +352,12 @@ static void workqueue_sysfs_unregister(struct workqueue_struct *wq); lockdep_is_held(&wq->mutex), \ "sched RCU or wq->mutex should be held") +#define assert_rcu_or_wq_mutex_or_pool_mutex(wq) \ + rcu_lockdep_assert(rcu_read_lock_sched_held() || \ + lockdep_is_held(&wq->mutex) || \ + lockdep_is_held(&wq_pool_mutex), \ + "sched RCU, wq->mutex or wq_pool_mutex should be held") + #define for_each_cpu_worker_pool(pool, cpu) \ for ((pool) = &per_cpu(cpu_worker_pools, cpu)[0]; \ (pool) < &per_cpu(cpu_worker_pools, cpu)[NR_STD_WORKER_POOLS]; \ @@ -551,7 +562,8 @@ static int worker_pool_assign_id(struct worker_pool *pool) * @wq: the target workqueue * @node: the node ID * - * This must be called either with pwq_lock held or sched RCU read locked. + * This must be called with any of wq_pool_mutex, wq->mutex or sched RCU + * read locked. * If the pwq needs to be used beyond the locking in effect, the caller is * responsible for guaranteeing that the pwq stays online. * @@ -560,7 +572,17 @@ static int worker_pool_assign_id(struct worker_pool *pool) static struct pool_workqueue *unbound_pwq_by_node(struct workqueue_struct *wq, int node) { - assert_rcu_or_wq_mutex(wq); + assert_rcu_or_wq_mutex_or_pool_mutex(wq); + + /* + * XXX: @node can be NUMA_NO_NODE if CPU goes offline while a + * delayed item is pending. The plan is to keep CPU -> NODE + * mapping valid and stable across CPU on/offlines. Once that + * happens, this workaround can be removed. + */ + if (unlikely(node == NUMA_NO_NODE)) + return wq->dfl_pwq; + return rcu_dereference_raw(wq->numa_pwq_tbl[node]); } @@ -632,6 +654,35 @@ static void set_work_pool_and_clear_pending(struct work_struct *work, */ smp_wmb(); set_work_data(work, (unsigned long)pool_id << WORK_OFFQ_POOL_SHIFT, 0); + /* + * The following mb guarantees that previous clear of a PENDING bit + * will not be reordered with any speculative LOADS or STORES from + * work->current_func, which is executed afterwards. This possible + * reordering can lead to a missed execution on attempt to qeueue + * the same @work. E.g. consider this case: + * + * CPU#0 CPU#1 + * ---------------------------- -------------------------------- + * + * 1 STORE event_indicated + * 2 queue_work_on() { + * 3 test_and_set_bit(PENDING) + * 4 } set_..._and_clear_pending() { + * 5 set_work_data() # clear bit + * 6 smp_mb() + * 7 work->current_func() { + * 8 LOAD event_indicated + * } + * + * Without an explicit full barrier speculative LOAD on line 8 can + * be executed before CPU#0 does STORE on line 1. If that happens, + * CPU#0 observes the PENDING bit is still set and new execution of + * a @work is not queued in a hope, that CPU#1 will eventually + * finish the queued @work. Meanwhile CPU#1 does not see + * event_indicated is set, because speculative LOAD was executed + * before actual STORE. + */ + smp_mb(); } static void clear_work_data(struct work_struct *work) @@ -1451,13 +1502,13 @@ static void __queue_delayed_work(int cpu, struct workqueue_struct *wq, timer_stats_timer_set_start_info(&dwork->timer); dwork->wq = wq; - /* timer isn't guaranteed to run in this cpu, record earlier */ - if (cpu == WORK_CPU_UNBOUND) - cpu = raw_smp_processor_id(); dwork->cpu = cpu; timer->expires = jiffies + delay; - add_timer_on(timer, cpu); + if (unlikely(cpu != WORK_CPU_UNBOUND)) + add_timer_on(timer, cpu); + else + add_timer(timer); } /** @@ -3425,17 +3476,6 @@ static struct pool_workqueue *alloc_unbound_pwq(struct workqueue_struct *wq, return pwq; } -/* undo alloc_unbound_pwq(), used only in the error path */ -static void free_unbound_pwq(struct pool_workqueue *pwq) -{ - lockdep_assert_held(&wq_pool_mutex); - - if (pwq) { - put_unbound_pool(pwq->pool); - kmem_cache_free(pwq_cache, pwq); - } -} - /** * wq_calc_node_mask - calculate a wq_attrs' cpumask for the specified node * @attrs: the wq_attrs of interest @@ -3488,6 +3528,7 @@ static struct pool_workqueue *numa_pwq_tbl_install(struct workqueue_struct *wq, { struct pool_workqueue *old_pwq; + lockdep_assert_held(&wq_pool_mutex); lockdep_assert_held(&wq->mutex); /* link_pwq() can handle duplicate calls */ @@ -3498,42 +3539,48 @@ static struct pool_workqueue *numa_pwq_tbl_install(struct workqueue_struct *wq, return old_pwq; } -/** - * apply_workqueue_attrs - apply new workqueue_attrs to an unbound workqueue - * @wq: the target workqueue - * @attrs: the workqueue_attrs to apply, allocated with alloc_workqueue_attrs() - * - * Apply @attrs to an unbound workqueue @wq. Unless disabled, on NUMA - * machines, this function maps a separate pwq to each NUMA node with - * possibles CPUs in @attrs->cpumask so that work items are affine to the - * NUMA node it was issued on. Older pwqs are released as in-flight work - * items finish. Note that a work item which repeatedly requeues itself - * back-to-back will stay on its current pwq. - * - * Performs GFP_KERNEL allocations. - * - * Return: 0 on success and -errno on failure. - */ -int apply_workqueue_attrs(struct workqueue_struct *wq, - const struct workqueue_attrs *attrs) +/* context to store the prepared attrs & pwqs before applying */ +struct apply_wqattrs_ctx { + struct workqueue_struct *wq; /* target workqueue */ + struct workqueue_attrs *attrs; /* attrs to apply */ + struct pool_workqueue *dfl_pwq; + struct pool_workqueue *pwq_tbl[]; +}; + +/* free the resources after success or abort */ +static void apply_wqattrs_cleanup(struct apply_wqattrs_ctx *ctx) +{ + if (ctx) { + int node; + + for_each_node(node) + put_pwq_unlocked(ctx->pwq_tbl[node]); + put_pwq_unlocked(ctx->dfl_pwq); + + free_workqueue_attrs(ctx->attrs); + + kfree(ctx); + } +} + +/* allocate the attrs and pwqs for later installation */ +static struct apply_wqattrs_ctx * +apply_wqattrs_prepare(struct workqueue_struct *wq, + const struct workqueue_attrs *attrs) { + struct apply_wqattrs_ctx *ctx; struct workqueue_attrs *new_attrs, *tmp_attrs; - struct pool_workqueue **pwq_tbl, *dfl_pwq; - int node, ret; + int node; - /* only unbound workqueues can change attributes */ - if (WARN_ON(!(wq->flags & WQ_UNBOUND))) - return -EINVAL; + lockdep_assert_held(&wq_pool_mutex); - /* creating multiple pwqs breaks ordering guarantee */ - if (WARN_ON((wq->flags & __WQ_ORDERED) && !list_empty(&wq->pwqs))) - return -EINVAL; + ctx = kzalloc(sizeof(*ctx) + nr_node_ids * sizeof(ctx->pwq_tbl[0]), + GFP_KERNEL); - pwq_tbl = kzalloc(nr_node_ids * sizeof(pwq_tbl[0]), GFP_KERNEL); new_attrs = alloc_workqueue_attrs(GFP_KERNEL); tmp_attrs = alloc_workqueue_attrs(GFP_KERNEL); - if (!pwq_tbl || !new_attrs || !tmp_attrs) - goto enomem; + if (!ctx || !new_attrs || !tmp_attrs) + goto out_free; /* make a copy of @attrs and sanitize it */ copy_workqueue_attrs(new_attrs, attrs); @@ -3547,75 +3594,111 @@ int apply_workqueue_attrs(struct workqueue_struct *wq, copy_workqueue_attrs(tmp_attrs, new_attrs); /* - * CPUs should stay stable across pwq creations and installations. - * Pin CPUs, determine the target cpumask for each node and create - * pwqs accordingly. - */ - get_online_cpus(); - - mutex_lock(&wq_pool_mutex); - - /* * If something goes wrong during CPU up/down, we'll fall back to * the default pwq covering whole @attrs->cpumask. Always create * it even if we don't use it immediately. */ - dfl_pwq = alloc_unbound_pwq(wq, new_attrs); - if (!dfl_pwq) - goto enomem_pwq; + ctx->dfl_pwq = alloc_unbound_pwq(wq, new_attrs); + if (!ctx->dfl_pwq) + goto out_free; for_each_node(node) { if (wq_calc_node_cpumask(attrs, node, -1, tmp_attrs->cpumask)) { - pwq_tbl[node] = alloc_unbound_pwq(wq, tmp_attrs); - if (!pwq_tbl[node]) - goto enomem_pwq; + ctx->pwq_tbl[node] = alloc_unbound_pwq(wq, tmp_attrs); + if (!ctx->pwq_tbl[node]) + goto out_free; } else { - dfl_pwq->refcnt++; - pwq_tbl[node] = dfl_pwq; + ctx->dfl_pwq->refcnt++; + ctx->pwq_tbl[node] = ctx->dfl_pwq; } } - mutex_unlock(&wq_pool_mutex); + ctx->attrs = new_attrs; + ctx->wq = wq; + free_workqueue_attrs(tmp_attrs); + return ctx; + +out_free: + free_workqueue_attrs(tmp_attrs); + free_workqueue_attrs(new_attrs); + apply_wqattrs_cleanup(ctx); + return NULL; +} + +/* set attrs and install prepared pwqs, @ctx points to old pwqs on return */ +static void apply_wqattrs_commit(struct apply_wqattrs_ctx *ctx) +{ + int node; /* all pwqs have been created successfully, let's install'em */ - mutex_lock(&wq->mutex); + mutex_lock(&ctx->wq->mutex); - copy_workqueue_attrs(wq->unbound_attrs, new_attrs); + copy_workqueue_attrs(ctx->wq->unbound_attrs, ctx->attrs); /* save the previous pwq and install the new one */ for_each_node(node) - pwq_tbl[node] = numa_pwq_tbl_install(wq, node, pwq_tbl[node]); + ctx->pwq_tbl[node] = numa_pwq_tbl_install(ctx->wq, node, + ctx->pwq_tbl[node]); /* @dfl_pwq might not have been used, ensure it's linked */ - link_pwq(dfl_pwq); - swap(wq->dfl_pwq, dfl_pwq); + link_pwq(ctx->dfl_pwq); + swap(ctx->wq->dfl_pwq, ctx->dfl_pwq); - mutex_unlock(&wq->mutex); + mutex_unlock(&ctx->wq->mutex); +} - /* put the old pwqs */ - for_each_node(node) - put_pwq_unlocked(pwq_tbl[node]); - put_pwq_unlocked(dfl_pwq); +/** + * apply_workqueue_attrs - apply new workqueue_attrs to an unbound workqueue + * @wq: the target workqueue + * @attrs: the workqueue_attrs to apply, allocated with alloc_workqueue_attrs() + * + * Apply @attrs to an unbound workqueue @wq. Unless disabled, on NUMA + * machines, this function maps a separate pwq to each NUMA node with + * possibles CPUs in @attrs->cpumask so that work items are affine to the + * NUMA node it was issued on. Older pwqs are released as in-flight work + * items finish. Note that a work item which repeatedly requeues itself + * back-to-back will stay on its current pwq. + * + * Performs GFP_KERNEL allocations. + * + * Return: 0 on success and -errno on failure. + */ +int apply_workqueue_attrs(struct workqueue_struct *wq, + const struct workqueue_attrs *attrs) +{ + struct apply_wqattrs_ctx *ctx; + int ret = -ENOMEM; - put_online_cpus(); - ret = 0; - /* fall through */ -out_free: - free_workqueue_attrs(tmp_attrs); - free_workqueue_attrs(new_attrs); - kfree(pwq_tbl); - return ret; + /* only unbound workqueues can change attributes */ + if (WARN_ON(!(wq->flags & WQ_UNBOUND))) + return -EINVAL; + + /* creating multiple pwqs breaks ordering guarantee */ + if (WARN_ON((wq->flags & __WQ_ORDERED) && !list_empty(&wq->pwqs))) + return -EINVAL; + + /* + * CPUs should stay stable across pwq creations and installations. + * Pin CPUs, determine the target cpumask for each node and create + * pwqs accordingly. + */ + get_online_cpus(); + mutex_lock(&wq_pool_mutex); + + ctx = apply_wqattrs_prepare(wq, attrs); + + /* the ctx has been prepared successfully, let's commit it */ + if (ctx) { + apply_wqattrs_commit(ctx); + ret = 0; + } -enomem_pwq: - free_unbound_pwq(dfl_pwq); - for_each_node(node) - if (pwq_tbl && pwq_tbl[node] != dfl_pwq) - free_unbound_pwq(pwq_tbl[node]); mutex_unlock(&wq_pool_mutex); put_online_cpus(); -enomem: - ret = -ENOMEM; - goto out_free; + + apply_wqattrs_cleanup(ctx); + + return ret; } /** @@ -4394,6 +4477,17 @@ static void rebind_workers(struct worker_pool *pool) pool->attrs->cpumask) < 0); spin_lock_irq(&pool->lock); + + /* + * XXX: CPU hotplug notifiers are weird and can call DOWN_FAILED + * w/o preceding DOWN_PREPARE. Work around it. CPU hotplug is + * being reworked and this can go away in time. + */ + if (!(pool->flags & POOL_DISASSOCIATED)) { + spin_unlock_irq(&pool->lock); + return; + } + pool->flags &= ~POOL_DISASSOCIATED; for_each_pool_worker(worker, pool) { |